Immigrant Identity in the Indus Civilization


Material Information

Immigrant Identity in the Indus Civilization a Multi-Site Isotopic Mortuary Analysis
Physical Description:
1 online resource (257 p.)
Valentine, Benjamin T
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Committee Chair:
Krigbaum, John
Committee Members:
Sassaman, Kenneth Edward, Jr
Davidson, James M
Brenner, Mark
Kenoyer, Jonathan Mark
Kamenov, George Dimitrov


Subjects / Keywords:
bioarchaeology -- harappa -- indus -- isotope -- lead -- mortuary -- strontium
Anthropology -- Dissertations, Academic -- UF
Anthropology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Despite widespread effort within academic, political, and popular arenas to understand the relationships between migration and urbanism,the study of migration in ancient urban contexts remains relatively understudied. This is particularly true for the Indus Civilization, a 3rdmillennium BC protohistoric urban society that spanned much of modern day Pakistan and northwest India. Yet migration almost certainly accompanied the sprawling trade networks connecting Indus lowland peoples with the hinterland cultures of the mineral-rich highlands; and the interregional flows of people likely challenged existing social contexts and redefined cultural relationships in a manner that resonates with modern and historical urban settings. Thus the application of isotopic provenience methods to Indus Civilization human remains offers unique insight into the ways that complex societies are entangled with the movement of bodies. Mortuary samples from the major urban center of Harappa,the eastern frontier town of Farmana, and the post-urban necropolis at Sanauli collectively provide a diachronic perspective on Indus migration. Individual life histories of mobility were gleaned from archaeological tooth enamel using isotopes of strontium, lead, oxygen, and carbon, and when considered within the broader context of mortuary material culture and South Asian geochemical variation, they suggest that urban era Indus cemetery burials were reserved exclusively for participants in a structured institution of immigration.Although their origins cannot be conclusively determined based on currently available evidence, the isotope data are consistent with birth outside of the alluvial plains including the adjacent resource-rich highlands. Further, intra-individual isotopic data suggest that many individuals, if not all, migrated as children. The implication is that child migration may have happened outside the context of familial residence change. Additional work is required to verify the ‘all immigrant’ hypothesis, but it is proposed as a framework for future investigations and to demonstrate the need for models of interaction that incorporate the complex social dimensions of the Indus landscape. An institution of fosterage is modeled based on ethnohistoric evidence from the adjacent Hindu Kush mountains. It is proposed that highland and lowland groups could have been economically united through kinship practices that reaffirmed their independent cultural identities.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Benjamin T Valentine.
Thesis (Ph.D.)--University of Florida, 2013.
Adviser: Krigbaum, John.
Electronic Access:

Record Information

Source Institution:
Rights Management:
Applicable rights reserved.
lcc - LD1780 2013
System ID:

This item is only available as the following downloads:

Full Text




2 2013 Benjamin Thomas Valentine


3 To Shannon


4 ACKNOWLEDGMENTS Truly, I have stood on the shoulders of my betters to reach this point in my career I could never have completed this dissertation without the unfailing support of my family, friends, and colleagues, both at home and abroad. I am grateful, most of all, for my wife, Shannon Chillingworth. I am humbled by the sacrifices she has made for dreams not her own. I c an never repay her for the gifts she has given me, nor will she ever call my debt due Shannon thank you. I am likewise indebted to the scholars and institutions that have facilitated my graduate resea rch these past eight years. Foremost among them is my faculty advisor, John Krigbaum, who took a chance on me, an aspiring researcher with little anthropological training, and welcomed me into the University of Florida (UF) Bone Chemistry Lab I have worke d hard not to fail him, as he has never failed me. Under mentorship, I have earned my chance to succeed in academe. During my time at UF, I have benefited from the efforts of many excellent faculty members, but I am especially grateful to James Davidson, Department of Anthropology and George Kamenov and Jason Curtis, Department of Geological Sciences. They have labored alongside me, in the classroom and laboratory, to ensure my progress. The hours they spent instructing, helping, or just li stening to me have directly contributed to the completion of this dissertation. My research would very likely have been on something else entirely, however, were it not for Jonathan Mark Kenoyer and his co directors at the Harappa Archaeological Research P roject, Richard Meadow and Rita Wright Jonathan Mark Kenoyer in particular, has supported my endeavors in Indus Civilization archaeology, offering vital counsel, collaboration, and, when needed, a firm push. His involvement


5 has been central to my success es thus far, and I cannot overstate how generous he has been. Likewise, the directors, researchers, and staff of the Arch aeological Survey of India have proven foundational to my efforts and I am forever grateful for the opportunities that they have prov ided. In particular, I must thank former Director Generals Guatam Sengupta and K.N. Srivastava as well as Directors R.S. Fonia and Subhra Pramanik They have opened the door to a world of archaeological scholarship that I am only just beginning to explore. The process of discovery, however, did not always go smoothly, and I frequently needed guidance in unfamiliar terrain. In this regard, I am especially thankful for the tireless efforts of D.V. Sharma and V.N. Prabhakar as they helped clear the path for th is lone traveler and wide eyed research er Their dedication to collaborative scholarship is inspiring. I received equally essential support from the faculties of Deccan College Post Graduate and Research Institute, Pune and Mahars hi Dayanand University, Ro htak I am truly indebted to Vasant Shinde and Manmohan Kumar for seeing what could come of this resear ch and repeatedly sharing their counsel and assistance I have also benefited greatly from the expertise and kindness of Veena Mushrif Tripathy Pramod Joglekar, Vijay Sathe, Amrita Sark ar Das, and Vivek Dangi who regularly shared their knowledge on matters both scholarly and mundane. Additional assistance was provided by former and present researchers affiliated with Centre for Archaeological Stu dies and Training, Eastern India, Kolkata as well as Research Institute for Humanity and Nature, Kyoto. I owe particular thanks to Suchira Roychoudhury, Akinori Uesugi, and Toshiki Osada. These and other international scholars have collectively welcomed me into their


6 academic community, and I am deeply appreciative of their respect for new ideas and new methods within archaeological research. Lastly I must acknowledge the financial support received through UF and external granting institutions. Within the university, the Office of Research, Graduate School, College of Liberal Arts and Sciences, Department of Anthropology, and Graduate Student Council have frequently assisted with the indirect costs o f this research. Additionally the American Institute of Indian Studies (AIIS) took me on as a Junior Fellow (2010 2011), during which time I conducted sample collection. Beyond monetary support, the dedication of the officers and staff at AIIS proved inva luable in overcoming the many challenges of field research. I offer my sincere thanks to Philip Lutgendorf, Purnima Mehta, Subir Sarkar, Anil Inamdar, and Mini Raji Kumar for their per sonal efforts on my behalf. Furthermore The Wenner Gren Foundation Diss ertation Fieldwork Grant (2011 2012) assisted directly with analytical expenses. Before embarking on this project, I imagined that research was something that people accomplished by themselves, either wearing a lab coat or writing insightful paragraphs la te into the night. I have since learned that the true privilege of scholarship is in working closely with a community of experts, and in bringing their diverse talents to bear on problems that fire the imagination. To all of th ose above, and to the many mo re that were not mentioned by name thank you. Were it not for you, this dissertation would not exist and my life would be substantially less rich.


7 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 10 LIST OF ABBREVIATIONS ................................ ................................ ........................... 12 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 OBJECTIVES AND OVERVIEW ................................ ................................ ............. 15 Mortuary Archaeology ................................ ................................ ............................. 19 Migration in Archaeology ................................ ................................ ........................ 22 2 INDUS CIVILIZATION IN CONTEXT ................................ ................................ ...... 29 Culture History ................................ ................................ ................................ ........ 33 Pre Urban/Regionalization Era ................................ ................................ ......... 36 Urban/Integration Era ................................ ................................ ....................... 37 Post Urban/Localization Era ................................ ................................ ............. 38 Integration Era Social Context ................................ ................................ ................ 40 Craft Economy ................................ ................................ ................................ .. 41 Food Production ................................ ................................ ............................... 4 3 Infrastructure and Logistics ................................ ................................ .............. 45 Trade and Interaction ................................ ................................ ....................... 47 Mortuary Variation ................................ ................................ ............................ 53 3 MATERIALS AND METHODS ................................ ................................ ................ 61 Principles of Isotope Analysis ................................ ................................ ................. 61 Heavy Stable Isotopes ................................ ................................ ..................... 62 Light Stab le Isotopes ................................ ................................ ........................ 67 Stable oxygen isotopes ................................ ................................ .............. 68 Stable carbon isotopes ................................ ................................ .............. 71 Osteological Development and Diagenesis ................................ ...................... 73 Laboratory and Field Methods ................................ ................................ ................ 76 4 GEOCHEMISTRY OF THE GREATER INDUS REGION ................................ ....... 82 Environmental Background ................................ ................................ ..................... 82


8 Isotopic Variation Heavy Isotopes ................................ ................................ .. 83 Isotopic Variation Light Stable Isotopes ................................ ......................... 86 Towards an Isotopic Baseline ................................ ................................ ........... 91 Results of the Baseline Analyses ................................ ................................ ............ 93 5 FARMANA ................................ ................................ ................................ ............ 108 Habitation Area ................................ ................................ ................................ ..... 110 Cemetery Area ................................ ................................ ................................ ...... 115 Results of the Analyses ................................ ................................ ........................ 121 6 HARAPPA ................................ ................................ ................................ ............. 136 Habitation Area ................................ ................................ ................................ ..... 137 Cemetery Area ................................ ................................ ................................ ...... 142 Results of the Analyses ................................ ................................ ........................ 144 7 SANAULI ................................ ................................ ................................ .............. 159 Localization Era Mortuar y Treatment ................................ ................................ .... 159 The Sanauli Cemetery ................................ ................................ .......................... 160 Primary Burials and the Osteological Sample ................................ ................ 162 Symbolic Burials ................................ ................................ ............................. 165 Results of the Analyses ................................ ................................ ........................ 167 8 DISCUSSI ON ................................ ................................ ................................ ....... 180 Integration Era Mobility ................................ ................................ ......................... 181 Immigrant Frequency ................................ ................................ ..................... 182 Immigrant Or igins ................................ ................................ ........................... 185 Mode of Migration ................................ ................................ ........................... 189 Localization Era Mobility ................................ ................................ ....................... 191 Dietary Variability ................................ ................................ ................................ .. 193 Towards an Integrated Model ................................ ................................ ............... 194 An Ethnographic Example ................................ ................................ .............. 197 The Indus Model ................................ ................................ ............................. 199 9 CONCLUSION ................................ ................................ ................................ ...... 210 LIST OF REFERENCES ................................ ................................ ............................. 214 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 257


9 LIST OF TABLES Table page 2 1 Indus Tradition chronology. ................................ ................................ ................ 58 3 1 Approximate crown formation time in years ................................ ........................ 81 4 1 Baseline samples results of the analyses ................................ ........................ 97 4 2 Baseline samples summary statistics ................................ ............ 100 5 1 Chronology, demographic data, and mortuary treatment at Farmana .............. 126 5 2 Mortuary elaboration at Farmana ................................ ................................ ..... 128 5 3 Farmana mortuary sample results of the analyses ................................ ........ 130 5 4 Farmana mortuary sample summary statistics -................................ 132 6 1 Chronology at Harappa ................................ ................................ .................... 149 6 2 Select demographic data and mortuary elaboration at Harappa ....................... 150 6 3 Harappa mortuary sample results of the analyses ................................ ......... 152 6 4 Harappa mortuary sample summary statistics -................................ 155 7 1 Demographic data and mortuary elaboration at Sanauli ................................ ... 171 7 2 Sanauli mortuary sample results of the ana lyses ................................ ........... 173 7 3 Sanauli mortuary sample summary statistics ................................ 176


10 LIST OF FIGURES Figure page 1 1 Map of study area including sites mentioned in the text. ................................ .... 28 2 1 Map of the maximum extent of the Indus Tradition culture area including major geographic features and adjacent cultural trad itions (in gray). ................. 59 2 2 Cultural maps of the Indus Tradition. A) Extent of the Harappan Phase during the Integration Era (ca. 2600 1900 BC) including regions with incomplete adoption of Harappan Phase materials. B) Extent of the phases of the Localization Era (ca. 1900 1300 BC). ................................ ................................ 60 4 1 Himalayan geology. ................................ ................................ .......................... 101 4 2 Riverine 87 Sr/ 86 Sr ranges ................................ ................................ .................. 102 4 3 Summer riverine 18 O ranges ................................ ................................ ........... 104 4 4 Baseline samples in Sr Sr space reciprocal of Sr concentration .................... 105 4 5 Baseline samples in Pb Pb space 208 Pb/ 204 Pb ................................ ............... 105 4 6 Baseline samples in Pb Pb space 207 Pb/ 204 Pb ................................ ............... 106 4 7 Baseline samples 18 O ................................ ................................ .................. 106 4 8 Baseline samples 13 C ................................ ................................ ................... 107 4 9 Baseline samples in Pb Sr space 87 Sr/ 86 Sr ................................ .................... 107 5 1 Farmana samples in Sr Sr space reciprocal of Sr concentration ................... 133 5 2 Farmana samples in Pb Pb space 208 Pb/ 204 Pb ................................ .............. 133 5 3 Farmana samples in Pb Pb space 207 Pb/ 204 Pb ................................ .............. 134 5 4 Farmana samples 18 O ................................ ................................ .................. 134 5 5 Farmana samples 13 C ................................ ................................ .................. 135 5 6 Farmana samples in Pb Sr space 87 Sr/ 86 Sr ................................ ................... 135 6 1 Harappa samples in Sr Sr space by sex reciprocal of Sr concentration ........ 156 6 2 Harappa samples in Pb Pb space 208 Pb/ 204 Pb ................................ ............... 156 6 3 Harappa samples in Pb Pb space by sex 207 Pb/ 204 Pb ................................ .... 157


11 6 4 Harappa samples 18 O ................................ ................................ .................. 157 6 5 Harappa samples 13 C ................................ ................................ ................... 158 6 6 Harappa samples in Pb Sr space by sex 87 Sr/ 86 Sr ................................ ......... 158 7 1 Sanauli samples in Sr Sr space reciprocal of Sr concentration ..................... 177 7 2 Sanauli samples in Pb Pb space 208 Pb/ 204 Pb ................................ ................. 177 7 3 Sanauli samples in Pb Pb space 207 Pb/ 204 Pb ................................ ................. 178 7 4 Sanauli samples 18 O ................................ ................................ .................... 178 7 5 Sanauli samples 13 C ................................ ................................ .................... 179 7 6 Sanauli in Pb Sr space 87 Sr/ 86 Sr ................................ ................................ .... 179 8 1 Local isotopic range at Farmana Pb Sr space ................................ ............... 202 8 2 Local isotopic range at Harappa Pb Sr space ................................ ................ 202 8 3 Mixing line at Farmana with non local anthropogenic Pb ................................ 203 8 4 Four source regions at Harappa ................................ ................................ ....... 20 3 8 5 Two mixing lines in the Greater Indus region suggesting distinct geological inputs Sr Sr space ................................ ................................ ......................... 204 8 6 Intertooth developmental sequences at Farmana in Pb Sr space showing childhood migration to the local area. A) Entire Farmana dataset with inset of Figure 8 6B. B) Closeup of Farmana intertooth variation. ................................ 205 8 7 Intertooth developmental sequences at Harappa inPb Sr space showing childhood migration to local or near local regions ................................ ............. 206 8 8 Two local range estimates and source regions at Sanauli Pb Sr space ........ 206 8 9 Intertooth developmental sequences at Sanauli in Pb Sr space showing variation in the timing and direction of migration ................................ ............... 207 8 10 Theoretical marital residence patterns that might produce entirely immigrant mortuary populations. A) Marital exchange system with three groups. B) Marital exchange system with four or more groups. ................................ ......... 208 8 11 Early childhood residence and migration by sex at Harappa in Pb Sr space characterized by non overlapping male and female distributions and male residence at a single non local locality ................................ ............................. 209


12 LIST OF ABBREVIATIONS C carbon CO 2 carbon dioxide HARP Harappa Archaeological Research Project HBr hydrobromic acid HCl hydrochloric acid HHC High Himalayan Crystallines HNO 3 nitric acid MC ICP MS multicollector inductively coupled plasma mass spectrometer NaOCl sodium hypochlorite (bleach) O oxygen P B lead S R stronti um TIMS thermal ionization mass spectrometer TRA time resolved analysis VPDB Vienna Pee Dee Belemnite


13 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for th e Degree of Doctor of Philosophy IMMIGRANT IDENTITY IN THE INDUS CIVILIZATION: A MULTI SITE ISOTOPIC MORTUARY ANALYSIS By Benjamin Thomas Valentine August 2013 Chair: John Krigbaum Major: Anthropology Despite widespread effort within academic, political, and popular arenas to understand the relationships between migration and urbanism, the study of migration in ancient urban contexts remains relatively understudied. This is particularly true for the Indus Civilization, a 3 rd millennium BC protoh istoric urban society that spanned much of modern day Pakistan and northwest India. Yet migration almost certainly accompanied the sprawling trade networks connecting Indus lowland peoples with the hinterland cultures of the mineral rich highlands; and the interregional flows of people likely challenged existing social contexts and redefined cultural relationships in a manner that resonates with modern and historical urban settings. Thus the application of isotopic provenience methods to Indus Civilization human remains offers unique insight into the ways that complex societies are entangled with the movement of bodies. Mortuary samples from the major urban center of Harappa, the eastern frontier town of Farmana, and the post urban necropolis at Sanauli col lectively provide a diachronic perspective on Indus migration. Individual life histories of mobility were gleaned from archaeological tooth enamel using isotopes of strontium, lead, oxygen, and carbon, and when considered within the broader context of mort uary material


14 culture and South Asian geochemical variation, they suggest that urban era Indus cemetery burials were reserved exclusively for participants in a structured institution of immigration. Although their origins cannot be conclusively determined based on currently available evidence, the isotope data are consistent with birth outside of the alluvial plains including the adjacent resource rich highlands Further, intra individual isotopic data suggest that many individuals, if not all, migrated as children. The implication is that child migration may have happened outside the context of familial residence change. is proposed as a framework for future investigations and to d emonstrate the need for models of interaction that incorporate the complex social dimensions of the Indus landscape An institution of fosterage is modeled based on ethnohistoric evidence from the adjacent Hindu Kush mountains. It is proposed that highland and lowland groups could have been economically united through kinship practices that reaffirmed their independent cultural identities.


15 CHAPTER 1 OBJECTIVES AND OVERVIEW T his dissertation examines the social outcomes of ancient urban migration through the isotopic mortuary analysis of individuals recovered from Indus Civilization cemeteries. Known best from sites in what is now Pakistan and northwe st India (Figure 1 1), the Indus Civilization contradicts the familiar narrative of statehood that begins with highly centralized power structures. The 3 rd millennium BC urban society was crosscut by heterarchical institutions in a m anner that is reminisce nt of the increasingly interconnected and decentralized modern world. Situated in a vast landscape estimated between 680,000 km 2 (Kenoyer 1991a) and 1,000,000 km 2 (Jansen 2002) Indus cemeteries provide essen tial data for understanding the roles of individuals in broader social institutions and the mechanisms that fostered large scale cultural integration T his study incorporates conventional archaeological data within an isotopic assessment of individual life histor ies I sotopes of lead, s trontium, oxygen, and carbon in human tooth enamel reveal aspects of past migration, climate, and diet Further, by sampling multip le teeth from given individuals, it was possible to access life histories spanning birth to early ad olescence. The resulting bioarchaeological data sets are used to address three research foci: How did early life migration vary within and between individua ls and groups ? What group identities can be inferred from patterns in the migration data ? How might intergroup relationships have been enmeshed within the broader social context of the Indus landscape? Interregional interaction was central to the developm ent and maintenance of the Indus Civilization (Kenoyer 1991a; Kenoyer 1995b) and thus interpretive emphasis for this study is placed on the identifi cation and sourci ng of immigrant individuals Much


16 effort has been made towards understanding the flow of goods throughout the subcontinent and beyond (Lamberg Karlovsky 1972; Fairservis 1975; Agrawa la and Kumar 1982; Allchin 1984; Gupta 1984; Chakrabarti 1990; Lahiri 1992; Asthana 1993; Hooja 1994; Kenoyer 1997; Law 2008) however, relatively little evidence is available regarding the flow of bodies (Kennedy et al. 1984; Hemphill et al. 1991; Kennedy 2000) Artifactual evidence for trade and culture contact implies the movement of people, but it is a necessarily incomplete picture that leaves open questions regarding the identity and mobility of various social groups. Some models of Indus mobility have been put forward that suggest mobile pastoralists served as intermediaries in interregional tra de (Possehl 1979; Mughal 1994b) and Kenoyer (2011a) suggested that far ranging systems of elite kinship could have promoted interregional alliances and cultural exchange Nevertheless, a more thorough investigation of Indus mortuary remains using tools of biogeoc hemistry is needed to refine understanding s of how Indus migration and mortuary practices were enmeshed wit hin a complex system of intergroup relationships. Well documented Indus burials are scarce and cemetery inhumations even more so (Possehl 2002b) I n part this paucity of human remains suggests that formal in terments represent individuals associated with a particular segment of society. Even at the large urban period cemeteries known from excavations of Harappa (Wheeler 1947; Sastri 1965; Mughal 1968; Miller 1991; Meadow and Kenoyer 1994) Kalibangan (Sharma 1999) and Farmana (Shinde 2011a) estimates of the number of burials at each site are in the low hundreds. Consequently, the archaeological problem is less about reconstructing demographic profiles and social hierarchies for internally sub divided mortuary popu lations, and more about discovering what the inhumed dead had


17 in common. Pollock (2008) analyzed cemeteries at Mehrgarh, Harappa, Kalibangan, and Lothal, and noted broad similarities in grave goods, burial orientation, and skeletal position at each site except Mehrgarh. The same general pattern holds for the cem eteries at Farmana (Shinde 2011a) and post urban Sanauli ( Prabhakar 2012) Considering that this mode of burial was widespread geographically and temporally but reserved for relatively few individuals, it seems likely that the Indus tradition of cemetery inhumations indexed shared ideological concepts and perp etuated certain relationships between individuals, groups, and institutions. This research will start to identify potential commonalities of identity in the sampled Indus cemeteries at Farmana, Harappa, and Sanauli. Novel lines of isotopic evidence are use d to assess early life patterns of migration among burials for which tooth enamel was available. L ead, strontium, and oxygen isotope data are analyzed to help assess the relative frequency of migration as well as the timing of migration events in an indivi ~14 years ). Probable geographic regions of origin are inferred whenever possible by comparison with sediments and fauna from other Indus sites including Allahdino, Balakot, Mehrgarh, and Nausharo Broader patterns of regional isotopic variation are inferred from the geological literature (e.g., Karim and Veizer 2000; Najman et al. 2000; Clift et al. 2002; Karim and Veizer 2002) and artifact provenience studies (e.g., Law 2008; Hoffman and Miller 200 9) F indings suggest that inhumation in formal cemeteries during the Indus urban florescence (ca. 2600 1900 BC) may have been exclusively reserved for individuals who migr ated to the area at a young age. This proposed institution of migration appears to ha ve disappeared or transformed in subsequent years given the lower frequency of immigrants in the post urban mortuary


18 population and the relatively stochastic nature of the data set. Inferred t rends in residential mobility are compared with archaeological m ortuary variation including dietary and demographic variables to identify and characterize potential social characteristics of Indus groups. Lastly, a model is developed suggesting how the structure of inter group relationships both derived from and contri buted to broader Indus Civilization social organization. P revious models of Indus Civilization social organization provide the fundamental underpinning for inferences about the structures and consequences of interregional interaction Chapter 2 outlines k ey aspects of Indus society including an overview of factional elite competition for highland resources Specific attention is paid to the influence of Indus kin groups on trade between peoples of the lowland study sites and adjacent potentially non Indus peoples in the highlands, piedmont, or other nearby areas (hereafter referred to simply as highlands) with strategic access to resources not normally found in the alluvial plains Kinship principles likely shaped group membership throughout Indus society (Kenoyer 1998) and could have provided a means of constructing relatedness between diverse peoples. Chapter 3 outlines research methods used to g enerate the isotopic data. Results are then presented in ch apters on isotopic variation in the Greater Indus region (Chapter 4) Farmana urban period humans (Chapter 5) Harappa urban period humans (Chapter 6) and Sanauli post urban humans (Chapter 7) Each chapter includes a site specific review of relevant arch aeological context. Chapter 8 synthesizes the data in a multi site isotopic mortuary analysis and presents a model explain ing how the construction of kinship between the inhumed dead and elite groups helped structure


19 interregional interaction. Chapter 9 concludes by highlighting the key contributions of this research and suggestions for future research. The entire orientation of this res earch, however, derives from a long history of archaeological approaches to death and migration in ancient societies Th e relevant literature is reviewed here to situate the current study within a broader theoretical trajectory. Mortuary Archaeology As it is practiced today, mortuary archaeology has moved well beyond the Saxe Binford paradigm (Saxe 1970; Binford 1971; Saxe 1971) so influential in processual understandings of social structure. Nevertheless, the discipline retains roots in the methods and lessons of that era. B rown (1971) and colleagues rejected the overt emphasis on description and chronology in earlier approaches, emphasizing the utility of mortuary variation for understanding systems of social organization (contra Kroeber 1927). Further, ethnography was transformed from a peripheral source of interpretive inspiration (sensu Ucko 1969) into the essential raw material for comparati ve analogy (Schuyler 1968) Though strict analogical use of ethnography is fraught with its own problems (Metcalf and Huntington 1991:1 7 19; Lyman and O'Brien 2001) ethnography regained an important role in the interpretation of past mortuary practices especially where historical continuity could be established (e.g., Pearson 1982; Dillehay 1995) No ethnographic scenario can be projected wholesale into the past, but broad concepts drawn from ethnographic accounts (such as categories of kinship and personhood ) may be relevant to inferences about archaeological societies that demonstrate some measure of structura l and historical continuity. (1984) e mphasis on the redundancy of mortuary variation. Not every distinction in mortuary


20 style has a corresponding social distinction, nor is every aspect of the mortuary program preserved. In fact, many mortuary practices are invisible in the archaeological rec ord, requiring that archaeologists look for redundancy in the burial program such that correlations between multiple indicators are presumed to convey the most significant social differences (O'Shea 1984:29 30; Weiss Krejci 2011) Those traits most readily observed by archaeologists such as the quantity and kind of grave goods or the age and sex of the interred may have been irrelevant or subordinate to the primary social d istinctions implied by the mortuary program. Archaeologists, therefore, have a greater chance of accurately identifying social distinctions that were observed in life by looking for redundant patterns of d ifferentiation across a full suite of archaeologica l, biological, and geochemical variables. A concern with the spatiality of mortuary contexts also has roots in (Goldstein 1981) Though limited at first to a representationist perspective i n which mortuary variability is interpreted as a reflection of social order (Peebles and Kus 1977; Goldstein 1980; Chapman 1981; Morris 1991) the post processual critique ushered in a concern for ideology and agency (Hodder 1982; Pearson 1982) Mortuary practices, it was argued, can be strategic means of crea ting relationships and building affiliations betwe en individuals and institutions (Cannon 1989; Pollock 1991; Gillespie 2001; Joyce 2001) Always, however, those practices are embedded in space a vital med ium of inclusion and exclusion. Mortuary practices are now increasingly perceived as historically contingent constructions that in turn serve to structure relationships betwe en people, th ings, and places (Arnold 2002; Charles and Buikstra 2002; Silverman 2002; Hodder 2012) For


21 example, Arnold (2002) suggested that the evolution, placement, and patterning of mortuary monuments central Europe was closely tied to changing perceptions and needs of the liv ing. Tumuli served as material reference points legitimizing territorial claims, and the pattern of their placement in the landscape was structured over time to make various and evolving assertions about the relationships be tween groups. Linear arrangement s of mortuary monuments next to settlements likely guided the approach of outsiders and residents alike, intentionally aligning the social power of the dead with that of local groups. Further, a ritually manipulated alliance between the living and the dead would necessarily have concretized a social distinction between the ritual allies and other, possibly competing groups. Thus construction of the ritual landscape itself was a means of contention, with socially powerful mo rtuary landscapes being co opted over time through the distinct yet structurally related mortuary practices of outsiders. Through such entangled relationships social landscapes are formed. Importantly, the social qualities of landscapes are not abstracted ephemera. T hey are inseparable f rom and constituted by the material interactions of people and things. To describe a mortuary landscape, like any other landscape, is to describe the ways in which bodies and objects, individuals and institutions, come to stand in reference to each other. Always in the process of becoming, mortuary landscapes are constantly being negotiated by differentially empowered (or differentially emplaced) actors who engage differen tly with the histories of a place (de Certeau 1984; Lefebvre 1991) Mortuary analysis must consider the spatiality of the subject matter. The locations of burials in relation to each other, of cemeteries to habitations, and of formally


22 similar cemeteries to each oth er all are dimensions of mortuary variation through which past relationships could have been influenced. However, spatiality need not be limited to the physical distances observe d by archaeologists. Bodies too are fundamentally spatial in that their histor ies have the potential to invoke different meanings and different places. The body of an immigrant possesses indexicality (Gell 1998) local origins, one is made mindful o f the non local place. The distant place is made present and the meanings and relationships associated with that place are brought to bear on subsequent interactions. The social significance of such associations is precisely what a mortuary analysis of imm igrants must investigate. Redundant patterns and correlations in the archaeological evidence have the potential to demarcate social boundaries and associated group identities that exist only with respect to each other. For archaeological interpretations to have any hope of accuracy, however, they must be grounded in the historical context that shaped the perceptions and motivations of the actors Through this approach a theory of migration can begin to take shape. Migration in Archaeology Migration is relat ively under theorized within archaeology partly because of an intellectual stigma associated with the migration concept. From early on in Anglophone archaeology, cultural types were conflated with ethnic groups such that the changing distributions of mater ial culture were assumed to reflect the changing residence of ethnic groups. M igration was proposed as an ad hoc explanation for cultural distributions rather than a process modeled in its own right (Adams et al. 1978) This advances in archaeological science and theory. For example, radiocarbon dating demonstrated that seemingly


23 abrupt transitions in material culture once attributed to population movem ents were actually gradual changes over the long term (Trigger 2006:383 384) Positivist internal responses to environmental factors as the most probable mechanism of culture change (Renfrew 1972; Renfrew 1973; Moseley 1975; Meacham 1977) Migration made a limited resurgence by the 1970s in the form of archaeogenetics (Renfrew 2000) Researchers attempted to understand large scale population movements such as the Austronesian dispersal in Southeast Asia (Bellwood 2004) through a combination of archaeological, linguistic, and eventually molecular data. The notion that groups of linguistically and economically similar migrants could be analyzed as a whole was not entirely dissimilar from previous culture histori cal concepts of ethnic groups. However, scholars tried to model specific mechan isms and processes of dispersal as in Amm erman and Cavalli (1971) demic diffusion of Neolithic agricult ural peoples across Europe. Outside of archaeogenetics, more recent calls for the study of migrat ion as a social process have emphasized the importance of ethnography (Anthony 1990; Anthony 1992) It was argued that e th nographically based classifications of the different types of migration could help explain past migrations by enabling an explicit focus on variab ility in the timing, scale, and permanence of migration events (Anthony 1997; Frachetti 201 1) Critiqued as direct analogies that ignore the specific contexts of prehistoric migrations (Chapman and Dolukhanov 1992) ethnography was recast by Burmeister (2000) as an investigative starting point from which to make comparisons and test for goodness of fit between theory and archaeological evidence.


24 Prospects for testing ethnographically derived models are limited, however, by the often coarse grained da ta available to archaeologists (Burmeister 2000) Several methods have been proposed to ameliorate these difficulties including the application of individual level data such as th ose gained from bioarchaeol ogical and biogeochemical methods (Harke 2007) along with the search for redundancy using multiple lines of evidence (Beekman and Christensen 2003) Whateve r the methods used to identify migration, explaining migration as a process necessarily requires a multi scalar approach in which diverse individual or group level actions are simultaneously embedded within and helping to construct macro level conditions (Hakenbeck 2008) By themselves, biogeochemical provenience data are insufficient for this task. Unless the link between individuals and particular geographic areas is embedded within a model of the social l andscape, t he movement of groups and individuals devolves into little more than lines connecting regions on a map. The lines that connect regions and in the most limited cases they are reduced to a description of how much information is flowing in which direction. The line might indicate who is migrating along with their departure point and destination, but it remains a generalized description of macro level processes. This understanding of migratio n does not allow for the many articulations between migrants and the world they inhabit. There is little room to discuss how macro level conditions such as the distribution of raw materials, food rich environments, and urban marketplaces intersect with mic ro level conditions such as the local politics of kinship practices.


25 One way forward is to apply the visual metaphor of weaving. The weave is s imilar to (2011:63 64) concept of meshwork in which t he world is composed of limitless intermingled flows of people and things rather than bounded entities in empty space Likewise, a woven tapestry cannot be made of images and patterns that exist in and of themselves I t is made of the warp and the weft, cr ossing and contrasting and connecting different places in the weave such that a pattern can emerge in relation to the surrounding threads. meshwork, however, a weave persists. It has history and i t holds time in such a way that new threads pull on, intersect with, diverge from, and modify the w ork that came before. A weave is in many ways evolutionary, and s o it is with archaeological cultures. Rather than thinking of bounded groups that exist in a nd of themselves, groups are defined only in opposition to something else, at the places where the many relationships between people and things are gathered and made material. The diverse flows of people and things come together in ways that are structured by material histories such that the outcomes are a product of past relationships Further, perceptions of the past change over time and vary between entities because weave is constantly evolving. The infinite extent of the meshwork limits its utility as an analytical tool, but a weave, a product of work, time, and perceptions, can be cut, tangled, or otherwise altered to stop flows in a particular configuration Such hybrid izing actions which conde nse diverse flows within a new medium (e.g., a body, thing, or institution), overcome t he interpretive challenge of limitless meshworks (Strathern 1996) T he intersection of diverse flows of peoples and thin gs can be concretized in an act of


26 creation that draws on different agents with different spatialities and different indexicalities. In this way, and only in this way, can social distinctions be perceived Groups are not differentiated from each other beca use they are composed of different people. G roups are differentiated because diverse flows are gathered (Ingold 2011:178 179) in hybrid loci that make a tension manifest. They hold a social contrast in plac e so that it can be considered, referenced, and invoked as a means of distinguishing its co nstituent flows from each other. Red becomes red at the place s where it is altered and hybridized into orange or purple An ethnic group becomes an ethnic group at t he place where its flows of bodies and things are altered by other flows much like the maintenance of social boundaries modeled by Barth (1969) and Cohen (1974) This meeting of flows, however, is not limited to the decisions of rational actors, and could be for example, a historical monument that gathers flows from the past and articulates them with the present in unanticipated and variable ways Similarly, a hybrid person can embody diverse flows -a human born of one culture and raised in another. The c onstituent flows ( diverse ways of life) that meet in a hybrid person may then be rendered as contrasting identities, thus providing one possible means of defining ethnic groups. A tangle d weave captures this process in a way that can be readily visualized and likened to archaeological flows of bodies and things. D ifferent threads laid down at different places and times are gathered together in a tangle The hybrid locus creates a new visual reference point by which old patterns can be redefined or eliminate d and new patterns can emerge. This contrasts with stubbornly implicit diffusionist tendencies in archaeological models of culture contact. Whereas migration is often assumed to lead


27 to a blending, graduating, and eventual homogenization of distinct cultur es, the tangled weave allows for cultural novelty. The hybrid locus creates the reference point by which new groups may be defined. Further, t he tangled threads do not pull equally on all aspects of the weave. It is only in proximity to such hybrid gatheri ngs that different regions in the weave can be differentiated from each other. Meanwhile, bodies flowing along more distant threads may barely feel the pull, may barely notice th e emerging boundary, and may perceive similarities to, rather than contrasts w ith th e gathered flows. N or is the ethnic boundary perceived equally by those bodies flowing closest to the tangle. Some of their threads are stretched tight, some loose, and still others hang slack. So it is with a weave of people and things. As people cr eate, they draw on the wider tapestry gathering diverse actions and objects into a hybrid structure that pulls differently on everyone and everything involved. By recasting migration as a weave gathered up in places by hybridizing tangles it becomes pos sible to discuss the variable perceptions held by the different parties involved, how their social environments and histories constrain the structure of their interaction, how small changes in the social environment might stabilize or destabilize such inte ractions in the long term, and what the macro level consequences of interaction might be. For any of this to occur, however, one must return to mortuary analysis as the means of identifying redundancies in mortuary practice indicative of group identities. This analytical process begins in C hapter 2 with a review of Indus Civilization archaeological context.


28 Figure 1 1. Map of study area including sites mentioned in the text.


29 CHAPTER 2 INDUS CIVILIZATION IN CONTEXT The term Indus Civilization refers colloquially to the urban zenith of northwestern South Asia during the third millennium BC. Archaeological chronologies tend to convey a more holistic view (Possehl 1977; Kenoyer 1991a; Shaffer 1992) with emphasis on the eight millennia long develo pment, convergence, and subsequent diversification of various cultural trajectories beginning with foraging lifeways of the early Holocene (ca. 10,000 BC). In either case, the emergence of cities and associated cultural integration across a region covering much of modern day Pakistan and northwestern India ( serves as an important reference point for explaining transformations of social complexity. Indeed, this dissertation is focused primarily on clarifying the nature of interregional relationships coincide nt with and immediately succeeding the urban climax An improved understanding of interregional relationships will provide new insights into the transformations of identity associated with this dynamic period of social differentiation and stratification (K enoyer 1998; Possehl 2002 ; Wright 2010 ) A diverse and relatively dispersed resource base likely played an important role in the development of interregional interaction in the Greater Indus region (Wright 2010:179 180) Indus sites were located primarily on the fertile alluvium of the western Indo Gangetic plains (Figure 2 1) Frequently situated on doabs, relative ly elevated interfluvial regions settlements were well positioned for ag ricultural production and riverine transportation Spaced roughly equidistant from each other, f our of the five major cities (Harappa, Mohenjo Daro, Ganweriwala, and Rakhigarhi) were located at strategic points along the Indus River, its tributaries (the Punjab region), or the Ghaggar Hakra (Mughal 1994a) Near confluenc es or mountain passes, residents


30 would have had unparalleled access to hinterland trade (Kenoyer 1998:43 ; Law 2006 ) They depended on trade for a variety of mineral and organic resources available only in the oceans, swamps, and jungles to the south and so utheast or the upland regions flanking the Indus Valley to the west, north, and east (Law 2008; Wright 2010:190, 211 212) Ranges of the Western Fold Belt mark the western highland border, perforated by mountain passes that access Iran and southern Afghan istan (Chakrabarti 1990; Thomas and Knox 1994) The mountains run approximately from north to south before turning to the southwest along the Makran coast. The northern boundary of the Punjab begins at the Potwar Plateau in the northwest and extends eastwa rd along the Himalayan foothills. The Hindu Kush and Karakorum ranges lie further to the north, overlooking the upper Indus drainage. T o the east of the Indus Valley, the Ganges River and its major tributary, the Yamuna demarcate the eastern limit of the Indus Civilization. The now seasonal Ghaggar Hakra flows south between the glacially fed Indus and Ganges systems before diverting to the southwest, skirting the northward projecting Aravalli mountains and disappearing into the Thar Desert. Coastal marshe s to the south and southeast characterize the Kutch region where the fifth major city, Dholavira, once had access to maritime routes (Bisht 1989; Kenoyer 1998:53) The Saurashtran peninsula lies beyond the wetlands to the southeast and marks the furthest p enetration of Indus sites into southern India. Each hinterland region was exploited for mineral wealth that was traded widely across the Greater Indus Region (Law 2008). Trade was not an equal opportunity endeavor, however, and elite groups likely competed over access to resources (Kenoyer 2000b)


31 The presence of elites is well established within the Indus Civilization as it manifests in many aspects of the archaeological record (Kenoyer 1998; Possehl 2 002b; Wright 2010) Hierarchy is most strongly inferred from the use and production contexts of various material symbols of ownership and regulation (e.g., Rao 1979; Vidale 1989) ; likewise, the logistical requir ements for the management of agricultural land in a diverse system of intensive food production and the construction and maintenance of expansive civic architecture suggest that certain kinds of people held authority over certain matters (e.g., Bisht 1991; Jansen 1993; Kenoyer 1993, Joshi 2003; Mi ller 2006b) However, power does not appear to have been concentrated within a central state institution. Instead, diverse groups seemed to coexist within the context of a shared ideology (Miller 1985) Corporat e groups may have sought advantages over each other by gaining preferential access to prestige goods (Kenoyer 1997a) Mineral wealth in particular was scarce in the alluvial plains, necessitating inter action between lowland settlements of the Indus tradition and non Indus highland peoples through either direct or indirect trade For e (2008) extensive provenience study of the mineral assembl age at Harappa indicate s far reaching acquisition routes with an emphasis on the adjacent highlands to the north. In many instances this would have necessitated direct or indirect interaction with Northern Neolithic peoples (Stacul 1992; Stacul 1994; Possehl 1999) Similarly, Indus groups from eastern sites like Farmana and Sanauli may have interacted with peoples of the Rajasthan Mawal Tradition in the Ar avalli Mountains to the south (Agrawala and Kumar 1982; Rizvi 2007)


32 Many aspects of the development of lowland highland interaction s in different Indus regions remain unclear Contrary t o views expressed by McIntosh (2002; 2008) violence was a fact of life for Indus peoples and was likely structured in some ways by group identity (Robbins Schug et al. 2012) Still, no evidence suggest s that interregional interaction between diverse groups was mediated through warfare or conquest to the degree documented in the formation of ancient Egyptian and Mesopotamian states (Akkermans and Schwartz 2003; Wenke 2009; Wright 2010) Relationships between people of different territories appear to have been established and main tained primarily through civilian activities. I nterregional exchange provided opportunities for the accumulation of power and was therefore a key motivation for people of different regions to come together (Kenoyer 1998:98 99). T he question then becomes ho w to explain the diverse outcomes of interregional interaction. Importantly, varying degrees of formal similarity in the assemblages at different sites across the Indus region suggest that interaction was neither uniform nor explainable by a single model. It is hoped that the mortuary analysis carried out here will begin to clarify Indus modes of interregional interaction as fundamentally social phenomena, thus moving Indus interaction studies beyond simply the trade of materials towards models that integra te the comple xities of the social landscape. T his dissertation will build from the Indus cultural context re viewed in this chapter. First, the Indus culture history is outlined with discussion of the different terminologies applied by various scholars. A r eview of the evidence for social differentiation is presented, after which the importance of interregional trade is established. Lastly, modes of interregional interaction proposed in the literature are


33 reviewed Each subtopic will be addressed with a focu s on the peoples of the Greater Indus Valley and the most relevant neighboring groups. External trade and interaction with more distant people are largely eschewed in an effort to deal concisely with the geochemical data. Culture History Indus Civilization chronology has been pieced together from a broad range of ceramic types and cultural traits variably expressed in different regions throughout the Greater Indus Valley. After the earliest development of food production in the shadow of the Western Fold Be lt at sites like Mehrgarh ( Jarrige 1995) the majority of Indus settlement occurred along the channels and tributaries of the Indus and Ghaggar Hakra river systems ( Possehl 1993; Mughal 1997 ) There is some formal diversity in ceramic assemblages along the se rivers and adjoining regi ons to the south, particularly prio r to full urban integration ca. 2600 BC ( Joshi 1984; Dales and Kenoyer 1986b; Possehl and Herman 1990; Mughal 1997; Kenoyer 2011b) Nevertheless, Indus types are regarded as such because they were either produced by people who participated within the broad cultural milieu out of which urban Indus forms emerged and existed, or they represent the region specific cultural variations that emerged directly out of the post urban transformation (Shaf fer 1992) A localized continuity of types is apparent in many cases such that gradual cultural transitions typically occur in a given order, but suites of pottery, architecture, food production, and interaction systems cannot be precisely seriated in a si ngle unifying evolutionary sequence. Still, the different cultural sub groupings can be ordered by their approximate beginnings even if they persist in some places and times alongside supposedly later cultural forms.


34 In his efforts to deal with the somewh at permeable boundaries of spatial and chronological periodization, Possehl (1977) used the term phase to refer to a collective of material culture and associated socio economic or political activities. For example, patterns of food production or trade are just as important as pottery in defining phases. Phases are grouped together in stages defined by maj or differences in socio economic organization the appearance of early food producing com munities and ends by 500 BC with phases of the early Iron Age. Possehl (2002b) emphasized that individual phases should be regarded as archaeological constructs that provid e a generalized sequence of cultural developments that are highly specific to distinct regional contexts. A phase might persis t into the succeeding stage in one region whereas many years earlier it may have disappeared from another region altogether. Shaffer (1992) used the term phase similarly in his assessment of the Indus Civilization, an adaption of the system Willey and Phillips (1958) originally developed for North American archaeology. Phases are subsumed under eras, roughly comparable to reas the entire cultural trajectory is defined as a tradition. The contemporaneous cultural complex in the highlands to the west of the lower Indus is classified as the Baluchistan Tradition. Subsequent cultural developments of this region, in particular t he Kulli culture area (Franke Vogt 2000; Franke Vogt 2005; Wright 2013) maintained a degree of cultural autonomy throughout the Indus urban florescence de spite ongoing interaction with the Indus lowlands but see Possehl ( 1992b) Other important traditions have been classified by Kenoyer (1991a; 2006) in his elaboration of Vi n dhya Tradition (incorporating the Ganges


35 river basin to the east) and the Malwa Rajasthan Tradition (inclusive of the Ganeshwar Jodhpura Phase in the Arava lli Mountains to the south of Farmana). The Indus Tradition chronology as laid out by Shaffer (1992) and Kenoyer (1991a) eriodization. However, it explicitly accounts for the Mesolithic and Microlithic Phases of the Foraging Era which began around 10,000 BC and persisted alongside other e ras until roughly 2000 BC (Table 2 1) Further, the Indus Tradition ended ca. 1300 BC, although as with all other categories there is substantial chronological variation between regions. Perhaps the most important conceptual contribution of the tradition era phase nomenclature is in its nuanced description of the history of settleme nt on the alluvial plains. Instead of making a simple chronological distinction like the Early and Mature Harappan stages, Shaffer and Kenoyer adopted era names indicative of trends in cultural interaction. The Regionalization, Integration, and Localizatio n Eras suggest the complexity and overall directi onality of cultural interaction. S imultaneously the use of eras resolves the conflicting usage of Harappan in reference to the site itself, the material culture, and the Indus chronology. Nevertheless Wri ght (2010) proposed a useful terminological simplification that scheme, the cultural trajectory of settlement on the alluvial plains is intuitively divided into Pre Urban, Urb an, and Post Urban. This straightforward distinction efficiently conveys the aspects of Indus institutional context most relevant to the assessment of social organization that is the focus of this dissertation


36 Pre Urban/Regionalization Era The Regionaliza tion Era encompasses a multitude of phases, although widespread imprecision within site reports precludes a comprehensive understanding of their spread and duration (Kenoyer 2011b) For example, the Ha kra phase as first used by Mughal (1982; 1997) to describe ceramic assemblages along the Ghaggar Hakra and Ravi river valleys has now been generalized to much of the Greater Indus R egion (e.g., Khan et al. 1990; Ajithprasad 2002; Khatri and Acharya 2005; Mallah 2008). In this way, it has become difficult to assess the development of modes of interaction, craft production, and social differentiation at a fine enough scale to begin explaining how Indus urbanism evolved Nevertheless, broader regional chronologies have been defined, as outlined in Table 2 1 Signs of incipient urbanism are found in pre urban material culture from across the Greater Indus Valley. Wright (2010 : 80 economic changes took hold. It is likely th at relatively elaborate social hierarchies began to emerge as implied by the stratified structure in many facets of material culture including settlement size, intra settlement architectural divisions, craft specialization and standardization, and the earl y usage of seals and script (Mughal 1990; Khatri and Acharya 1995; Kenoyer and Meado w 1999; Kenoyer 2000; Kenoyer and Meadow 2008) Further, some groups must have exercised increased administrative authority to build and maintain the fortifications and massive platforms known from this time (e.g., Bisht 1991; Kenoyer 1993, Joshi 2003). Lastly, the extensive long distance trade routes that characterize the urban era have their roots hundreds of years earlier in the systems of exc hange establishe d by pre urban peoples (Law 2008 ).


37 Urban/Integration Era It is widely acknowledged that by ca. 2600 BC the nascent cultural developments of the pre urban era had intensified, diversified, and spread to the point that Indus society had fund amentally transformed (Shaffer and Lichtenstein 1989; Kenoyer 1998; Ratnagar 2001; Possehl 2002; Wright 2010) Population growth continued to increase ( sensu Christaller 1966) anchored by five major cities: Mohenjo Daro, Harappa, Rakhigarhi, Ganweriwala, and Dholavira. Several authors have suggested that these centers served as semi autonomous cultural and economic foci for the surrounding settlements and that they may best be likened to city states or peer polities (Joshi 1984; Kenoyer 1997a; Possehl 1997; Smith 1997) The major cities were roughly equidistant from each other in such a way that each likely served as a hub for resource acquisition from adjacent resource rich highlands (Fentress 1976; Lahiri 1992; Asthana 1993; Possehl 1993; Mughal 1994a; Law 2008) Further, their distinct resource catchments provided unique economic oppo rtunities for the peoples of each area and may have helped to differentiate their political systems from one other. Despite such differences, people from across the Greater Indus region appear remarkably integrated in their material culture (Figure 2 2A) Whether in the Indus or Ghaggar Hakra basins or as far south as the Saurashtra peninsula, common ways of living and thinking are apparent from the relative homogeneity in the archaeological record (Miller 1985) Shared traits include but are not limited to Harappan Phase ceramic assemblages (Dales and Kenoyer 1986b; Wright 2010:187) the widespread use of Indus script (presently undecip hered) in stamp seals, tablets, and pottery inscriptions (Mahadevan 1977; Frenez and Tosi 2005) a standardized system of


38 weights and measures (Kenoyer 1998:98), and participation in a comp lex network of internal and external trade routes (Chakrabarti 1990; Lahiri 1992; Tosi 199 3; Kenoyer 1995b; Law 2008) In short, the Harappan Phase likely correlates to generalized acceptance of a common ideology and participation within the same system of political legitimization. However, incomplete adoption of Harappan culture in some re gions suggests variable penetration and acceptance of the Harappan way of life. For example, settlements with Kot Diji assemblages persist into the Harappan phase along the northern edge of the Harappan Phase site distribution (Allchin 1984). Further to th e south west Kulli sites exhibit only a partial adoption of Harappan material culture (Possehl 1992b ; Wright 2013 ). A similar situation may have existed at the southernmost regions of the Saurashtra peninsula in what Possehl (1992a) labeled the Sorath Harappan. Variable acceptance of Harappan Phase material culture suggests a compelling need to better understand how interaction was structured differently in instances of apparent ideological resistance. Po st Urban/Localization Era By ca. 1900 BC, the urban expression of the Indus Civilization went into decline and a process of localized cultural differentiation took place. Cities gradually fell into disrepair, settlements became smaller and more numerous, a nd much of the population along the Indus River basin dispersed to other regions especially towards the upper reaches of the Ghaggar Hakra and the western extent of the Gangetic River system (Wheeler 1968; Possehl 1997; Gangal et al. 2010) Further, the disappearance of seals and inscriptions suggests that Harappan elite groups lost much of their administrative power concordant with widespread decrease s in interregional trade (Kenoyer 2005b)


39 Many of the integrative behaviors that supported the adoption and perpetuation of the Harappan Phase way of life ceased to operate, effectively setting each regi on on semi independent cultural trajectories. The post urban phases of the Indus Civilization are less well documented than their antecedents, but they are clearly distinct from and continuous with the preceding urban material culture (Figure 2 2B ) The Jh ukar Phase emerged along the lower reaches of the Indus River where settlements shrank and became less organized before disappearing altogether over a few hundred years (Mughal 1992; Mi ller 2005) Similar changes occurred at former Harappan sites in and around the Saurashtra peninsula (e.g., Rao 1985; Dhavalikar 1992), but at many other regional sites there was a general reorganization and transformation rather than a decline (Shaffer and Lichtenstein 1999) In particular, Saurashtran sites (Rangpur Phase) remained viable (Possehl 1997) The Cemetery H Phase encompasses post urban variation spanning the upper Indus and Ghaggar Hakra, reaching as far east as the Ganges Yamuna doab. From excavations at Harappa (Meadow et al. 2001) and various regional surveys (Stein 1931; Stein 1937; Mughal 1997; Nath 1998; Nath 1999; Dangi 2009) it is clear that people were reorganizing rather than abandoning the area altogether. Though many diagnostic traits of Harappan urbanism had disappeared, it would be inaccurate to characterize th is phase as stagnant For ex ample, Meadow and coworkers (1996) documented a variety of technical advances in craft production during this phase at the site of Harappa. Harappa eventually diminished in size, although perhaps only sligh tly (Kenoyer 2005b), coincident with regional climatic changes To that end, settlements gradually shifted


40 northward and eastward towards regions with more dependable monsoonal precipitation (Kumar 2009; Gangal et al. 2010; Giosan et al. 2012) Early interpretations suggested that the entire post urban transformation was the product of conquest and population replacement resulting from the so invasi (Vats 1940; Sastri 1965) there is no biological evidence to suggest major phenotypic discontinuities (Hemphill et al. 1991; Kennedy 2000). Instead, recent research suggests that a long period of increasing aridification played a major, but non deterministic role in the trajectory of the Indus Tradition as a whole (Fuller and Madella 2001; Madella and Fuller 2006; Wright et al. 2008; Giosan et al. 2012) M id Holocene monsoonal wea kening may have reduced the intensity of flooding and created more productive conditions for floodplain agriculture. In the face of ongoing aridification, however, a threshold was eventually passed by ca. 2000 BC making flood dependent agriculture unreliab le in general especially so for the lower reaches of the Ghaggar Hakra basin (Giosan et al. 2012) Regional responses to aridification were most likely variable, but evidence for increasingly diverse cropping regimes including selection for more dr o ught to lerant plants such as millets suggests that people were actively seeking to cope with environmental change (Weber and Belcher 2003) Integration Era Social Context The story of Indus urbanization is in many ways a story of social differentiation. Indus scholars generally agree that the processes driving social differentiation were entangled with the intensification and diversification of production and consumption such that people, groups, and institutions became increasingly interdependent across an increasingly large region (Kenoyer 1998; Possehl 2002b ; Wright 2010). Though


41 integration and intensification across the Indus region was complex and likely cannot be explained by a single factor, the processes when se t in motion would have fundamental ly altered the economic landscape by creating new opportunities for craft specialists and elite oversight (Morrison 1994) As discussed below, increased conflict over labo r and land resulting from a broad range of specialized modes of food production would have necessarily resulted in hierarchical relationships of power (Wright 2010). Further the logistical needs of an urbanized landscape would have contributed to the ri se of administrative groups or institutions that would likely have ratcheted up demand for prestige items in their efforts to validate relations of inequality (Kenoyer 198 9; Kenoyer 1995a) Craft Economy Artifacts made of exotic materials are particularly important indicators of social stratification because they typically represent only a small fraction of any given artifact class. Several authors have argued that Indu s consumption was organized along axes of relative value such that a common ideology was promoted by formally similar artifacts but that status ranking was associated with the rarity of the material and the complexity of the technology used in production (Kenoyer 1992; Vidale 1992; Vidale and Miller 2000) The logic is that exclusivity corresponds with high value, high value i tems would only have been attainable by elites, elites are a minority, and therefore artifacts made using less accessible materials or technologies should be fewer in number. Wright (2010) found some support for this distribution across artifact classes an d sites, suggesting the structure of hierarchies can be glimpsed in detailed artifact classifications. Further, Vidale and Miller (2000) suggested that with the growth of a bureaucratic middle class, ubiquitous artificial materials such as faience offered a


42 means of expanding the middle tiers in hierarchies of value to accommodate the expanding social strata. Again, some artifact distributions support this idea, but Miller (2008) has suggested that interp retations of past value must also account for factors beyond exclusivity. For example, softer precious stones are nearly unrepresented among bangles. Instead, harder natural materials may have been acquired or artificial ones manufactured so that bangles c ould provide an auditory experience. No matter how one interprets the spectrum of craft value, however, the wide range of styles in personal dress and ornamentation apparent from inscriptions, figurines, and statuary are suggestive of socio economic and et hnic differentiation (Kenoyer 1995a) The various modes of production evident in the Greater Indus Valley are also suggestive of social diversification. Artisans must have dedicated themselves to years of study to acquire t he inferred from high quality Indus crafts (Vidale and Miller 2000) As demonstrated below using examples from Indus bangle production, specialist knowledge was like ly seen as proprietary and subject to internal and external administrative controls (Halim and Vidale 1984; Vidale 1989) Further, ethnicities likely emerged at the intersection of production and kinship, with many professions potentially organized by kinship (Kenoyer 1989). For example, h undreds of years of continuity in a cluster of segregated kilns and workshops at Harappa have been interpreted as evidence for kin structured prod uction of ceramics relatively free from centralized oversight (Dales and Kenoyer 1990a; Wright 1991) However, certain other crafts were likely subject to administrative control T he production context of s toneware bangles suggests tha t elites closely regulated the manufacturing process (Halim and Vidale 1984; Vidale 1989) S toneware bangles were produced in strictly


43 segregated workshops located only at Harappa and Mohenjo Daro and a t different stages in the manufacturing process, sealing s and inscriptions were applied presumably as a regulatory measure (Halim and Vidale 1984; Blackman and Vidale 1992) Further, stoneware bangles vanished at the end of the Integration era along with other administrative paraphe r n a lia (e.g., Indus script and stamp se als), suggesting their production was contingent on the social complexity and class distinctions of an urbanized environment. Most craft production however, was not as rigidly controlled as stoneware bangles. Evidence suggests a diversity of production co ntexts, as in the case of shell bangle w orkshops. At sites such as Balakot, Nageswar, and Kuntasi, production areas appear to have been variously organized at different points on a spectrum between small scale household operations geared for local consumpt ion and large scale elite run workshops internally segregated by task and oriented towards export (Bhan and Kenoyer 1983; Bhan 1986; Dh avalikar 1992; Vidale 2000) Taken as a whole, the evidence for Indus craft economy suggests a complexity of organization that is difficult to explain without reference to elites or hierarchy Food Production Other aspects of Indus production seem to su pport similar conclusions. With regard to the diversified, intensified, and specialized forms of food production recognized archaeologically (e.g., Fuller and Madella 2001; Madella 2003; Weber 2003, Miller 2004), conflicting land usage by food producers wo uld likely have required complex (and potentially hierarchical) institutions of mediation (Wright 2010: 213 ) Because large scale, labor intensive irrigation is unknown in Indus contexts, water management practices probably did not give rise directly to cen tralized power structures


44 (Scarborough 2003) H owever, evidence for traction (Miller 2004) and extra household crop processing (Fuller and Madella 200 1 ; Weber 2003) implies that agropastoral land use was organized by diverse groups and institutions (Wright 2010: 206 Further, Miller (Miller 1991) suggested that groups with large tracts of corporately owne d land would have best been able to cope with the shifting nature of river courses in the Indus plains. Miller (2006b) later emphasized that kin based groups would have been more effective at allocati ng and organizing land use given the potential for Indus farmers to easily relocate if pressed too hard by elites. In her model, Miller proposed that land owning kin groups were another plausible source of social control that could have crosscut power stru ctures predicated on authority over non kin Indeed, Kenoyer (1989, 1998) proposed an enduring role for kin groups within urban Indus power structures based on models of the organization of craft production. Additionally, groups other than kin structured staple crop agriculturalists would have had competing interests. A variety of Indus crops such as cotton, grapes, dates, and hemp would have required dedicated plots of land and specialist labor (Madella 2003; Weber 2003; Wright 2010) Different groups of pastoralists would also have needed access to pasturage. At Harappa, there is limited zooarchaeological evidence consistent with dairying ( Miller 2004). Further, cattle would have been used to pull carts and plows (Wheeler 1947; Miller 2003) while secondary products from cattle and sheep may also have been important (Meadow 1989b; Miller 2004) Sedentary and mobile pastoralists would likely have had different perspectives on land use which could have brought them into conflict with farmers or each other; alt hough sedentary pastoralism and agriculture are known to coexist within ethnographic South Asian kin groups ( Miller


4 5 2003). Much like with the Indus craft economy, the complex social context of food production probably created conflicts and inequalities nec essitating some degree of hierarchical organization. Infrastructure and Logistics Whatever the means of production, the process of getting products to consumers created opportunities for intervention by other Indus groups. With respect to fish at Harappa, at least, indirect distribution mechanisms were more prevalent during the urban era implying yet another manifestation of specialized labor and possible source of conflict and inequality (Belcher 2003) The transportation of goods may also have opened the door to regulation by elites. In an excavated warehouse at Lothal for example there were dozens of clay sealings bearing multiple impressions by different seals (Rao 1979) Kenoyer (1998) suggested that they may indicate the regulation of cargo by bureaucratic officials, and further transported goods may have been subject to taxation as indicated by the frequency and location of classically Harappan chert weights. Specifically, the standardized weights are fewer in number than might be expected if merchants used them for everyday exchanges. Weights also tend to be associated with gates choke points where taxes on g oods could most readily be levied. Even t heir production might have been subject to elite oversight given that they were predominantly crafted using a specific kind of banded chert from the Rohri Hills in the lower Indus region (Kenoyer 1991) Perhaps some of the most compelling evidence for hierarchy is in urban architecture. Significant logistical oversight would have been needed to coordinate public works project s such as the foundational mudbrick platf orms and extensive system of wells at Mohenjo Daro (Jansen 1984; Jansen 1993) The platforms were truly


46 massive and according to Possehl (2002 b :103), amounted to 4 million person day s of labor. Also from Mohenjo Daro comes evidence that the town was laid out according to astrological data (Wanzke 1984) suggesting coordination between different specialists and manual laborers. Elites were not l imited to such grand endeavors, however, and may well have organized the care of ubiquitous public drains and sump pits. Wright (2010:238, 242) suggested that extensive sanitation features would have demanded the coordination of maintenance, and if central ly administered, would have required a substantial civic labor force (e.g., Jansen 1993). Even if public sanitation was not centrally administered, those considered fit for such an unpleasant task would almost certainly have been low er status individuals. Further, the role of urban centers as hubs of Indus life must have conferred a certain prestige on sophisticated urban dwellers. As is often true of cities today, inhabitants of the surrounding rural areas would have been instantly recognizable upon enter ing the dynamic urban settings of Harappan cities. Social life among urbanites was also stratified as suggested by the spectrum of prosperity apparent in a comparison of urba n households (e.g., Jansen 1993 ) Larger domiciles with more private spaces and im proved access to drains almost certainly belonged to relatively prosperous families. Given the complex patterns of social differentiation considered in the preceding pages, it seems clear that the lived Indus landscape was enmeshed in relations of conflict and control. Where so many diverse interests intersect ed some groups must surely have been systematically favored over others; but since the earliest systematic excavations by Vats (1940) and Marshall (1931 ), archaeologists have noted the lack of grand monuments or elaborate tombs exalting a specific few over the many. Nor is


47 there evidence for palatial structures or military conquest associated with specific hereditary leaders or groups (Kenoyer 1998:99 100) Whatever the nature of Indus hierarchies, they were not unassailable. They could not withstand the potential backlash caused by such blatant assertions of authority. Present evidence cannot identify the specific groups in Indus hierarchies, but their int eractions with each other were almost certainly crosscut by multiple identities and allegiances subject to situational changes and active manipulation. As the ultimate source of food, fertility, and life, land would have conferred authority on those who co ntrolled access to it B ut land is also steeped in history and blood. Short of military force, the power derived from land would not have been easily stripped from political rivals. Mineral wealth, however, came from outside the places of kin and forebears Power born of highland resources was a power from outside their histories and lineages and therefore uniquely accessible to the ambitious and entrepreneurial (Helms 1988; Helms 1993) For th is reason, inferences about Indus identity must begin with an understanding of economic exchange. Trade and Interaction Indus people maintained extensive internal and external trade routes, transporting all kinds of natural resources by bullock cart and fl at bottomed boat (Miller 2004; Miller 2006a) Trade was driven in part by the reliance of elites on prestige goods as a source of legitimation (Kenoyer 2000) However, the alluvial plains of the Greater Indus Valley are poor in most mineral resources, and a broad variety of raw materials w as acquired from highland regions throughout the Indus Tradition for mor e utilitarian pu rposes (Law 2008 ). The demand for non local resources would have created a powerful incentive for people and groups to compete for preferential access to highland minerals (Kenoyer 2000). The mineral rich highlands surrounding the Greater Indus


48 region very often contain multiple sources of any given metal a fact that might have encouraged competition and prevented any single group from gaining an unchallenged advantage (Kenoyer and Miller 1999) The strong est artifactual evidence for elite competition comes from the comprehensive assessment of mineral provenience at Harappa conducted by Ra ndall Law (2008 ), the interpretations of which are briefly reviewed here. Through the application of isotopic and compos itional analyses, Law (2008) suggested that at least some raw materials were restricted to certain communities at Harappa Certain bead making communities for example, seem to have had proprietary access to uniquely that is variably argued to be an unknown mineral (Kenoyer and Vidale 1992) or a particular kind of heat treated fl int clay (Law 2008 :appendix 4.5). In either case, Law suggested the sheer abundance of Gujarat, far to the south of Harappa. At Harappa, an extremely hard variety of garnet, vesuvianite grossular, shares a similarly restricted distribution as it probably could be worked only grossular are uniquely associated with alabaster bangle production and exclusive sources of grindingstone altho ugh the sample size is small for the latter two materials. Additionally, Integration Era settlements used Rohri Hills chert almost exclusively despite the prevalence of other sources among pre urban peoples. According to Law (2008) this might have resulte d from an elite Finally, Law suggested that competition to control resources might be underrepresented


49 at Harappa given that much of the benefit from monopolizing mine rals would have derived from trading them to other communities. Therefore patterns of exclusive acquisition might be obscured by patterns of distribution. Thus it cannot be assumed that elites did not compete for access to a particular resource based only on the dataset produced by Law (2008) W idely available materials such as copper must have created a highly competitive environment for Indus elites and highland suppliers alike. Unlike chert, for which a single source provided an incomparable product, qu ality copper was available from a variety of regions including Baluchistan and Afghanistan to the west, Oman and possibly Iran via maritime routes, Himalayan deposits to the north, and the Aravalli Mountains to the east (Ahmad 1969; Bazin an d Hbner 1969; Nandan et al. 198 1; Geological Survey of India 1994; ESCAP 1995; Chakrabarti and Lahiri 1996; Kazmi and Jan 1997; Kenoyer and Miller 1999; Weeks 20 04; Peters et al. 2007; Law 2008 ). The Aravalli deposits are of particular interest for this study as the northernmost mineral rich fewer than 150km south of Farmana and other eastern Harappan sites. Their importance however, should not be o verstated. Despite a plethora of claims that most Indus copper came from the Khetri deposits (e.g., Sana Ullah 1940; Allchin and Allchin 1982; Agrawala 1984 a ; Dhavalikar 1997; Kenoyer and Miller 1999), recent work suggests that the Khetri region was one copper source amon g many (Law 2008; Hoffman and Miller 200 9) Additional analyses are needed to resolve the details of the Indus copper trade, but of primary interest here is that the decision to acquire copper from any given region would have depended very much on the different highland suppliers. One can en vi sion


50 for the best bargain. They might have considered factors like the length or reliability of the supply route, the number of involved, whether negotiations would be with many factional producers or a few collective representatives, whether or not other products could be acquired from the same source, whether or not the highland groups had conflicting allegiances to competing Indus elites, and of course the price of the copper itself. Undoubtedly, much effort went into negotiating the most favorable terms, and these would have been subject to change depending on any number of factors. The decisions that were made by all parties surely had wide ranging repercussions for how each group perceived the o ther and how their relationships were maintained and changed over time. Trade in copper and other materials must have played an important role in shaping the cultural dynamics between ethnic groups of protohistoric south Asia. If archaeologists are to use exchange as an entry poin t for explaining process in a region, there must also be a consideration of how exchange was mediated and not just the location of the parties involved Much more work is nee ded to firmly establish the myriad source regions and cul tural associations involved throughout the Indus region, but inferred relationships of exchange must be modeled to guide ongoing studies of interregional interaction Fortunately, there are hints in the ar chaeological record as to what such relationships may have looked like. Some Harappan settlements appear to have been built as strategic outposts fa r away from the alluvial plains where they would have served as trading posts in the resource rich hinterlands. Possehl (1980) suggested that Lothal served as a that the relatively compact fortifications were constructed according to a single plan.


51 According to Possehl, Lothal was a base of operations from which traders could acquire goods to ship back to larger urban markets in the north. From comparative statistical analyses of Indus skeletal remains, it was suggested that individuals interred at Lothal were phenotypically intermediate betw een people buried at alluvial sites and those from Langhnaj, a hunter gatherer site fewer than 150 km to the north (Possehl and Kennedy 1979; Kennedy et al. 1984) The author s suggested that the relationships between urban traders and hunter gatherers went beyond material exchange to include significant patterns of gene flow. Economic symbiosis between sedentary agriculturalists and hunter gatherers is well known from ethnogra phic accounts of South Asian foragers (Possehl 2002a) and something similar may well have been happening during Indus times. In exchange for providing forest resources, people at Langhnaj may have been c ompensated with carbohydrate rich staple crops as suggested by Lukacs and Pal (1993) based on an uncommonly high occurrence of dental caries. If these inferences are correct, it suggests that economic interact ion could have been partly shaped by the creation of kin relations. Genetic relatedness does not necessarily correlate with social kinship, but at the very least it is suggestive of cultural interaction at a deeper level than a straightforward extraction o f resources by lowland groups from highland regions. Unfortunately, there is no skeletal record at other outposts to the north and west, but several factors suggest their primary function was to acquire resources from exotic locales. In the northern highlands, archaeologists discovered Shortugai, a small urb an era settlement laid out according to typical Harappan designs (Francfort et al. 1989) Further, pottery and beads at the site were skillfully crafted in Harappan styles using


52 local goods. Various precious m aterials are abundant in the area including gold, silver, copper, tin, and lapis lazuli, all adjacent to important trade routes (Kohl 1978) Lapis lazuli is limited in distribution, and the residents of Shortugai we re probably responsible for the lapis found at Nausharo, Ghazi Shah, Nagwada, and other Indus settlements (Sonawane 1992; Flam 1993; Wright 2010) Based on the ab ove evidence, Wright (2010) suggested that Shortugai was consistent with a trade mission rather than a large scale migration of settlers. Outpost settlements like Shortugai were probably structured by the commercial ambitions of Harappan urban dwellers, b ut they were also variably enmeshed in local ways of life. Their different fates in the post urban era might indicate a greater or lesser degree of cultural integration with local groups With the decline of urbanism, Harappan traits at Shortugai were grad ually replaced by those of the neighboring Bactria Margiana Archaeological Complex (BMAC) (Francfort et al. 1989) whereas Sutkagen dor and Sotka koh far to the southwest were abandoned altogether (Dales and Lipo 1 992) The latter settlements were separated from Harappan sites by those of the Kulli culture which like the BMAC, spread to many formerly Harappan sites in the post urban era. Kulli people had maintained a variety of independent cultural features in th e face of Harappan influence throughout much of the Integration Era, but seemingly grudgingly gave way to a cultural hybridization before their post urban resurgence (Dales 1976; Franke Vogt 2000) Perhaps l inked by trade in lead (Law 2008 ), Kulli peoples seem to have had a qualitatively different experience with Harappan traders than many other groups did.


53 Another group strongly implicated in trade but relatively less well understo od is the Northern Neolithic people who liv ed in the highlands, north of Harappa. They probably provided many of the mineral resources used by residents at Harappa and elsewhere (Law 2008 ), but their means of interaction remains unclear. Perhaps groups at late occurring Kot Diji settlements situated between Harappan and Northern Neolithic peoples facilitated exchange between their neighbors, but this has yet to be tested. The nature of interaction with people of the Ganeshwar Jodhpura Phase in the northe rn Aravallis is similarly uncertain. One possible explanation is that nomadic pastoralists served as intermediaries (Possehl 1979; Mughal 1994), but again, there is little direct evidence for this proposal. In summary a great deal is unknown about how Ha rappans interacted with their highland neighbors, but at the very least it would seem that different groups were engaging with each other in different ways One aspect of highland exchange that is relatively certain is that Harappan involvement was shaped by the interests of competi ng elite groups (Kenoyer 2000) Further work is needed to ascertain whether or not exchange was carried out directly with highland groups or via intermediaries as this has implication s for the politics of identity. Mortuary Varia tion Widespread consistency in the Indus program of inhumation provides another important means of understanding cultural integration in the Great Indus region. With the exception of the formal disposal area at Mehrgarh (Jarrige et al. 1995) urban era cemeteries have many similarities. The mortuary program is st rongly normative with little evidence for internal subdivisions. Burials are rectangular or oval and oriented along a north south axis with the head to the north (Wheeler 1947; Sastri 1965; Mughal


54 1968; Rao 1973; Rao 1979; Rao 1985; Dales and Kenoyer 1989a; Jarrige et al. 1995 ; Nath 1998; Nath 1999; Sharma 1999; Trivedi 2009; Shinde 2011a ). When included as grave goods ceramics are most often laid near the head and consist largely of types recovered from habitation deposits. Pottery is sometimes also found at the feet, along the sides or underneath the body but usually in addition to vessels placed near the head. Further, pottery was sometimes buried first before the body was lowered into the grave (Dales and Kenoyer 1989a) In most cases, only modest quantities of ceramics were interred although a few graves have more than 40 pots (e.g., Dales and Kenoyer 1989a; Sharma 1999) Apart f rom the pottery, grave goods are limited to personal ornaments and toilet objects including jewelry and hand mirrors. Bodies generally lay extended and supine, and were occasionally wrapped in shrouds, enclosed in coffins, or sheltered by a brick or clay grave lining The graves were typically organized in the same way e ven in the case of secondary burials and cenotaphs (so called symbolic burials) that lack some or all skeletal elements (Shinde 2011a) Though few correlations have been observed between different mortuary vari ables, Kenoyer (2011a) suggested certain structuring principles at Cemetery R 37. For example, m ales are usually associated with less elaborate graves. Also, shell bangles and certain types of stone pendant s ten d to be associated with females A rare devia tion from the normative grave type is found in the pot burials at Kalibangan (Sharma 1999). U nlike Cemetery H, the post urban cemetery at Harappa, the Kalibangan pot burials contain no human remains. Further, they are set apart from t he other graves, perha ps indicative of a separate mortuary tradition. Nevertheless, the Kalibangan pot burials are not replicated at other sites, and thus are not suggestive of


55 significant internal divisions with in the Indus mortuary program of inhumations. Indeed, the majority of urban era burials show only modest variation in layout with minor redundancies in aspects of mortuary treatment The overall impression that emerges is of a specific category of person. T he inhumed perhaps, were perceived largely in terms of a particu lar group identity, with o nly secondary importance placed on individual differences. The collective identity posited for the Indus inhumed is based partly on the scarcity of cemetery inhumations. In fact, formal disposal areas during the urban period are only well documented at Mehrgarh (Jarrige et al. 1995), Harappa (Dales and Kenoyer 1989a ; Mughal 196 8 ; Sastri 1965; Wheeler 1947), Kalibangan (Sharma 1999), Lothal (Rao 1973; Rao 1979; Rao 1985) Rakhigarhi (Nath 1998; Nath 1999) Tarkhanewala Dera (Trivedi 2009) and Farmana (Shinde 2011a) The total number of burials is small such that excavations at the largest known cemetery (Cemetery R 37 at Harappa) have produced skeletal remains for fewer than 400 ind ividuals ( Possehl 2002b :Table 9.4 ). Instead, the vast majority of Indus people must have received mortuary treatments that did not preserve archaeologically such as cremation, exposure, or immersion. Thus, i t is generally agreed that cemetery burial was re served for members of a particular group and that cemetery populations do not represent the population at large (Kenoyer 1998: 122 ; Possehl 2002b :171 ; Wright 2010: 263 ). There is less consensus however, on the social identity once held by the inhumed. It is telling that Indus burials have been variously characterized as belonging to the lower (Rao 1 979:143, Sharma 1999:14) and upper classes (Kenoyer 1998:122 124 ) despite overall similarities between cemetery inhumations. Different aspects of the


56 archaeo logical and osteological record can be interpreted as evidence of either reverence or disregard although current evidence may not be able to resolve the issue T he occasional incorporation of particularly well crafted ornaments and pottery as is the case for some burials in Cemetery R 37 at Harappa (Kenoyer 1998:122 124), suggests a willingness to invest in mortuary ritual The osteological data from Harappa (Hemphill et al. 1991, Kennedy 2000) are especially compelling in that they portray relatively hea lthy, well cared for individuals. Paleopathologies are few, including a m oderate prevalence of osteoarthritis in the cervical vertebrae (possibly attributable to biomechanical occupational stresses) and a moderate ly higher incidence of caries among female s. However, the preponderance of data suggests the indivi duals inhumed at Harappa enjo yed relatively healthy lives with ties to more prosperous social groups. This impression is reinforced by comparison with t he disturbed post urban remains from Area G at Harappa that have a much higher incidence o f trauma ( Robbins S c hug et al. 2012). Even though urban era inhumations imply relative prosperity, several aspects of the mortuary program are more ambiguous with regard to the status of the inhumed, and they could potentially be interpreted as evidence for disregard Cemeteries were invariably located hundreds of meters outside the habitation area, and at Harappa, the cemetery was closely associated with a disposal area for more conventional domestic refuse (Dales and Kenoyer 1989a) Further, b urials were frequentl y crosscut by other grave cuts in a haphazard way and disturbed remains often were casually pushed aside or i ncorporated into the grave fill rather th an reinterred ( Wheeler 1947; Dales and Kenoyer 1989 a ; Shinde 2011a ) One possible explanation for the practice could be that


57 individuals belonging to a separate, and presumably ambivalent, class or group were tasked with digging graves (Kenoyer 1998:122). Another interpretation could be that cemeteries were places to carry out a necessary task without unduly inconveniencing the living and that the i nhumed occupied a middle ground between affiliation and alienation. Neither exceptionally privileged nor disadvantaged the inhumed might have held a liminal social role on the periphery of high status groups. Unfortunately, the matter likely cannot be resolved until more detailed reports of cemetery excavations are made available. The challen ge for Indus mortuary archaeology is to better understand the social role of the deceased and to model how they could have been enmeshed within a broader cultural context The problem is daunting, but isotopic bone chemistry methods can contribute to impor tant research questions by beginning to trace the connections between the Indus social environment and the physical landscape. M igration is assessed as a potential mode of interaction between Indus regions and p articular emphasis is placed on the likeliho od of highland lowland mobility as influenced by the needs of compet ing elites.


58 Table 2 1. Indus Tradition chronology. [ adapted from Kenoyer J M 2011b. Regional Cultures of the Greater Indus Valley: The Ravi and Kot Diji Phase Assemblages of Harappa, Pakistan. In Cultural Relations between the Indus and the Iranian Plateau during the Third Millennium BCE. Toshiki Osada and Michael Witzel, eds. Pp. 165 217 (Page 166, Table 1). Cambridge: Department of South Asian Studies, Harvard University.] Era Phases Date s Foraging Mesolithic, Microlithic ca. 10000 2000 BC Early Food Producing Mehrgarh ca. 7000 5000 BC Regionalization Hakra, Ravi, Sheri Khan Tarakai, Balakot, Amri, Kot Diji, Nal, Sothi, etc. ca. 5500 2500 BC Integration Harappan ca. 2600 1900 BC Localization Punjab, Jhukar, Rangpur ca. 1900 1300 BC


59 Figure 2 1. Map of the maximum extent of the Indus Tradition culture area including major geographic features and adjacent cultural traditions (in gray). [adapted from Law, R. W. 2008 Inter Regional Interaction and Urbanism in the Ancient Assemblage. PhD dissertation (Page 41 Figure 2.6 ). University of Wisconsin Madison]


60 Figure 2 2. Cultural maps of the Indu s Tradition. A) Extent of the Harappan Phase during the Integration Era ( ca. 2600 1900 BC) including regions with incomplete adoption of Harappan Phase materials. B) Extent of the phases of the Localization Era ( ca. 1900 1300 BC). [adapted from Law, R. W. 2008. Inter Regional Interaction and Urbanism in the Ancient Indus Valley: A PhD dissertation (Page 41 Figure 2.6 ). University of Wisconsin Madison]


61 CHAPTER 3 MATERIALS AND MET HODS Principles of Isotope Analysis Isotopic data derived from osseous remains provide a proxy measure for the isotopic composition of substances ingested during the life of individual organisms. By understanding how the different isotopes of an element be have in different geochemical environments, archaeologists can infer the likely source or sources for a given element during the period in which the analyzed sample was formed. Differential ingestion or incorporation of various environmental sources can in dicate differences in behavior or environmental exposure that are useful for reconstructing certain aspects of individual life histories such as migration (e.g., Dupras and Schwarcz 2001; Bentley et al. 2007; Schroeder et al. 2009; Chenery et al. 2010; Price et al. 2010), diet (e.g., Richards et al. 2000; Ambrose et al. 2003; Krigbaum 2003; Hu et al. 2009; Kinaston et al. 2013 ), and climate (e.g., Fricke et al. 1995 ; White et al. 2004b ; Daux et al. 2005; Touzeau et al. in press ) This mode of archaeological inquiry, like many bioarchaeological methods, has the advantage of associating s pecific behaviors or environmental conditions with specific individuals, thus improving the chances of identifying and characterizing different sub groups within a study population. In the aggregate then, trends in isotopic life history data can be used to reconstruct some of the ways regularized behaviors and institutions were entangled with different kinds of social bodies. Isotopic methods are especially powerful tools for inv estigating life history in that different skeletal elements represent differen t periods of exposure depending on their initial formation times and predisposition to turnover I sotopic data are thereby accessible for discrete periods of early life via tooth enamel apatite and from a relatively


62 time averaged period incorporating late life exposure via bone apatite, bone collagen, and tooth dentine. Virtually any other biological substance can be analyzed for isotopic values, but due to their high mineral content, bones and teeth tend to preserve most often in the archaeological record Consequently, this chapter will focus on the analysis of osteological remains with an emphasis on apatite. Even for osseous materials, however, the post mortem environment can threaten the integrity of the isotopic values established during life. Therefor e, any archaeological isotope study must account for the potentially confounding process of diagenesis. The following sections outline the geochemical principles underlying the bioarchaeological analysis and interpretation of strontium (Sr), lead (Pb), car bon (C), and oxygen (O) isotopes. Each of these elements is present in bone and tooth mineral, sometimes substituting for calcium in the case of strontium and lead or as the main components of the carbonate phase (CO 3 ) which substitutes for the hydroxyl group (OH). Further, biological apatites are dynamic minerals in life as well as death; therefore this chapter will include a brief discussion of osteological formation proce sses and post mortem alteration Lastly, an outline of the research design will ad dress the sampling process, mechanical and chemical sample preparation and methods of mass spectrometry. Heavy Stable Isotopes Two factors make heavy stable isotopes of elements like strontium and lead useful for archaeological provenience studies. First, the isotopes of heavier elements have relatively small mass differences. Consequently, they undergo negligible fractionation during chemical reactions and phase changes; or to put it another way, the isotope ratios of a given heavy element remain the same as they pass from bedrock into


63 the biosphere and eventually into human bones and teeth. Second, at least one of the isotopes of a given element is the radiogenic daughter isotope of a radioactive parent isotope. In the case of strontium, rubidium ( 87 Rb) d ecays into 87 Sr. For lead, radioactive isotopes of uranium and thorium ( 238 U, 235 U, and 232 Th) decay into 208 Pb, 207 Pb, and 206 Pb respectively. Thus rocks formed at different times have different ratios of isotopes depending on the initial quantity of the parent isotope and the time that has elapsed (Faure and Mensing 2005) As a result of these two properties, the isotope ratios in an analyzed substance can potentially be matched to the isotope ratios from the geological source. This is complicated, however, when the subject of investigation has acquired geochemical input from multiple sources. This is true of biological samples in the same way that it is true for metallurgical samples. For example, a copper artifact might be made from ingots smelted from different ore bodies, it could be made from recycled metal artifacts of different provenience, or it might b e intentionally alloyed with other elements. In all cases, the artifact will have weighted mean isotope ratios determined by the relative contributions from the different geological sources. Likewise, bulk samples of bones and teeth archive the average iso topic input derived from the initial process of mineralization, subsequent regenerative processes, and post mortem processes of chemical alterati on or contamination (Montgomery 2002 ; Bentley 2006) Much of the challenge for isotopic provenience work lies in understanding how different sources were averaged and finding appropriate c omparative samples that characterize the different geological sources and biogeochemical reservoirs i n the study environment. There can be a great deal of isotopic variation within the environment; so


64 much so, that in many cases it can be difficult to find adjacent geographic regions in which the isotope ranges d o not overlap. After all, even different mi nerals in the same rock can have widely differing isotopic ratios (Fullagar et al. 1971) Further, different plants or even different parts within a single plant can show considerable isotopic variation depend ing on the depth and extent of roots as well as the relative ability of foliage to capture aeolian particulate (Klaminder et al. 2008; Reynolds et al. 2012) A bsolute ranges in rocks or plants however, are not necessarily the best proxy for the averaged ratios acquired by organisms higher up the food chain. For one thing, whole rock isotope ratios are not necessarily representative of the values incorporated into the biosphere Weathering processes preferentially release the soluble fraction of strontium from bedrock and sediments, and it is this biologically available portion that ends up in water, plants, and animals (Sillen et al. 1998; Price et al. 2002) E ven leachates of sediments i.e., weak acid solutions used to extrac t the most soluble fraction of st r ontium and lead can vary considerably within a relatively small area depending on micro level variatio n s in sediment composition (Capo et al. 1998) Fortunately, this variability is averaged somewhat by plants and to an even greater degree in bones and teeth because of the long term uptake of strontium from a relatively large catchment area (Burton et al. 1999) Broadly similar principles apply to lead, but it is somewhat less mobile than strontium and undergoes even greater biopurificati on through trophic processes as relatively little soil lead is taken up by plants (Klaminder et al. 2005) Therefore lead levels (and consequently isotopic rati os) in animals tend to be more influenced by t he ingestion and inhalation of dust than by the lead content of ingested foods (Elias et al. 1982; McBride 1994:337; Kohn et al. 2013)


65 Indeed, the relatively expansive dietary catchments of animals provide the best estimates of regional isotopic variation for use in provenience studies. Animals living in close proximity to humans should be used whenever possible in the study of human migr ation as they provide an independent measure of local isotopic ranges (Price et al. 2002; Evans and Tatham 2004) The most appropriate faunal m aterial varies from context to context, however, such that pigs have been used in the study of prehistoric Germany (Bentley and Knipper 2005) and guinea pig bones for archaeological Peru (Knudson et al. 2004) Some authors suggest that animals with very small territories be used to define isotopic variability (e.g., snails), although this might underestimate the local isotopic ranges of species that have relatively far r eaching dietary catchments such as humans (Price et al. 2002). When faunal samples are unavailable, modern plants, water samples, sediment leachates, and even knowledge of the local lithology can be used to provide estimates of isotopic variability (e.g., Beard and Johnson 2000; Hodell et al. 2004; Bataille and Bowen 2012) Multiple sample types may best approximate regional variation because no single sam ple type offers the perfect proxy measure. An accurate estimate of regional isotopic variability may even be obtained from members of the study population that can be independently inferred not to have migrated. For example, when a dataset is highly struct ured with discretely clustering subsets, the cluster displaying values similar to those of local sediment leachates likely approximates the local dietary catchment. For all proxy measures, however, there is the potential for disagreement because each has t aken up lead and strontium from different sources across a different


66 geographic area. Furthermore, not all sources contribute equal amounts of strontium or lead to osseous mineral. Sea salt, for example, has a relatively high ratio of strontium to calcium (Sr/Ca) and thus contributes more strontium to bones and teeth than many other foods (Burton and Wright 1995) Given its propensity to be traded across long distances in many archaeological contexts, sea salt c an potentially skew the isotopic range displayed by individuals that actually lived and died within the study area (e.g., Wright 2005). Also, lime solutions used by many New World cultures to process maize contri bute very large quantities of strontium, effectively obscuring isotopic ratios from the rest of the diet. Even without such unusually rich sources of Sr a relatively small number of plant food sources will tend to dominate 87 Sr/ 86 Sr given Sr/Ca differences among dietary items (Burton and Wright 1995). Nevertheless, only the regular consumption of mineral enriched foods from specific non local areas is likely to alter isotope ratios in the average diet relative to local values. In vivo anthropoge nic lead contamination must also be considered as certain cultural practices can result in disproportionately high exposure to lead from specific geological sources. In modern contexts, particular kinds of surma, kohl, and other mineral bearing cosmetics h ave b een linked with elevated blood Pb levels (e.g., Parr y and Eaton 1991 ; Gorospe et al. 2008 ). Generally speaking, however, the risk of Pb contamination is impacted by multiple fac tors including hand to mouth activity, inhalation of resuspended soil Pb, and occupational Pb exposure (Gogte et al. 1991; Kadir et al. 2008; Qu et al. 2012; Zahran et al. 2013) In particular, the parental transport of Pb bearing dust from metallurgical occupations to domestic contexts poses an increased risk of Pb poisoning (Roscoe et al. 1999) Pervasive Pb exposure in archaeological


67 the convergence of Pb isotope ratios and a corresponding spike in Pb concentrations across individuals in a population (Montgomery et al. 2005) Marine aerosols and other kinds of aeolian deposition deriv ed from non local areas can also significantly influence the heavy is otope ratios of a region (Whipkey et al. 2000; Komrek et al. 2008; Evans et al. 2010) The potential for aeolian influence is particularly important to consider when interpreting lead isotope data from osteological remains given that ingested dust rat her than the food itself, contributes much of the lead in animal tissues ( Elias et al 1982; McBride 1994 ; Kohn et al. 2013 ). Unfortunately, airborne particulate from industrial activities has swamped the natural background lead even in remote re gions of t he globe (Bindler 2011; Klaminder et al. 2011) Modern contaminants like lead in industrial pollution and strontium in fertilizers can affect the isotope ratios of modern samples used to evaluate regional isotope varia bility (Bentley 2006). Because of globally ubiquitous lead contamination, modern plants and animals should not be used to approximate background values for lead isotopes. S ediment samples taken from lower in the soil profile however, can be used given the tendency of lead to bind with organic content in the upper levels (Klaminder et al. 2006) In a similar vein, samples exposed to certain fertilizers and other strontiu m rich sources should be avoided when estimating natural isotopic variability. Light Stable Isotopes In contrast to heavy isotopes, the isotope ratios of lighter elements like carbon and oxygen undergo fractionat ion when they are involved in physical, che mical, and biological processes The mass difference between isotopes of light element is relatively large, causing o ne isotope to be preferentially selected over the other as they undergo


68 chemical reactions and phase changes. This change in ratios is reported in parts per mil (Coplen 1994) and because of the systematic nature of fractionation processes, isotope measurements can be used to make inferences about certain constraining environmental factors. For example, oxygen isotope fractionation is mediated by the temperature of different water sources and evaporation, and varies with respect to related changes in altitude and latitude (Dansgaard 1964) In this way, oxygen isotope values from biological apatites and other carbonate sources reflect climat e conditions. Interpreting light stable isotope data is not strictly a matter of determining weighted a verage values for a range of environmental inputs. One must also account for the fractionation processes of different elements. Stable o xygen i sotopes The stable isotopes of oxygen relevant for archaeological purposes are 18 O and 16 O; their ratios in bones and teeth are the end result of a long chain of fractionation processes typically beginning with the evaporation of ocean water from tropical latitudes. Water molecules containing lighter 16 O isotopes preferentially evaporate resulting in water vapor with 18 O Subsequently, condensation and precipitation favor the heavier 18 O 18 O compared to water vapor. This fractionation factor increases at lower temperatures resulting in increased depletion of 18 O for water vapor. Further, as the moist air mass water vapor even more depleted in 18 O Thus, rainfall derived from a given air mass exhibits progressively lower 18 O 18 O for a given air mass and its precipitat ion decreases with lowered temperature, high precipitation, high latitude, high 18 O increases with higher


69 temperatures, low precipita tion, low latitude, low altitude, and proximity to the ocean (Bowen and Wilkinson 2002; Gat 1996; Kendall and Coplen 2001) In tropical latitudes, variation in 18 O whereas temperature has a greater influence on seasonality at higher latitudes (Dansgaard 1964; Rozanski et al. 1993) Adding to the complexity, s easonal changes in weather patterns may bring moisture from different regions, causing further differentiat ion in the isotopic values of intra annual precipitation (Fricke and O'Neil 19 96; Scholl et al. 2009) Plants and animals acquire water from different sources with potentially different 18 O For example, rivers originating in the highlands, meteoric water, ground water, and lake or pond water from the same region might all have 18 O because their isotopic compositions have evolved under variable regimes of condensation, evaporation, and source mixing. The water management practices of humans are similarly diverse, potentially drawing from multiple environmental sources storing water in evaporation prone vessels and reservoirs, or boiling water during cooking (Knudson 2009) Animals also acquire and conserve water using multiple strategies. For example, some animals get the bul k of their water through vegetation wh ereas others must drink 18 O of leaf water is strongly influenced by processes of evaporative enrichment. In the case of obligate drinkers, however 18 O more closely tracks the variability in environmental water sources (Levin et al. 2006) Once ingested, oxygen isotopes fractionate further as a consequence of m etabolic processes 18 O is the aggregate of isotopic input from drinking water, food, and air modified by fractionation in output through urination,


70 perspiration, and exhalation. Biological apatites in bones and teeth form in isotopic equilibrium with body water and act as an 18 O capturing an aggregate measure of complex environmental and physiological fractionation mechanisms (Longinelli 1984; Luz et al. 1984; Luz and Kolodny 1985) Enamel formed during early childhood, in particular, records the isotopic input from breastmilk An isotopic shift associated with weaning can be observed between early and late forming dentition because breastmilk is derived from body water, a reservoir relatively enriched in 18 O compared to environmental sources The magnitude of the shift has been measured in (Wr ight and Schwarcz 1998) although the shift may be obscured by cultural practices including the consumption of relatively 18 O wa ter such as that in boiled foods (Daux et al. 2008) 18 O also vari es geographically, it can be used to provenience osseous remains in much the same way as heavy stable isotopes (e.g., Sch warcz et al 1993; Dupras and Schwarcz 2001; White et al. 2004 b ; Prowse et al. 2007; Wright 2012). Geographic comparison is less direct than is the case for strontium and lead, however, because climate and hydrology are less stable than bedrock composition Therefore determining an appropriate sample type to represent 18 O ranges is more challenging. For one thing, modern samples may not capture the range of past values Paleoclimat e records derived from inorganic carbonates sidestep this problem bec ause they formed under the direct influence of past climate conditions, but they fail to account for fractionation in the biosphere. To facilitate the comparison of biological and inorganic values, several methods have been proposed to predict environmenta 18 O from 18 O (e.g., Longinelli 1984; Luz et al. 1984; Levinson et al. 1987; Daux et al.


71 2008) but they have an error of approximately 1 (Pollard et al. 2011) 18 O from human bone and enamel can be compared with faunal data following the same logic used for determining local variation in heavy stable isotope ratios. Unfortunatel 18 O derived from humans and other animals may not be compa rable because of variation introduced by cultural behaviors (White et al. 2004a) Therefore comparisons with archaeological fauna should be attempted when possible, but highly structured data sets may offer the best route to identifying immigrants. Stable c arbon i sotopes Stable carbon isotopes 13 C and 12 C also undergo fractionation in the biosphere prior to being incorporated into bone and tooth enamel. Carbon isotope values in plant tissues show large diffe rences from the value in atmospheric CO 2 Depending on the photosynthetic pathway used, plants differentially discriminate against the heavier 13 C isotope as a function of enzymatic differences. Most cool season grasses, trees, and herbs use the Calvin cyc le (C 3 plants) and have a mean 13 C of approximately (Kohn 2010) whereas warm season grasses, sedges, and some dicotyledons use the Hatch Slack cycle (C 4 plants) and have a mean 13 C of (O'Leary 1988) Certain arid adapted succulents and epiphytes employ crassulacean acid metabolism (CAM plants) resulting in intermediate values, although such plants rarely form a major constituent of human diets. A variety of environmental variables cause small scale variation (~3 3 plants, although there is relatively little intraspecific variation in 13 C among C 4 plants (Tieszen 1991 ; Brookman and Ambrose 2013) Many such changes result from variation in the partial pressure of CO 2 Plants increasingly discriminate against 13 C as intercellular pressure goes up, causing a corresponding decrease in 13 C (Farquhar et


72 al. 1982) Increases in pressure can result from decreased irradiance, whereas relatively arid conditions can trigger reduced stomatal conductance, decreasing the partial pressure of CO 2 in some plants. These and other factors show broad geographic and climatic trends such that 13 C increases slightly with altitude and mean annual temperature while decreasing slightly at higher latitudes (Francey and Farquhar 198 2; Farquhar et al. 1989) 13 C in understory environments is caused by the recycling of isotopically lighter CO 2 from decomposing leaf litter (van der Merwe and Medina 1989; van der Merwe and Medina 1991) In animals, metabolic processes cause further carbon fractionation The process is not uniform for all carbon bearing compounds however, as carbon from the protein portion of diet preferentially contributes to bone collagen, and carbon from the whole diet contributes to bone and tooth apatite. Consequently, isotopic values for the different compartments of bone have different offsets from dietary 13 C Collagen is isotopical ly heavier than the food source by ~3.5 for bone apatite carbonate ranges between ~9 and (Krueger and Sullivan 1984; Ambrose and Norr 1993; Tieszen and Fagre 1993; Howland et al. 2003; Jim et al. 2004) Notably, the magnitude of the offset between the food source and bone apatite or collagen varies among organisms For apatite carb onate, for example, herbivorous ungulates are offset by ~12 (e.g., Cerling and Harris 1999; Balasse 2002) and lab rodents between ~9 (e.g., De N iro and Epstein 1978; Jim et al. 2003). The interspecific variation in offset values is important for data interpretation in that the difference in 13 C between C 3 plants and C 4 (Passey et al. 2005) Taxonomic differences are partly explained by dietary differences as low meat intake does not allow for the direct


73 routing of ingested amino acids into protein synthesis and thus leads to fractionation assoc iated with the synthesis of proteins from carbohydrates (Schwarcz 2002) Further, carnivorous diets are high in isotopically light lipids, but diet alone fails to explain the full range of interspecific varia tion. Differences in digestive physiology likely contribute to the variability; production and absorption of isotopically light methane may well explain lower 13 C for ruminants (Hedges 2003) The offset for apa tite in humans is variably a range that might be partially explained by omnivory and cultural variation in diet (Ambrose and Krigbau m 2003; Harrison and Katzenberg 2003) Finally, modern era industrial emissions of isotopically light CO 2 require that an additional offset of 13 C to make comparisons with modern data (Friedli et al. 1986) Therefore an archaeological human with a pure C 4 diet will have depending on local values for C 3 and C 4 plants. Isotope ratios from apatite carbonate alone give an approximate sense of C 3 vs. C 4 plant consumption and are most useful in detecting broad relative differences in diet although dietary inferences are limited without corresponding collagen 13 C More precise reconstruction of diet from carbon isotopes becomes possible when the organic content of bones has been preserved and apatite carbonate 13 C can be compared with collagen 13 C (Kellner and Schoening er 2007) In ideal circumstances, a variety of faunal specimens in the local food web can also be analyzed to provide constraints on regional isotopic variation. Osteological Development and Diagenesis Many tissues can be analyzed for isotopes, but osse ous materials tend to be the best preserved in archaeological contexts. Despite their lack of collagen, teeth have developmental and structural qualities that make dental enamel a more suitable choice


74 for isotopic analysis than bone in many contexts. For o ne thing, dental enamel forms an incremental record of environmental conditions at the time of mineralization. Mineralization spans several years per tooth in humans, after which the tooth becomes largely impermeable to further chemical exchange. Additiona lly, different tooth types systematically and predictably mineralize at different chronological ages in childhood. Consequently, any given tooth permanently stores isotopic data from a specific time period 1) (Hillson 1996) Through intertooth sampling then, bone chemistry tools can help reconstruct aspects of early life mobility, diet, and climate that are otherwise difficult to infer. Indeed, the increased interpretive pot ential of intertooth sampling has increasingly led to explicit intertooth sampling protocols within isotopic research designs (e.g., Wright 2012; Giblin et al. 2013). Any analysis of e arly permanent dentition such as first molars must also consider the potential implications of breastfeeding and weaning. As discussed earlier, stable oxygen isotopes demonstrate a trophic shift and can be used to identify the transition to drinking water. Likewise, stable carbon isotopes may indicate the transition to solid foods if the foods used for weaning have an isotopic composition distinct from breast milk (Wright and Schwarcz 1998) Weaning may also be i ndicated in lead isotope ratios of permanent teeth when prepartum maternal residence is isotopically distinct from postpartum residence (Gulson et al. 2003; Manton et al. 2003) The skele ton acts as a reservoir for lead which may be mobilized along with bone calcium during lactation, although less of the maternal lead burden is released into the blood when the diet is calcium rich (Gulson et al. 2 004) In one study on modern immigrant mothers, breast milk lead accounted for 36 80% of infant blood lead for the first 90 days of life (Gulson


75 et al. 1998) although Manton and coworkers (2000) found that dust contributed far more to infant blood lead than breast milk. Presumably, a similar process to that described in Gulson et al. (2004) could happen with maternal strontium, but biopurification against strontium in the mammary glands suggests that other sources of strontium would have a greater influence on infant isotopic ratios (Wasserman et al. 1958; Blakely 1989) Apart from documenting life history, tooth enamel has the additional advantage of resisting diagenetic alteration. Bone mineral is relatively porous and permeable to the labile ionic content of surrounding sediments. By contrast dental enamel is made of larger crystals arranged in a tighter lattice that resists penetration by diagenetic agents (Driessens and Ve rbeeck 1990; Budd et al. 2000a; Budd et al. 2000b; Chiaradia et al. 2003) In the case of strontium and lead at least, what little diagenetic alteration occurs seems to be confined to the outermost surface (Budd et al. 1998) Some authors have suggested that successive weak acid pretreatments could recover biogenic strontium isotope ratios from diagenetically altered bone (Sillen 1986; Sillen 1989) but later work (Hoppe et al. 2003; Trickett et al. 2003) Thus dental enamel remains the most reliable material for heavy stable isotope analysis with little preparation required beyond abrasion of the surface enamel. Nevertheless, weak acetic acid pretreatments can help remove weakly adsorbed carbonates, and the fact that different pretreatment protocols yield comparable analyses for strontium and lead (Valentine et al. 2008) suggests that there is no harm in pretreating teeth with weak acetic acid for less than 16 hours. Given that the source of isotopically analyzed carbon and ox ygen for this study was structural carbonate, however, pretreatment was necessary to remove adsorbed


76 contaminants for light stable isotope analysis. Care must be taken to limit the strength and duration of acetic acid treatment to prevent the dissolution a nd reprecipitation of structural carbonate (Koch et al. 1997; Garvie Lok et al. 2004) Apatite phosphate may also be used for stable oxygen isotope analysis and yields systematical ly related data, but expenses for the preparation of phosphate are significantly greater (Sponheimer and Lee Thorp 1999; Chenery et al. 2012) The enhanced molecular strength of apatite phosphate and its increased resistance to diagenesis make it particularly appropriate for research on a geological time scale (Bunton et al. 1961; Lecuyer et al. 1999) Lab oratory and Field Methods Sample collection was guided by a few basic principles. Whenever possible, bulk enamel samples were collected from the first, second, and third molars of human burials to ascertain isotopic life history in three stages: from birth to age three, age three to six, and age eight to twelve. In a few instances when first or second molars were unavailable, teeth with roughly similar mineralization times were sampled instead. For purpose s of visual presentation, different tooth types are grouped into cohorts (first, second, and third molar cohorts) based on similar enamel formation times. Incisors are grouped with first molars, whereas canines and p remolars are classified as second molars. Further, undomesticated or commensal faunal were s elected along with sediments to provide proxy measures for regional variation. Every effort was made to fully document teeth before destructive sampling was undertaken. Unless stated otherwise, photographs or scans were taken and dental impressions were ma de for all teeth using 3M ESPE Express Vinyl Polysiloxane Impression Material regular body. Impression material was applied to the tooth crown using a Garant Hand Dispenser, after which the tooth was inverted on a smooth surface until dry.


77 Once fully do cumented, approximately 50 mg of tooth enamel spanning cusp tip to cementoenamel junction was collected in one of two ways. Teeth taken from the Harappa Archaeological Research Project were brought to the University of Florida Bone Chemistry Lab and sample d under 10x magnification using a Brassler dental drill with a diamond bit. Teeth collected from the Archaeological Survey of India and Deccan College Post Graduate and Research Institute were sampled using a rotary tool with a diamond cutting wheel. Sampl e collection proceeded in four phases. In 2009, the Harappa Archaeological Research Project provided access to regional fauna and sediments from the following sites: Allahdino (Sus, N=5), Balakot (sediment, N=3), Harappa (Canis, N=3; Sus, N=8), Mehrgarh (C anis, N=2; sediment, N=2), and Nausharo (Equus, N=4; Gazella, N=2). No dental impressions were made during this initial pilot phase. During the 2010 field season, Farmana human teeth were sampled from 21 individuals (N=37) curated at Deccan College Post Graduate and Research Institute, Pune, India. Tentative estimates of age and sex were made following Buikstra and Ubelaker (1994), but few reliable identifications could be made given the extremely poor preservation and general scarcity of skeletal elements. Most skeletal remains were still encased in soil, some of which was collected (N=3) for isotopic analysis Additional sediment sampl es (N=3) were collected from the site of Sanauli. A clean soil profile was exposed using a shovel and approximately 200 mg of sediment collected from roughly 70 cm below the surface to avoid recent anthropogenic lead bound in the upper levels (Klaminder et al. 2006)


78 In 2011, human teeth from Sanauli were sampled for 33 individuals (N=67) curated by Archaeological Survey of India at Purana Qila, New Delhi, India. Again, a paucity of skeletal elements hindered accurate estimation of age and sex beyond an adult/sub adult distinction. Faunal samples (Sus, N=8) were also collected from the Rakhigarhi assemblage curated at Deccan College. In 2012, an opportunistic sample of teeth from 44 individuals at Harappa (N=51) was collected from the Harappa Archaeological Research Project. Most had been previously sampled for carbon, oxygen, and strontium isotope analysis by Drs. Jonathan Mark Kenoyer and Douglas Price University of Wisconsin, Madison. No additional documentation was undertaken before sampling. Unpublished age and sex data were made available by Dr. Kenoyer. At the University of Florida Department of Anthropology Bone Chemistry Lab, the surfaces of all enamel samples were abraded and all dentine was removed using a dental drill. S amples were then ground into powder with an agate mortar and pestle before being pretre ated with 2 ml of reagent grade 50% Na O Cl solution in a centrifuge tube for 16 hours to remove organic content. The supernate was pipetted out, after which samples were rinsed with 2x distilled water until neutral. The same procedure was then carried out u sing 0.2N acetic acid to remove adsorbed carbonate. The pretreated samples were finally freeze dried for 72 hours and stored in a desiccator At the University of Florida Department of Geological Sciences clean lab facilit y approximately 20 30 mg of pretr eated enamel powder for each sample was dissolved in pre cleaned Teflon vials by heating for 24 hours in 8N nitric acid (HNO 3 ) (optima). The vials were then opened and evaporated to dryness in a laminar flow hood. For each


79 sediment sample, approximately 10 0 mg of sediment was leached in pre cleaned Teflon vials for 2 hours using 4 ml of 0.1N acetic acid. The leachate was pipetted off, the sample evaporated to dryness, an additional 4 ml of 2N HCl added, and the leachate pipetted off once more. Both leachate s were evaporated to dryness in preparation for lead and strontium separation. Ion chromatography was then used to separate strontium and lead from single aliquots. Lead was purified using conventional hydrobromic acid (HBr) procedures on Dowex 1X 8 resin, and the washes were collected for further strontium separation as the latter element is not absorbed on the resin. The dried residues were dissolved in 2 ml of 8N nitric acid, producing bromine gas from any residual hydrobromic acid which would otherwise interfere with strontium separation. Once dried, the samples were redissolved in 3.5N nitric acid and loaded onto cation exchange columns packed with strontium selective crown ether resin (Sr spec, Eichrom Technologies, Inc.) to separate strontium from oth er ions, following the procedure by Pin and Bassin (1992) All samples were analyzed using the mass spectrometry facilities at the University of Florida Department of Geological Sciences. Lead isotopic ratios were collector inductively coupled plasma mass spectrometer (MC ICP MS) following the thallium normalization technique of Kamenov and coworkers (2004) The data are reported relative to the following values of NBS 981: 206 Pb/ 204 Pb = 16.937 0.004 ( 207 Pb/ 204 Pb = 15.490 0.003 ( 208 Pb/ 204 Pb = 36.695 0.009 ( Pilot strontium data for samples from the 2009 phase were also analyzed by MC ICP MS followi ng the time resolved analysis method of Kamenov et al. (2006) The


80 long term reproducibility of the TRA measured 87 Sr/ 86 Sr of NBS 987 is 0.710246 0.000030 ( 87 Sr/ 86 Sr tungsten filaments, the samples were run at 1.5V for 100 ratios whenever possible, and the resulting data normalized to 86 Sr/ 88 Sr = 0.1194. The long term reproducible 87 Sr/ 86 Sr of NBS 987 is 0.710240 0.000023 ( preparation device. No duplicate samples were prepared for teeth that had already been run by Drs. Kenoyer and Price. All others were reacted in a common acid bath at 90 C and water was cryogenically removed in a methanol slush. Evolved CO 2 gas was measured online and all isotope result s are reported in standard delta notation relative to Vienna Pee Dee Belemnite (VPDB). Analytical precision (1 standard deviation of 13 18 O


81 Table 3 1. Approximate crown form ation time in years 1 Incisor Canine Premolar First Molar Second Molar Third Molar Initial Formation < 1 < 1 2 6 0 2.5 5 7 11 Crown Completion 3.5 5 4 7 5 9 2 4 6 9 10 17 1 estimated ranges derived from minimum and maximum values in [ Hillson S. 1996. Dental Anthropology ( Table 5 1 ), Cambridge: Cambridge University Press]


82 CHAPTER 4 GEOCHEMISTRY OF THE GREATER INDUS REGION Environmental Background Stretching between the Himalayas and the Arabian Sea, the Indus settlement area is bounded o n the west by mountain ranges of the Western Fold Belt and on the east by the Aravallis. The Indo Gangetic plains derive largely from Quaternary Himalayan detritus overlying and abutting the Indian Shield to the south, thus regional geochemistry can be des cribed primarily in terms of Himalayan geological units. Even sediments of the Makran coast and the Kirthar and Sulaiman Ranges to the west have their ultimate origins in rivers draining the Himalayas. However, the Himalayas are composed of distinct thrust sheets, each with its own lithology, resulting in a geochemically diverse landscape. Approximately 55 million years of orogeny have dramatically altered the Indian plate and created the vast alluvial plains of the Himalayan peripheral foreland basin in w hich Indus cities thrived (Kazmi and Jan 1997 ; Ramakrishnan and Vaidyanadhan 2010). The subduction and subsequent collision of the South Asian subcontinent with the Eurasian plate caused the sequential uplift and ex humation of the major Himalayan units (Figure 4 1) The Trans Himalaya is northernmost and represents the leading edge of the Eurasian plate. It is composed of several major sub units including the Hindu Kush and Karakorum Ranges. The Trans Himalaya is sep arated from the Indian formations to the south by the Indus Tsangpo S uture Z one (ITSZ) which channels the upper course of the Indus River. At its western end, the suture diverges around the Kohistan Arc, an ancient island arc sandwiched between Asia and th e subcontinent. Running parallel to the Trans Himalaya along the southern edge of the ITSZ is the


83 Tibetan Sedimentary Series (TSS) or Tethys Himalaya. To the south and likewise running from east to west are the High Himalayan Crystallines (HHC), the Lesser Himalaya, and the Sub Himalaya. The latter encompasses various sub basins of which the Neogene strata are collectively termed the Siwalik Group. All together, the Himalayas have a diverse lithology ranging from continental crust to ophiolites (obducted oc eanic crust) and young igneous formations to heavily metamorphosed strata. Different rocks have provided sediments to the foreland basin at different times as various Himalayan units and drainages evolved. Importantly for archaeological research, however, portions of the Indo Gangetic plains have been partially homogenized as a consequence of complexities of fluvial and aeolian transportation (Tripathi and Rajamani 1999; Tripathi e t al. 2007) This results in a relatively large scale geochemical mosaic conducive to isotopic studies of interregional mobility. Isotopic Variatio n Heavy Isotopes The Sr and Pb isotope systematics of the major Himalayan units are well characterized, allowing sediment provenience in some cases despite the isotopic overlap between units (Najman et al. 200 0; Clift et al. 2002) Further, the different isotope systems are independent of each other and their isotopic ratios need not correlate. Different geological units influence Sr and Pb isotope ratios in biological materials as a function of their solubi lity and Sr and Pb concentrations. Carbonates, for example, are easily weathered and tend to have high Sr/Ca but low concentrations of Pb. As a result, carbonates have a disproportionate effect on 87 Sr/ 86 Sr in archaeological tooth enamel. Although much geo logical data is derived from silicates and therefore not directly comparable with the soluble Sr and Pb archived in enamel, silicate data still indicate relative differences in the biologically available Sr and Pb isotope ratios. Based


84 on Sr and neodymium (Nd) isotope data, Najman and coworkers proposed that from most radiogenic to least, the isotopic end members contributing to the foreland basin are the Lesser Himalaya, the HHC, the TSS, and geologically young elements of the ITSZ including the Kohistan A rc. Clift et al (2002) depict similar influences on Pb isotope geochemistry in the Indus basin except their analysis pointed to Asian portions of the Trans Himalaya rather than the TSS. Few data are available for Pb isotopes in the Himalayan foreland bas in beyond the study of detrital grains by Clift et al. (2002). Nevertheless, the analysis by Clift and coworkers suggests that the Trans Himalaya and ITSZ are the primary controls on Pb isotope composition in the Indus main channel, whereas the highly radi ogenic values of the foreland tributaries (e.g., Chenab, Ravi, Sutlej) are largely derived from the HHC and Lesser Himalaya. The latter two geological units are expected to exert an even greater isotopic influence on Pb in river basins to the east as a fun ction of increased elevation and weathering. The Sr isotope literature is comparatively rich and includes analyses of the soluble fraction of Sr in river waters and sediments (Figure 4 2 ) (e.g., Karim and Veizer 2000; Tripathi et al. 2013). Within the Indus basin, 87 Sr/ 86 Sr da ta suggest three end members the HHC, the carbonate rich Western Fold Belt, and the Kohistan Arc (Pande et al. 1994; Karim 1999; Karim and Veizer 2000) In the upper reaches, the Indus drains mainly the low 87 Sr/ 86 Sr terrain of the Kohistan Arc and has ratios of ~ 0.710 The main channel becomes gradually more radiogenic (~0 .712) with increased contributions from the HHC. By the middle reaches of the Indus River however, 87 Sr/ 86 Sr dips slightly from the influx of high concentration, low 87 Sr/ 86 Sr waters draining


85 the Western Fold Belt. The Sr isotope ratios then remain relatively stable (~0.711) through the lower reaches as the main channel is joined by the more radiogenic waters of the foreland tributaries. Significantly for provenience studies, the main Indus channel exhibits lower 87 Sr/ 86 Sr than mos t of the foreland tributaries which drain the uniquely radiogenic carbonate terrain of the HHC (Palmer and Edmond 1992) Indeed, Kenoyer and coworkers (2013) e stimate high biologically available 87 Sr/ 86 Sr (~0.720) in the area near Harappa. Ratios taper off gradually to the west and south, whereas higher values are most likely found in the highlands to the north and northeast. The situation is similar for other r ivers with headwaters that drain the HHC. The lowland waters of the Yamuna are broadly comparable to those of the radiogenic Indus foreland tributaries (Dalai et al. 2003) whereas the most radiogenic 87 Sr/ 86 Sr values for any Himalayan river are found in the Ganges where ratios can be as high as 0.741 (Krishnaswami et al. 1992; Chakrapani 2005) Limited Sr isotope data are also available for sediments and leachates in regions near Farmana and Sanauli (Tripathi et al. 2004; Tripathi et al. 2013) Samples from the semi a rid region between the Yamuna and Ghaggar Hakra likely approximate the geochemical environment at Farmana, wh ereas those near the Yamuna less than 100 km to the east provide a 87 Sr/ 86 Sr range estimate for Sanauli. Despite their proximity, the two areas a ppear to be isotopically distinct. Leachates of Yamuna alluvium have 87 Sr/ 86 Sr between 0.715 and 0.718, significantly higher than the 0.711 0.716 range exhibited by dust and sand leachates in the Ghaggar Yamuna interfluve. Bulk sediment analyses confirm that the Yamuna alluvium is more radiogenic, but the most striking comparison is between aeolian sediments of the Ghaggar Yamuna interfluve and the


86 Thar Desert fringe near the Aravalli foothills less than 150 km to the south. Samples from the two areas have essentially identical ratios an unsurprising similarity given their potentiall y shared origins. Current evidence suggests that aeolian sediments from the lower Indus contribute significantly to the Thar Desert and adjacent terrain (Alizai et al. 2011) These sediments likely retain high co ncentrations of low 87 Sr/ 86 Sr as in the lower Indus River, and may therefore swamp the relatively radiogenic ratios derived from local alluvial sources. Beyond the Himalayan foreland basins, the Indus settleme nt area extended southward into Saurashtra Iso topically speaking, the region is not well described, but provisional estimates of low 87 Sr/ 86 Sr (< 0.710) can be made based on regional lithology (Chamyal et al. 2003) The presence of young volcanic rocks, lime stones, and even aeolian sands from the Thar Desert suggest s a generally less radiogenic environment, although it is difficult to estimate the more radiogenic contribution from the Precambrian rocks of the adjacent Aravallis. Isotopic Variation Light Stab le Isotopes The South Asian hydrological landscape is also conducive to provenience research using stable isotopes of oxygen 18 O ) Summer and winter rains originating in the southern oceans and the Mediterranean, respectively, serve as the primary source s of precipitation in the Indus watershed (Wright et al. 2008) Further, different vapor sources have different 18 O as exemplified in climatological studies of nearby Oman (Weyhenmeyer et al. 2000; Weyhenmeyer et al. 2002) showing that northern sources range between 10 and the primacy of summer monsoon precipitat ion results in relatively low average 18 O associated with many parts of th e subcontinent. By contrast, relatively high 18 O winter


87 storms move along the Western Fold Belt and into the western Himalayas (Wright et al 2008). Many regions of the Indus Civilization, including the city of Harappa, relied on both winter and summer rains to support seasonal multi cropping. It is expected, th en, that the relative contributions of summer and winter precipitation will be reflected in 18 O of archaeological tooth enamel apatites, much as it is today in the waters of the Indus tributaries (Karim and Veizer 20 02) Even in those regions unaffected by Mediterranean precipitation, systematic variation in the summer monsoon creates gradients of 18 O (Gupta and Deshpande 2005; Lambs et al. 2005) Th is complicated phenomenon can be explained in part by four factors: isotopically distinct vapor sources, rainout, evaporative effects, and altitude effects. Summer monsoon begins as the annual northward migration of the Intertropical Convergence Zone (ITCZ ) creates low pressure systems in the northwestern subcontinent. High pressure oceanic air masses then move onto land, releasing rains that reflect the isotopic composition of the vapor source. For the Indo Gangetic p lains, this primarily means relatively low 18 O in the Bay of Bengal monsoonal branch. The northwest travelling air mass becomes gradually more depleted in 18 O with greater distance from the ocean because of rainout (Chapter 3). The Bay of Bengal branch alone, however, cannot explain modern var iation in South Asian surface waters. The Arabian Sea branch of the summer monsoon has relatively higher 18 O ( ~ contributes measurably to the precipitation budget of Pakistan and Northwest India (Gupta et al. 2005; Sengupta and Sarkar 2006) Further 18 O enrichment occur s at local level s depending on evaporation.


88 Lastly, high altitude drainages channel relatively depleted waters from the Himalayas down to the plains. The headwate rs of the Indus (Karim and Veizer 2002), Yamuna (Dalai et al. 2002) and Ganges (Ramesh and Sarin 1992) for example, have lower 18 O than do lowland channels. By comparison, rain fed rivers draining the Himalayan foothills should have relatively high 18 O The most significant rain fed river in the Greater Indus Valley during the third millennium BC was the Ghaggar Hakra. At times identified w ith the perennial glacier fed Sarasvati of Vedic tradition, this now seasonal river in the Sutlej Yamuna interfluve is associated with numerous Indus settlements (Mughal 1997; Possehl 2 002b) Recent work, however, confirms that it was a perennial monsoon fed river (Giosan et al. 2012) and as such, would likely have had relatively high 18 O The data from modern water sources cannot be compa red directly to archaeological tooth enamel, but they still help estimate the relative differences that existed in different regions during the third millennium BC. In particular, studies of South Asian river water provide the most comprehensive 18 O data set from which to infer variation across the Greater Indus Valley (Figure 4 3 ) Broadly speaking, the Indus, Ghaggar Hakra, and Yamuna Rivers represent distinct isotopic regions. The Indus can be further subdivided into the upper reaches, the western tribu taries, and the combined region covered by the foreland tributaries and the lower reaches of the main Indus channel. The latter region includes the major urban centers of Harappa and Mohenjo ion. During the summer months the lower Indus and foreland tributaries share similar 18 O Far greater variability in 18 O is found in the middle and upper Indus because of the countervailing


89 influences of high altitude and relatively enriched winter rain s. For example, the high altitude reaches of the Indus main channel have lower 18 O compared to the foreland 18 O han the foreland tributaries a diffe rence presumably attributable to greater winter precipitation (Karim and Veizer 2002) Similarly high values might also have been found in the far southeastern regions of the Indus Civilization as inferred from modern 18 O in the Narmada and Tapti rivers d raining western India (Lambs et al. 2005) A more modest increase in 18 O can be found to the east of the Indus heartland. As it emerges from the Himalayas, for example, the lower reaches of the Yamuna main channel exhibit 18 O ~ 2 han the foreland tributaries a difference that may be partly attributable to summer precipitation and the proximity of the Bay of Bengal. Even higher 18 O may characterize the Ghaggar Hakra basin given the rel atively low altitude headwaters and increased contribution from the Arabian Sea monsoon. Collectively then, modern river data suggest a complex hydrological environment in which the highest 18 O is found along the western highlands and southeastern coast followed by the Ghaggar Hakra basin ( F igure 4 3 ) The next highest values come from the Yamuna basin, the combined lower Indus foreland tributary region, and lastly the middle and upper reaches of the Indus excluding those tributaries draining western peak s. The riverine data suggest certain broad trends in 18 O but the potential complexities of local hydrology must always be considered. River water tracks the isotopic composition of seasonal precipitation regimes (e.g., Dalai et al 2002), wh ereas


90 variat ion in regional groundwaters can be even more complex depending on source mixing and evaporative effects (Gupta and Deshpande 2005; Lorenzen et al 2012). Considering the potential for diverse archaeological wa ter management practices, it is difficult to say which modern data sets best approximate protohistoric 18 O Large scale climatic changes in the late Holocene further complicate this matter. Precipitation gradually declined throughout the urban era and int o the second millennium BC, but western sources of winter precipitation dwindled more slowly (Wright et al 2008). It is uncertain how significantly major climate changes might have alter ed the broad relative differences identified across the region Never theless, spatial patterning in 18 O is significantly influenced by topography which has remained unchanged since Indus times. Furthermore, river water and tooth enamel samples both represent a spatial and chronological average of environmental water sources suggesting they may correlate closely despite the complexities of local hydrology. A final environmental consideration is the pattern of millet consumption (a C 4 crop) as reflected in stable carbon isotope values 13 C) of tooth enamel. There has been a tendency in the past to emphasize broad regional similarities and large scale chronological trends in Indus agriculture (e.g., Meadow 1989a; Meadow 1996; Weber 199 9 ). Weber and colleagues (2010) however, recently advocated a departure from a grand narrative in which a predominantly winter cropping Indu s heartland turned to millets during t he drier, post urban era. Instead, they emphasized regional variation in subsistence regimes and call ed for high resol ution ecological modeling. The view of Weber et al. serves as a reminder that agriculture in general, and millet consumption in pa rticular cannot be presented as either/or scenarios Therefore no simple predictions


91 can be made for 13 C at Harappa, Farmana, and Sanauli. Even given the high quality archaeobotanical data recovered from Harappa (Weber 2003) the non representative nature of Indus mor tuary populations precludes straightforward modeling of 13 C. Towards an Isotopic Baseline Prior to isotopic mortuary analyses at the primary study sites, sediments and faunal tooth enamel were used to evaluate interregional isotopic differences in the biologically available fraction of strontium, lead, and oxygen. Sample collection included materials from secure chronological contexts at Rakhigarhi, Allahdino, Balakot, Mehrgarh, and Nausharo as well as the mortuary sites of Farmana, Harappa, and Sanauli. As observed above, there is a general lack of regional data for the biologically availab le fraction of environmental Pb, however, i t is possible to provide estimates for 87 Sr/ 86 Sr and 18 O Allahdino an d Balakot, for example, are coastal sites near the Indus delta and Makran coast respectively. Each should reflect relatively less radiogenic contributions from the Western Fold Belt, as well as a potential marine influence on 87 Sr/ 86 Sr in the form of mari ne aerosols or through the consumption of marine food resources (Bentley 2006). In such case s one would expect 87 Sr/ 86 Sr intermediate between that of the lower Indus (~0.711) and the ocean (~0.709) (McArthur et al 2012) Mehrgarh and Nausharo, on the other hand, lie near the Bolan River which drains the Western Fold Belt. Therefore, local geochemistry should be consistent with winter storms (relatively high 18 O ) and carbonate rich highlands (~0.708 0.709). Far higher 87 Sr/ 86 Sr is expected at Harappa and Sanauli (~0.718 0.720) based on their proximity to Himalayan rivers, wh ereas Rakhigarhi and Farmana should fall somewhere in the middle due to carbonate rich aeolian contributions from the lower Indus region. Harappa 18 O should be broadly consistent with that of Allahdino and Balakot to the south,


92 slightly lower than Sanauli to th e east, and even lower than semi arid Rakhigarhi and Farmana to the southeast. The above regional estimates of isotopic variability were tested using sediment leachates and tooth enamel from archaeological fauna All environmental samples taken for this s tudy approximate the biologically available component of local isotope systems, but any proxy for human dietary catchments is problematic. One reason for this is that interspecific differences in diet and drinking water can result in isotopic variation wit hin a given locale. Pigs ( Sus ), dogs ( Canis ), equids ( Equus ) and gazelles ( Gazella ) consume vegetation under different regimes of evapotranspiration a dietary variable known to affect 18 O (Kohn et al. 1996; Schoeninger et al. 2000; Sponheimer and Lee Thorp 2001) Further, obligate drinkers ( e.g., pigs, dogs, equids, and humans) tend to have lower 18 O than browsing animals that receive their water pri marily through vegetation they consume ( e.g., gazelles). Other complications arise when considering the mobility of fauna or t heir food sources. The diet of non domesticated mobile animals may include foods from regions not exploited by nearby humans and c ould therefore be non representative of local geochemistry. Even domesticated and commensal animals need not have lived in the same place from birth to death, nor must they have been consumed or disposed of in the same location. Pigs have been shown to hav e low 87 Sr/ 86 Sr variation in certain contexts (Bentley and Knipper 2005) but they can have widely varying die t or be transported relatively long distances in other contexts (Hide 2003) Sediment leachates are similarly imprecise as they represent only a very small portion of a large human dietary catchment. Also, the fraction of Sr and Pb leachable by


93 methods used in this study may not precisely represent the soluble fraction taken up by biological organisms. For these reasons, any isotopic proxy for local human dietary catchments must necessarily be imprecise, and likely overestimates local ranges. Results of the Baseline Analyses The results of the baseline analyses presented in Table 4 1 show considerable interregional isotopic variation Values of 87 Sr/ 86 Sr range from 0.70792 to 0.72112 with Sr concentrations f rom 444 to 1098 ppm. Lead isotope ratios are similarly wide ranging for all three ratios : 208 Pb/ 204 Pb from 38.371 to 39.511 207 Pb/ 204 Pb from 15.655 to 15.823 and 206 Pb/ 204 Pb from 18.468 to 19.392 Faunal tooth enamel yielded 18 O of 6.1 to 13 C of 13.3 to Means and st andard deviations are presented on a site by site basis in Table 4 2 As a cost saving measure, MC ICP MS was used rather than TIMS for the Sr isotope analysis of environmental samples during the pilot phase of this study. Using TIMS, however, permits the simultaneous determination of Sr concentration. Consequently, no conce ntration data are available for most environmental samples. Only the subset of environmental samples analyzed using TIMS during the second phase of research are presented in Figure 4 4 No Sr concentrations were determined for sediment leachates as such da ta are non representative of the human dietary catchment and irrelevant to archaeological provenience studies. The Sr concentrations for fauna at Harappa and Rakhigarhi appear roughly equivalent with no obvious patterning. Also, there is a clear interregio nal difference in 87 Sr/ 86 Sr despite the overlap resulting from two outliers Mean 87 Sr/ 86 Sr at Harappa is 0.718 6 5 0.00270 ( distinctly higher than 0.716 17 0.00262 ( at Rakhigarhi t (19) = 4.12, p < 0 .001


94 All fauna and sediments are shown in the Pb Pb plots (Figures 4 4 & 4 5 ). Both distributions plot along an elongated triangular field suggesting three end members. Farmana and Sanauli plot at the uppermost corner of the triangular distribution, implying a distinct source of Pb in the east. Allahdino, Balakot, Mehrgarh, and Nausharo have the lowest values and are indistinguishable from each other. Therefore the south and west of the Greater Indus Valley likely share a common Pb source. In Figure 4 5 Harappa data are nearly identical to those from southwestern sites, but based on Figure 4 6 it may be that Harappa receives a distinct geochemical contribution from a third end member. Generally speaking, Pb isotope composition is more radiogenic at northeastern sites and less radiogenic in the southwest, but the slight offset in data from Harappa ( towards low 206 Pb/ 204 Pb and moderate 207 Pb/ 204 Pb ) suggests that Harappa is inf luenced by a particular Himalayan lithology that is not strongly represented either in th e lower reaches of the Indus River or in the more easterly alluvial plains Further, the data in Figure 5 3 plot in a broadly similar distribution to the regional lithological units depicted by Clift and coworkers (2002). Unfortunately, the large degree of overlap between the Pb Pb fields of Clift et al. preclude s precise provenience, and only tentative sourcing is possible without additional isotopic analyses of biologically available Pb throughout the Greater Indus region. One possibility is that the low isotope ratios of the lower Indus are derived from the ITSZ, whereas the high ratios at Harappa, Rakhigarhi, Farmana, and Sanauli are derived from the HHC and Lesser Himalaya. At Harappa, the relatively low 206 Pb/ 204 Pb as compared to sites further east may reflect proportionately greater input from the Lesser Himalaya.


95 Distributions of 18 O for faunal tooth enamel at the baseline sites are consistent with the expectations outlined previously (Figure 4 7 ). Very similar values at Harappa ( 3.1 2.3 [ ] ) an d Allahdino ( 2.9 2.4 [ ] ) show relative homogeneity within the Indus heartland, wh ereas slightly higher 18 O is found at Rakhigarhi ( 1.8 4.5 [ ] ) along the Ghaggar Hakra catchment ( t (19) = 2.188, p = 0 .041) The highest values are found at Mehrgarh (3.0 ) and Nausharo ( 2.5 3.9 [ ] ) as anticipated based on their proximity to the Western Fold Belt drainage. All fauna sampled are obligate drinkers with the exception of two gazelles recovered from Nausharo. The latter two samples show expected 18 O enrichment ( 7.8 1.5 [ ] ) and the data are considered separately fro m the rest of the faunal sample Distributions of faunal 13 C suggest no marked interregional differences, although equids at Nausharo have a very strong C 4 signal (1.0 0.8 [ ] ) as one would expect f or grazers in an arid environment ( Figure 4 8 ). Further, fauna l 13 C at Allahdino and Harappa exhibit a bimodal distributi on that is not explained by interspecific differences. Both pigs and dogs have high and low 13 C perhaps reflecting differences between wild and domesticated fauna or different provisioning practices. Their differences aside the majority of faunal 13 C d ata suggest a predominantly C 3 diet with slight to moderate contributions from C 4 foods. Far more geographic patterning is apparent in the heavy isotope ratios (Figure 4 9 ). The Sr isotope data partition largely as expected with southwestern sites domina ted by low 87 Sr/ 86 Sr sources and northern sites influenced by highly radiogenic Himalayan sediments. Slightly less radiogenic values at Farmana and Rakhigarhi may reflect low 87 Sr/ 86 Sr aeolian contributions from the southern Indus plains. The extremely low


96 87 Sr/ 86 Sr at Mehrgarh and Nausharo is probably controlled by the carbonate rich Western Fold Belt, and the modest increase in 87 Sr/ 86 Sr at Allahdino and Balakot likely comes from a co mbination of terrains: theTrans Himalaya and HHC sources in the middle and upper reaches of the main Indus channel as well as the foreland tributaries draining the Lesser Himalaya. Whatever the ultimate source, each region has distinct heavy isotopic range s. Further, the faunal isotopic distributions from Harappa and consistent with a primarily local dietary catchment supplemented by the long distance transportatio n of faunal resources. Consequently, the smaller clusters may represent local dietary catchments more accurately than the inclusive faunal data set. Lastly, sediment leachates and fauna from the same site are broadly consistent, suggesting that both proxie s are useful in determining local ranges of isotopic variation in the Greater Indus region. In summary analyses of archaeological fauna and sediment leachates support inferences from the literature that the Indo Gangetic Plains and adjacent areas are geoc hemically distinct and suitable for archaeological provenience studies. Though much additional work is required, th is developing data set appears to be systematically related to proximal sediment sources. Even where environmental data are presently unavail able, this suggests that some constraints can be put on the provenience of individuals who do not precisely match the current comparative data set.


97 Table 4 1 Baseline s amples results of the a nalyses Sample Site Leachate/ Taxon 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C AS1 Allahdino Sus 0.70876 38.702 15.674 18.555 1.5 10.9 AS2 Allahdino Sus 0.71082 38.840 15.689 18.647 2.4 10.8 AS3b Allahdino Sus 0.70873 39.103 15.705 18.795 2.6 11.7 AS4 Allahdino Sus 0.71088 38.945 15.697 18.702 3.0 10.7 AS5 Allahdino Sus 0.70870 38.696 15.668 18.565 4.8 0.7 BS3a Balakot acetic 0.70890 38.950 15.706 18.676 BS3h Balakot HCl 0.70894 38.990 15.704 18.698 BS5a Balakot acetic 0.70888 39.132 15.723 18.785 BS5h Balakot HCl 0.70896 39.146 15.717 18.792 BS13a Balakot acetic 0.70869 39.161 15.727 18.796 BS13h Balakot HCl 0.70883 39.186 15.727 18.810 F2Sa Farmana acetic 0.71553 39.421 15.823 19.339 F2Sh Farmana HCl 0.71565 39.452 15.822 19.341 F20Sa Farmana acetic 0.71559 39.393 15.817 19.297 F20Sh Farmana HCl 0.71565 39.402 15.813 19.307 F53Sa Farmana acetic 0.71554 39.425 15.821 19.339 F53Sh Farmana HCl 0.71594 39.449 15.820 19.343 HC1 Harappa Canis 0.71828 38.860 15.720 18.687 2.3 2.7 HC2 Harappa Canis 0.71797 1098 39.071 15.733 18.874 3.4 8.7 HC3 Harappa Canis 0.71828 39.027 15.728 18.762 3.8 7.8 HS1 Harappa Sus 0.71913 39.042 15.733 18.779 2.4 10.2 HS2 Harappa Sus 0.72112 38.939 15.733 18.684 1.4 7.2 HS3 Harappa Sus 0.71796 38.919 15.730 18.718 3.8 11.2 HS4 Harappa Sus 0.72084 498 39.087 15.737 18.791 2.7 9.3 HS5 Harappa Sus 0.71569 444 38.818 15.741 18.674 4.1 1.6 1 light stable isotope data from [ Kenoyer J. M., T. D. Price, and J. H. Burton 2013. A New Approach to Tracking Connections between the Indus Valley and Mesopotamia: Initial Results of Strontium Isotope Analyses from Harappa and Ur. Journal of Archaeological Science 40(5):2286 2297 (Page 2294, Table 2)]


98 Table 4 1 Continued Sample Site Leachate/ Taxon 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C HS6 Harappa Sus 0.71855 38.580 15.728 18.468 4.7 10.1 HS7 Harappa Sus 0.71795 38.979 15.736 18.796 3.2 12.9 HS8 Harappa Sus 0.71908 853 38.955 15.737 18.735 4.2 11.5 HA05 1 Harappa Sus 0.71892 663 39.149 15.749 18.845 0.7 7.6 HA06 1 Harappa Sus 0.71869 837 39.189 15.749 18.858 3.9 10.7 MC1 Mehrgarh Canis 0.70815 38.886 15.696 18.659 3.0 9.9 MC2 Mehrgarh Canis 0.70802 38.814 15.685 18.622 MS85a Mehrgarh acetic 0.70795 39.082 15.708 18.795 MS85h Mehrgarh HCl 0.70803 39.115 15.708 18.798 MS132a Mehrgarh acetic 0.70792 39.023 15.706 18.751 MS132h Mehrgarh HCl 0.70812 39.070 15.707 18.765 NE1 Nausharo Equus 0.70827 39.105 15.708 18.782 4.0 0.6 NE2 Nausharo Equus 0.70825 38.934 15.697 18.662 2.0 0.8 NE3 Nausharo Equus 0.70821 38.750 15.681 18.638 0.0 1.6 NE4 Nausharo Equus 0.70820 39.051 15.694 18.748 4.1 1.2 NG1 Nausharo Gazella 0.70811 38.888 15.690 18.705 7.3 7.5 NG2 Nausharo Gazella 0.70816 38.955 15.691 18.699 8.3 13.2 RS1 Rakhigarhi Sus 0.71574 617 39.054 15.781 18.971 1.2 8.8 RS2 Rakhigarhi Sus 0.71585 767 39.188 15.799 19.053 2.5 9.7 RS3 Rakhigarhi Sus 0.71568 894 38.996 15.781 18.920 3.3 9.4 RS4 Rakhigarhi Sus 0.71471 872 38.815 15.715 18.660 6.1 6.8 RS5 Rakhigarhi Sus 0.71903 483 39.070 15.763 18.970 1.1 5.9 RS6 Rakhigarhi Sus 0.71702 510 39.020 15.759 18.887 0.7 9.6 RS7 Rakhigarhi Sus 0.71582 514 39.078 15.765 19.008 2.0 6.1 RS8 Rakhigarhi Sus 0.71556 584 38.371 15.655 18.584 0.9 7.8 1 light stable isotope data from [ Kenoyer J. M., T. D. Price, and J. H. Burton 2013. A New Approach to Tracking Connections between the Indus Valley and Mesopotamia: Initial Results of Strontium Isotope Analyses from Harappa and Ur. Journal of Archaeological Science 40(5):2286 2297 (Page 2294, Table 2)]


99 Table 4 1 Continued Sample Site Leachate/ Taxon 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C SXD3a Sanauli acetic 0.71762 39.412 15.812 19.333 SXD3h Sanauli HCl 0.71842 39.461 15.820 19.369 SXD4a Sanauli acetic 0.71823 39.444 15.812 19.363 SXD4h Sanauli HCl 0.71967 39.491 15.817 19.381 SCSa Sanauli acetic 0.71752 39.486 15.820 19.392 SCSh Sanauli HCl 0.71900 39.511 15.819 19.390 1 light stable isotope data from [ Kenoyer J. M., T. D. Price, and J. H. Burton 2013. A New Approach to Tracking Connections between the Indus Valley and Mesopotamia: Initial Results of Strontium Isotope Analyses from Harappa and Ur. Journal of Archaeological Science 40(5):2286 2297 (Page 2294, Table 2)]

PAGE 100

100 Table 4 2 Baseline sample s summary statistics Site 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C Allahdino 0.709580.00232 38.8570.344 15.6870.030 18.6530.200 2.92.4 9.09.2 Balakot 0.708870.00019 39.0940.197 15.7170.021 18.7600.115 Farmana 0.715650.00030 39.4230.048 15.8190.007 19.3280.040 Harappa 0.718650.00270 732492 38.9700.318 15.7350.017 18.7440.212 3.12.3 8.66.6 Mehrgarh 0.708030.00018 38.9980.241 15.7020.018 18.7310.148 3.0 9.9 Nausharo ( Equus ) 0.708230.00006 38.9600.314 15.6950.022 18.7070.137 2.53.9 1.00.8 Nausharo ( Gazella ) 1 0.708110.00007 38.9210.094 15.6900.001 18.7020.009 7.81.5 10.48.2 Rakhigarhi 0.716170.00262 655333 38.9490.512 15.7520.092 18.8820.338 1.84.5 8.03.2 Sanauli 0.718410.00165 39.4670.072 15.8160.008 19.3710.044 1

PAGE 101

101 Figure 4 1. Himalayan geology. [adapted from Critelli, S. and Garzanti, E. 1994. Provenance of the Lower Tertiary Murree Redbeds (Hazara Kashmir Syntaxis, Pakistan) and Initial Rising of the Himalayas. Sedimentary Geology 89(3 4):265 284 (Page 266 Figure 1 )]

PAGE 102

102 Figure 4 2 Riverine 87 Sr/ 86 Sr ranges [ data from Krishnaswami S., J. R. Trivedi, M. M. Sarin, R. Ramesh, and K. K. Sharma 1992. Strontium Isotopes and Rubidium in the Ganga Brahmaputra River System: Weathering in the Himalaya, Fluxes to the Bay of Bengal and Contributions to the Evolution of the Oceanic 87Sr/86Sr. Earth and Planetary Science Letters 109(1 2):243 253 (Page 246 Table s 1 & 2 ); Pande, K., M. M. Sarin, J. R. Trivedi, S. Krishnaswami, and K. K. Sharma 1994. The Indus River System (India Pakistan): Major Ion Chemistry, Uranium and Strontium Isotopes. Chem ical Geology 116(3 4):245 259 (Page 255 Table 4 ); Karim, A. and J. Veizer 2000. Weathering Processes in the Indus River Basin: Implications from Riverine Carbon, Sulfur, Oxygen, and Strontium Isotopes. Chemical Geology 170(1 4):153 177 (Page 158 160 Tabl e 1 ); Dalai, T. K., S. Krishnaswami, and A. Kumar 2003. Sr and 87Sr/86Sr in the Yamuna River System in the Himalaya: Sources,

PAGE 103

103 Fluxes, and Controls on Sr Isotope Composition. Geochimica et Cosmochimica Acta 67(16):2931 2948 (Page 2936 2937 Table 1 ); Chakra pani, G. J. 2005. Major and Trace Element Geochemistry in Upper Ganga River in the Himalayas, India. Environmental Geology 48(2):189 201 (Page 197 Table 4 )]

PAGE 104

104 Figure 4 3 Summer riverine 18 O ranges [data from Ramesh, R. and M. M. Sarin 1992. Stable I sotope Study of the Ganga (Ganges) River System. Journal of Hydrology 139(1 4):49 62 (Page 54 55 Table 1 ); Dalai, T. K., S. K. Bhattacharya, and S. Krishnaswami 2002. Stable Isotopes in the Source Waters of the Yamuna and its Tributaries: Seasonal and Alt itudinal Variations and Relation to Major Cations. Hydrological Processes 16(17):3345 3364 (Page 3350 3351 Table 1 ); Karim A. and J. Veizer 2002 Water Balance of the Indus River Basin and Moisture Source in the Karakoram and Western Himalayas: Implicati ons from Hydrogen and Oxygen Isotopes in River Water. Journal of Geophysical Research: Atmospheres 107(D18):4362 (Table 1 )]

PAGE 105

105 Figure 4 4 Baseline samples in Sr Sr space reciprocal of Sr concentration Figure 4 5 Baseline samples in Pb Pb space 208 Pb/ 204 Pb

PAGE 106

106 Figure 4 6 Baseline samples in Pb Pb space 207 Pb/ 204 Pb Figure 4 7 Baseline samples 18 O

PAGE 107

107 Figure 4 8 Baseline samples 13 C Figure 4 9 Baseline samples in Pb Sr spac e 87 Sr/ 86 Sr

PAGE 108

108 CHAPTER 5 FARMANA The site of Farmana is located in the modern Indian state of Haryana among the easternmost regions of Indus settlements. The site sits ~ 30 km off the Chautang River, a now dry tributa ry of the seasonal Ghaggar River. The site occupied a strategic position between the urban center that is Rakhigarhi and the highlands of the Aravalli Range to the south. The entire region is part of the interfluve between the Indus and Ganges basins, and as such has distinct ecologies and geochemical makeup. The local sediments are large ly a product of alluvial deposition through drainage of the Lesser Himalayas to the north and aeolian input from the Thar Desert to the southwest. The Thar Desert is similarly alluvial in nature but derives primarily from the Indus River and therefore h elps differentiate the southern Ghaggar basin from the northern areas (Thussu 1995) Archaeologically, the entire region is noteworthy for a high concentration of Indus Civilization sites during the Integration Er a (2600 1900 BC), after which declining monsoonal precipitation may have spurred eastward and northward migration towards relatively more wet and stable ecological zones ( Giosan et al. 2012 ). Indus tradition deposits at Farmana date to ca. 2500 2200 BC, su ggesting residents lived during a period with ready access to water (Paleo Labo AMS Dating Group 2011) Their precise water management practices however, remain uncertain. There is scarce evidence wi thin the Indus Civilization for large scale irrigation canals and people likely relied on a range of practices including rain fed or inundated fields supplemented by small scale extension canals and well o r lift irrigation (Miller 2006b) At Farmana, surface collection of several burnt wedge shaped bricks typically used in well construction hint at the latter

PAGE 109

109 method, but streams and lakes may have contributed to the local wa ter supply as well (Shinde 2011b) T he ecological setting suggests an environment amenable to isotopic comparison s with other contemporary sites. For example, there is nothing to suggest that water sources at Farmana were inordinately e xp osed to evaporative processes, thereby 18 O relative to that of local precipitation. Further, the scale of variation in geological features is suitable to questions of interregional mobility. Whether immigrants came from the Aravallis, the Thar Desert, the Indus tributaries to the west, the Himalayan piedmont, or the Yamuna and Ganges to the east, there is every reason to expect that their non local origins will be reflected in geochemical data derived from tooth enamel ap atite Though much future work is needed to fully characterize isotopic variation within the region, the Farmana environment is well suited to archaeological investigation using isotopic methods. Further, a rchaeological evidence from three consecutive fie ld seasons of excavation at Farmana (2006 2009) (Shinde et al. 2008a; Shinde et al. 2008b; Shinde et al. 2011a) offers an important comparison against relatively comprehensive archaeological records at major urban centers. Archaeologists increasingly acknowledge that so called peripheral areas often serve as dynamic cultural frontiers and regions in which individuals of diverse backgrounds are most like ly to interact. In a frontier, the social milieu of evolving relationships conditions the processes of acculturation and ethnogenesis relevant to the cultural politics of modern nation states (Lightfoot et al. 1998; Stein 2002; Naum 2010) Farmana sits near the eastern limit of the Indus Civilization site distribution and as such, is a vital source of data for

PAGE 110

110 understanding how these processes operated in the contex t of ancient urbanized polities. Given the sites proximity to mineral rich highlands and sites of other cultural traditions (e.g., Ganeshwar Jodhpura phase ) it can be hypothesized that Farmana served as a kind of outpost in which material (and therefore cultural) exchange regularly took place (Shinde 2011 b ). The likelihood that people regularly moved materials through Farmana invites a fuller investigation of how people themselves migrated into the site during its occupation. The strontium, lead, carbon, and oxygen isotope values and strontium concentrations presented in this chapter provide compelling evidence for consistently patterned migration among those individuals in the excavated mortuary population with preserved tooth enamel. A review of the archaeological context is needed, however, to better understand the significance of the isotopic data. This chapter summarizes the results of excavation in the ha bitation area and the cemetery followed by a presentation of the biogeochemical data derived from human tooth enamel. Habitation Area The habitation area at Farmana is made up of a single mound spread over 18 ha a relatively large area given that the vas t majority of Integration Era sites are less than 5 ha in size ( Possehl 2002b ; Shinde 2011 b ). It is clear from size alone that Farmana must have been viewed as a relatively significant place on the Indus landscape. Unfortunately, active use of the surround ing agricultural lands has destroyed more recent archaeological deposits postdating 2300 BC and may have destroyed additional material in the peripheral portions of the site that could provide a more precise estimate of settlement area. Though modern land use has complicated work on more recent archaeological material at Farmana, excavators have described twelve stratigraphic

PAGE 111

111 layers of anthropogenic deposits that show the culture history of typical Indus traits at the site. For example, the earliest strata document the use of a round pit feature that likely had a wooden superstructure along with three pots of what Shinde (Shinde 2011c) label ed the Hakra phase. Similar pit structures have been documented at the nearby sites of Bhiranna (Rao et al. 2004) Girawad (Shinde et al. 2011b) and Kunal (Khatri and Acharya 1995) in contexts t hat suggest they were used for domestic purposes. Shinde (2011c) regard ed these founding deposits as Early Harappan, in part because they immediately precede strata containing typical Harappan, or Integration Era, materials and architecture such as a grid layout of buildings composed of mud bricks formed with the characteristic proportions of 1:2:4. Further, Hakra ceramics are known to occur in pre urban contexts, although the use of the term Hakra requires additional e history is internally consistent but terminologically complicated by the lack of coordination across the India (1992) b ) chronologies use Hakra as exemplified in the work of Mughal (1997) to indicate the pottery phase preceding Kot Diji ceramics of the Early Harappan period in Cholistan. Across the border in the contiguous regions of the Ghaggar basin that include Farm ana variation in Early Harappan ceramic styles has led to the designation Sothi Hakra phase as diagnostic of the Early Harappan. In a pointed effort to disentangle the Indus culture history nome nclature, Kenoyer (2011b) suggested that Hakra legitimately refers to the Early Harappan period but that the term has been applied so broadly as to obscure the interregional differences and interacti ons so critical to understanding the

PAGE 112

112 incipient development of integrated urbanism during the Regionalization Era. Sharing a concern for regional differences, Uesugi (2011b) note d in his alternative chronol ogy of the Farmana ceramic assemblage that there are close similarities between regional styles of pottery and the Sothi Siswal phase Though Shinde and Uesugi use different methods and reach different conclusions about Farmana chronology, both observe the persistence of regional ceramic styles alongside the Harappan phase ceramics pottery commonly associated with urbanization and the Integration Era. Uesugi specifically suggested that no particular trend in the frequencies of Harappan versus non Harappan t ypes could be observed either vertically or in different excavation areas. However, potential mixing of the strata and incomplete stratigraphic sequences in certain excavation areas as a consequence of modern agricultural activity at Farmana complicates an y relative chronology. As a result, s tatements about the cultural sequence need to be used cautiously and tested through further excavation at undisturbed areas of the site ( Kenoyer personal communication) Therefore it is unclear tation of the horizontal and stratigraphic distribution of pottery types reflects the pre taphonomic reality. Despite uncertainties within the Farmana culture history the different pottery styles identified at the site broadly corroborate the radiocarbon dates in suggesting that the surviving cultural deposits formed over a few hundred years. Harappan and regional styles are found in significant quantities, making it clear that Farmana residents had external interactions at the regional and Greater Indus s cales. Harappan pottery is defined in part by its formal similarities to types described at Mohenjo Daro along the lower Indus Valley (Marshall 1931; Mackay 1938; Dales and Kenoyer 1986a) and

PAGE 113

113 Harappa along the upper Indus Valley (Wheeler 1947; Jenkins 1994) Further, Harappan pottery is thrown entirely using fast wheel rotation a nd its painted motifs, when present, are distinctive. Non Harappan pottery consists of styles with a more limited regional distribution and a smaller range of vessel forms. The painted motifs are distinct from those of Harappan ceramics and various technic al features including an appliqu ring base and discontinuous striations suggest greater use of hand molding methods (Uesugi 2011 b) participation in interregional networks (Konasukawa et al. 2011) Beads and bangles are made of a range of locally unavailable materials including marine shell, fired steatite, carnelian, copper, and gold. Administrative technologies such as diagnos tic chert broader Indus urbanism. Even the ubiquitous triangular terracotta cake implies the use of classic Indus pyrotechnology, and the architectural context is similarl y indicative of Integration Era habitation. The rectangular, multi room, mud brick complexes contain central courtyards and orient along a grid plan of streets all features of the Integration Era. Interestingly, differences in the quality of architecture m ay suggest differential access to labor or material by different groups. Certainly the wide variety of local and non local materials used to craft common forms of jewelry suggests a hierarchy of value consistent with social inequality. Even the diversity of foodstuffs discovered in the archaeobotanical analyses (Sugiyama 2011; Weber et al. 2011) suggests the possibility that dietary choices may have been guided in part by group affiliation. The overall assemblage of plants revolved

PAGE 114

114 around cereals and pulses and incorporated both summer and winter crops. Rice is absent, but there is combined use of Southwest Asian cereals like barley and wheat with indigenous millet s. By dividing the botanical sample into two groups representing upper strata and lower strata, Weber and colleagues inferred a decrease in wheat and barley through time that could have resulted from decreasing winter rains. They also recovered direct evid ence for consumption of millets, barley, and gram in dental calculus from a small sample of human teeth. To summarize, various lines of evidence confirm the fact that Farmana was part of the larger Indus Tradition from the outset, contributing to regional identity in the early phases and eventually integrated into the Harappan Phase by around 2600 BC Though chronology remains imprecis e, very broad trends appear in comparison s of the earliest and latest strata. The first residents likely lived in pit dwell ings and used Early Harappan material culture, but those Early Harappan strata are immediately succeeded by typically urban Integration Era architecture suggesting cultural transformation rather than abandonment and resettlement. Additionally, documentatio n of Early Harappan assemblages across the Greater Indus region shows that the Early Harappan period is the evolutionary forebear of Harappan period urbanism (e.g., Kenoyer and Meadow 2000; Bisht 1997; Nath 1998 ; Nath 2001). Change at Farmana is not limited to ceramics and architecture however. Archaeobotanical evidence suggests decreased reliance over time on cereals of Southwest Asian origin such as whea t and barley. This may have resulted from changing patterns of precipitation, such that climatic and dietary changes should register in carbon and oxygen isotope values of human dental enamel.

PAGE 115

115 T he diversity of raw materials and foodstuffs at Farmana also suggests possible behavioral elements of social differentiation. This aggregate view of variability at Farmana may actually encompass different lifeways of different social groups, raising the question of how representative the mortuary sample is of the population at large. A clearer understanding of mortuary variability is needed to determine precisely how the bioarchaeological data re late to trends inferred from the habitation area. Cemetery Area The Farmana cemetery is located ~ 900 m northwest of the habitation area, and like the settlement itself, is disturbed by agr icultural activity (Shinde 2011a ). The excavated area covers less than 1 ha although initial surveys suggest the entire cemetery may cover ~3 h a It was excavated over two seasons (Shinde et al. 2008b, 2011) after having been accidentally discovered in the 2007 2008 season. The excavators discovered 70 burials, 58 of whi ch were fully excavated by the end of the 2008 2009 field season. The graves were readily identified by their darker backfill, and it is clear that all of the burials were fully surveyed in the excavation area measuring 35x21 m 2 As at the habitation area, the relative chronology for the cemetery re mains problematic. Shinde (2011a ) placed the burials into three periods, the Early (2A), Middle (2B), and Late (2C) phases of the Mature Harappan Period (2600 1900 BC), wh ereas Uesugi (2011a) divided the burials across four sequential phases b etween ca. 2500 2200 BC (Table 5 1). Further, there is substantial disagreement between the two approaches to seriation of pottery resulting in incompatible chronologies. Pro blematically, the burials at Farmana cannot be linked stratigraphically or across space because the overlying strata that would show where the burial pits are located

PAGE 116

116 were removed by later agricultural activity. This makes it impossible to determine their precise chronological relationships. The widely different results from the two attempts at periodization suggest that pottery is not necessarily the best way to seriate the graves except at a very broad level of earlier and later (Kenoyer personal communic ation). For this reason, the isotopic mortuary analysis of Farmana will treat the mortuary sample as a single chronological population. Unfortunately, similar logic precludes the reliable use of biological sex in the mortuary analysis. Preservation of the Farmana skeletal remains was extremely poor, resulting in highly friable material that retains few reliable osteological indicators of sex. Osteological assessment of the material using the methods of Buikstra and Ubelaker (1994) was followed with only a handful of cranial traits scored for any given individual. Many diagnostic regions were represented by sections of bone that were fractured, deformed, heavily eroded or otherwise damaged through generally poor preservation. A more thorough analysis was performed by the Deccan College Post Graduate and Research Institute Biological Anthro pology Lab that resulted in different sex assignments (Table 5 1) (Mushrif Tripathy personal communication). Given the poor sample preservation and high inter observer error, the most conservative approach is to ignore biological sex as a category in the F armana mortuary analysis. T he relatively better preservation of tooth enamel however, aided in age estimations. Distinctions between adults and sub adults were made for all sampled individuals, and conservative sub adult biological age ranges can be infer red based on enamel formation times and tooth eruption sequences (Table 5 1) (Hillson 1996; Reid and Dean 2006) Though the sample size is modest, the lack of individuals younger than two y ears old suggests that

PAGE 117

117 infants were not included in the program of inhumation. Because not all skeletal elements exposed by the excavators were preserved for curation, photographic documentation of all 58 burials provides an additional, albeit less precise line of evidence suggesting that infants were not buried in the Farmana cemetery (Shinde et al. 2008b ; Shinde 2011a ). I nhumations are surprisingly scarce throughout the Indus Civilization and formal cemeteries even more so. As is the case for most Integr ation Era cemetery based inhumations, the Farmana graves are rectangular pits oriented around a north south axis (Shinde 2011a) The basic burial program remained the same throughout the life of the cemetery and included many of the elements common to m ost Integration Era cemeteries. Skeletal remains are deposited in primary and secondary contexts with the cranium towards the north end of the grave abutting a variable number of ceramic vessels. Primary inhumations lie extended and supine and sometimes are adorned with a small number of personal ornaments or other seemingly minor artifacts. Not all burials contain skeletal remains, but most such symbolic burials or cenotaphs contain ceramics arranged in similar ways. Lastly, a few graves have an additional c lay lining. The intermittent use of containment features such as wooden coffins and clay or brick linings has been noted at other Integration Era cemeteries like that of Harappa (Dales and Kenoyer 1989a) In short, the Farmana burials seem to conform to a broader tradition of Integration Era mortuary practice that was consistently reserved for a minority of the population. Mortuary variation for each burial is disc ussed in detail by Shinde (2011a ), and there fore only a summary of the differences in mortuary treatment at Farmana is

PAGE 118

118 presented here. The quantity of ceramics is broadly comparable to that of other sites with an average of 7 pots per burial and a maximum of 27. Five burials had more than 21 pots an d twelve burials had no pots although it is not always clear whether or not pots were removed or destroyed through post depositional activity. Further, the majority of burials have mixed ceramic assemblages consisting of Harappan and regional non Harappan types. Only six burials have exclusively non Harappan types, and ten burials have exclusively Harappan types. Within the isotopic mortuary sample, these include burials 52, 66, and 67 for non Harappan vessels and burials 18, 20, and 23 for Harappan vessel s (Table 5 2). Harappan types are distin guished by the exclusive use of fast wheel rotation distinct forms, more finely tempered and levigated clay and flat bottoms. By contrast, non Harappan ceramics are generally more variable in terms of production met hods including a greater variety of surface treatments and burnishing techniques. A handful of burials contain painted non Harappan vessels including burials 39 and 65 in the isotopic mortuary sample. There is no obvious structure in terms of pottery with in the mortuary sample as burials containing predominantly Harappan or non Harappan types both exhibit variation in skeletal age, orientation, presence or absence of clay linings, and personal ornamentation. T he overall impression is that mortuary variatio n was randomly distributed within a culturally prescribed spectrum. For example, some burial pits are lined in clay including burials 26, 52, 53, 54, 58, 62, 64, and 67 (Table 5 1 ) The orientation of the pit varies slightly around a NNW SSE axis with a few oriented along a NE SW axis including burials 20 and 65. Some burials contain additional items like jewelry made of shell, copper, or semi precious

PAGE 119

119 stones, but no single burial is particularly ostentatious ( Table 5 2 ). Within the osteological sample, beads of semi precious stones or steatite are associated with burials 1, 3, 14, 20, 26, 41, 47, 50A, and 65. Some individuals, including both adults and subadults, are adorned with one or more copper bangles (burials 5, 20, 41, 50B and 52), and four indivi duals (adult and subadult) are wearing shell bangles (burials 3, 20 50B, and 65). Burial 20, one of the more heavily ornamented individuals, also has copper bangles, a copper earring, and the aforementioned microsteatite beads. Lastly, burial s 14 32, and 34 are associated with bone and shell tools There may be an association between the quantity of items and grave orientation. Three total burials are oriented along a NE SW axis, and one of these, burial 20, is well adorned in jewelry. Burial 65 has no je welry but does have 22 pots, a fairly large number for the Farmana assemblage. Burial 22 which is not represented in the osteological sample is also oriented NE SW but has only ten pots and a bivalve shell. These three individuals are all above average i n terms of the quantity of grave goods, and atypical grave orientation may have been exploited to accentuate their differences from other individuals. However, with only three data points and the presence of more elaborate burials among the conventionally oriented graves, this trend remains inconclusive. In general then, there is relatively little systematic differentiation among the Farmana burials. Certainly there is variability, but it does not appear to be consistently patterned across multiple media in a way that indicates groups of individuals are qualitatively distinct from each other. The overall impression from the mortuary program is that individuals are inhumed largely on the basis of their similarities with less concern for their differences.

PAGE 120

120 Fi nally, the distribution of burials is somewhat haphazard with five instances of crosscutting and overlapping grav e shafts. This tendency has been observed at Harappa, leading Kenoyer (1998) to suggest that those who dug the burial pits belonged to a separa te ethnic group or otherwise did not identify with the deceased. The modern grave backfill is starkly visible against the surrounding soil which supports the idea that there was a general disregard for older burials, although the visual contrast may have been less apparent more than four millennia ago. Nevertheless, given the similarities with Harappa, the spatial disorganization and crosscutting suggests that inhumation was a relatively short term affair with little emphasis on enduring commemoration. The situation at Farmana is consistent with the possibility that the inhumed constituted a different kind of person who inspired little deference among those responsible for carrying out the mortuary program. In summary, the mortuary program at Farmana indica tes a degree of social distance between the dead and the living and relatively less distance between those who were inhumed. Whatever their individual differences, the inhumed were treated as a socially distinct group that required a specific and extremely uncommon form of mortuary treatment. Further, the mortuary variability fails to show a consistent and repeated pattern of differentiation within the mortuary population as a whole. Though some of the dead (or those who mourned them) might have held positi ons of relative prestige, the differences between individual inhumations appear subordinate to the overall impr ession of similarity. Taken together, the mortuary context is consistent with the possibility that the act of inhumation referenced a shared and in some ways socially distinct group identity that had been lived by the deceased.

PAGE 121

121 Results of the Analyses Isotopic data and Sr concentrations resulting from the analyses are presented in Table 5 3 Values of 87 Sr/ 86 Sr range from 0.7 1529 to 0.72 038 with Sr concentrations f rom 246 to 1388 ppm. Lead isotope ratios also exhibit heterogeneity with 208 Pb/ 204 Pb from 36.279 to 39. 431 207 Pb/ 204 Pb from 15. 572 to 15.823, and 206 Pb/ 204 Pb from 16.506 to 19.3 11 T 18 O of 5.0 to 2.1 1 3 C of 12.0 to 6.8 Means and st andard deviations are presented for each tooth cohort in Table 5 4 Only isotope ratios of Sr and Pb were obtained for the acid leachates of sediment samples as Sr concentrations and light stable isotope values from soils are irrelevant to the objectives of this dissertation. Recall from Chapter 3 that the only available faunal material at the time of sample collection was from the Rakhigarhi material curated at Deccan College Post Graduate and Research Institute. Therefor e the eight analyzed samples of pig enamel are more representative of the geochemistry in the dietary catchment of Rakhigarhi located ~ 30 km northwest of Farma na. Regardless, the faunal and leachate data are quite similar and together constitute the availa ble proxy for local isotope values at Farmana. During human tooth enamel sampling, some samples were taken from deciduous or incompletely mineralized teeth (burials 5, 15, and 52) to assess the likelihood of diagene s is relative to f ully mineralized enamel from permanent dentition. Because of visual or quantitative observations, the data from these samples w ere disregarded. Note that human enamel samples from burials 5 and 52 are from deciduous dentition. The enamel structure of deciduous teeth is known to be less regular with larger crystals than that of permanent dentition and presumably more susceptible to diagenesis. Because of the very high Sr concentration and the close overlap between sample 5 and the

PAGE 122

122 sediment leachates, this sample is regarded as con taminated and is disregarded in the subsequent presentation and discussion of data. Even though sample 52 does not have similar values to the sediment leachates, the possibility remains that an unquantifiable degree of diagenetic alteration has occurred an d so the data from this burial are also rejected. Sample 15 was an extremely thin slice of incipiently mineralized cusp tips, and it was observed to begin dissolution in the 0.1N acetic acid solution used for pretreatment. Therefore it is assumed that the natural pH of the burial environment could also have contributed to alteration of isotopic values. Consequently, data from sample 15 are also rejected. The data from sample 39 must also be disregarded because of indeterminate molar position. At the time o f sampling, it was anticipated that if there were any meaningful differences in isotope values across teeth for the same individual then it was most likely to occur in later childhood, just prior to full mineralization of the third molar. However, given th e interpretive significance of differences in first and second molar isotope values, the data from sample 39 must be excluded from this analysis. They may still be relevant to other research questions, but they cannot resolve the patterns of early childhoo d migration that turned out to be so significant to this isotopic mortuary analysis. Lastly, light stable isotope values for samples 18 3 and 67 2 could not be obtained because of small sample size. After the removal of dentine in preparation for pretreat ment and analysis, the remaining amount of enamel was smaller than initially apparent on visual inspection. Given the priority of heavy isotope data over light isotope data for studies of mobility and the larger sample size requirements for heavy isotope

PAGE 123

123 a nalysis, it was decided to use the entire sample for Sr and Pb isotope analysis to maximize the likelihood of acquiring useful data. Strontium isotope data are first presented in comparison with 1/Sr concentration following the method of Montgomery and cow orkers (2007) (Figure 5 1). Even where Sr isotope ratios for a given region are very similar because they are heavily influenced by a single environmental source, concentration data can be used to infer loca tion along an environmental Sr mixing line between two sources. The data in F igure 5 1 likely indicate a relatively radiogenic endmember with low Sr concentration and a high concentration endmember with low 87 Sr/ 86 Sr. The dataset appears unstructured with regard to tooth type as first, second, and third molar cohorts have similar values. Unlike the Sr isotope data, the Pb data show much more pronounced systematic variation by tooth type (Figures 5 2 & 5 3). There may also be a simple mixing system at work b etween a less radiogenic source and a more radiogenic source. T he striking pattern however, is the clustering of first molar values away from second and third molar values. There is also fairly good correlation between third molar values and the approxima te intersection of Farmana sediment leachates and Rakhigarhi faunal enamel. Oxygen isotope values show a clear trend based on tooth type (Figure 5 4). First molar mean 18 O ( 1.2 [ ] ) is higher than that of second ( 3 5 1.4 [ ] ) ( t (18) = 2.618, p = 0 .017) or third molars ( 1.5 [ ] ) ( t (17) = 3.019, p = 0 .008), whereas no significant difference exists between second and third molar cohorts. A change in water source for the individuals in the study samples occurred largely betw een first and second molar mineralization times. Because tooth enamel formation

PAGE 124

124 times are standardized across populations with a species, this means the change in fluid uptake occurred around the chronological age of three. Carbon isotope values differ fr om those of Pb and O in that they do not change significantly with age (Figure 5 5). There is a fair amount of variation in the relative contribution of C 3 and C 4 foods at the population level ( 13 C variation is not systematically c orrelated with tooth type. The means for first, second, and third molars are 10.1 2.3 ( 10.1 1.7 ( and 10.1 2.6 ( respectively. In other words, there is variation in the diets of individuals but the relative contribution of C 3 and C 4 foods is not a factor of life history. Lastly, plotting isotope ratios of Pb against those of Sr increases the discriminatory power of the two isotope systems and shows the differences within and between tooth types more clearly (Figure 5 6). Figure 5 6 shows a clustering of third molar values near the intersection of Farmana sediment leachates and Rakhigarhi faunal enamel. Second molar data substantially overlap with third mola r data but have a slightly greater range therefore indicating a small contrib ution of environmental Pb and Sr from less radiogenic sources early in the mine ralization process. First molar data plot apart from other tooth types. The isotope data clearly detail a change in environmental sources of Pb and O through time. In general, t here is a convergence through time towards the sediment leachate and faunal isotope values suggesting that most individuals shared a common geochemical environment (i.e., Farmana) by late childhood. Most of the change seems to occur within an environment o f fairly homogenous Sr sources although some individuals are exposed to fairly radiogenic strontium sources throughout childhood.

PAGE 125

125 There is very little change in 13 C by tooth type suggesting that throughout childhood, individuals consumed similar types of food with respect to C 3 and C 4 vegetation. The isotope data will be interpreted with regard to human behavior in Chapter 8 at which point all isotopic and mortuary data will be synthesized.

PAGE 126

126 Table 5 1. Chronology, demographic data, and mortuary treatment at Farmana Burial No. Period 1 Phase 2 Sex 3 Sex 4 Age Burial Type Harappan Non Harappan Orientation Clay Lining Pot Qty. Pot Qty. 1 4 M adult primary 23 N S absent 2 F F adult primary NW SE absent 3 4 subadult secondary 3 N S absent 5 2 6 secondary NW SE absent 6 2 F F adult secondary 15 1 NW SE absent 8 2C 2 M adult primary 2 4 NW SE absent 10 2C 2 F adult primary 3 6 NW SE absent 11 2C 3 M adult primary 3 1 NW SE absent 14 2B F M adult primary NW SE absent 15 2C 2 4 secondary 1 1 NW SE absent 18 2C 4 F F adult primary 7 N S absent 20 2C 1 F F adult primary 2 NE SW absent 23 2B 3 12 18 primary 9 N S absent 25 2C 4 M adult secondary 8 8 N S absent 26 2C 4 F F adult primary 1 1 NW SE present 28 2A 1 subadult primary 8 N S absent 32 2A 3 F adult primary 21 5 NW SE absent 34 2C 4 M adult secondary 1 8 N S absent 39A 2B 2 F adult secondary 10 6 N S absent 39B 2B 2 M adult secondary 10 6 N S absent 41 2C 4 5 9 primary 1 7 N S absent 45 2B 12 18 primary NW SE absent 47 2C 5 9 secondary NW SE absent 48 2B 3 M adult primary 1 9 NW SE absent 50A 2B 2 M adult primary 6 15 N S absent 1 Shinde 2011a 2 Uesugi 2011a 3 4 Mushrif Tripathy personal communication

PAGE 127

127 Table 5 1. Continued Burial No. Period 1 Phase 2 Sex 3 Sex 4 Age Burial Type Harappan Non Harappan Orientation Clay Lining Pot Qty. Pot Qty. 50B 2B 2 F adult primary 4 1 NW SE absent 52 2C 2 6 secondary 6 NW SE present 53 2C 4 F adult primary 3 7 NW SE present 54 2B 2 M M adult primary 3 3 NW SE present 58 2B adult secondary N S present 62 2C 3 9 14 primary 2 6 N S present 64 2C 1 M adult primary 12 1 NW SE present 65 2C 2 adult primary 2 20 NE SW absent 66 2C M M adult secondary 1 N S absent 67 2C F M adult primary 1 NW SE present 1 Shinde 2011a 2 Uesugi 2011a 3 4 Mushrif Tripathy personal communication

PAGE 128

128 Table 5 2. Mortuary elaboration at Farmana Burial Pottery Copper Shell Semi Precious Steatite Other No. Qty. Bangles Bangles Beads Ornaments Items 1 23 3 carnelian, 2 copper 30 disc beads 2 3 1 1 agate 5 2 6 16 8 6 10 11 4 14 1 agate 2 bone tools 15 2 18 7 20 2 5 3 microbeads 1 copper earring 23 9 25 26 2 1 agate 28 32 1 shell spoon 34 3 bone tools 39A 16 39B 16 41 8 1 1 jasper 45 47 1 carnelian 48 50A microbeads

PAGE 129

129 Table 5 2. Continued Burial Pottery Copper Shell Semi Precious Steatite Other No. Qty. Bangles Bangles Beads Ornaments Items 50B 1 3 52 6 2 53 54 6 58 62 8 64 65 22 1 3 carnelian 66 1 67 1

PAGE 130

130 Table 5 3 Farmana mortuary sample results of the analyses Sample No. Burial Tooth 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C F2 1 2 LM 1 0.71529 1167 38.623 15.734 18.553 3.9 11.2 F2 2 2 LM 2 0.71529 1260 39.351 15.805 19.197 4.5 11.2 F2 3 2 LM 3 0.71541 1163 39.401 15.809 19.222 5.0 10.7 F5 d1 1 5 Rdm 1 0.71557 1206 39.431 15.818 19.311 4.1 10.0 F6 1 6 LM 1 0.71582 704 38.871 15.727 18.725 3.3 8.3 F6 2 6 LM 2 0.71584 778 39.200 15.775 19.118 3.5 9.5 F6 3 6 LM 3 0.71581 881 39.286 15.796 19.171 3.7 11.2 F11 2 11 LM 2 0.71581 881 39.351 15.803 19.233 3.6 10.9 F11 3 11 LM 3 0.71583 857 39.413 15.814 19.293 3.4 9.1 F14 1 14 RM 1 0.71551 1030 38.842 15.723 18.750 2.4 9.6 F14 2 14 LM 2 0.71549 971 39.315 15.786 19.218 2.6 8.8 F14 3 14 RM 3 0.71560 1072 39.211 15.783 19.195 2.8 10.6 F15 2 1 15 RM 2 0.71574 1336 39.296 15.796 19.195 2.3 9.3 F18 1 18 LM 1 0.71589 758 38.133 15.679 18.166 2.1 9.4 F18 2 18 LM 2 0.71587 739 38.929 15.753 18.908 2.6 9.9 F18 3 18 RM 3 0.71578 702 39.044 15.766 19.028 3 3 F20 1 20 RM 1 0.71572 626 37.950 15.673 18.037 2.3 10.6 F20 3 20 RM 3 0.71571 630 39.041 15.775 19.020 4.0 10.4 F23 1 23 RM 1 0.71592 788 38.546 15.700 18.502 2.6 10.4 F23 2 23 RM 2 0.71584 984 38.871 15.741 18.842 2.9 9.9 F23 3 23 RM 3 0.71579 816 39.150 15.775 19.090 3.6 10.7 1 enamel is deciduous or poorly mineralized 2 indeterminate molar position 3 insufficient sample for analysis

PAGE 131

131 Table 5 3 Continued Sample No. Burial Tooth 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C F26 2 26 RM 2 0.71589 542 39.387 15.801 19.896 4.1 10.3 F26 3 26 RM 3 0.71600 595 39.184 15.777 19.172 3.6 7.3 F39B 1/2 2 39B RM 1/2 0.71586 1061 39.249 15.785 19.192 3.1 6.8 F41 1 41 RM 1 0.71576 983 36.279 15.572 16.506 3.1 10.4 F45 p3 45 RP 3 0.71756 504 39.190 15.782 19.086 4.5 11.4 F47 p4 47 RP 4 0.71588 1388 39.148 15.773 19.075 4.0 10.4 F52 d2 1 52 Rdm 2 0.71577 1016 38.252 15.705 18.315 3.2 9.8 F54 1 54 RM 1 0.71620 828 39.161 15.765 19.100 4.9 10.6 F58 2 58 RM 2 0.71570 933 38.975 15.744 18.903 3.1 9.1 F62 1 62 LM 1 0.71572 991 38.027 15.687 18.137 2.7 9.6 F62 3 62 LM 3 0.71590 820 39.280 15.797 19.204 2.8 9.3 F65 1 65 RM 1 0.71578 855 38.899 15.727 18.776 2.3 12.0 F65 2 65 RM 2 0.71588 814 39.086 15.757 19.010 3.1 9.6 F66 3 66 RM 3 0.71704 605 39.200 15.766 19.160 4.7 10.2 F67 2 67 LM 2 0.71975 342 39.228 15.798 19.299 3 3 F67 3 67 LM 3 0.72038 246 38.814 15.738 19.013 3.0 11.9 1 enamel is deciduous or poorly mineralized 2 indeterminate molar position 3 insufficient sample for analysis

PAGE 132

132 Table 5 4 Farmana mortuary sample summary statistics -Cohort 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/204Pb 206 Pb/ 204 Pb 18 O 13 C All 0.716070.00210 861515 38.9351.189 15.7570.100 18.9081.104 3.31.6 10.02.3 1st molar 0.715710.00042 865366 38.4860.789 15.7060.049 18.4560.602 2.71.2 10.12.3 2nd molar 0.716230.00247 845600 39.1700.347 15.7760.047 19.1490.551 3.51.4 10.11.7 3rd molar 0.716300.00283 762500 39.1840.346 15.7810.044 19.1420.184 3.71.5 10.12.6

PAGE 133

133 Figure 5 1 Farmana samples in Sr Sr space reciprocal of Sr concentration Figure 5 2 Farmana samples in Pb Pb space 208 Pb/ 204 Pb

PAGE 134

134 Figure 5 3 Farmana samples in Pb Pb space 207 Pb/ 204 Pb Figure 5 4 Farmana samples 18 O

PAGE 135

135 Figure 5 5 Farmana samples 13 C Figure 5 6 Farmana samples in Pb Sr space 87 Sr/ 86 Sr

PAGE 136

136 CHAPTER 6 HARAPPA Harappa was a major urban center situated ~ 8 km from the Ravi River in the Punjab of present day Pakistan. It is uniquely well studied among Indus sites, and a wealth of data resulting from decades of excavation has strongly influenced how the entire Indus Civilization is viewed. R ediscovered by antiquarian Charles Masson in 1829 (Masson 1844) the first systematic excavations di d not take place until the 1920s and 19 30 s decades after large swaths of the uppermost strata had been looted for bricks used in railroad construction (Possehl 2002b). Nevertheless, excavations under the Archaeological Survey of India (Sahni 1921; Sahni 1926; Sahni 1927; Vats 1933; Vats 1935; Vats 1936; Wheeler 1947) and Department of Archaeology and Museums, Government of Pakistan (Mughal 1968) revealed much of the habitation mounds, fortifications, and cemetery areas. More recent collaborative ventures between the Pakistani agency and United States archaeologists (Dales and Kenoyer 1986b; Dales and Kenoyer 1987; Dales and Kenoyer 1988; Dales and Kenoyer 1989b ; Dales and Kenoyer 1990b; Meadow and Kenoyer 1992; Meadow and Kenoyer 1993; Meadow et al. 1994; Meadow et al. 1995; Meadow et al. 1996; Meadow et al. 1997; Meadow et al. 1998; Meadow et al. 1999; Meadow et al. 2001) have provided the most comprehensive dataset from any Indus site using modern methods of excavation and chronometric dating. Many of the interpretations of Indus social organization presented in Chapter 2 are derived in large part from the Harappa literature and will not be revisited here. I nstead, the first part of this chapter emphasizes site specific culture history and mortuary variability as they pertain to individual life histories and broader social contexts

PAGE 137

137 Habitation Area The protohistory of Harappa is divided into five periods repr esented by six distinct archaeological phases (Table 6 1). The earliest levels pre date many diagnostic traits of Indus urbanism (e.g., wheel thrown potte ry and baked brick architecture), but the limited excavations of these strata suggest important contin uities between the small Ravi Phase village and the large city that it was to beco me. For example, symbols etched in sherds may represent incipient Indus script (Kenoyer and Mea dow 1996; Meadow and Kenoyer 2008) Likewise, many of the long distance trade routes that later proved integral to the development and maintenance of Indus urbanism were already in place at this time, providing exotic raw materials for the local product ion of prestige goods (Kenoyer and Meadow 1999; Kenoyer and Meadow 2000; Kenoyer 2005a; Law 2008) Excavations of the relatively expanded Kot Diji levels ( ~ 25 ha) revealed a trend of increasing complexity (Kenoyer 1998). Harappans had access to greater quantities and varieties of imported goods which may have been regulated through the increased use of script, stamp seals, chert weights and perimeter walls (Meadow and Kenoyer 2001; Meadow and Kenoyer 2005; Meadow and Kenoyer 2008) The settlement b egan to take on much of its eventual urban character during this time, being clearly divided into separate walled mounds using mud bricks with standardized proportions. Importantly, Period 2 levels did not exist in a cultural vacuum. Many of the architectu ral developments from Harappa were mirrored at other Regionalization Era sites including Kalibangan (Joshi 2003) Dholavira (Bisht 1991) Rehman Dheri (Durrani 1988; Durrani et al. 1995) and Kunal (Khatri and Acharya 1995), wh ereas Kot Diji ceramics are found

PAGE 138

138 at sites spanning ~ 1000 km northeast southwest along the Indus River Valley (Law 2008) By 2600 BC, the Kot Dijian way of life had gradually transformed into a fully urbanized existence represented by three sub phases. Much of the diagnostic suite of Indus material culture is present throughout Periods 3A, 3B, and 3C, but interpretations of the Harappa Phase cannot be reduced to a single trajectory of growth or decay. Instead, urban development was variable and fluctuated through time (Meadow and Kenoyer 2005). During Period 3A, Kot Dijian revetments were replaced and elaborated on using bak ed bricks, while settlement remained largely confined to the existing boundaries of Mounds AB and E. Further, civic architecture deteriorated in some regions of Mound E, suggesting a measure of stagnation before the urban renewal of Period 3B (Kenoyer 1991b) Eventually, crumbling structures were repaired and new habitation mounds were founded. One of the new mounds, Mound ET, has been ecause the two areas shared their perimeter wall and a city gate (Meadow and Kenoyer 1997) Mound F, on the other hand, had its own wall and became the site for new large scale civic projects including the so (Vats 1940; Fentress 1984; Meadow and Kenoyer 2008) Period 3C strata are the best studied levels at Harappa and serve as a primary referenc e for the normative mode of Indus urbanism so much so that Kenoyer (personal communication) suggested that phase is the pointed base goblet, a mass produced ceramic that is ubiquitous within Period 3C

PAGE 139

139 deposits (Meadow and Kenoyer 2001). The presence of these and other diagnostic finds show that the Mature Harappan settlement was at its largest, extending beneath the present day city and beyond the southernmost gate (Meadow and Kenoyer 1993) to cover an area of ~ 150 ha (Dales and Kenoyer 1989a) Though growing in size, not all areas were equally prosperous. Mound E fell into disrepair once more, wh ereas evidence suggests the inhabitants of Mounds AB and F were relatively affluent (Meadow and Kenoyer 2005). Coincident with the post urban regional demographic shift discussed in Chapter 2, P eriod 4 at Harappa marks the beginning of a process of profound cultural transformation (Kenoyer 2005b) Strictly speaking, habitation at Harappa between ca. 1900 and 1700 BC retained certain urban qual ities, but many of the traits associated with the fully integrated and spatially expansive Indus Civilization phenomenon (e.g., Indus script, chert weights, stamp seals) disappeared. Subsistence practices also changed during this time as wheat production d eclined and summer crops became increasingly common (Weber 2003). There were even technological innovations in bead drilling (and possibly in glass making) p erhaps to compensate for the loss of various crafts as exchange networks deteriorated (Kenoyer 2005 b). By ca 1700 BC, however, the unique styles and motifs of the Cemetery H Phase replaced the Harappa Phase as the predominant archaeological culture. Through much of this transformation, Harappa remained a vibrant population center, and the settled area m ay have covered ~ 100 ha during Period 5 before its decline sometime prior to 1300 BC. To summarize, the cultural trajectory at Harappa is not readily described by a single generalized trend. Instead, different habitation areas seem to ebb and flow at

PAGE 140

140 diff erent times and only at the most general level can the Harappan site history be account for diverse groups with differing perspectives and possible conflicts. At the same time, cultural phases were widely shared and cultural transitions were gradual rather than abrupt, suggesting that diversity was accommodated within a common ideological system. Future high resolution analyses of intra site patterning (akin to the min eral provenience study by Law (2008) are needed to better understand the social dynamics in this complex system of interdependence. The dietary catchments at Harappa may have been similarly complex, and variability in the food supply or certain technologic al practices could potentially impact lead, strontium, oxygen, and carbon For example, the transportation of food from regions with distinct isotopic compositions could alter consumer isotopic values relative to local fauna and se diment leachates. Marine foods are one possible source as the remains of marine fish have been documented at Harappa (Belcher 1991; Belcher 2003) The evidence is extremely lim ited, however, suggesting that neither marine fish nor the salt used for their preservation were significant sources of dietary Sr. Likewise, there is little evidence for widespread anthropogenic Pb input in the form of smelting with the exception of a sin gle large slag (Kenoyer personal communication). Finished goods such as bronze vessels were a more plausible cause of technological Pb exposure. Makeup may also have included Pb like artifact of granulat ed material found in a burial of Cemetery R 37. Galena crystals were identified within the rod which may have been a solidified cosmetic similar to surma or kohl. Another small jar that

PAGE 141

141 contained black residue and was found in habitation contexts ha d level s of Pb detectable by ICP MS whereas seven other jars and sherds yielded no conclusive evidence for Pb. The possibility of Pb exposure via makeup remains an open question given the possibility of Pb contamination associated with modern surma use as discus sed in Chapter 3. I t may be however, that cosmetics were predominantly carbon based products as proposed by Mackay (1938) Whatever the case, any impact on Pb isotope ratios in humans with access to the bustl ing markets of Harappa may be characterized by high inter individual variation, as the Pb fraction in metal wares and other products at the site has widely varying provenience (Law 2008; Hoffman and Miller 2009 ). Water management practices at Harappa may have been similarly varied. Different water sources could have been exploited by different people at different times, 18 O. The glacially fed Ravi River may have flowed cl oser to the site during its protohistoric habitation and theoretically could have supplied residents with potable water (Pendall and Amundson 1990) Its waters are derived from snow and ice melt at high altitude s and should therefore yield relatively low 18 O compared to lowland precipitation. A large culturally sterile depression at the center of the site might also have served as a rain fed reservoir (Kenoyer 1998: 59). Harappans relying on an uncovered reservo 18 O because of evaporative effects, wh ereas those drawing groundwater from wells might have intermediate values given the potential for mixing of river water and precipitation along with a reduction in evaporation. Only eight wells have been discovered at Harappa to date, however, so the exact nature of urban water consumption remains uncertain. In

PAGE 142

142 general, it is surmised that Harappan cultural practices impacted the isotopic composition of human tooth enamel, but no specific inferences can be made on the basis of archaeological evidence. Cemetery Area Cemetery R 37, the urban era cemetery, is located ~ 200 m south of the habitation area on a low rise. The area served simultaneously as a cemetery and disposal area for Harappan Phase debris (Dales and Kenoyer 1991) Five radiocarbon dates from charcoal in good context corroborate the relative date, putting the period of active use from ca. 2550 2030 BC (Meadow and Kenoyer 1994) Cemetery excavations occurred at different times under different directors, and no comprehensive publication on the burials and their contents has been published. R eports by Vats (1940), Wheeler (1947 ), Mughal (1968), Dales and Kenoyer (1989a, 1991), an d Meadow and Kenoyer (1994) lack sufficient detail for a detailed consideration of mortuary variability Only a cursory analysis is possible pending the comprehensive cemetery report by the Harappa Archa eological Research Project but the available data are summarized below for those burials with relatively complete descriptions (Table 6 2 ). Table 6 2 is a non representative sample as many burials were more fully described precisely because of their uniqu e features. Problems aside, the excavators at Cemetery R 37 confirmed a strongly normative mode of burial (Vats 1940; Wheeler 1947; Sastri 1965; Mughal 1968; Dales and Kenoyer 1991; Kenoyer 1998). Nearly all burials are extended and supine with the head to the north, and pottery is most often, but not exclusively, located at the head of the grave. Pottery vessels included with individuals number between one and fifty per grave, and occasionally are placed alongside the feet, sides of the body or underneath

PAGE 143

143 the body. Non pottery grave goods were generally limited to personal ornaments and toilet objects and included rings, earrings, and bangles of copper or shell, numerous beads of semi precious stone or steatite, shell spoons, and copper mirrors. Also, coni cal stone pendants and shell bangles are confined largely, but not exclusively, to women. Overall, Harappa has some of the richest graves known from the Indus Tradition, but the graves remain relatively modest and do not contain grave wealth beyond what i ndividuals could have adorned themselves with at one time. The body itself was occasionally interred in a wooden coffin or wrapped in a shroud, and a great many were disturbed in antiquity. Wheeler (1947) observed that eighteen burials crosscut earlier bur ials, and eight of those were crosscut themselves a tendency confirmed by subsequent excavations (Dales and Kenoyer 1991). Most of the burials in good condition are confined to a relatively undisturbed east west ridge, wh ereas fragmentary remains are freq uently found along the eroding slopes of the site or in the backfill of grave shafts. Contrary to the relative homogeneity in mortuary treatment, osteological analysis suggests certain gender based distinctions. For example, frequencies of caries and enam el hypoplasias suggest women likely had poorer childhood nutrition and increased access to cariogenic foods (Hemphill et al. 1991). Also, principal component analysis for comparisons of metric and non metric craniodental traits suggests that males and fema les derive from genetically distinct populations (Hemphill et al. 1991) Because the urban and post urban female populations retain closer genetic affinity to each other than do the urban and post urban males, this gender difference has been interpreted as evidence for matrilocality (Kennedy 2000)

PAGE 144

144 Lastly, the isotopic sample represents a relatively small fraction of all excavated burials, and as such, may not be representative of the large r mortuary population (Table 6 3 ). The present sample (n=51) comes f rom 40 features excavated in the 1987, 1988, and 1994 field seasons of the Harappa Archaeological Research Project (HARP). Samples include one or more teeth from 17 primary burials in various states of preservation, a single secondary burial, and the remai nder from disturbed contexts. Results of the Analyses Isotopic values and Sr concentrations resulting from the analyses are presented in Tabl e 6 3 Values of 87 Sr/ 86 Sr show a considerable range from 0.7 1113 to 0.72 802 with relatively modest Sr concentr ations ranging between 137 and 642 ppm. Lead isotope ratios also exhibit heterogeneity with 208 Pb/ 204 Pb from 38.013 to 39. 377 207 Pb/ 204 Pb from 15. 650 to 15.770 and 206 Pb/ 204 Pb from 18.054 to 19. 220 T ooth 18 O of 7.8 to 2.8 13 C of 14.0 to 8.7 andard deviations are presented fo r each tooth cohort in Table 6 4 As was the case for certain faunal teeth in the Harappa sample ( Chapter 4), many human teeth were also analyzed for 87 Sr/ 86 Sr previously on a Nu Plasma HR a high resolution multi collector, double focusing, plasma source mass spectrometer (MC ICP MS) at the University of Illinois Urbana Champaign. Standard deviation for 87 Sr/ 86 Sr was generally better than 0.00001 (Kenoyer et al. 2013), providing precision com parable to that of the TIMS used in the University of Florida Department of Geological Sciences (Chapter 3). Whenever possible, previously sampled teeth were re run using the TIMS (n=35) to maintain an internally consistent dataset and verify the MC ICP MS results. Further, repeat analysis on the TIMS was the most cost efficient way to acquire both 87 Sr/ 86 Sr and Sr concentrations without consuming additional

PAGE 145

145 sample beyond the amount required for Pb column chemistry. Except for two pairs of analyses with an absolute difference greater than 0.001, the TIMS results corroborate the MC ICP MS data. After excluding burials H87/85 74a and H88/161 170, the mean absolute difference between MC IC P MS and TIMS analyses is 0.00018 0.000 32 ( ) Inter laboratory differ ences might reflect variation in analytical procedures or sample preparation, or they might reflect real biological variation. Any given bulk enamel sample has a unique composition of appositional growth layers that mineralized at different times in that i different parts of a tooth crown might reflect the different geochemical environments experienced by an individual in life. Caution is advised when comparing datasets from study regions havi ng relatively little isotopic variability without first reviewing differences in laboratory methods. Regardless, the magnitude of 87 Sr/ 86 Sr differences reported here is largely irrelevant to a study of interregional migration given the diverse geochemical environments of the Greater Indus Valley. Therefore, the more complete TIMS dataset is shown in Table 6 3 and it serves as the basis for subseq uent figures and interpretations Isotopic data from burials H87/85 74a and H88/161 170 are excluded as the discrepancy in 87 Sr/ 86 Sr casts doubt on their provenience. Likewise, burials H87/85 49d.1, H88/185 186, and H88/206 208a are disregarded because the tooth position could not be det ermined. Lastly, data from premolars and canines are graphically presented as second molars and data from anterior dentition are classified as first molars because of their similar enamel mineralization times. A plot of 87 Sr/ 86 Sr against the reciprocal of Sr concentration is presented in Figure 6 1. Overlaid fields indicate the range of values for male s and female s There is

PAGE 146

146 substantial variation in the dataset, which appears to be controlled by at least three end members. Individuals appear to fall on one of two mixing lines between the local range estimated from faunal samples and either a more radiogenic low concentration source(s) or a less radiogenic source(s) of roughly similar Sr concentration. Male and female distributions overlap but are not cote rminous. Males predominantly plot along the less radiogenic mixing line whereas females plot on both. There is no clear segregation by tooth type, although first molars may be excluded from the local range estimated from the intersection of faunal and hum an data. Pb isotope ratios display similar evidence for mixing lines and segregation by sex (Figures 6 2 & 6 3). The variation is most apparent in the 207 Pb/ 204 Pb data presented in Figure 6 3. The triangular distribution of data suggests a geochemical envi ronment controlled by three end members. As with 87 Sr/ 86 Sr, males and females have overlapping but non identical distributions in Pb Pb space. Data from males, females, and fauna converge near one end of the distribution, suggesting the bulk of the data fa ll on mixing lines between the local or near local environment and one or more end members. The data are relatively homogenous with respect to tooth type, except that first molars seem to plot outside of the local range. Some first molar data points are si milar to faunal values, but they remain distinct from the tightest and slightly more radiogenic cluster of human and faunal values. Oxygen isotope variation also appears to correlate with sex and tooth type, although extremely small sample sizes preclude a ny statistical comparison wit h male and female third molars. 18 O across tooth types for the entire population, but male and female sub groups appear to show inverse trends over

PAGE 147

147 developmental time (Figure 6 4). The difference bet ween m ean 18 O for female and male first molars is not statistically significant, but a qualitative assessment suggests that female second molar 18 O increase s relative to first molars. The decrease between male first molar mean 18 O ( [ ] ) and second molar mean 18 O ( 1.2 [ ] ) is significant ( t (13) = 2.522, p = 0 .025) thereby indicating a change in water source around two or three years of age. 13 C although there is substantial variability in the data set (Figure 6 5). While there were certainly inter individual differences in diet, it does not appear to have been structured by age or sex. All sub samples have mean 13 ~ Lastly, Pb and Sr isotope systems are combined in Figure 6 6. As shown in the preceding heavy isotope plots, the data are structured by age and sex. Male and female distributions are different from one an other, but they converge with the faunal data at a point not well defined by a single end member. The increased resolution afforded by multiple isotope systems suggests a possible fourth end member with relatively radiogenic Pb isotope ratios and moderate values of 87 Sr/ 86 Sr. Additionally, first molars pl ot outside of the presumably local environment indicated at the human faunal convergence. As with the Farmana data set, the isotope data demonstrate changing environmental inputs over developmental time. Isotope ratios of Pb and Sr in conjunction with Sr c oncentrations suggest that individuals of all ages were exposed to a diverse and complex geochemical environment, but that first molars archived non local Sr and Pb inputs exclusively. Oxygen isotope values are similarly structured by

PAGE 148

148 tooth type, suggestin g a change in water sources around three years of age. Further, males and females tended to be exposed to different geochemical environments. Only 13 C shows no strong correlation with age or sex. T here is however, a wide range of 13 C values suggesting s ubstantial heterogeneity in diet between individuals.

PAGE 149

149 Table 6 1. Ch r onology at Harappa 1 Period Phase Dates 1 Ravi ca. >3300 2800 BC 2 Kot Diji ca. 2800 2600 BC 3A Harappa Phase A ca. 2600 2450 BC 3B Harappa Phase B ca. 2450 2200 BC 3C Harappa Phase C ca. 2200 1900 BC 4 Harappa/Late Harappa Transitional ca. 1900 1800 BC? 5 Cemetery H ca. 1800? <1300 BC 1 [adapted from Meadow R. H., and J. M. Kenoyer 2001 Recent Discoveries and Highlights from Excavations at Harappa: 1998 2000. Indo Koko Kenkyu [Indian Archaeological Studies] 22:19 36]

PAGE 150

150 Table 6 2 Select demographic data and mortuary elaboration at Harappa 1 Burial No. MNI Age Sex Orientation Position Pottery Qty. Ornaments Other Items H779a f, H783 2 7 4 adults 3F/1M N S supine H782 2 1 N S supine 2 faunal ribs H793 2 1 adult M N S left side steatite discs H793/A 2 1 adult M N S prone H793/B 2 1 adult M N S right side H794 2 1 adult M N S right side steatite discs faunal rib H795/A & B 2 1 adult F N S right side H798 2 6 3 adults 3M N S H798/A 2 1 adult F N S supine H798/B 2 1 subadult N S H799 2 1 N S left side H800, H800/A & B 2 1 N S supine faunal rib H801/A 2 1 adult F N S supine faunal rib H801/B 2 2 subadult M 2 faunal bones H803 2 1 NNE SSW right side faunal bone H804 2 1 adult F N S right side steatite discs, shell bangle H805/A 2 1 N S prone H806 2 2 adult M N S supine faunal ribs H808 2 1 N S supine 2 gold & carnelian beads 1 incomplete descriptions of material culture 2 from Vats 1940 3 from Wheeler 1947

PAGE 151

151 Table 6 2 Continued Burial No. MNI Age Sex Orientation Position Pottery Qty. Ornaments Other Items H814 2 2 N S supine H816, H816/A 2 2 adult F SE NW prone H817 2 1 adult F N S supine avian bones, lamp H818 2 1 adult M N S supine H820 2 3 3 adults 2F/1M N S left side 1 3 1 adult M N S supine 21 steatite disc chert flake 2 3 1 adult F N S supine 21 steatite disc copper mirror 3 3 1 21 4 3 1 0 5 3 1 adult F N S supine 37 copper ring, 3 shell rings coffin 6 3 1 3 faunal bones 7 3 1 3 8 3 1 2 1 faunal bone 9 3 1 15 10 3 1 adult M N S 2 brick lining 1 incomplete descriptions of material culture 2 from Vats 1940 3 from Wheeler 1947

PAGE 152

152 Table 6 3 Harappa mortuary sample results of the analyses Sample No. Burial Tooth 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C H18a 1 1 H87/ 25 18a LM 1 0.71485 390 38.013 15.683 18.054 5.4 11.7 H34a 2 1 H87/ 40 34a LM 2 0.71459 284 38.905 15.714 18.723 5.1 12.3 H34a2 1 1 H87/ 40 34a.2 RM 1 0.71360 246 38.865 15.728 18.658 3.5 12.9 H34b1 1 1 H87/ 40 89 34b.1 RM 1 0.71236 370 38.867 15.723 18.662 2.8 11.2 H49b1 3 H87/ 71 49b1 RM 3 0.71343 358 38.672 15.681 18.566 5.4 13.1 H49c 1 1 H87/ 71 49c M 1 0.71585 209 38.822 15.717 18.670 6.0 11.8 H49c 3 H87/ 71 49c M 3 0.71828 290 39.030 15.735 18.814 4.0 11.9 H49ha i H87/ 72 49h LI 1 0.71321 321 38.863 15.721 18.736 5.6 11.9 H49ha c 1 H87/ 72 49h LC 0.71741 366 39.033 15.732 18.794 5.9 11.7 H49hb 1 H87/ 72 49h M 1 0.71754 358 38.894 15.724 18.715 5.1 11.3 H49hc 3 1 H87/ 72 49h M 3 0.72090 177 39.114 15.738 18.836 4.6 11.3 H49b 1 1 H87/ 72 49b RM 1 0.71415 345 38.700 15.683 18.531 4.8 12.7 H49d1 m 12 H87/ 85 49d1 M 0.71389 334 38.654 15.684 18.526 4.6 12.5 H49ga 2 1 H87/ 85 49g LM 2 0.71922 404 38.746 15.712 18.565 4.0 10.9 H49gb 2 1 H87/ 85 49g LM 2 0.72190 238 39.066 15.736 18.825 4.3 12.1 H49hc 2 1 H87/ 85 49h RM 2 0.71760 388 39.081 15.743 18.868 5.1 11.4 H74a 2 1 H87/ 85 74a RM 2 0.71528 642 38.530 15.696 18.393 5.6 12.6 H80a 2 H87/ 91 80a RM 2 0.71659 481 38.821 15.727 18.724 5.0 11.8 H81a 3 H87/ 92 81a LM 3 0.71817 457 38.684 15.713 18.498 5.1 10.1 H126a 3 H87/ 108 126a RM 3 0.71880 467 39.095 15.742 18.856 4.7 12.3 1 light stable isotope data from [ Kenoyer J. M., T. D. Price, and J. H. Burton 2013. A New Approach to Tracking Connections between the Indus Valley and Mesopotamia: Initial Results of Strontium Isotope Analyses from Harappa and Ur. Journal of Archaeological Science 40(5):2286 2297 (Page 2294, T able 2)] 2 indeterminate molar position

PAGE 153

153 Table 6 3 Continued Sample No. Burial Tooth 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C H128a 1 H87/ 116 128a LM 1 0.71305 320 38.731 15.687 18.621 5.3 12.7 H147a 1 1 H87/ 136 147a LM 1 0.71272 323 38.802 15.700 18.648 4.9 12.6 H152a 1 1 H87/ 141 152a LM 1 0.71962 205 38.721 15.708 18.559 4.2 12.5 H156a 1 1 H87/ 145 156a LM 1 0.71853 372 38.515 15.705 18.405 4.2 10.8 H203a 2 1 H87/ 200 203a RM 2 0.71691 256 39.377 15.768 19.099 4.5 9.8 H127a 1 1 H88/ 114 127a LM 1 0.71230 469 38.699 15.700 18.545 5.2 12.3 H147a 1 1 H88/ 130 147a LM 2 0.71202 422 39.007 15.725 18.793 6.0 12.3 H170 2/3 12 H88/ 161 170 LM 2/3 0.71738 297 39.367 15.753 19.220 4.2 8.8 H121 p3 1 H88/ 162 121 RP 3 0.72123 265 39.046 15.735 18.804 3.5 12.3 H133a10 2 1 H88/ 173 133a.10 LM 2 0.71930 338 38.767 15.709 18.608 5.4 11.1 H126b2a 1 1 H88/ 174 126b.2 RM 1 0.71546 232 38.846 15.705 18.658 5.3 11.8 H126b2b 3 H88/ 174 126b.2 RM 3 0.71543 235 39.043 15.730 18.798 6.3 12.3 H170a17a 1 H88/ 180 170a.17 LM 1 0.71367 315 38.857 15.701 18.665 5.3 13.2 H170a17b 3 H88/ 180 170a.17 LM 3 0.71286 318 39.000 15.719 18.778 7.0 13.1 H185c1 2 1 H88/ 191 185c.1 RM 2 0.72707 167 39.054 15.748 18.901 4.0 11.4 H185f 1 1 H88/ 191 185f LM 1 0.72802 137 38.827 15.770 18.845 4.1 12.7 H196a 2 1 H88/ 194 196a LM 2 0.71275 235 38.935 15.713 18.724 4.9 11.4 H200a 2 H88/ 198 200a LM 2 0.71980 295 38.656 15.690 18.559 2.8 12.6 H204a 1 H88/ 201 204a RM 1 0.71951 367 38.878 15.710 18.697 4.4 11.4 1 light stable isotope data from [ Kenoyer J. M., T. D. Price, and J. H. Burton 2013. A New Approach to Tracking Connections between the Indus Valley and Mesopotamia: Initial Results of Strontium Isotope Analyses from Harappa and Ur. Journal of Archaeological Science 40(5):2286 2297 (Page 2294, T able 2)] 2 indeterminate molar position

PAGE 154

154 Table 6 3 Continued Sample No. Burial Tooth 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C H208a mx 12 H88/ 206 208a M 0.71842 314 38.972 15.730 18.779 3.9 11.3 H208a my 12 H88/ 206 208a M 0.72128 239 39.035 15.744 18.839 4.7 11.6 H219a 1 1 H88/ 216 219a LM 1 0.71896 264 39.167 15.751 18.980 4.1 8.7 H220a 2 1 H88/ 217 220a LM 2 0.71300 430 38.737 15.689 18.576 6.1 11.7 H4b 2 1 H88/ 439 4b LM 2 0.71258 340 38.753 15.710 18.604 5.9 12.3 H5#2 2 H94/ 243 5#2 LM 2 0.71113 557 38.625 15.678 18.612 5.2 12.0 H27 2 1 H94/ 243 27 LM 2 0.71604 411 38.934 15.735 18.856 3.8 12.3 H7 1 1 H94/ 245 7 LM 1 0.71475 309 38.613 15.697 18.536 4.9 12.6 H17 1 1 H94/ 250 17 RM 1 0.71480 358 38.744 15.694 18.604 4.6 12.8 H18 1 1 H94/ 253 18 LM 1 0.71274 353 38.149 15.650 18.530 4.0 12.5 H18 2 1 H94/ 253 18 LM 2 0.71248 360 38.172 15.650 18.535 7.8 14.0 1 light stable isotope data from [ Kenoyer J. M., T. D. Price, and J. H. Burton 2013. A New Approach to Tracking Connections between the Indus Valley and Mesopotamia: Initial Results of Strontium Isotope Analyses from Harappa and Ur. Journal of Archaeological Science 40(5):2286 2297 (Page 2294, T able 2)] 2 indeterminate molar position

PAGE 155

155 Table 6 4 Harappa mortuary sample summary statistics -Cohort 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C 1st molar: all 0.715090.005 320145 38.7050.566 15.7060.046 18.5930.382 4.71.6 11.92.1 1st molar: female 0.716050.005 311182 38.6770.618 15.7050.041 18.5490.489 4.91.3 11.72.5 1st molar: male 0.713850.004 33385 38.7410.531 15.7060.055 18.6500.138 4.42.0 12.21.4 2nd molar: all 0.716530.009 321180 38.9890.349 15.7280.043 18.7960.289 5.01.7 11.71.5 2nd molar: female 1 0.721670.011 272251 39.0180.146 15.7400.015 18.8610.076 4.00.5 11.90.9 2nd molar: male 0.714610.005 340149 38.9790.407 15.7240.048 18.7720.328 5.41.2 11.61.7 3rd molar: all 1 0.718210.005 234113 39.0620.091 15.7340.008 18.8160.038 5.02.3 11.81.0 3rd molar: female 1 0.716860.004 26378 39.0370.019 15.7320.007 18.8060.022 5.13.2 12.10.7 3rd molar: male 1 0.72090 177 39.114 15.738 18.836 4.6 11.3 1

PAGE 156

156 Figure 6 1 Harappa samples in Sr Sr space by sex reciprocal of Sr concentration Figure 6 2 Harappa samples in Pb Pb space 208 Pb/ 204 Pb

PAGE 157

157 Figure 6 3 Harappa samples in Pb Pb space by sex 207 Pb/ 204 Pb Figure 6 4 Harappa samples 18 O

PAGE 158

158 Figure 6 5 Harappa samples 13 C Figure 6 6 Harappa samples in Pb Sr space by sex 87 Sr/ 86 Sr

PAGE 159

159 CHAPTER 7 SANAULI Localization Era Mortuary Treatment Localization Era cemeteries are even less well documented than those of the Integration Era, thus the mortuary site of Sanauli provides one of the few available opportunities for a diachronic isotopic perspective on mobility within the Indus Civilization. Sanauli is the only known Localization Era cemetery apart from Cemetery H and Area G at Harappa, but the disturbed remains from Area G demonstrates little continuity with broader Indus mortuary practices (Vats 1940) T he formalized inhumations of Cemetery H provide the best contemporaneous parallel for mortuary practices at Sanauli. Excavated by M. S. Vats between 1928 1934 (1940) and Sir Mortimer Wheeler in 1946 (1947), 26 inhumations were discovered in the lower levels (Stratum II) of Cemetery H which wer in the uppermost levels (Stratum I) (Gupta et al. 1962). Archaeologists have variously emphasized the similarities of Stratum II inhumations with those of Cemetery R 37 (Kenoyer 1998: 174 175) as well as th e dissimila rities (Wright 2011: 266) Though no exhaustive descriptions of Cemetery H burials have been published, ratum II inhumations (1940: 220 240) depicts a mortuary program that is broadly similar to that of the urban er a. Some individuals are buried in a flexed posture, but others have the extended supine position known from urban era cemeteries. Likewise, the same broad functional classes of pottery are present at Cemetery H (e.g., dish on stand) Though pottery was som etimes deposited at the feet, at other times it was placed near the head and occasionally buried below the body as frequently occurred at urban cemeteries. Other parallels can be seen in the relatively

PAGE 160

160 modest display of grave wealth which is limited primar ily to pottery and a few personal ornaments. Cemetery H burials are differentiated from those of the urban era by the inclusion of some faunal remains and diverse grave orientations, but the relatively greater variability of post urban mortuary practices does not appear to be strong ly patterned ) selective description. The lack of redundancy in mortuary traits implies that the different mortuary treatments are best regarded as variations on a common theme. For example, demographic categ ories like age and sex are not consistently associated with any particular burial practices, and spatial variables such as grave orientation show no obvious correlation with other mortuary traits. The provisional conclusion, then, is that the post urban pr ogram of inhumations dealt primarily with those aspects of identity shared by the deceased rather than those that differentiated them from each other. Indeed, a similar inference was made in the preceding chapters for urban era inhumations, suggesting that the general criteria for inhumation in a cemetery may have remained the same even when affiliations with specific social institutions or cultures changed One possibility to consider is that migration or a social identity contingent on residence change wa s a prerequisite for inclusion in the highly restrictive program of Indus cemetery inhumations. The Sanauli Cemetery Located between the Yamuna and Hindon Rivers in what today is western Uttar Pradesh, Sanauli is one of the few well documented post urban mortuary sites of the Indus Tradition. D.V. Sharma led excavations from 2005 2006 and discovered 94 burials and 22 additional associated features across a 1565 m 2 area. The full extent of the site is unclear, but local residents report finding further arch aeological materials in

PAGE 161

161 the surrounding fields, suggesting the site coul d be much larger (Prabhakar 2012 ). No associated habitation area has been discovered yet, but Indus inhabitants likely would have enjoyed relatively reliable agricultural production co mpared to people living in more arid regions of the Indo Gangetic divide. Modern precipitation gradients and lithostratigraphic transects for this ecologically transitional region suggest that summer precipitation declined steeply in areas to the west and south (Saini and Mujtaba 2012) allowing Sanauli residents to conceivably grow more drought susceptible cereals like wheat, barley, and rice. Further, modern topographical observations suggest the site may have bee n adjacent to an ox bow lake, a now dry remnant of the Ya muna that is presently located ~ 7 km to the west (Prabhakar 2012 ). In fact, local ecologies were probably important influences in the redistribution and transformation of Indus settlements across the subcontinent during the post urban era, and no uniform statements can be made about post urban agricultural production and social organization (Weber et al. 2010 ; Wright and Schuldenrein 2008 ). Unfortunately, such post urban heterogeneity coupled with the lack of habitation data permit only the most tenuous speculations on the broader socio economic conditions at the site. It might even be the case that Sanauli was a stand alone necropolis used by multiple communities in the area. M odern residential and ag ricultural land use however, might be concealing the habitation area at the site. Domestic context aside, the mortuary material culture at Sanauli has much in common with other post urban Indus sites. Though some specific types are unique to the site, much of the ceramic corpus consists of well fired, wheel made red ware in forms known from other post urban sites including Bhagwanpura, Hulas, Mitathal,

PAGE 162

162 Mandi, and Cemetery H at Harappa (Sharma et al. 2005, 2007; Prabhakar 2012 ). area reliable chronology is unavailable however, V.N. Prabhakar (201 2) u s ed a modified version of the classification system advocated by Dales and Kenoyer (1986 a ) and estimated that the cemetery was active during the first half of the 2nd millennium BC. Unfortunately, d isturbed topsoil makes stratigraphic comparison unreli able. Multiple phases of burial activity are apparent, though not precisely defined, suggesting relatively long term use of the site (Prabhakar personal communication). Lastly, the presence of multiple gold grave goods further supports a post urban chronol ogy because the precious metal is only rarely found in urban era mortuary contexts. Primary Burials and the Osteological Sample Though Sanauli can provide vital insight into post urban Indus life, any interpretation of the mortuary program must be precede d by a discussion of bone preservation. The vicissitudes of monsoon weather and extensive disturbance over the millennia have heavily impacted the cemetery. The skeletal remains in particular have been damaged, and even the most complete skeletons are miss ing various elements. In many instances, only highly fragmentary crania or long bones r emain. Robust elements of the legs and pelvis are particularly well represented, and this bias can give the impression of secondary mortuary activity. In each case, howe ver, the preservation can be attributed to disturbances such as modern agricultural activity or crosscutting burials (Prabhakar personal communication). The latter disturbances appear to be a common feature of the broader Indus mortuary program and at Ceme tery R 37, at least, ha ve resulted in numerous skeletal fragments either isolated or in grave shaft backfill (Dales and Kenoye r 1989 a ). Several recorded features at Sanauli are little more than small

PAGE 163

163 piles of human or faunal bone fragments, and they are pe rhaps best interpreted as an unintentional byproduct of repeated digging. However, secondary mortuary activity may best explain some aspects of the skeletal record It was noted that the tibia of burial 21 is longer than the associated femur and one clavi cle of burial 37 is shorter and more robust than the other (Sh arma et al. 2007; Prabhakar 2012 ). Attributed to pathology and occupational stress, these are nevertheless very unusual conditions and it is difficult to rule out either a technical error or the possibility that the skeletons were misassembled during secondary mortuary treatment. Given the limited information available however, the most conservative approach for this analysis is to follow the classification of Prabhakar and assume all but four o f the burials are primary inhumations. The remaining four burials and are discussed below Pooled demographic data from the osteological field analysis and my own assessment of the isotopic sample suggest the mortuary population is likely no t a representative cross section of a breeding population (Table 7 1). Adults are nearly twice as prevalent relative to sub adults among the 35 individuals for whom age estimation was possible. This inversion of the mortality profile commonly found in stab le archaeological population s suggests either that the majority of sub adult skeletons were not preserved or more plausibly, that sub adults were less frequently inhumed in the cemetery No such bias appears to have existed based on sex, however, as males and females are present in nearly equal proportions. Also, observations from the field analysis suggest that females are relatively robust and of slightly greater stature than males. Though sample sizes are small, this finding could suggest genetic distance

PAGE 164

164 between the sexes i.e., males and females derived from different populations as has been inferred at Harappa from biodistance studies (Hemphill et al. 1991; Kennedy 2000). The relatively small number of sub adults suggests the mortuary population was structured by cultural practices, but there is little correlation between demographic variables and particular mortuary treatments (Table 7 1). One possible exception, however, is the inclusion of beads. Out of 13 burials with b eads, six are associated with sub adults, three are associated with adults, three are from burials with no age estimate, and one comes from a double burial containing the remains of one adult and one partial skeleton with no a ge estimate. In fact, the rich est burial in the cemetery is a sub adult adorned with four go ld bangles and elaborate gold and copper bead necklaces. O ne possible explanation for the observed distribution of wealth items such as beads is that burial was an expensive process reserved for more prominent kin. Wealthier families, however, may have been able to bury younger individuals as readily as adults, leading to a higher proportion of sub adults buried with beads. Nevertheless, this remains speculative and mortuary samples at other site s are needed to confirm the trend. The substantial majority of burials at Sanauli is less elaborate and conforms to the typical Indus pattern extended and supine skeletal remains oriented in a northerly direction with pottery near the head and modest grav e wealth, if any. The local expression of the mortuary program varies across a range of traits, most of which are known from other cemetery sites. For example, burials are oriented around a NW SE axis as is the case for Farmana instead of the N S trend o bserved at Cemetery R 37.

PAGE 165

165 Four inhumations (16, 18, 25, and 73) deviate furthest from the N S axis and orient nearly west to east, although this alignment does not appear to be associated with any other mortuary trait in particular. Like grave orientation, associated ceramics vary in ways that are similar to other Indus cemeteries. Pots were sometimes interred at a level lower than the body (n=5), suggesting a multistep inhumation process like that known from Cemetery R 37 (Dales and Kenoyer 1991). Pots wer e also sometimes placed underneath the pelvic area, along the sides, or at the feet rather than just near the head (n=19). No obvious correlation is discernible between these alternative modes of pottery placement and other mortuary variables, although com parisons are limited by the highly disturbed skeletal sample and the nature of the ceramic analysis. No discussion is presented here on the proportions of Harappan vs. non Harappan ceramics as was the case in C hapter 5 because the Sanauli reports (Sh arma et al. 2007; Prabhakar 2012 ) make no systematic burial specific distinction between pots of the Indus Tradition (e.g., Cemetery H or Punjab phase) and pots made in regional styles. Symbolic Burials The most unusual burials at Sanauli are the so 106, and 116). T he symbolic burials clearly constitute a distinct class although a spectrum of continuity with other burials in terms of their overall orientation, pottery forms, and pottery distribution might suggest a hybridizat ion of mortuary practices. Further, intermediate burial forms might indicate interregional interaction between local residents and non locals with diverse cultural affiliations. Unfortunately, skeletal material is absent having been substituted for by seve ral unique ceramic types and various ritual objects. Burial 14 contains 18 vessels, three of which are a type of vase found nowhere else in the Sanauli assemblage. In place of a skeleton, a copper sheath lies

PAGE 166

166 across a dish on stand, and situated amongst th e pots at the head of the burial with swords are known from archaeological caches of copper artifacts including celts, weapons, anthropomorphs and other items that co llectively are referred to as copper hoards. They are most often found in and around the Ganga Yamuna doab, but they have been recove red from many parts of India u sually as isolated finds without associated context (Lal 1951; Gupta 1989) Excavations at Sapai, however, have shown them to be contemporaneous with Ochre Colored Pottery (OCP), an archaeological complex of the 2nd millennium BC variously deemed a distinct archaeological culture or a regional manifestation of the post urban Indus Tradition (Lal 1972; Dikshit 1979; Singh 2008:216 218) In either case, the antenna sword of burial 14 sugg ests that Sanauli residents occupied a culturally diverse landscape. s that symbolic burials constitute a distinct mortuary type expressed partly in terms of the normative mode of inhumation. For example, burial 2 8 consists of 28 copper anthropomorphs laid out in the shape of a violin and bounded by a copper strip. The composite copper display is directly adjacent to two dishes on stands of a unique type that nevertheless seem to reference the way that a dish on st and supports the pelvis in some extended burials. Burial 106 invokes both the violin shaped display of burial 28 and the copper sheath of burial 14. No ceramics are present, but multiple rows of steatite inlays outline a violin shape with a along an E W orientation. A copper sheath lies in the center of the inlays and t he entire display is flanked by a mud brick wall to the north which could possibly be one wall of a burial chamber. Two such walls are present at the northern corners of

PAGE 167

167 burial 116 which is made up of 15 ceramic vessels adjacent to a copper display A bowl sits in the middle of a violin shaped display made from approximately 600 faience beads. Though highly disturbed, burials 66 and 73 further support the idea of a hybrid mortuary program. Both contain fragmentary skeletal remains and are laid out in a manner comparable to other inhumations, but they also contain copper items like those of the symbolic burials. Burial 66 includes a display of six copper anthropomorphs arran ged similarly to those of burial 28, and burial 73 has a copper sheath. Collectively, the symbolic burials and burials 28 and 73 suggest that participation in the Sanauli mortuary program was not subject to the same ideological controls that have been infe rred for urban era cemeteries based on their relative homogeneity ( Pollock 2008 ). At least some residents had leeway to hybridize different modes of burial, such that they were very likely aware of the diverse challenges and opportunities in a variable soc ial landscape. Hybridized burials may have been one medium by which interregional relationships between OCP producing peoples and other post urban groups were negotiated. Further, migration may have been structured by the interactions between OCP sites fro m the east and south and other post urban Ind us sites to the north and west. No tooth enamel was recovered from the hybridized burials, but evidence for interaction along these lines may still be observable in the broader isotopic sample. Results of the A nalyses Isotopic data and Sr concentrations are presented in Table 7 2. Values of 87 Sr/ 86 Sr range from 0.7 1254 to 0.72 714 with Sr concentrations f rom 93 to 359 ppm. Lead isotope ratios also exhibit heterogeneity with 208 Pb/ 204 Pb from 38.156 to 39. 478 207 P b/ 204 Pb from 15. 677 to 15.827 and 206 Pb/ 204 Pb from 18.227 to 19.439 T ooth

PAGE 168

168 18 O of 6.6 to 1.7 13 C of 13.5 to 4.4 andard deviations are presented for each tooth cohort in T able 7 3 No faunal teeth were available at t he time of sampling, and sediment leachates serve as the only proxy for local environmental variability of Sr and Pb isotope ratios. The results of all analyses are given, but data from deciduous teeth (burials 36, 67, 74B, 95, and 99) are excluded from su bsequent figures and interpretations because deciduous enamel is presumed to be more susceptible to diagenetic alteration. Data from burials 34, 55, and an individual of unknown burial number are similarly excluded because of their indeterminate molar position. A plot of Sr ratios against the recipr ocal of Sr concentration is presented in Figure 7 1. Unlike the Farmana sample, the Sanauli distribution suggests a random scatter rather than a distinct mixing line between two endmembers This could mean that the geochemical environment was relatively ho mogenized with respect to Sr isotopes. On the one hand, variability is substantial but does not seem to be represented by a simple mixing line On the other hand, high inter individual variation in Sr concentration caused by dietary differences might obscu re a linear trend between a more radiogenic low concentration source and a less radiogenic high concentration source. The Pb isotope data, however, show a clear linear trend (Figures 7 2 & 7 3). The patterning suggests that i ndividuals at Sanauli were expo sed to environmental sources with diverse Pb isotope ratios although the range of values cannot be used to infer geographic distance without additional baseline analyses The data may indicate a two component mixing system controlled by a more radiogenic and a less radiogenic

PAGE 169

169 source. Further, local soil ratios cluster at one end of the distribution whereas the full range of ratios is represented by all three tooth types. Oxygen isotope values show a clear trend based on tooth type (Figure 7 4). The f irst molar mean 18 O ( 1.8 [ ] ) is not significantly higher than that of second molars ( 0 1.6 [ ] ) but i s significantly higher than the third molar mean 18 O ( 1.3 [ ] ) ( t (34 ) = 2. 867 p = 0 .0 07 ). The shift was ongoing or not yet begun at the time of second molar mineralization suggesting a change in water source typically occurred late in the mineralization process at ~5 years of age Carbon isotope values differ from 18 O in that the mean values remain stable across developmental time (Figure 7 5). M ean 13 C for first molars is 1.5 ( for second molars is 1.4 ( 1.0 ( Further, t here is relatively less intra popula tion variation in the consumption of C 3 vs. C 4 13 C In other words, there is some variation in the diets of individuals but the relative contribution of C 3 and C 4 foods is not systematically associated with partic ular stages in life. Lastly, plotting isotope ratios of Pb against those of Sr shows a more complicated pattern of variability than suggested by either isotope system alone (Figure 7 6). Figure 7 6 confirms that environmental sources of Sr and Pb are not s ystematically related to developmental stage s as is apparent for individuals at Farmana and Harappa. The only possible exception is that the overall range in third molar data is slightly less than that for first and second molars, which could indicate fewe r sources of environmental Sr and Pb in late childhood. Overall, the data cluster most tightly near the soil values, suggesting most samples were formed under exposure to the local environment. Lastly,

PAGE 170

170 at least three linear distributions radiate from the l suggesting at least three additional sources of exposure to environmental Sr and Pb. The heavy isotope data suggest that most individuals shared a common geochemical environment independent of their age. Further, the fan sha ped distribution of data points seems to radiate away from locally available sediment leachates towards less radiogenic Pb and more radiogenic Sr Nevertheless, a substantial portion of the dataset can likely only be explained by additional exposure to one of at least three 18 O correlates with tooth type, suggesting most individuals experienced a shift in water sources around age five. Lastly, there is very little change in 13 C by tooth type suggesting that the same range of foods underlies isotopic exposure throughout childhood. Further syntheses and interpretations are given in the following chapters.

PAGE 171

171 Table 7 1. Demographic data and mortuar y elaboration at Sanauli Burial Age Sex Orientation Pottery Qty. Beads Bangles Other Items 2 3 4 NW SE 2 1 agate 4 adult F NW SE 2 copper 18 adult F W E 8 20 adult NW SE 21 adult F NW SE 2 22 8 12 NW SE 2 23 adult M NW SE 5 2 carnelian 25 adult F W E 2 27A adult M NW SE 6 1 steatite, 1 agate 36 4 5 NW SE 5 37 adult M NW SE 1 38 adult M NW SE 42 adult M NW SE 9 1 gold caplet 44A 2 adults NW SE 4 48 1 3 NW SE 2 agate 2 copper 49 adult M NW SE 57 adult M NW SE 60 adult M NW SE 2 61 12 15 NW SE 67 1 2 NW SE 2 5 69 3 adults F/M/ NW SE 25 71 5 9 NW SE 74 adult/1 3 F/ NW SE 80 12 15 NW SE 85 adult F NW SE 10 93 6 7 NW SE 5 5 2 copper

PAGE 172

172 Table 7 1. Continued Burial Age Sex Orientation Pottery Qty. Beads Bangles Other Items 95 2 4 NW SE 6 8 carnelian 4 gold gold & copper necklaces 97 adult NW SE 98 adult F NW SE 1 99 4 5 NW SE 3 112 adult NW SE 1

PAGE 173

173 Table 7 2 Sanauli mortuary sample results of the analyses Sample No. Burial Tooth 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C SYC 1 YC1/2 hearth RM 1 0.72078 180 39.317 15.798 19.259 4.7 11.8 SYC 2 YC1/2 hearth RM 2 0.72042 217 39.310 15.792 19.265 5.9 11.6 SU3 1/2 1 unknown 3 RM 1/2 0.72064 155 39.261 15.799 19.115 4.7 12.7 S20 p4 20 LP 4 0.72714 171 39.079 15.781 19.166 3.6 13.3 S20 3 20 LM 3 0.72213 211 38.913 15.748 18.917 4.8 12.4 S22 1 22 RM 1 0.71709 254 39.387 15.807 19.312 4.1 11.1 S22 2 22 RM 2 0.71693 295 39.298 15.793 19.228 3.9 11.1 S25 1 25 LM 1 0.71938 203 39.435 15.816 19.353 4.1 10.3 S25 2 25 LM 2 0.71957 234 39.329 15.812 19.248 3.9 11.5 S25 3 25 LM 3 0.71984 208 39.291 15.793 19.225 4.1 12.2 S29 p3 29 LP 3 0.71962 215 39.169 15.763 19.250 3.1 11.5 S34 1/2 1 34 RM 1/2 0.71984 195 39.282 15.784 19.232 4.9 11.6 S36 d2 2 36 LdM 2 0.71881 193 39.296 15.787 19.221 4.0 11.9 S36 1 36 RM 1 0.71876 219 39.444 15.813 19.346 4.4 11.9 S37 1 37 LM 1 0.72182 93 38.971 15.750 19.025 3.7 11.6 S37 2 37 LM 2 0.72216 98 39.284 15.798 19.205 4.0 10.5 S37 3 37 LM 3 0.72213 126 39.330 15.803 19.227 4.9 11.4 S38 1 38 RM 1 0.71925 157 38.883 15.730 18.985 4.5 13.3 S38 2 38 RM 2 0.71906 208 39.401 15.801 19.290 5.0 12.7 S44A 1 44A RM 1 0.71923 133 39.274 15.779 19.211 4.6 12.6 S44A 2 44A RM 2 0.71900 174 39.356 15.797 19.267 5.4 12.1 S44A 3 44A RM 3 0.71891 156 38.589 15.677 18.752 5.3 11.7 S44B 1 44B LM 1 0.72164 127 39.145 15.776 19.008 4.1 12.8 S44B 2 44B LM 2 0.72119 209 39.171 15.773 19.073 4.5 11.6 S44B 3 44B LM 3 0.72175 170 39.111 15.775 18.992 5.0 11.3 1 indeterminate molar position 2 enamel is deciduous or poorly mineralized

PAGE 174

174 Table 7 2 Continued Sample No. Burial Tooth 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C S44C 1 44C RM 1 0.71757 181 39.302 15.786 19.237 4.6 12.3 S44C 2 44C RM 2 0.71762 245 39.375 15.807 19.316 5.2 11.3 S49 2 49 RM 2 0.72101 158 39.347 15.802 19.287 4.1 12.2 S54 1 54 LM 1 0.71894 189 39.026 15.747 19.016 3.3 11.6 S54 2 54 RM 2 0.71807 198 39.379 15.805 19.294 4.4 10.3 S55 1/2 1 55 LM 1/2 0.71254 236 38.882 15.716 18.902 2.7 4.4 S58 i2 58 LI 2 0.71863 141 38.156 15.725 18.227 5.4 12.5 S61 1 61 RM 1 0.71866 191 39.318 15.827 19.439 3.3 12.6 S61 2 61 RM 2 0.71851 203 39.255 15.791 19.216 5.3 12.8 S61 3 61 RM 3 0.71895 241 39.387 15.801 19.294 6.6 11.8 S67 d2 2 67 RdM 2 0.72209 158 39.405 15.808 19.330 4.3 12.4 S69A 1 69A LM 1 0.71983 225 39.089 15.770 19.075 4.3 12.1 S69A 2 69A LM 2 0.71950 300 39.095 15.764 19.095 4.4 11.6 S69A 3 69A LM 3 0.71899 312 39.293 15.787 19.222 4.7 11.9 S69B 1 69B LM 1 0.71816 226 39.314 15.792 19.346 1.7 11.3 S69C 2 69C LM 2 0.71953 254 38.755 15.721 18.805 4.0 11.8 S69C 3 69C LM 3 0.71909 233 39.244 15.780 19.196 4.7 12.0 S71 1 71 LM 1 0.71974 167 39.313 15.791 19.230 2.5 12.1 S71 2 71 LM 2 0.71978 183 39.251 15.786 19.187 4.1 11.7 S74A 1 74A LM 1 0.71900 160 39.378 15.801 19.289 3.6 13.5 S74A 2 74A LM 2 0.71910 183 39.351 15.800 19.261 2.5 12.5 S74A 3 74A LM 3 0.72005 164 39.292 15.787 19.191 4.7 11.8 S74B d2 2 74B RdM 2 0.71893 101 39.110 15.765 19.059 3.9 12.7 S80 1 80 RM 1 0.71940 206 39.334 15.791 19.254 5.6 12.8 S80 2 80 RM 2 0.71991 278 39.365 15.803 19.257 4.6 11.8 1 indeterminate molar position 2 enamel is deciduous or poorly mineralized

PAGE 175

175 Table 7 2 Continued Sample No. Burial Tooth 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C S80 3 80 RM 3 0.71875 264 39.444 15.817 19.357 5.1 12.9 S85 1 85 LM 1 0.72133 179 39.180 15.787 19.068 3.7 11.4 S85 2 85 LM 2 0.72069 213 39.005 15.727 19.122 4.1 11.8 S93 1 93 LM 1 0.71953 198 39.345 15.798 19.269 3.4 11.9 S95 d2 2 95 RdM 2 0.71933 153 39.373 15.801 19.293 4.3 11.5 S97 1 97 LM 1 0.71905 182 39.064 15.754 19.130 4.3 12.8 S97 2 97 LM 2 0.71946 200 39.003 15.747 19.016 5.0 12.4 S97 3 97 LM 3 0.71940 180 38.891 15.732 18.966 4.6 12.4 S98 1 98 LM 1 0.72021 163 39.372 15.804 19.275 3.7 13.1 S98 2 98 LM 2 0.72048 243 39.277 15.794 19.200 4.0 12.1 S98 3 98 LM 3 0.72137 217 39.362 15.801 19.259 4.0 12.7 S99 d2 2 99 LdM 2 0.71990 136 39.229 15.790 19.172 4.4 11.2 S99 1 99 LM 1 0.72008 178 39.314 15.796 19.237 4.9 11.5 S112 1 112 LM 1 0.71846 300 39.478 15.827 19.377 5.1 12.3 S112 2 112 RM 2 0.71858 290 39.472 15.824 19.369 5.0 12.0 S112 3 112 LM 3 0.71962 359 39.421 15.821 19.316 4.9 12.8 1 indeterminate molar position 2 enamel is deciduous or poorly mineralized

PAGE 176

176 Table 7 3 Sanauli mortuary sample summary statistics Cohort 87 Sr/ 86 Sr Sr ppm 208 Pb/ 204 Pb 207 Pb/ 204 Pb 206 Pb/ 204 Pb 18 O 13 C All 0.719650.00351 200105 39.2210.450 15.7840.060 19.1760.366 4.31.6 11.92.3 1st molar1 0.719410.00238 18588 39.2100.558 15.7850.056 19.1720.488 4.11.8 12.11.5 2nd molar 0.719880.00403 21797 39.2420.336 15.7860.053 19.2010.244 4.41.6 11.81.4 3rd molar 0.720080.00260 218130 39.1970.508 15.7790.079 19.1470.363 4.91.3 12.11.0

PAGE 177

177 Figure 7 1. Sanauli samples in Sr Sr space reciprocal of Sr concentration Figure 7 2. Sanauli samples in Pb Pb space 208 Pb/ 204 Pb

PAGE 178

178 Figure 7 3. Sanauli samples in Pb Pb space 207 Pb/ 204 Pb Figure 7 4. Sanauli samples 18 O

PAGE 179

179 Figure 7 5. Sanauli samples 13 C Figure 7 6. Sanauli in Pb Sr space 87 Sr/ 86 Sr

PAGE 180

180 CHAPTER 8 DISCUSSION The isotope data presented in chapters 5, 6, and 7 are structured and several patterns emerge that require further discussion. In particular, isotope data from Farmana and Harappa are structured by tooth cohort and sex. At Farmana, the first molar cohort is completely segregated from all other human, faunal, and sediment samples by Pb isotope ratios (Figures 5 2 & 5 3) whereas th ere is little variation in 87 Sr/ 86 Sr ( Figure 5 1 ) The first molar cohort at Harappa overlaps with both the second and third molar cohort with the except ion that no first molar data plot in the local range inferred from the intersection of faunal and human heavy isotope data (Figure 6 6 ) Further, Pb, Sr, and O isotope datasets from Harappa are structured by sex. In general, females exhibit a wider range of variation in Pb Sr space (Figure 6 6 ) and appear to 18 O between first and secon d molars (Figure 6 4 ) contra ry to the expected decrease 18 O associated with weaning ( Wright and Schwarcz 1998 ) One possible interpretation is that inhumation at the Integration Era cemeteries of Harappa and Farmana was reserved for immigrants who left their natal groups at a young age, presumably without the company of their genetic kin. Further, provenience data are consistent with origins in the highlands, piedmont, or other hinterland regions outside of th e foreland alluvium. As such, immigrants may have originated from non Indus sites although the present evidence is insufficient to unequivocally determine their cultural affiliations. If confirmed by future research, however, immigration from non Indus si tes might help explain their distinct cultural trajectories within a broader interregional system of economic exchange. Multiple lines of isotopic evidence are presented in support of the following claims :

PAGE 181

181 Nearly all individuals inhumed in Integration Era cemeteries were immigrants. The immigrants came from hinterland sites. Initial residence change occurred early in childhood. By contrast, t he isotope data from Sanauli sugge st the mortuary population was less mobile and that migration largely occurred wi th in the alluvial lowlands C ertain elements of the urban era institution of migration may have persisted into the post urban era, but if so, the relationship between institutionalized immigration and cultural identity had fundamentally transformed. As mentioned above, a dditional analyses are needed to resolve issues of equifinality in the geographic provenience and cultural affiliation of urban era individuals, however, a model of highland lowland interaction is proposed predicated on the possibility t hat immigrants originated with non Indus groups that were trading with lowland elites. The model provide s an investigative framework for future hypothesis testing, and more broadly, demonstrates the need to account for diverse social dimensions when interpreting migration and exchange. The mere delineation of trade routes and migration paths does little to advance our understanding of past societies unless such provenience data are embedded within the social landscape. Integration Era Mobility Before moving forward with behavioral explanations for the isotope data, the validity of the data must be addressed. If the isotopic values had been altered at any stage in the post mortem environment, it would compromise the utility of the data for modeling past human mobility. Fortunately, the structure of the data set provides evidence that the reported isotope values are repre sentative of biogenic exposure rather than post mortem diagenesis First and foremost, the isotopic data would not be

PAGE 182

182 patterned by tooth type if diagene sis w ere a major concern; instead, all teeth would display similar isotope values The urban cemeteries show a trend towards non local isotope valu es among the first molar cohort whereas values of the second and third molar cohorts tend to plot near local sediments and fauna. Even at Sanauli where there is no single pattern of convergence, intertooth variation appears to be structured along a series of mixing lines that cannot be readily explained by isotopic variation in local sediments. Therefore the isotope data from Farmana, Harappa, and Sanauli appear to preserve the in vivo signal that reflects ingested environmental sources It is conceivable that some partial post mortem alteration occurred thus shifting values towards those of th e burial environment but not enough to eradicat e the intra individual patterns. It is difficult to disprove this possibility for any given sample, but as discussed in Chapter 3, the preponderance of data from analyses in a wide range of archaeological cont exts suggest s that enamel provides a high fidelity record of isotopic exposure. Therefore, it is assumed that the isotopic values for Indus inhumations accurately represent weighted averages of environmental input at the time of enamel mineralization for e ach individual Immigrant Frequency Multiple lines of isotopic evidence suggest that only first generation immigrants received inhumation in the urban era cemeteries at Farmana and Harappa. When plotted by tooth cohort isotope ratios of Pb and Sr show that all or part of early life enamel mineralization occurred during a time when the individuals were exposed to non local environmental inputs. The distinction between local and non local is complicated by the inherent imprecision in any proxy measure of the hu man dietary catchment (Chapter 3) In this case, however, the internal structure of multi isotope

PAGE 183

183 datasets allows for a relatively precise estimate of local ratios. In scatter plots of Pb and Sr isotope ratios, t he convergent distributions of fauna, sediment leachates, and late developing human tooth enamel suggest that a relatively small range of ratios characterized the local environments at Farmana (Figure 8 1) and Harappa (Figure 8 2) Further, it is unnecessarily conservative to adopt the full is otopic range exhibited by fauna as the local estimate Determin ation of the local range using a multi proxy multi isotope approach remains a visually subjective process in that no precise quantitative measure distinguishes between local and non local withi n continuous distributions of baseline data; however, the increased interpretive power justifies the loss of quantitative rigor. Several possible explanations apart fr om migration can explain non local values in first molar cohor t s For example, exposure in childhood to anthropogenic sources of Pb could account for intertooth isotopic differences. This possibility will be discussed below with regard to the Farmana mortuary population. At Harappa, however, it is highly unlikely that anthro pogenic exposure c ould explain sex based differences in the isotope data. First molars of males and females are distinguished by Pb and Sr isotope ratios (Figures 6 1 6 2 6 3, & 6 4 ) whereas males and females may exhibit opposite shifts in 18 O between first and second molars. Each isotope system is governed by different environmental fac tors such that o nly the systematic sex based application of dramatically different water management practices, differential exposure to industrial activity or metal bearing goods, and differential consumption of imported foods could account for the patterning in the data set. Thus anthropogenic sources of non local isotope values can be ruled out.

PAGE 184

184 A nother possible source of non local environmental input in earl y life is the maternal contribution imparted by immigrant mothers during gestation and breast feeding In modern contexts, for example, Gulson et al. (2003) ha ve documented the mobilization of the maternal skel etal Pb burden into breast milk and its subsequent uptake in t he deciduous dentition of children A similar process could happen with Pb and Sr in archaeological popu lation s, potentially altering their respective isotope ratios which would otherwise reflect local maternal residence. Again, however, the sex based structure of the Harappa data set suggests that it was the individuals themselves who moved rather than their mothers. It is highly unlikely that Cemetery R 37 contains only th e sons of mothers from one region and the daughters of mothers from another region. Further, 18 O of enamel formed in utero or during nursing reflects only the isotopic 18 O. Y et 18 nearly opposite shifts (~ This difference is only possible if males and females inhumed at Harappa were actually raise d in hydrologically distinct regions Unfortunately, the lack of reliable sex estimates means that similar interpretive logic cannot be applied at Farmana. Further, the 87 Sr/ 86 Sr (Figure 5 1 ) and 18 O (Figure 5 4 ) data are equivocal placing the burden of proof solely on the Pb isotope ratios. For many individuals, homogenous 87 Sr/ 86 Sr across tooth types is consistent either with local origins or with origins from elsewhere in a region of homogenous 87 Sr/ 86 Sr. The latter possibility is consistent with the published data for the Thar Desert fringe (Chapter 4), although additional analyses of sediment leachates and archaeological fauna are

PAGE 185

185 needed to confirm this trend. 18 O data are similarly inconclusive in that the relative homoge neity anticipated for much of the Ghaggar Yamuna interfluve precludes the 18 O between tooth types, it is impossible to disentangle the effect of weaning on enamel 18 O from the effects of inter annual climatic variation spatial climatic variation, and behavioral variation in water management or water metabolism. Without corroborating evidenc e from Sr and O isotope systems, it is difficult to rule out the possibility that young individuals were systematically exposed in a local setting t o anthropogenic Pb of non local origin. In particular, surma could have been applied to the eyes of childre n as it is today in many parts of the world. identification s of Pb bearin g makeup at Harappa and Mohenjo Daro are correct, then it cannot be ruled out that individuals at Farmana may have been exposed to similar makeup at a young age. As discussed in Chapter 3, the use of surma and kohl is in some cases associated with increase d blood lead concentration and can therefore alter the Pb isotope ratios recorded in tooth enamel. Another possibility is that the first molar cohort recorded the skeletal Pb burden mobilized in the breast milk of immigrant mothers, although the expectatio n is that their breast milk should eventually equilibrate with local Pb sources as a consequence of bone turnover. A third hypothesis is that ant hropogenic Pb exposure indicates non local residence at sites polluted by heavy smelting activity. This possibi lity is discussed in more detail below. Immigrant Origins The first molars at Farmana have less radiogenic heavy isotope ratios than other tooth types and their distribution in Pb Pb space suggests a mixing line between local values and an endmember that is unique among the environmental samples analyzed

PAGE 186

186 for this study (Figure 8 3 ) No direct association can be made with the natural isotopic back ground of a given region, but there is some correlation in Pb isotope ratios between Farmana first molars and sa mples taken from large archaeological slag heaps at Singhana and Ganeshwar, two sites of the Ganeshwara Jodhpura Cultural Complex associated with intensive copper production. The region has long been considered a primary source of copper for I ndus people, and the non Indus inhabitants very likely traded ex t ensively with their closest Indus neighbors at sites like Farmana less than 150km to the north. If immigrants at Farmana came from this region, particulate from intensive smelting could potentially explai n the isotopic similarity between human teeth and copper slags. Alternatively, the low Pb isotope ratios may simply reflect an uncharacterized background source derived from the Aravallis. T he Sr and O isotope data do not directly support an immigrant hypothesis, but neither do they contradict it. The inferred homogeneity in biologically available 87 Sr/ 86 Sr for the Thar Desert fringe and nearby areas is consistent with the small range of variation in the Farmana data set Further, limited analyses of g roundwater on a transect between Delhi and the Aravalli foothills (Bhattacharya et al. 1985) suggest 18 O. Additional environmental analyses are needed to confirm the plausibili ty of relatively large scale homogeneity in Sr and O but migration between copper smelting highland sites and agricultural lowland sites remains a viable explanation for the intertooth differences at Farmana Some individuals at Farmana are more clearly of non local origin. Rather than low Pb isotope ratios and relatively stable 87 Sr/ 86 Sr, t hree individuals represented only by second and third molars have more radiogenic Pb or Sr (Figure 8 1 ). Their

PAGE 187

187 distribution with respect to Sanauli sediment leachates and some of the more peripheral Rakhigarhi fauna suggests a residence influenced more by radiogenic Himalayan alluvium to the north or northeast than the aeolian sands of the Thar Desert to the southwest Lastly, no distinction can be made between individuals exposed to more or less radiogenic environments on the basis of mortuary material culture. The small number of more radiogenic individuals (burials 26, 66, and 67) precludes statistical analysis, but t here is no clear relationship between isotope ratios and variation in ceramics, ornaments, burial type, or burial orientation The radiogenic burials have modest quantities of pottery, appear with and without beads, have multiple burial types, and multiple burial orientations. The isotope data from Harappa more clearly constrain the possible origins of migrants. When compared with the environmental samples in Pb Sr space (Figure 8 4 ) the Harappa data suggest four primary non local regions T hree data poin ts exhibit intermediate values between those of Harappan fauna and the sediment leachates from Farmana and Sanauli which may indicate origins somewhere along the Ghaggar Hakra basin This is supported by a cl ose identification with some of the more periphe ral Rakhigarhi fauna. Two more data points extend in the opposite direction, towards much less radiogenic 206 Pb/ 204 Pb and 87 Sr/ 86 Sr of approximately 0.71 5 No faunal data characterize this region, although the Harappan data points exhibit isotope ratios v ery similar to those of first molar s at Farmana. Incidentally, t his suggests that the anthropogenic Pb exposure posited for Farmana individuals may be confined to a relatively homogenous Sr province and does not result from a widespread cultural practice. A third source region at Harappa is characterized by highly radiogenic 87 Sr/ 86 Sr

PAGE 188

188 and likely has geochemical origins to the north and northeast in the Himalayan foothills. The radiogenic area likely begins at the outer fring es of the Harappan provisioning catchment as indicated by three peripheral faunal data points. The fourth source region and by far the most significant is characterized by 87 Sr/ 86 Sr between 0.712 and 0.716 with Pb isotope ratios very similar to those of f auna at southern sites such as Mehrgarh, Nausharo, and Allahdino (Figure 8 4) This strongly suggests an origin along the main Indus catchment upstream of the Western Fold Belt carbonate influx It is difficult to establish precise geographic parameters ba sed on the current data set, but a conservative estimate spans the middle reaches of the Indus River, beginning upstream of the conflu ence with foreland tributaries and extending well into the Himalayas. Importantly, t his region includes many of the source areas from which the Harappan mineral assemblage was derived (Law 2008) along with sites of the Northern Neolithic and la te occurring Kot Diji cultures. Figure 8 5 adds further support to a western and northwestern provenience for the fourth isotopic reg ion. Two mixing lines are suggested between a single less radiogenic end member with high Sr concentration and two low concentration end members with relatively high 87 Sr/ 86 Sr. Farmana, Sanauli, and most Harappa data points above 0.716 plot along a single mixing line that is potentially defined at one end by radiogenic c ontributions to the Indus foreland from the HHC and Lesser Himalaya The other end is characterized by low 87 Sr/ 86 Sr aeolian carbonates derived ultimately from the Western Fold Belt. The less radiogenic ratios from Harappa (<0.716) plot along a separate mixing line presumably between the Western Fold Belt carbonates and a different radiogenic source including the Karakoram drainages. Faunal data from

PAGE 189

189 Harappa and Rakhigarhi do not appear entirely comparable to human values, but their distribution is broadly consistent with the easternmost aeolian mixing system Thus Harappan immigrants came primarily from regio ns in or adjacent to the northwestern highlands A smaller influx of migrants originated from various places in a broad swath of territory stretching east and south from the northern foothills into the Ghaggar Hakra basin and Thar Desert fringe. Interestin gly, males came almost exclusively from the western and northwestern area whereas females originated in each of the proposed source regions. Additionally, isotopic distinction between the sexes is consistent with morphometric studies of population affinit y at Harappa that suggest genetic distance between males and females (Hemphill et al. 1991; Kennedy 2000). T he available archaeological context is insufficient to assess whether or not there is a correlation between isotopic orig in and sex although Ke noye r (2011a) has suggested that males typically had less elaborate graves. This could potentially be related to their origins outside of the Indus heartland but no conclusive statement can be made using the available evidence Mode of Migration The timing of migration events in the Farman a and Harappa mortuary populations is less clearly constrained than the geographic provenience. If the low Pb isotope ratios exhibited by first molars from Farmana are attributed to non local exposure then the data set strongly indicates that migration events bega n in early childhood The sequential distribution from first to second and then tightly clustered third molars is best explained by a c hange in Pb exposure early in the mineralization phase of second molars around three or four years of age (Figure 8 6A) Note that most individuals had converged on the local range at some point prior to the mineralization of

PAGE 190

190 their third molars which suggests local residence a nd normalized adult Pb exposur e by the age of seven or eight (Figure 8 6B) Intertooth sequences in Pb Sr space (Figure 8 6 A ) show that four individuals originate in more radiogenic terrain, and at least two of them had not yet reached Farmana at the point of third molar mineralization. However, intertooth sequences also suggest that at least two of the more radiogenic individuals had changed residence one or more times during childhood. The other two radiogenic individuals are each represented by only a si ngle data point Nine individuals, eight of whom have multiple data points, were exposed to less radiogenic Pb exclusively in early childhood (Figure 8 6B) Again assuming that shifts in Pb isotope ratios reflect residence change, the preponderance of data suggests that all individuals first changed residence at an early age. Most significantly, no individual displays low Pb isotope ratios into adulthood, suggesting that only children migrated from less radiogenic environments. Such young migrants would pre sumably need to be cared for by a local family or kingroup as there is no evidence in the isotopic sample for older individuals having made the same journey. The opportunistic nature of the Harappa sample precludes a thorough analysis of developmental sequences. Only two such sequences can be confidently determined based on the available context, and both are consistent with an early child hood migration event (Figure 8 7 ). The s equences indicated on Figure 8 7 are of a first a nd third molar and an incisor and canine, and in both cases, the later mineralizing tooth plot s closer to the local range at Harappa. The intertooth isotopic shift occurs for ratios of Sr and Pb, confirming that the sequence depicts actual residence change rather than

PAGE 191

191 anthropogenic Pb exposure. Additional sampling is required to determine whether or not the entire mortuary population was mobile in early life. However, the oxygen isotope composition of female first and second molars provides one additional c lue. The shift 18 O in second molars (discussed above) runs counter to the typical 18 O and is therefore consistent with an early life migration event. Small sample sizes and a lack of developmental sequences li mit the interpretive power of individual isotopic measures at Farmana and Harappa, but taken collectively, t hey are highly suggestive of a pattern of early life migration from adjacent highlands associated in many instances with non Indus cultures. Locali zation Era Mobility The structure of the Sanauli data set sugg ests significant changes in the Indus mortuary program during the post urban era. Unlike the all immigrant mortuary profile proposed for Farmana and Harappa, Sanauli includes individuals born locally. Sediment leachates plot next to a large cluster of data points in Pb Sr space, suggesting that many individuals lived near the mortuary site (Figure 8 8 ). The local range cannot be defined with precision without additional environmental samples, but the central hub of the radial distribution of Sanauli data points l ikely coincides with local isotope ratios. Depending on how conservatively one estimates the local range, between 11 and 23 individuals in the isotopic sample of 33 individuals spent all or part of their early lives outside the immediate vicinity of the site. The potential source regions can be categorized into three broad groups. Two individuals come from a region with l ow 87 Sr/ 86 Sr (<0.718) and high 206 Pb/ 204 Pb ( >19.2) that plots near the most radiogenic Farmana data points. This distribution might represent parts of the Ghaggar Yamuna interfluve to the north and east of the Tha r Desert fringe. E ight individuals have high

PAGE 192

192 87 Sr/ 86 Sr (>0.72) and a relatively wide ra nge of 206 Pb/ 204 Pb (>18.9), which suggests increased proximity to the relatively radiogenic terrain of the Himalayan foothills to the north or perhaps the Gangetic River to the east and northeast. Another eight individuals have intermediate 87 Sr/ 86 Sr (0.71 8 0.720) and low 206 Pb/ 204 Pb (<19.1). Those individuals with the least radiogenic Pb isotope ratios plot close to some of the peripheral Harappan fauna, suggesting possible origins to the west and northwest in the upper reaches of the foreland basin. Overa ll, Sanauli individuals show considerable intertooth variability suggesting that migration in childhood was not uncommon (Figure 8 9 ) T here is however, no specific trend in the timing of migration. Some individuals show the most change between first and second molars, some between second and third molars, and some show relatively little change across all three tooth types. Further, first, second, and thi rd molars often do not plot along a linear trajectory. It is unclear whether this indicates multiple changes of residence or simply the range of isotopic variation for a given region. The relatively wide scatter of Sanauli data and lack of structure in the data set is mos t consistent with a typical breeding population. The presence of non local individuals might be attributable to exogamy, although any specific expla nations must remain speculative No further clues are found in the mortuary material culture as n o obvious correlations with the isotope data can be detected Sex identifications are too few to ascertain any patterning, and each potential source region includes individuals of all ages, ceramics of every quantity, and grave goods of all kinds. Ev en b urials with more than one individual exhibit a range of isotopic values suggesting the deceased

PAGE 193

193 were not united by common geographic origin. Unfortunately, the most unusual burial type, symbolic burials, lacked appropriate skeletal material for isotopi c analysis In summary the Sanauli mortuary program was insensitive to or only indirectly concerne d with geographic origin. The frequency of migration is still relatively high, and therefore the kind of person who received inhumation may have tended to be more mobile, but the proposed urban era correlation between migration and inhumation had changed. Further, the migration patterns at Sanauli suggest that residence change served as an internally integrative mechanism, connecting Sanauli to other Indus sit es rather than settlements in the highlands Dietary Variability The carbon isotope data do not show significant shifts between tooth types or any sex based differences at the sampled sites ( Figures 5 5 6 5 & 7 5 )) Likewise, they do not appear to be str uctured by mobility Instead, they are suggestive of agricultural adaptations to regional climatic conditions. Harappa and Sanauli both 13 C of 11.9 ereas the mean value at Farmana is 10.0 These data contradict a simplified narrative of increasing millet consumption over time, and instead confirm that agricultural practices varied at the regional scale. Very few millets or other C 4 crops were consumed by individuals in the mortuary populations of Sanaul i and Harappa, but millets contribut ed significantly to diet at Farmana. Their hardy nature and relatively low labor requirements means they could ha ve been used to minimize the agricultural risks associated with semi arid conditions in the Thar Desert fringe. 13 C variation provi des additional insight into the mode of millet consumption at Farmana. Whereas i for a given developmental sequence, 13 C at for the same individual (Tables 5 3 & 7 2 )

PAGE 194

194 At Farm ana then, the decision to eat millets varied from year to year, probably as a function of interannual climatic fluctuations. Overall, the data support the notion put forward by Weber and colleagues (2010) that Indus agriculture is best understood in terms of ecological adaptations at the regional and sub regional scales. Towards an Integrated Model Though additional research is needed to fully support the interpretations of migration made in the preceding sections it is worth exploring the potential impli cations for broader Indus society. A model is needed to explain the presence of immigrant cemeteries in order to go beyond a simple presentation of data and provide a framework for future investigation. T his section builds on the propos ition that all indiv iduals sampled at Harappa and Farmana were immigrants separated from t heir natal groups in the non Indus highlands early in childhood. It is further proposed that during the urban era, the uncommon and relatively standardized practice of cemetery inhumatio n was part of a social institution revolving around particular types of immigrant s To reiterate the point made above, non Indus origins cannot be unequivocally determined at this point. Nevertheless, the model outlined here illustrates the untapped potent ial for Indus provenience studies and serves as an important example of the analytical and inferential process. Further, it may be validated by future research as it fits well with the non isotopic evidence. For example, published studies focused on the physical anthropology at Harappa and Lothal ( Possehl and Kennedy 1979; Kennedy et al. 1984; Hemphill et al.1991; Kennedy 2000 ) suggest genetic discontinuities that can be explained by systematic migration. The long term maintenance of genetic distance bet ween the sexes at Cemetery R 37 and Cemetery H may be better explained by an i nflux of immigrants

PAGE 195

195 rather than ongoing marital exchanges between local and non local groups. Further, skeletal s imil arities between the Lothal assemblage and contemporaneous h un ter gatherers in the southern Aravallis have been interpreted as the result of gene flow. It may be, however, that the Lothal skeletons belonged to immigrants from hunter gatherer populations rather than their locally born offspring. Th e all immigrant hypo thesis also fits well with conventional archaeological evidence. Given the scarcity of Indus in humations, it is difficult to model the mortuary populations as cross sections of interconnected breeding populations. There simply are too few inhumations in to o few cemeteries for the collective urban era mortuary population to represent an ethnic group. Yet close similarities between inhumations from different cemeteries suggest strong ideological controls and common membership in an Indus group or institution. Additionally, the exclusive mortuary treatment carried out despite years of local residence implies a profound social distinction lasting even into death Th is distinct social identity could be interpreted as membership within a peripheral class of first generation immigrants attached to or otherwise affiliated with local groups. The affiliate approach to Indus immigration is poorly explained by conventional relations of marriage and consanguinity. For a given kinship system to fit the isotope data, it must result exclusively in a mortuary population of individuals that spent their early childhood in at least one non local location. Further, it must account for gender based differences in the isotopi c data set Both of these results can be achieved given several assumptions. First, t he mother of the deceased would have had to marry into the local community from non local locations. On becoming pregnant she would have

PAGE 196

196 had to return to her paternal gro up for childbirth and nursing, not leaving for several years Eventually a mothers and her offspring woul d resume living with the husband s group but the mother would return on e last time to her non local paternal group for burial. Locally born female of fspring must follow a similar pattern by marrying outside of the local group but returning home to the father s community for childrearing and eventual burial. Male offspring must also marry outside of the local population, but not to the same group a s the ir sisters. Further, they would be buried at their adult residence. I f this residence pattern was maintained between three or more communities, any given mortuary population would contain only males and females with different non local isotopic signatures in their early mineralizing dentition. The hypothetical marital residence patterns (represented diagrammatically in Figure 8 10 ) result in distinct patterns of residence change that can be tested against the isotopic data. Following the three population s ystem in Figure 8 10 A for local mortuary population B, male dentition will reflect the isotope values at location C in infancy, location A in childhood, and finally location B (the adult residence) if they move d to their affinal residence before their thir d molars had finished mineralizing. Female dentition would reflect the sequence A, B, and potentially C depending on when they move d to their affinal residence. This scenario suggests that male second molars should coincide with female first molars, a pattern not supported by the heavy iso tope data at Harappa (Figure 8 11 ). A residence system composed of four or more popul ations as in Figure 8 10B would result in the same migration sequence for females (A, B, and possibly C) but a different pattern for males. Note that males and females in Figure 8 10B never live in the same locations in early life, resulting in a male

PAGE 197

197 isotopic sequen ce of D, E, and potentially B. Again, this pattern is not apparent in the Harappa data set (Figure 8 11 ) as males appear only to move between one non local region and the local Harappan catchment. Further analyses will permit a more refined test of these scenarios, but two additional complications suggest they are not applicable to the present data set First, affines all without having reached the stage in life where they woul d have moved there in their own right. Further, decades of excavation in South Asia have yet to reveal cemeteries that could plausibly represent the complementary kin groups indicated in these mate exchange systems. Isotope data aside too many special arg uments ar e required to make a system of marital residence fit the mortuary profile at Harappa. An institution of fictive or created kinship, however, is consistent with the isotopic data set. It can explain a mortuary population composed entirely of indiv iduals who moved as children into lowland settlements from their natal groups in the Indus hinterland. It provides an ethnographically realistic context by which genetically unrelated children from non Indus cultures can be fostered and raised under safe a nd healthy conditions within the Indus social system. And lastly, it explains how despite years of participation within Indus society, those individuals could be perceived as socially distinct persons even in death as evinced by their inclusion within an extraordinarily uncommon mortuary program. An Ethnographic Example Fosterage, often literally or figuratively tied to the act of wet nursing, has been widely used as a strategic means of building alliances in many parts of the world including the Middle Ea st and Eurasia (Altorki 1980; Goody 1982; Khatib Chahidi 1992)

PAGE 198

198 The practice was used among t he historic kingdoms of the Hindu Kush an area that was once the source for many of the raw materials imported to Harappa in Indus times. The rights and obligations generated through fosterage helped create the social foundation necessary for hierarchical political alliances between fractious peripheral polities (Gwynn 1913; Parkes 2001) Parkes (2001) extensively reviewed the late 19th century practice of fosterage in the Hindu Kush region in which fosterage most often involved the transfer of elite chi ldren to junior households. Foster families owed intermarry. Should the elite child rise to power, foster families were compensated with land and positions of authority. There was variability in the practice, however, and Parkes (2001: 20 21) suggested that successive iterations of milk kinship could create relations of dependency between previously intermarrying descent groups. Eventually, the prohibitions against intermarriag e would completely separate the groups such that the initially junior descent group could become a dependent status grade. People of the subordinate grade might then be bonded and transferable, potentially reversing the physical flow of bodies observed in prior relations of milk kinship. It is unreasonable to suppose that the many means of creating kinship recorded historically in South Asia and elsewhere ever existed in identical form within the Indus Civilization. Projecting ethnography unchanged into the past has long been rejected as a valid approach for understanding archaeological peoples. However, a historical understanding of the frequency and diversity of ethnographic practices highlights common themes that could readily have applied to protohistori c South Asia. The ethnographic cases defy a simplistic understanding of the archaeological past, and the

PAGE 199

199 often implicit assumptions that people in the past were immobile or that archaeological cemeteries always contain breeding populations must be abandone d (Cobb 2005; Jackes 2011) Adding to the complexity is the established premise that archaeological people, just like modern ones, are capable of appropriating and using non local mater ial culture. For this reason, no one to one correlation can be assumed between pottery types and ethnic groups (Jones 1997:122 124) In short, the notion that past individuals who participated in a vast, technologically sophisticated, and incredibly complex urban society were any less capable of transferring children between ethnic groups as a means of constructing kinship is wholly invalid. Shedding such implicit assumptions about the relationships betwee n past mobility, mortuary practices, and material artifacts opens the door to a far richer understanding of Indus inhumations. The Indus Model In the pre market economy of protohistoric South Asia, highland and lowland groups may have turned to the rights and obligations of kinship as a way of mediating economic exchange. E lite Indus groups likely sought economic advantages over each other, and they had strong incentives to secure preferential access to exotic raw materials. Likewise, highland groups must have desired access to the trade routes and sophisticated crafts of Indus merchants This kind of economic interdependence came to define the way of life among Ganeshwar Jodhpura peoples in the northern Aravallis such that settlements were structured arou nd the production of copper for trade with lowland communities (Agrawala 1984b; Rizvi 2007}}{{47 Agrawala,R.C. 1984 a; Rizvi 2007) A relationship structured by fosterage could have e stablished the framework within which goods were exchanged, although it need not have been a relationship of equality.

PAGE 200

200 Just as kin groups in the historic Hindu Kush kingdoms established asymmetric alliances through fosterage, elite Indus groups may have d esired a form of kinship that prohibited marriage and consequently prevented inheritance claims to th eir accumulated wealth. Marriage creates a shared identity and presupposes relative equality between groups that once were socially distinct, but a fostere d individual generates a different kind of bond. The fostered individual is a constant reminder of the obligations between two groups precisely because the individual is a living reference to those qualities, those things, and those ways of living that mak e two groups distinct. into a social hybrid that embodies qualities of both groups but can never fully become a member of either In a way, they become a living contract, but the contract lasts only as long as t he individual. On death, the tension of hybridity is resolved, dissolving obligations and requiring a new act of fosterage. Offsp ring of the fostered individual cannot restore the tension of mutual obligation because t hey are not truly of both cultural worlds They are one or the other, but they are not both; and when dead, they are treated as such an d granted the corresponding mortuary treatment. I n the Indus region, o nly those who were fostered, the social hybrids, received inhumation in a cemetery. With their usefulness gone and h aving never acquired the full social identity of their Indus hosts, they were disposed of in a unique way that reflected their liminal status. Yet the need for a social hybrid remained, and for hund reds of years, new individuals were fostered to re establish the identities.

PAGE 201

201 The asymmetric relationships of fosterage may have influenced broader c ultural perceptions muc h like modern cultural stereotype s of immigrants and encouraged members of both groups to perceive the other as socially distinct. This could have reinforced social boundaries, resulting in relatively independent cultural trajectories for the highland gro ups linked in this way. In fact, this scenario is consistent with the limited archaeological evidence. Cultures in the highland regions connected by fosterage (e.g., late occurring Kot Diji, Northern Neolithic, Ganeshwar Jodhpura ) seemingly adopted fewer I ndus traits over time than other highland cultures such as the Kulli (Chapter 2). Trade and migration in archaeology are often assumed to be integrative, but a close consideration of the structures of these phenomena suggests ways to explain alternate outc omes that move closer to the nuanced anthropological understandings of modern migration.

PAGE 202

202 Figure 8 1. Local isotopic range at Farmana Pb Sr space Figure 8 2. Local isotopic range at Harappa Pb Sr space

PAGE 203

203 Figure 8 3. Mixing line at Farmana with non local anthropogenic Pb [slag data from Law, R. W. 2008. Inter Regional Interaction and Urbanism in the Ancient Assemblage. PhD dissertation (Page 744, Appendix 12.8). University o f Wisconsin Madison] Figure 8 4. Four source regions at Harappa

PAGE 204

204 Fi gure 8 5. Two mixing lines in the Greater Indus region suggesting distinct geological inputs Sr Sr space

PAGE 205

205 Figure 8 6. Intertooth developmental sequences at Farmana in Pb Sr space showing childhood migration to the local area A) Entire Farmana dataset with inset of Figure 8 6B. B) Closeup of Farmana intertooth variation.

PAGE 206

206 Figure 8 7. Intertooth developmental sequences at Harappa in Pb Sr space showing childhood migration to lo cal or near local regions Figure 8 8. Two local range estimates and source regions at Sanauli Pb Sr space

PAGE 207

207 Figure 8 9. Intertooth developmental sequences at Sanauli in Pb Sr space showing variation in the timing and direction of migration

PAGE 208

208 Figure 8 10. Theoretical marital residence patterns that might produce entirely immigrant mortuary populations A) Marital exchange system with three groups B) Marital exchange system with four or more groups

PAGE 209

209 Figure 8 11. Early childhood residence and migration by sex at Harappa in Pb Sr space characterized by non overlapping male and female distributions and male residence at a single non local locality

PAGE 210

210 CHAPTER 9 CONCLUSION If confirmed by future work, the model presented above has significant consequences for the study of the Indus Civilization and, more broadly, the study of mortuary populations and migration This research challenges the implicit assumption in many bioarchaeo logical studies that cemetery remains are representative of local population s Instead, the social identities of the deceased must be assessed before inferences can be made about a particular group This further requires that bioarchaeological studies go b eyond the application of analytical methods and are instead explicitly informed b y models of social organization. In particular, the possibility must be considered that a given mortuary population has little genetic affinity with or only indirect associati on with local archaeological cultures. Though such biological and social distance may be uncommon, it must nevertheless be ruled out before claims can be made about the health, lifestyle, or demography of local cultural groups. Further, this research emph asizes a more versatile understanding of the social consequences of migration than is typically employed in archaeological studies. Migration is often implicitly understood to connect groups and individuals in a way that leads to cultural homogenization. A fter all, an intuitive interpretation of contexts like those at Harappa and Farmana in which immigrants are associated with local material culture is that non locals were culturally assimilated. By extension, it might be assumed that the sending and receiv ing societies were blending together or otherwise hybridizing. T he model presented above however, highlights the importance of the social structure of migration and the specific kinds of relationships t hat were indicated by immigrant bodies, both living a nd dead. Although different groups may be united in

PAGE 211

211 certain ways by the flow of bodies between sending and receiving societies, the context of that movement could have reinforce d perceptions of alterity. Migration can potentially lead to more diverse outco mes than suggested by the implicit diffusionism still present in the archaeology of migration, and students of migration must adopt theoretical frameworks that accommodate a wider range of cultural outcomes. Additionally, the specific model presented above for Integration Era mortuary populations must be tested against future research. Further work is needed to better understand the nature of isotopic variation in the Greater Indus region and beyond so that immigrants can be more accurately identified and p rovenienced. Ultimately, many more baseline samples including sediment leachates and tooth enamel from archaeological fauna must be analyzed from a wide range of sites and taxa. The creation of a regional isotopic database will benefit not on l y Indus Civil ization research, but any isotopic inquiry into past human behavior and ecology within the region. Most immediately, the faunal baseline samples from this project must be analyzed for Sr concentration to test the above inference that immigrants to Harappa came from one of two broad isotopic regions: either the middle and upper reaches of the Indus River Valley or the remainder of the western foreland basin. Future sampling must also emphasize comparability, such that full regional datasets are developed for each sample type (e.g., sediment leachates, Sus tooth enamel). Hypothesized source regions such as the northern Aravallis and northwestern Himalayas are especially vital if the all immigrant model is to be tested In addition to isotopic baseline analyses, the Harappa mortuary population must be systematically analyzed for intertooth variation within individuals. If it is determined

PAGE 212

212 that all individuals migrated early in life as suggested by the Farmana dataset and limited developmental sequences at Harappa then it lends strong support to the premise that mortuary populations were composed of dependent individuals rather than independent groups. Similar analyses need to be conducted for other Integration Era mortuar y populations. The Kalibangan cemetery is particularly promising given recent osteological work on the mortuary population by Robbins Schug (personal communication) More details on mortuary variation are required to validate the primary interpretations of mortuary material culture discussed above that urban cemetery mortuary populations are not strongly internally differentiated and are representative of a specific sub population Forth coming comprehensive reports on excavations of cemetery R 37 (Kenoyer p ersonal communication) will provide the information needed for a more nuanced assessment of Harappan mortuary variability. Another possibility that must be considered is that the means by which the living invested in mortuary ritual are poorly preserved. F easting or some other form of communal commemoration might have been observed, although there is little corroborative evidence in the relatively modest mortuary assemblages. R esidue analysis of mortuary ceramics, perhaps, could assess the nature of consump tion associated with mortuary events. Further, additional mortuary details might help reconstruct the post depositional treatment of remains in secondary contexts. More deliberate ritually oriented treatment of human remains might suggest a lasting social role for the deceased potentially leading to new inferences about the social role of the inhumed in life

PAGE 213

213 Though the conclusions reached by this research require further validation, they have yielded a model that will guide future hypothesis testing. If n ew evidence contradicts the specific interpretations presented above, the degree of isotopic variability in the datasets nevertheless demands novel social approaches to interregional interaction that emphasize the role of diverse individuals, groups, and i nstitutions in structuring the Indus phenomenon. Though valuable in their own right, studies limited to the provenience of trade goods and human remains fall short of understanding the ways in which interregional interaction helped shape the Indus Civiliza of study. Instead, f uture work on the Indus Civilization must strive with renewed vigor towards a theory of communities and generate new testable models of the relationship s between groups in the highly diverse and complex social landscape of the Indus Tradition

PAGE 214

214 LIST OF REFERENCES Adams, William Y., Dennis P. Van Gerven, and Richard S. Levy 1978 The Retreat from Migrationism. Annual Review of Anthropology 7:483 532. Agrawala, R. C. 1984 Aravalli, the Major Source of Copper for the Indus Civilization and Indus Related Cultures. In Frontiers of the Indus Civilization. B. B. Lal and S. P. Gupta, eds. Pp. 157 162. New Delhi: Indian Archaeological Society. 1984 Ganesh war Culture A Review. Journal of the Oriental Institute, Baroda 34(1 2):89 95. Agrawala, R. C., and V. Kumar 1982 Ganeshwar Jodhpura Culture: New Traits in Indian Archaeology. In Harappan Civilization. G. L. Possehl, ed. Pp. 125 134. New Delhi: Oxford & IBH Publishing Company. Ahmad, Z. 1969 Directory of Mineral Deposits of Pakistan. Records of the Geological Survey Pakistan 15(3):1 220. Ajithprasad, P. 2002 The Pre Harappan Cultures of Gujarat. In Indian Archaeology in Retrospect, Volume 2 (Proto history: Archaeology of the Harappan Civilization). S. Settar and R. Korisettar, eds. Pp. 129 158. New Delhi: Indian Council of Historical Research. Akkermans, Peter M. M. G., and Glenn M. Schwartz 2003 The Archaeology of Syria: From Complex Hunter Gath erers to Early Urban Societies (Ca. 16,000 300 BC). Cambridge: Cambridge University Press. Alizai, Anwar, Andrew Carter, Peter D. Clift, Sam VanLaningham, Jeremy C. Williams, and Ravindra Kumar 2011 Sediment Provenance, Reworking and Transport Processes in the Indus River by U Pb Dating of Detrital Zircon Grains. Global and Planetary Change 76(1 2):33 55. Allchin, F. R. 1984 The Northern Limits of the Harappan Culture Zone. In Frontiers of the Indus Civilization. B. B. Lal and S. P. Gupta, eds. Pp. 51 54. New Delhi: Books and Books. Allchin, R., and B. Allchin 1982 The Rise of Civilization in India and Pakistan. 3rd ed. Cambridge: Cambridge University Press.

PAGE 215

215 Altorki, Soraya 1980 Milk Kinship in Arab Society: An Unexplored Problem in the Ethnography of Marriage. Ethnology 19(2):233 244. Ambrose, Stanley H., Jane Buikstra, and Harold W. Krueger 2003 Status and Gender Differences in Diet at Mound 72, Cahokia, Revealed by Isotopic Analysis of Bone. Journal of Anthropological Archaeology 22 (3):217 226. Ambrose, Stanley H., and John Krigbaum 2003 Bone Chemistry and Bioarchaeology. Journal of Anthropological Archaeology 22(3):193 199. Ambrose, Stanley H., and L. Norr 1993 Experimental Evidence for the Relationship of the Carbon Isotope R atios of Whole Diet and Dietary Protein to those of Bone Collagen and Carbonate. In Prehistoric Human Bone:Archaeology at the Molecular Level. J. B. Lambert and G. Grupe, eds. Pp. 1 37. New York: Springer Verlag. Ammerman, A. J., and L. L. Cavalli Sforza 1971 Measuring the Rate of Spread of Early Farming in Europe. Man 6(4):674 688. Anthony, David W. 1990 Migration in Archeology: The Baby and the Bathwater. American Anthropologist 92(4):895 914. 1992 The Bath Refilled: Migration in Archeology Again American Anthropologist 94(1):174 176. 1997 Prehistoric Migration as Social Process. In Migration and Invasion in Archaeological Explanation. John Chapman and H. Hamerow, eds. Pp. 21 32. Oxford: Archaeopress. Arnold, Bettina 2002 A Landscape of Ancestors: The Space and Place of Death in Iron Age West Central Europe. Archeological Papers of the American Anthropological Association 11(1):129 143. Asthana, Shashi 1993 Harappan Trade in Metals and Minerals: A Regional Approach. In Harappan Civiliz ation. G. L. Possehl, ed. Pp. 271 285. Delhi: American Institute of Indian Studies. Balasse, Marie 2002 Reconstructing Dietary and Environmental History from Enamel Isotopic Analysis: Time Resolution of Intra Tooth Sequential Sampling. International Jou rnal of Osteoarchaeology 12(3):155 165.

PAGE 216

216 Barth, Fredrik 1969 Introduction. In Ethnic Groups and Boundaries. Fredrik Barth, ed. Pp. 9 38. Boston: Little Brown. Bataille, Clment P., and Gabriel J. Bowen 2012 Mapping 87 Sr/ 86 Sr Variations in Bedrock and W ater for Large Scale Provenance Studies. Chemical Geology 304 305:39 52. Bazin, D., and H. Hubner 1969 Copper Deposits in Iran. Geological Survey of Iran 13:1 195. Beard, B. L., and C. M. Johnson 2000 Strontium Isotope Composition of Skeletal Materia l can Determine the Birth Place and Geographic Mobility of Humans and Animals. Journal of Forensic Sciences 45(5):1049 1061. Beekman, Christopher S., and Alexander F. Christensen 2003 Controlling for Doubt and Uncertainty through Multiple Lines of Evidence: A New Look at the Mesoamerican Nahua Migrations. Journal of Archaeological Method and Theory 10(2):111 164. Belcher, William R. 1991 Fish Resources in an Early Urban Context at Harappa. In Harappa Excavations 1986 1990. Richard H. Meadow, ed. Pp. 107 120. Madison: Prehistory Press. 2003 Fish Exploitation of the Indus Valley Tradition. In Indus Ethnobiology: New Perspectives from the Field. Steven A. Weber and William R. Belcher, eds. Pp. 95 174. Lanham: Lexington Books. Bellwood, Peter 2004 The Origins and Dispersals of Agricultural Communities in Southeast Asia. In Southeast Asia: From Prehistory to History. Ian Glover and Peter Bellwood, eds. Pp. 21 40. London: Routledge Curzon. Bentley, R. Alexander 2006 Strontium Isotopes from the Earth to the Archaeological Skeleton: A Review. Journal of Archaeological Method and Theory 13(3):135 187. Bentley, R. A lexander ., and C. Knipper 2005 Geographical Patterns in Biologically Available Strontium, Carbon and Oxygen Isotope Signatures in Pr ehistoric SW Germany*. Archaeometry 47(3):629 644. Bentley, R. Alexander, Hallie R. Buckley, Matthew Spriggs, Stuart Bedford, Chris J. Ottley, Geoff M. Nowell, Colin G. Macpherson, and D. Graham Pearson 2007 Lapita Migrants in the Pacific's Oldest Cemet ery: Isotopic Analysis at Teouma, Vanuatu. American Antiquity 72(4):645 656.

PAGE 217

217 Bhan, K. K., and Jonathan Mark Kenoyer 1983 Nageswar, an Industrial Centre of Chalcolithic Period. Puratattva 12:115 119. Bhan, Kuldeep K. 1986 Recent Explorations in the Jam nagar District of Saurashtra. Man and Environment 10:1 21. Bhattacharya, S. K., S. K. Gupta, and R. V. Krishnamurthy 1985 Oxygen and Hydrogen Isotopic Ratios in Groundwaters and River Waters from India. Proceedings of the Indian Academy of Sciences Eart h and Planetary Sciences 94(3):283 295. Bindler, Richard 2011 Contaminated Lead Environments of Man: Reviewing the Lead Isotopic Evidence in Sediments, Peat, and Soils for the Temporal and Spatial Patterns of Atmospheric Lead Pollution in Sweden. Environmental Geochemistry and Health 33(4):311 329. Binford, Lewis R. 1971 Mortuary Practices: Their Study and their Potential. Memoirs of the Soci ety for American Archaeology 25 :6 29. Bisht, R. S. 1989 A New Model of the Harappan Town Planning as Revealed at Dholavira in Kutch: A Surface Study of its Plan and Architecture. In History and Archaeology. B. Chatterjee, ed. Pp. 397 408. Delhi: Ramanand Vidhya Bhawan. 1991 Dholavira: A New Horizon of t he Indus Civilization. Puratattva 20:71 82. 1997 Dholavira Excavations: 1990 94. In Facets of Indian Civilization: Recent Perspectives (Essays in Honour of Prof. B. B. Lal). Jagat Pati Joshi, ed. Pp. 107 120. New Delhi: Aryan Books International. Bla ckman, M. James, and Massimo Vidale 1992 The Production and Distribution of Stoneware Bangles at Mohenjo Daro and Harappa as Monitored by Chemical Characterization Studies. In South Asian Archaeology 1989. C. Jarrige, ed. Pp. 37 44. Madison: Prehistory P ress. Blakely, Robert L. 1989 Bone Strontium in Pregnant and Lactating Females from Archaeological Samples. American Journal of Physical Anthropology 80(2):173 185. Bowen, Gabriel J., and Bruce Wilkinson 18 O in Meteoric Precipitation. Geology 30(4):315 318.

PAGE 218

218 Brookman, Tom H., and Stanley H. Ambrose 2013 Kangaroo Tooth Enamel Oxygen and Carbon Isotope Variation on a Latitudinal Transect in Southern Australia: Implications for Palaeoenvironmental Reconstruction. Oecologi a 171(2):403 416. Brown, James A., ed. 1971 Approaches to the Social Dimensions of Mortuary Practices. Vol. 25. Washington: Society for American Archaeology. Budd, P., J. Montgomery, A. Cox, P. Krause, B. Barreiro, and R. G. Thomas 1998 The Distribut ion of Lead within Ancient and Modern Human Teeth: Implications for Long Term and Historical Exposure Monitoring. Science of the Total Environment 220(2 3):121 136. Budd, P., J. Montgomery, J. Evans, and B. Barreiro 2000 Human Tooth Enamel as a Record o f the Comparative Lead Exposure of Prehistoric and Modern People. Science of the Total Environment 263(1 3):1 10. Budd, P ., J. Montgomery, B. Barreiro, and R. G. Thomas 2000 Differential Diagenesis of Strontium in Archaeological Human Dental Tissues. Applied Geochemistry 15(5):687 694. Buikstra, Jane E., and Douglas H. Ubelaker, eds. 1994 Standards for Data Collection from Human Skeletal Remains: Proceedings of a Seminar at the Field Museum of Natural History. Fayetteville: Arkansas Archaeological Survey. Bunton, C. A., D. R. Llewellyn, C. A. Vernon, and V. A. Welch 1961 The Reactions of Organic Phosphates. Part IV. Oxygen Exchange between Water and Orthophosphoric Acid. Journal of the Chemical Society (Resumed):1636 1640. Burmeister, Stefan 2 000 Archaeology and Migration: Approaches to an Archaeological Proof of Migration. Current Anthropology 41(4):539 567. Burton, James H., T. Douglas Price, and William D. Middleton 1999 Correlation of Bone Ba/Ca and Sr/Ca due to Biological Purification of Calcium. Journal of Archaeological Science 26(6):609 616. Burton, James H., and Lori E. Wright 1995 Nonlinearity in the Relationship between Bone Sr/Ca and Diet: Paleodietary Implications. American Journal of Physical Anthropology 96(3):273 282.

PAGE 219

219 C annon, Aubrey 1989 The Historical Dimension in Mortuary Expressions of Status and Sentiment. Current Anthropology 30(4):437 458. Capo, Rosemary C., Brian W. Stewart, and Oliver A. Chadwick 1998 Strontium Isotopes as Tracers of Ecosystem Processes: The ory and Methods. Geoderma 82(1 3):197 225. Cerling, Thure E., and John M. Harris 1999 Carbon Isotope Fractionation between Diet and Bioapatite in Ungulate Mammals and Implications for Ecological and Paleoecological Studies. Oecologia 120(3):347 363. Ch akrabarti, Dilip K. 1990 The External Trade of the Indus Civilization. New Delhi: Munshiram Manoharlal Publishers. Chakrabarti, Dilip K., and Nayanjot Lahir 1996 Copper and its Alloys in Ancient India. New Delhi: Munshiram Manoharlal. Chakrapani, G. J. 2005 Major and Trace Element Geochemistry in Upper Ganga River in the Himalayas, India. Environmental Geology 48(2):189 201. Chamyal, L. S., D. M. Maurya, and Rachna Raj 2003 Fluvial Systems of the Drylands of Western India: A Synthesis of Late Quaternary Environmental and Tectonic Changes. Quaternary International 104(1):69 86. Chapman, Robert P Tombs in Prehistoric Europe. In The Archaeology of Death. Robert Chapman, I an Kinnes and Klavs Randsborg, eds. Pp. 71 81. Cambridge: Cambridge University Press. Chapman, John, and Pavel M. Dolukhanov 1992 The Baby and the Bathwater: Pulling the Plug on Migrations. American Anthropologist 94(1):169 174. Charles, Douglas K., and Jane E. Buikstra 2002 Siting, Sighting, and Citing the Dead. Archeological Papers of the American Anthropological Association 11(1):13 25.

PAGE 220

220 Chenery, Carolyn A., Vanessa Pashley, Angela L. Lamb, Hilary J. Sloane, and Jane A. Evans 2012 The Oxygen Isotope Relationship between the Phosphate and Structural Carbonate Fractions of Human Bioapatite. Rapid Communications in Mass Spectrometry 26(3):309 319. Chenery, Carolyn, Gundula Mldner, Jane Evans, Hella Eckardt, and Mary Lewis 2010 Strontium and S table Isotope Evidence for Diet and Mobility in Roman Gloucester, UK. Journal of Archaeological Science 37(1):150 163. Chiaradia, M., A. Gallay, and W. Todt 2003 Different Contamination Styles of Prehistoric Human Teeth at a Swiss Necropolis (Sion, Vala is) Inferred from Lead and Strontium Isotopes. Applied Geochemistry 18(3):353 370. Christaller, W. 1966 Central Places in Southern Germany. Englewood Cliffs: Prentice Hall. Clift, Peter D., Jae Il Lee, Peter Hildebrand, Nobumichi Shimizu, Graham D. Lay ne, Jerzy Blusztajn, Joel D. Blum, Eduardo Garzanti, and Athar Ali Khan 2002 Nd and Pb Isotope Variability in the Indus River System: Implications for Sediment Provenance and Crustal Heterogeneity in the Western Himalaya. Earth and Planetary Science Lett ers 200(1 2):91 106. Cobb, Charles R. 2005 Archaeology and the "Savage Slot": Displacement and Emplacement in the Premodern World. American Anthropologist 107(4):563 574. Cohen, Abner 1974 Introduction: The Lesson of Ethnicity. In Urban Ethnicity. Abner Cohen, ed. Pp. ix xxiv. London: Tavistock Publications. Coplen, T. B. 1994 Reporting of Stable Hydrogen, Carbon, and Oxygen Isotopic Abundances. Pure and Applied Chemistry 66:273 276. Critelli, Salvatore, and Eduardo Garzanti 1994 Provenance of the Lower Tertiary Murree Redbeds (Hazara Kashmir Syntaxis, Pakistan) and Initial Rising of the Himalayas. Sedimentary Geology 89(3 4):265 284. Dalai, Tarun K., S. K. Bhattacharya, and S. Krishnaswami 2002 Stable Isotopes in the Source Waters of the Ya muna and its Tributaries: Seasonal and Altitudinal Variations and Relation to Major Cations. Hydrological Processes 16(17):3345 3364.

PAGE 221

221 Dalai, Tarun K., S. Krishnaswami, and Anil Kumar 2003 Sr and 87 Sr/ 86 Sr in the Yamuna River System in the Himalaya: Sour ces, Fluxes, and Controls on Sr I sotope Composition. Geochimica e t Cosmochimica Acta 67(16):2931 2948. Dales, George F. 1976 Shifting Trade Patterns between the Iranian Plateau and the Indus Valley in the Third Millennium B.C. In Le Plateau Iranien e t l'Asie Central Des Origines a l a Conquete Islamique. Pp. 67 78. Paris: Colloques Internationaux du Centre National de la Recherche Scientifique. 1991 Some Specialized Ceramic Studies at Harappa. In Harappa Excavations 1986 1990. Richard Meadow H., ed. Pp. 61 70. Madison: Prehistory Press. Dales, George F., and Jonathan Mark Kenoyer 1986 Excavations at Mohenjo Daro, Pakistan: The Pottery. Philadelphia: University Museum Press. 1986 Preliminary Report on the First Season of Research at Harappa, Pakistan. Report submitted to the Director General of Archaeology and Museums, Government of Pakistan, Karachi. 1987 Preliminary Report on the Second Season of Research at Harappa, Pa kistan: January to April 1987. Report submitted to the Director General of Archaeology and Museums, Government of Pakistan, Karachi. 1988 Preliminary Report on the Third Season (January March) of Research at Harappa, Pakistan. Report submitted to t he Director General of Archaeology and Museums, Government of Pakistan, Karachi. 1989 Excavation at Harappa 1988. Pakistan Archaeology 24:68 176. 1989 Preliminary Report on the Fourth Season at Harappa, Pakistan, January 15 March 31, 1989. Report submitted to the Director General of Archaeology and Museums, Government of Pakistan, Karachi. 1990 Excavation at Harappa 1989. Pakistan Archaeology 25:241 280. 1990 Preliminary Report on the Fifth Season at Harappa, Pakistan, January 1 March 31, 1990. Report submitted to the Director General of Archaeology and Museums, Government of Pakistan, Karachi. 1991 Summaries of Five Seasons of Research at Harappa (District Sahiwal, Punjab, Pakistan) 1986 1990. In Harappa Excavations 1986 1990: A M ulti Disciplinary Approach to 3rd Millennium Urbanism. Richard H. Meadow, ed. Pp. 185 262. Madison: Prehistory Press.

PAGE 222

222 Dales, George F., and Carl P. Lipo 1992 Explorations on the Makran Coast, Pakistan: A Search for Paradise. Berkeley: Archaeological Research Facility, University of California, Berkeley. Dangi, Vivek 2009 Archaeology of the Ghaggar Basin: Settlement Archaeology of Meham Block, Rohtak, Haryana, India. In Occasional Paper 8: Linguistics, Archaeology, and the Human Past. Toshiki Osada and Akinori Uesugi, eds. Pp. 1 111. Kyoto: Indus Project, Research Institute for Humanity and Nature. Dansgaard, W. 1964 Stable Isotopes in Precipitation. Tellus 16(4):436 468. Daux, Valerie, Christophe LeCuyer, Frederic Adam, Francois Martineau, and F rancoise Vimeux 2005 Oxygen Isotope Composition of Human Teeth and the Record of Climate Changes in France (Lorraine) during the Last 1700 Years. Climatic Change 70(3):445 464. Daux, Valrie, Christophe Lcuyer, Marie Anne Hran, Romain Amiot, Laurent S imon, Franois Fourel, Franois Martineau, Niels Lynnerup, Herv Reychler, and Gilles Escarguel 2008 Oxygen Isotope Fractionation between Human Phosphate and Water Revisited. Journal of Human Evolution 55(6):1138 1147. de Certeau, Michel 1984 The Practice of Everyday Life. Berkeley: University of California Press. DeNiro, Michael J., and Samuel Epstein 1978 Influence of Diet on the Distribution of Carbon I sotopes in Animals. Geochimica e t Cosmochimica Acta 42(5):495 506. Dhavalikar, M. K. 199 2 Kuntasi: A Harappan Port in Western India. In South Asian Archaeology 1989. C. Jarrige, ed. Pp. 73 82. Madison: Prehistory Press. 1997 Meluhha, the Land of Copper. South Asian Studies 13:275 278. Dikshit, K. N. 1979 The Late Harappan Culture in Ind ia. In Essays in Indian Protohistory. D. P. Agrawal and Dilip K. Chakrabarti, eds. Pp. 123 133. Delhi: B. R. Publishing. Dillehay, Tom D. 1995 Mounds of Social Death: Araucanian Funerary Rites and Political Succession. In Tombs for the Living: Andean Mo rtuary Practices. Tom D. Dillehay, ed. Pp. 281 314. Washington: Dumbarton Oaks.

PAGE 223

223 Driessens, F. C. M., and R. M. H. Verbeeck 1990 Biominerals. Boca Raton: CRC Press. Dupras, Tosha L., and Henry P. Schwarcz 2001 Strangers in a Strange Land: Stable Isotop e Evidence for Human Migration in the Dakhleh Oasis, Egypt. Journal of Archaeological Science 28(11):1199 1208. Durrani, F. A. 1988 Excavations in the Gomal Valley: Rehman Dheri Excavation Report no. 1. Ancient Pakistan 6:1 232. Durrani, F. A., I. Ali, and G. Erdosy 1995 New Perspectives on Indus Urbanism from Rehman Dheri. East and West 45(1 4):18 96. Elias, Robert W., Yoshimitsu Hirao, and Clair C. Patterson 1982 The Circumvention of the Natural Biopurification of Calcium Along Nutrient Pathways by Atmospheric Inputs of Indus trial Lead. Geochimica e t Cosmochimica Acta 46(12):2561 2580. ESCAP (Economic and Social Commission for Asia and the Pacific) 1995 Geology and Mineral Resources of Afghanistan. In Atlas of Mineral Resources of the ESCAP Reg ion, Volume 2. New York: United Nations. Evans, J. A., J. Montgomery, G. Wildman, and N. Boulton 2010 Spatial Variations in Biosphere 87 Sr/ 86 Sr in Britain. Journal of the Geological Society 167(1):1 4. Evans, J. A., and S. Tatham ocal to Twelfth Century Anglo Saxon Population Living on a Jurassic Clay Carbonate Terrain, Rutland, UK. Geological Society, London, Special Publications 232(1):237 248. Fairservis, W. A. 1975 The Roots of Ancient India. Chicago: University of Chicago Press. Farquhar, G. D., J. R. Ehleringer, and K. T. Hubick 1989 Carbon Isotope Discrimination and Photosynthesis. Annual Review of Plant Physiology and Plant Molecular Biology 40(1):503 537. Farqu har, G. D., Marion H. O'Leary, and J. A. Berry 1982 On the Relationship between Carbon Isotope Discrimination and the Intercellular Carbon Dioxide Concentration in Leaves. Functional Plant Biology 9(2):121 137.

PAGE 224

224 Faure, Gunter, and Teresa M. Mensing 2005 Isotopes: Principles and Applications. 3rd ed. Hoboken: John Wiley and Sons. Fentress, M. A. 1976 Resource Access, Exchange Systems and Regional Interaction in the Indus Valley: An Investigation of Archaeological Variability at Harappa and Mohenjod aro. Ph.D. dissertation, Department of Oriental Studies, University of Pennsylvania. In Studies in the Archaeology and Palaeoanthropology of South Asia. Kenneth A. R. Kennedy and Gregory L. Possehl, eds. Pp. 89 97. New Delhi: Oxford & IBH. Flam, Louis 1993 Excavations at Ghazi Shah, Sindh, Pakistan. In Harappan Civilization: A Recent Perspective. Gregory L. Possehl, ed. Pp. 457 467. Oxford: Oxford & IBH & American I nstitute of Indian Studies. Frachetti, Michael D. 2011 Migration Concepts in Central Eurasian Archaeology*. Annual Review of Anthropology 40(1):195 212. Francey, R. J., and G. D. Farquhar 1982 An Explanation of 13 C/ 12 C Variations in Tree Rings. Natur e 297(5861):28 31. Francfort, H. P, C. Boisset, L. Butchet, J. Desse, J. Echallier, A. Kermorvant, and G. Willcox 1989 Foui lles d e Shortughai: Recherches s ur l 'Asie Centrale Protohistorique. Paris: Diffusion de Boccard. Franke Vogt, U. 2000 The Arch aeology of Southeastern Balochistan. Electronic document, accessed 5/24, 2012. 2005 Excavations at Sohr Damb/Nal: Results of the 2002 and 2004 Seasons. In South Asian Archaeology 2003. U. Franke Vogt and H. J Weisshaar, eds. Pp. 63 76. Aachen: Linden Soft. Frenez, D., and M. Tosi 2005 Lothal Sealings: Records from an Indus Civilization Town at the Eastern End of the Maritime Trade Circuits Across the Arabian Sea. In Studi in Onore d i Enrica Fia ndra, Contributi di Archeologia Egea e Vicinorientale. Massimo Pernaa, ed. Pp. 65 103. Paris: Diffusion de Boccard.

PAGE 225

225 Fricke, Henry C., and James R. O'Neil 1996 Inter and Intra Tooth Variation in the Oxygen Isotope Composition of Mammalian Tooth Enamel Phosphate: Implications for Palaeoclimatological and Palaeobiological Research. Palaeogeography, Palaeoclimatology, Palaeoecology 126(1 2):91 99. Fricke, Henry C., James R. O'Neil, and Niels Lynnerup 1995 SPECIAL RE PORT: Oxygen Isotope Composition of Human Tooth Enamel from Medieval Greenland: Linking Climate and Society. Geology 23(10):869 872. Friedli, H., H. Lotscher, H. Oeschger, U. Siegenthaler, and B. Stauffer 1986 Ice Core Record of the 13 C/ 12 C Ratio of Atm ospheric CO 2 in the Past Two Centuries. Nature 324(6094):237 238. Fullagar, Paul D., Robert E. Lemmon, and Paul C. Ragland 1971 Petrochemical and Geochronological Studies of Plutonic Rocks in the Southern Appalachians: I. the Salisbury Pluton. Geologica l Society of America Bulletin 82(2):409 416. Fuller, Dorian Q., and Marco Madella 2001 Issues in Harappan Archaeobotany: Retrospect and Prospect. In Indian Archaeology in Retrospect, Volume 2 (Protohistory: Archaeology of the Harappan Civilization). S. Settar and R. Korisettar, eds. Pp. 317 390. New Delhi: Manohar. Gangal, K., M. Vahia, and R. Adhikari 2010 Spatio Temporal Analysis of the Indus Urbanization. Current Science 98:846 852. Garvie Lok, Sandra J., Tamara L. Varney, and M. Anne Katzenberg 2004 Preparation of Bone Carbonate for Stable Isotope Analysis: The Effects of Treatment Time and Acid Concentration. Journal of Archaeological Science 31(6):763 776. Gat, J. R. 1996 Oxygen and Hydrogen Isotopes in the Hydrologic Cycle. Annual Review of Earth and Planetary Sciences 24(1):225 262. Gell, Alfred 1998 Art and Agency: An Anthropological Theory. Oxford: Clarendon Press. Geological Survey of India 1994 Detailed Information on Copper Lead Zinc Ores in Rajasthan Gujarat (India). Calcutta: Geological Survey of India.

PAGE 226

226 Giblin, Julia I., Kelly J. Knudson, Zsolt Bereczki, Gyrgy Plfi, and Ildik Pap 2013 Strontium Isotope Analysis and Human Mobility during the Neolithic and Copper Age: A Case Study from the Great Hungarian Plain. Journal of Archaeological Science 40(1):227 239. Gillespie, Susan D. 2001 Personhood, Agency, and Mortuary Ritual: A Case Study from the Ancient Maya. Journal of Anthropological Archaeology 20(1):73 112. Giosan, Liviu, Peter D. Clift, Mark G. Macklin, Dorian Q Fuller, Stefan Constantinescu, Julie A. Durcan, Thomas Stevens, Geoff A. T. Duller, Ali R. Tabrez, Kavita Gangal, Ronojoy Adhikari, Anwar Alizai, Florin Filip, Sam VanLaningham, and James P. M. Syvitski 2012 Fluvial Landscapes of the Harappan Civilizat ion. Proceedings of the National Academy of Sciences 109(26):E1688 E1694. Gogte, S. T., Nandita Basu, Seeta Sinclair, O. P. Ghai, and N. K. Bhide 1991 Blood Lead Levels of Children with Pica and Surma Use. The Indian Journal of Pediatrics 58(4):513 519. Goldstein, Lynne 1980 Mississippian Mortuary Practices: A Case Study of Two Cemeteries in the Lower Illinois Valley. Scientific Papers 4. 1981 One Dimensional Archaeology and Multi Dimensional People: Spatial Organization and Mortuary Analysis. In The Archaeology of Death. Robert Chapman, Ian Kinnes and Klavs Randsborg, eds. Pp. 53 69. Cambridge: Cambridge University Press. Goody, E. 1982 Parenthood and Social Reproduction. Cambridge: Cambridge University Press. Gorospe, Emmanuel C., and Shawn L. Gerstenberger 2008 Atypical Sources of Childhood Lead Poisoning in the United States: A Systematic Review from 1966 2006. Clinical Toxicology 46(8):728 737. Gulson, Brian L., Karen J. Mizon, Jacqueline M. Palmer, Michael J. Korsch, Alan J. Taylor, a nd Kathryn R. Mahaffey 2004 Blood Lead Changes during Pregnancy and Postpartum with Calcium Supplementation. Environmental Health Perspectives 112(15):1499 1507. Gulson, Brian L., C. William Jameson, Kathryn R. Mahaffey, Karen J. Mizon, Nicole Patison, Alistair J. Law, Michael J. Korsch, and Mary A. Salter 1998 Relationships of Lead in Breast Milk to Lead in Blood, Urine, and Diet of the Infant and Mother. Environmental Health Perspectives 106(10):pp. 667 674.

PAGE 227

227 Gulson, Brian L., Karen J. Mizon, Michael J. Korsch, Jacqueline M. Palmer, and John B. Donnelly 2003 Mobilization of Lead from Human Bone Tissue during Pregnancy and Lactation a Summary of Long Term Research. Science of the Total Environment 303(1 2):79 104. Gupta, S. K., and R. D. Deshpande 2005 Groundwater Isotopic Investigations in India: What has been Learned? Current Science 89(5):825 835. Gupta, S. P. 1984 Internal Trade of the Harappans. In Frontiers of the Indus Civilization. B. B. Lal and S. P. Gupta, eds. Pp. 417 424. New Delhi: Books and Books. 1989 Copper Hoard. In An Encyclopaedia of Indian Archaeology. A. Ghosh, ed. Pp. 92. Delhi: Munshiram Manoharlal. Gupta, S. K., R. D. Deshpande, S. K. Bhattacharya, and R. A. Jani 18 Peninsula: Influence of the Arabian Sea and the Bay of Bengal Branches of the Summer Monsoon. Journal of Hydrology 303(1 4):38 55. Gwynn, E. J. 1913 Fosterage. In Encyclopaedia of Religion and Ethics. J. Hastings, ed. Pp. 104 109. Edinburgh: T. and T. C lark. Hakenbeck, Susanne 2008 Migration in Archaeology: Are We Nearly T here Yet? Archaeological Review from Cambridge 23(2):9 26. Halim, Mohammad A., and Massimo Vidale 1984 Kilns, Bangles, and Coated Vessels: Ceramic Production in Closed Containers at Mohenjodaro. In Interim Reports, Volume 1: Reports on Field Work Carried Out at Mohenjo Daro, Pakistan 1982 83. Michael Jansen and Gunter Urban, eds. Pp. 63 97. Aachen: RWTH IsMEO. Harke, Heinrich 2007 Ethnicity, 'Race' and Migration in Mortuary Archae ology: An Attempt at a Short Answer. In Early Medieval Mortuary Practices. S. Semple and H. WIlliams, eds. Pp. 12 18. Oxford: Oxbow. Harrison, Roman G., and M. Anne Katzenberg 2003 Paleodiet Studies using Stable Carbon Isotopes from Bone Apatite and Collagen: Examples from Southern Ontario and San Nicolas Island, California. Journal of Anthropological Archaeology 22(3):227 244.

PAGE 228

228 Hedges, R. E. M. 2003 On Bone Collagen Apatite Carbonate Isotopic Relationships. International Journal of Osteoarchaeology 13(1 2):66 79. Helms, Mary W. 1988 Ulysses' Sail: An Ethnographic Odyssey of Power, Knowledge, and Geographical Distance. Princeton: Princeton University Press. 1993 Craft and the Kingly Ideal: Art, Trade, and Power. Austin: University of Texas Pre ss. Hemphill, Brian E., John R. Lukacs, and Kenneth A. R. Kennedy 1991 Biological Adaptations and Affinities of Bronze Age Harappans. In Harappa Excavations 1986 1990: A Multidisciplinary Approach to Third Millennium Urbanism. Richard H. Meadow, ed. Pp. 137 182. Madison: Prehistory Press. Hide, Robin 2003 Pig Husbandry in New Guinea: A Literature Review and Bibliography. Canberra: Australian Centre for International Agricultural Research. Hillson, Simon 1996 Dental Anthropology. Cambridge: Cambridge University Press. Hodder, Ian 1982 The Identification and Interpretation of Ranking in Prehistory: A Contextual Perspective. In Ranking, Resource, and Exchange: Aspects of the Archaeology of Early European Society. A. C. Renfrew and S. J. Shen nen, eds. Pp. 150 154. Cambridge: Cambridge University Press. 2012 Entangled: An Archaeology of the Relationships between Humans and Things. Chichester: John Wiley and Sons. Hodell, David A., Rhonda L. Quinn, Mark Brenner, and George Kamenov 2004 Spatial Variation of Strontium Isotopes ( 87 Sr/ 86 Sr) in the Maya Region: A Tool for Tracking Ancient Human Migration. Journal of Archaeological Science 31(5):585 601. Hoffman, Brett C., and Heather M. L Miller 2009 Production and Consumption of Copper B ase Metals in the Indus Civilization. Journal of World Prehistory 22:237 264. Hooja, R. 1994 Ganeshwar and the "Frontier Theory": A Possible Framework for Understanding Ganeshwar Harappan Interaction. Shodhak 23(69):127 136.

PAGE 229

229 Hoppe, K. A., P. L. Koch, and T. T. Furutani 2003 Assessing the Preservation of Biogenic Strontium in Fossil Bones and Tooth Enamel. International Journal of Osteoarchaeology 13(1 2):20 28. Howland, M. R., L. T. Corr, S. M. M. Young, V. Jones, S. Jim, N. J. Van Der Merwe, A. D. Mitchell, and R. P. Evershed 2003 Expression of the Dietary Isotope Signal in the Compound Specific 13 C Values of Pig Bone Lipids and Amino Acids. International Journal of Osteoarchaeology 13(1 2):54 65. Hu, Yaowu, Hong Shang, Haowen Tong, Olaf Nehlich Wu Liu, Chaohong Zhao, Jincheng Yu, Changsui Wang, Erik Trinkaus, and Michael P. Richards 2009 Stable Isotope Dietary Analysis of the Tianyuan 1 Early Modern Human. Proceedings of the National Academy of Sciences 106(27):10971 10974. Ingold, Tim 2011 Being Alive: Essays on Movement, Knowledge, and Description. Abingdon: Routledge. Jackes, Mary 2011 Representativeness and Bias in Archaeological Skeletal Samples. In Social Bioarchaeology. Sabrina C. Agarwal and Bonnie A. Glencross, eds. Pp. 107 1 46. Chichester: Wiley Blackwell. Jansen, Michael 1984 Preliminary Results on the "Forma Urbis" Research at Mohenjo Daro. In Interim Reports, Volume 2: Reports on Field Work Carried Out at Mohenjo Daro, Pakistan 1983 84 by the IsMEO Aachen University Mission. Michael Jansen and Gunter Urban, eds. Pp. 9 22. Aachen: Forschungsprojekt "Mohenjo Daro". 1993 Mohenjo Daro: City of Wells and Drains, Water Splendour 4500 Years Ago. Bonn: Bergisch Gladbach Frontinus Gesellschaft. 2002 Settlement Networks of the Indus Civilization. In Indian Archaeology in Retrospect, Volume 2 (Protohistory: Archaeology of the Harappan Civilization). S. Settar and R. Korisettar, eds. Pp. 105 128. New Delhi: Manohar. Jarrige, C., J. F Jarrige, Ri chard H. Meadow, and G. Q uivron 1995 Mehrgarh: Field Reports 1974 1985, from Neolithic Times to the Indus Civilization. Karachi: The Department of Culture and Tourism, Government of Sindh. Jarrige, J. F 1995 Introduction. In Mehrgarh: Field Reports 1974 1985. C. Jarrige, J. F Jarrige, Richard H. Meadow and G. Quivron, eds. Pp. 51 103. Karachi: The Department of Culture and Tourism, Government of Sindh.

PAGE 230

230 Jenkins, P. C. 1994 Continuity and Change in the Ceramic Sequence at Harappa. In South Asian Archaeology 1993. Asko Parpol a and P. Koskikallio, eds. Pp. 315 328. Helsinki: Suomalainen Tiedeakatemia. Jim, Susan, Stanley H. A mbrose, and Richard P. Evershed 2003 Natural Abundance Stable Carbon Isotope Evidence for the Routing and De Novo Synthesis of Bone FA and Cholesterol. Lipids 38(2):179 186. 2004 Stable Carbon Isotopic Evidence for Differences in the Dietary Origin of Bone Cholesterol, Collagen and Apatite: Implications for Their U se in Palaeodietary Reconstruction. Geochimica e t Cosmochimica Acta 68(1):61 72. Jones Sian 1997 The Archaeology of Ethnicity: Constructing Identities in the Past and Present. London: Routledge. Joshi, Jagat Pati 1984 Harappa Culture: Emergence of a New Picture. Puratattva 13 14:51 54. 2003 Structures. In Excavations at Kalibangan: The Early Harappans (1960 1969). B. B. Lal, B. K. Thapar, Jagat Pati Joshi and Madhu Bala, eds. Pp. 51 94. New Delhi: Archaeological Survey of India. Joyce, Rosemary A. 2001 Burying the Dead at Tlatilco: Social Memory and Social Identities. Archeological Papers of the American Anthropological Association 10(1):12 26. Kadir, Muhammad Masood, Naveed Zafar Janjua, Sibylle Kristensen, Zafa r Fatmi, and Nalini Sathiakumar 2008 Status of Children's Blood Lead Levels in Pakistan: Implications for Research and Policy. Public Health 122(7):708 715. Kamenov, George D., P. A. Mueller, A. Gilli, S. Coyner, and S. H. H. Nielsen 2006 A Simple Method for Rapid, High Precision Isotope Analyses of Small Samples by MC ICP MS. Eos Transactions AGU 87 Fall M eeting Supplement:V21A 0542. Kamenov, George D., Paul A. Mueller, and Michael R. Perfit 2004 Optimization of Mixed Pb Tl Solutions for High Precision Isotopic Analyses by MC ICP MS. Journal of Analytical Atomic Spectrometry 19(9):1262 1267. Karim, Ajaz 1999 Hydrochemistry and Isotope Systematics of the Indus River Basin. Ph.D. dissertation, Department of Earth Science, University of Ottawa.

PAGE 231

231 Karim, Ajaz, and Jan Veizer 2000 Weathering Processes in the Indus River Basin: Implications from Riverine C arbon, Sulfur, Oxygen, and Strontium Isotopes. Chemical Geology 170(1 4):153 177. 2002 Water Balance of the Indus River Basin and Moisture Source in the Karakoram and Western Himalayas: Implications from Hydrogen and Oxygen Isotopes in River Water. Jo urnal of Geophysical Research: Atmospheres 107(D18):4362, doi:10.1029/2000JD000253. Kazmi, A. H., and M. Q. Jan 1997 Geology and Tectonics of Pakistan. Karachi: Graphic Publishers. Kellner, Corina M., and Margaret J. Schoeninger 2007 A Simple Carbon Isotope Model for Reconstructing Prehistoric Human Diet. American Journal of Physical Anthropology 133(4):1112 1127. Kend all, Carol, and Tyler B. Coplen 2001 Distribution of Oxygen 18 and Deuterium in River Waters Across the United States. Hydrological Processes 15(7):1363 1393. Kennedy, Kenneth A. R. 2000 God Apes and Fossil Men : Paleoanthropology of South Asia. Ann Arbor: University of Michigan Press. Kennedy, Kenneth A. R., John Chiment, Todd Disotell, and David Meyers 1984 Principal Components Analysis of Prehistoric South Asian Crania. American Journal of Physical Anthropology 64(2):105 118. Kenoyer, Jon athan Mark 1989 Socio Economic Structures of the Indus Civilization as Reflected in Specialized Crafts and the Question of Ritual Segregation. In Old Problems and New Perspectives in the Archaeology of South Asia. Jonathan Mark Kenoyer, ed. Pp. 183 192. Madison: Department of Anthropology, University of Wisconsin, Madison. 1991 The Indus Valley Tradition of Pakistan and Western India. Journal of World Prehistory 5(4):331 385. 1991 Urban Process in the Indus Tradition: A Preliminary Model from Harappa. In Harappa Excavations 1986 1990. Richard H. Meadow, ed. Pp. 29 60. M adison: Prehistory Press. 1992 Harappan Craft Specialization and the Question of Urban Segregation and Stratification. Eastern Anthropologist 45(1 2):39 54.

PAGE 232

232 1993 Excavations on Mound E, Harappa: A Systematic Approach to the Study of Indus Urbanis m. In South Asian Archaeology 1991. A. J. Gail and G. J. R. Mevissen, eds. Pp. 165 194. Stuttgart: F. S. Verlag. 1995 Ideology and Legitimation in the Indus State as Revealed through Symbolic Objects. The Archaeological Review 4(1/2):87 131. 1995 Interaction Systems, Specialized Crafts, and Culture Change: The Indus Valley Tradition and the Indo Gangetic Tradition in South Asia. In The Indo Aryans of Ancient South Asia: Language, Material Culture, and Ethnicity. G. Erdosy, ed. Pp. 213 257. Berlin: deGruyter. 1997 Early City States in South Asia: Comparing the Harappan Phase and Early Historic Period. In The Archaeology of City States: Cross Cultural Approaches. D. L. Nichols and T. H. Charlton, eds. Pp. 51 70. Washington, D.C.: Smithson ian Institution Press. 1998 Ancient Cities of the Indus Valley Civilization. Oxford: Oxford University Press. 2000 Wealth and Socioeconomic Hierarchies of the Indus Valley Civilization. In Order, Legitimacy, and Wealth in Early States. J. Richards and M. Van Buren, eds. Pp. 88 109. Cambridge: Cambridge University Press. 2005 Bead Technologies at Harappa, 3300 1900 BC: A Comparative Summary. In South Asian Archaeology 2001. C. Jarrige and V. Lefevre, eds. Pp. 157 170. Paris: Editions Recerche s ur les Civilisations. 2005 Culture Change during the Late Harappan Period at Harappa: New Insights on Vedic Aryan Issues. In The Indo Aryan Controversy: Evidence and Inference in Indian History. Edwin F. Bryant and Laurie L. Patton, eds. Pp. 21 49. Oxon: Routledge. 2006 Cultures and Societies of the Indus Tradition. In Historical Roots in the Making of 'the Aryan'. Romila Thapar, ed. Pp. 41 97. New Delhi: National Book Trust. 2011 Changing Perspectives of the Indus Civilization: New Discoveries and Challenges. Puratattva 41:1 18. 2011 Regional Cultures of the Greater Indus Valley: The Ravi and Kot Diji Phase Assemblages of Harappa, Pakistan. In Cultural Relations betwee n the Indus and the Iranian Plateau during the Third Millennium BCE. Toshiki Osada and Michael Witzel, eds. Pp. 165 217. Cambridge: Department of South Asian Studies, Harvard University. Kenoyer, Jonat han Mark, and Richard H. Meadow 1996 New Inscribe d Objects from Harappa. Lahore Museum Bulletin 9(1):1 29.

PAGE 233

233 1999 Harappa: New Discoveries on its Origins and Growth. Lahore Museum Bulletin 12(1):1 12. 2000 The Ravi Phase: A New Cultural Manifestation at Harappa. In South Asian Archaeology 1997. M. T addei, ed. Pp. 55 76. Rome: IsIAO. 2008 The Early Indus Script at Harappa: Origins and Development. In Intercultural Relations between South and Southwest Asia: Studies in Commemoration of E. C. L. during Caspers (1934 1996). E. Olijdam and R. H. Spoor, eds. Pp. 124 131. Oxford: Archaeopress. Kenoyer, Jonathan Mark, and Heat her M. L Miller 1999 Metal Technologies of the Indus Valley Tradition in Pakistan and Western India. In The Archaeometal lurgy of the Asian Old World. V. C. Pigott, ed. Pp. 107 151. Philadelphia: University Museum, University of Pennsylvania. Kenoyer, Jona than Mark, and Massimo Vidale 1992 A New Look at Stone Drills of the Indus Valley Tradition. In Materials Issues in Ar t and Archaeology, III. P. Vandiver, J. R. Druzik, G. S. Wheeler and I. Freestone, eds. Pp. 495 518. Pittsburgh: Materials Research Society. Kenoyer, Jonathan Mark, T. Dou glas Price, and James H. Burton 2013 A New Approach to Tracking Connections between the Indus Valley and Mesopotamia: Initial Results of Strontium Isotope Analyses from Harappa and Ur. Journal of Archaeological Science 40(5):2286 2297. Khan, F ., J. R. Knox, and K. D. Thomas 1990 Sheri Khan Tarakai: A Neolithic Village in Bannu District, NWFP. In South Asian Archaeology 1987. M. Taddei and P. Callieri, eds. Pp. 111 127. Rome: IsMEO. Khatib Chahidi, J. 1992 Milk Kinship in Shi'Ite Islamic Iran. In The Anthropology of Breast Feeding. V. Maher, ed. Pp. 109 132. Oxford: Berg. Kha tri, J. S., and M. Acharya 1995 Kunal: A New Indus Saraswati Site. Puratattva 25:84 86. 2005 Kunal Excavations: New Light on the Origin of the Harappan Civilization. In In Search of Vedic Harappan Relationship. A. Agrawal, ed. Pp. 104 117. New Delhi: Aryan Books International. Kinaston, Rebecca, Hallie Buckley, Andrew Gr ay, Ben Shaw, and Herman Mandui 2013 Exploring Subsistence and Cultural Complexes on the South Coast o f Papua New Guinea using Palaeodietary Analyses. Journal of Archaeological Science 40(2):904 913.

PAGE 234

2 34 Klaminder, Jonatan, Richard Bindler, O ve Emteryd, and Ingemar Renberg 2005 Uptake and Recycling of Lead by Boreal Forest Plants: Quantitative Estimates fro m a Site in Northern Sweden. Geochimica e t Cosmochimica Acta 69(10):2485 2496. Klaminder, Jonatan, Richard Bindler, Hjalmar Laudon, Kevin Bishop, Ove Emteryd, and Ingemar Renberg 2006 Flux Rates of Atmospheric Lead Pollution within Soils of a Small Catc hment in Northern Sweden and their Implications for Future Stream Water Quality. Environmental Science & Technology 40(15):4639 4645. Klaminder, Jonatan, Richa rd Bindler, and Ingemar Renberg 2008 The Biogeochemistry of Atmospherically Derived Pb in the Boreal Forest of Sweden. Applied Geochemistry 23(10):2922 2931. Klaminder, Jonatan, John G. Farmer, and Angus B. MacKenzie 2011 The Origin of Lead in the Organic Horizon of Tundra Soils: Atmospheric Deposition, Plant Translocation from the Mineral Soil Or Soil Mineral Mixing? Science of the Total Environment 409(20):4344 4350. Knudson, K. J. 2009 Oxygen Isotope Analysis in a Land of Environmental Extremes: The Complexities of Isotopic Work in the Andes. International Journal of Osteoarchaeology 19(2): 171 191. Knudson, K. J., T. D. Price, J. E. Buikstra, and D. E. Blom 2004 The use of Strontium Isotope Analysis to Investigate Tiwanaku Migration and Mortuary Ritual in Bolivia and Peru. Archaeometry 46(1):5 18. Koch, Paul L., Noreen Tuross, and Marilyn L. Fogel 1997 The Effects of Sample Treatment and Diagenesis on the Isotopic Integrity of Carbonate in Biogenic Hydroxylapatite. Journal of Archaeological Science 24(5):417 429. Kohl, P. L. 1978 Archaeological Reconnaissances in Eastern Afghanistan, 1975 1976. Estratto Da ANNALI Dell'Istituto Orientale d i Napoli 38:64 74. Kohn, Matthew J. 2010 Carbon Isotope Compositions of Terrestrial C3 Plants as Indicators of (Paleo)Ecology and (Paleo)C limate. Proceedings of the National Academy of Sciences 107(46):19691 19695.

PAGE 235

235 Kohn, Matthew J., Jennifer Morris, and Paul Olin 2013 Trace Element Concentrations in Teeth a Modern Idaho Baseline with Implications for Archeometry, Forensics, and Palae ontology. Journal of Archaeological Science 40(4):1689 1699. Kohn, Matthew J., Margaret J. Schoeninger, and John W. Valley 1996 Herbivore Tooth Oxygen Isotope Compositions: Effects of D iet and Physiology. Geochimica e t Cosmochimica Acta 60(20):3889 3896 2008 Lead Isotopes in Environmental Sciences: A Review. Environment International 34(4):562 577. Konasukawa, Ayumu, H itoshi Endo, and Akinori Uesugi 2011 Minor Objects from the Settlement Area. In Excavations at Farmana: District Rohtak, Haryana, India, 2006 2008. Vasant Shinde, Toshiki Osada and Manmohan Kumar, eds. Pp. 369 529. Kyoto: Indus Project, Research Institute for Humanity and Nature. Krigbaum, John 2003 Neolith ic Subsistence Patterns in Northern Borneo Reconstructed with Stable Carbon Isotopes of Enamel. Journal of Anthropological Archaeology 22(3):292 304. Krishnaswami, S., J. R. Trivedi, M. M. Sar in, R. Ramesh, and K. K. Sharma 1992 Strontium Isotopes and Rubidium in the Ganga Brahmaputra River System: Weathering in the Himalaya, Fluxes to the Bay of Bengal and Contributions to the Evolution of Oceanic 87 Sr/ 86 Sr. Earth and Planetary Science Letters 109(1 2):243 253. Kroeber, A. L. 1927 Disposal of the Dead. American Anthropologist 29(3):308 315. Krueger, Har old W., and Charles Sullivan H. 1984 Models for Carbon Isotope Fractionation between Diet and Bone. In Stable Isotopes in Nutrition. Judith R. Turnlund and Phyllis E. Johnson, eds. Pp. 205 220. Washington: American Chemical Society. Kumar, Manmohan 2009 In Occasional Paper 7: Linguistics, Archaeology, and the Human Past. Toshiki Osada and Akinori Uesugi, eds. Pp. 1 75. Kyoto: Indus Project, Research Institute for Hum anity and Nature. Lahiri, Nayanjot 1992 The Archaeology of Indian Trade Routes Up to C. 200 BC: Resource use, Resource Access and Lines of Communication. Delhi: Oxford University Press.

PAGE 236

236 Lal, B. B. 1951 Further Copper Hoards from the Gangetic Basin an d a Review of the Problem. Ancient India 7:20 39. 1972 A Note on the Excavation at Sapai. Puratattva 5:46 49. Lamberg Karlovsky, C. C. 1972 Trade Mechanisms in Indus Mesopotamian Interrelations. Journal of the American Oriental Society 92(2):222 229. L ambs, L., K. Balakrish na, F. Brunet, and J. L. Probst 2005 Oxygen and Hydrogen Isotopic Composition of Major Indian Rivers: A First Global Assessment. Hydrological Processes 19(17):3345 3355. Law, Randall William 2006 Moving Mountains: The Trade and Transport of Rocks and Minerals within the Greater Indus Valley Region. In Space and Spatial Analysis in Archaeology. E. C. Robertson, J. D. Seibert, D. C. Fernandez and M. U. Zender, eds. Pp. 301 313. Calgary: Universi ty of Calgary Press. 2008 Inter Regional Interaction and Urbanism in the Ancient Indus Valley: A Geologic Provenience Study of Harappa's Rock and Mineral Assemblage. Ph.D. dissertation, Department of Anthropology, University of Wisconsin Madison. Lec uyer, Christophe, Patricia Gran djean, and Simon M. F. Sheppard 1999 Oxygen Isotope Exchange between Dissolved Phosphate and Water at cal Fractionations. Geochimica e t Cosmochimica Acta 63(6):855 862. Lefebvre Henri 1991 The Production of Space. Oxford: Blackwell. Levin, Naomi E., Thure E. Cerling, Benjamin H. Passey, John M. Harris, and James R. Ehleringer 2006 A Stable Isotope Aridity Index for Terrestrial Environments. Proceedings of the National Acade my of Sciences 103(30):11201 11205. Levinson, A. A., Boaz Luz, and Yehoshua Kolodny 1987 Variations in Oxygen Isotopic Compositions of Human Teeth and Urinary Stones. Applied Geochemistry 2(4):367 371. Lightfoot, Kent G., Antoine tte Martinez, and Ann M Schiff 1998 Daily Practice and Material Culture in Pluralistic Social Settings: An Archaeological Study of Culture Change and Persistence from Fort Ross, California. American Antiquity 63(2):199 222.

PAGE 237

237 Longinelli, Antonio 1984 Oxygen Isotopes in Mammal Bone Phosphate: A New Tool for Paleohydrological and Paleoclima tological Research? Geochimica e t Cosmochimica Acta 48(2):385 390. Lorenzen, Gunnar, Christoph Sprenger, Paul Baudron, Dinesh Gupta, and Asaf Pekdeger 2012 Origin and Dynamics of Grou ndwater Salinity in the Alluvial Plains of Western Delhi and Adjacent Territories of Haryana State, India. Hydrological Processes 26(15):2333 2345. Lukacs, John R., and J. N. Pal 1993 Mesolithic Subsistence in North India: Inferences from Dental Attribu tes. Current Anthropology 34(5):745 765. Luz, Boaz, and Yehoshua Kolodny 1985 Oxygen Isotope Variations in Phosphate of Biogenic Apatites, IV. Mammal Teeth and Bones. Earth and Planetary Science Letters 75(1):29 36. Luz, Boaz, Yehosh ua Kolodny, and Mic hal Horowitz 1984 Fractionation of Oxygen Isotopes between Mammalian Bone Phosphate and Environmen tal Drinking Water. Geochimica e t Cosmochimica Acta 48(8):1689 1693. Lym an, R. L., and MichaelJ O'Brien 2001 The Direct Historical Approach, Analogical R easoning, and Theory in Americanist Archaeology. Journal of Archaeological Method and Theory 8(4):303 342. Mackay, E. J. H. 1938 Further Excavations at Mohenjo Daro. Delhi: Government of India. Madella, Marco 2003 Investigating Agriculture and Enviro nment in South Asia: Present and Future Contributions of Opal Phytoliths. In Indus Ethnobiology: New Perspectives from the Field. Steven A. Weber and William R. Belcher, eds. Pp. 199 250. Lanham: Lexington Books. Madel la, Marco, and Dorian Q. Fuller 200 6 Palaeoecology and the Harappan Civilisation of South Asia: A Reconsideration. Quaternary Science Reviews 25(11 12):1283 1301. Mahadevan, I. 1977 The Indus Script: Texts, Concordance and Tables. New Delhi: Archaeological Survey of India.

PAGE 238

238 Mallah, Q. 2008 Research Report on Surface and Subsurface Analysis of Archaeological Sites Around the Lakes of the Bui Mirway Desert in Sindh, Pakistan. In Linguistics, Archaeology, and the Human Past: Occastional Paper 4. Toshiki Osada and Akinori Uesugi, eds. Pp. 1 81. Kyoto: Indus Project, Research Institute for Humanity and Nature. Manton, W. I., C. R. Angle, K. L. Stanek, D. Kuntzelman, Y. R. Reese, and T. J. Kuehnema nn 2003 Release of Lead from Bone in Pregnancy and Lactation. Environmental Research 92(2):139 151. Manton, W. I., C. R. Angle, K. L. Stanek, Y. R. Reese, and T. J. Kuehnemann 2000 Acquisition and Retention of Lead by Young Children. Environmental Research 82(1):60 80. Marshall, J. H. 1931 Mohenjo Daro and the Indus Civilization. Lon don: Arthur Probsthain. Masson, Charles 1844 Narrative of various Journeys in Balochistan, Afghanistan, the Panjab, & Kalt, during a Residence in those Countries: To which is Added an Account of the Insurrection at Kalat, and a Memoir on Eastern Baloch istan. Vol. 4. London: Richard Bentley. McArthur, J. M., R J. Howarth, and G. A. Shields 2012 Strontium Isotope Stratigraphy. In The Geologic Time Scale 2012. Felix M. Gradstein, James G. Ogg, Mark D. Schmitz and Gabi M. Ogg, eds. Pp. 127 144. Oxford: Elsevier. McBride, Murray B. 1994 Environmental Chemistry of Soils. Oxford: Oxford UniversityPress. McIntosh, Jane R. 2002 A Peaceful Realm: The Rise and Fall of the Indus Civilization. Boulder: Westview Press. 2008 The Ancient Indus Valley: New Perspectives. Santa Barbara: ABC CLIO. Meacham, William 1977 Continuity and Local Evolution in the Neolithic of South China: A Non Nuclear Approach. Current Anthropology 18(3):419 440.

PAGE 239

239 Mea dow, Richard H. 1989 Continuity and Change in the Agricultu re of the Greater Indus Valley. In Old Problems and New Perspectives in the Archaeology of South Asia. Jonathan Mark Kenoyer, ed. Pp. 61 74. Madison: University of Wisconsin. 1989 Prehistoric Wild Sheep and Sheep Domestication on the Eastern Margin of t he Middle East. In Animal Domestication and its Cultural Context. P. J. Crabtree, D. V. Campana and K. Ryan, eds. Pp. 24 36. Philadelphia: University Museum, University of Pennsylvania. 1996 The Origins and Spread of Agriculture and Pastoralism in th e Northwestern South Asia. In The Origins and Spread of Agriculture and Pastoralism in Eurasia. D. Harris, ed. Pp. 390 412. London: UCL Press. Meadow, Richar d H., and Jonathan Mark Kenoyer 1992 Harappa Archaeological Project 1992: End of Season Report. Report submitted to the Director General of Archaeology and Museums, Government of Pakistan, Karachi. 1993 Harappa Archaeological Project: 1993 Excavations. Report submitted to the Director General of Archaeology and Museums, Government of Pakistan, Karachi. 1994 Excavations at Harappa 1993: The City Walls and Inscribed Materials. In South Asian Archaeology, 1993. Asko Parpola and P. Koskikallio, eds. Pp. 451 470. Helsinki: Suo malainen Tiedeakatemia. 1997 Excavations at Harappa 1994 1995: New Perspectives on the Indus Script, Craft Activities and City Organization. In South Asian Archaeology 1995. B. Allchin and F. R. Allchin, eds. Pp. 139 172. New Delhi: Oxford & IBH. 2 001 Recent Discoveries and Highlights from Excavations at Harappa: 1998 2000. Indo Koko Kenkyu [Indian Archaeological Studies] 22:19 36. 2005 Excavations at Harappa 2000 2001: New Insights on Chronology and City Organization. In South Asian Archaeolog y 2001. C. Jarrige and V. Lefevre, eds. Pp. 207 226. Paris: Editions Recherche sur les Civilisations. 2008 Harappa Excavations 1998 1999: New Evidence for the Development and Manifestation of the Harappan Phenomenon. In South Asian Archaeology 1999. E M. Raven, ed. Pp. 85 109. Groningen: Egbert Forsten. Meadow, Richard H., Jonathan M ark Kenoyer, and Rita P. Wright 1994 Harappa Archaeological Research Project: Harappa Excavations 1994. Report submitted to the Director General of Archaeology and Museums, Government of Pakistan, Karachi.

PAGE 240

240 1995 Harappa Archaeological Research Project: Harappa Excavations 1995. Report submitted to the Director General of Archaeology and Museums, Government of Pakistan, Karachi. 1996 Harappa Archaeological Resea rch Project: Harappa Excavations 1996. Report submitted to the Director General of Archaeology and Museums, Government of Pakistan, Karachi. 1997 Harappa Archaeological Research Project: Harappa Excavations 1997. Report submitted to the Director Gene ral of Archaeology and Museums, Government of Pakistan, Karachi. 1998 Harappa Archaeological Research Project: Harappa Excavations 1998. Report submitted to the Director General of Archaeology and Museums, Government of Pakistan, Karachi. 1999 Harappa Archaeological Research Project: Harappa Excavations 1999. Report submitted to the Director General of Archaeology and Museums, Government of Pakistan, Karachi. 2001 Harappa Archaeological Research Project: Harappa Excavations 2000 and 20 01. Report submitted to the Director General of Archaeology and Museums, Government of Pakistan, Karachi. Metcalf Peter, and Richard Huntington 1991 Celebrations of Death: The Anthropology of Mortuary Ritual.Cambridge University Press. Miller, Heather M. L 1991 Urban Palaeoethnobotany at Harappa. In Harappa Excavations 1986 1990: A Multi Disciplinary Approach to 3rd Millennium Urbanism. Richard H. Meadow, ed. Pp. 121 126. Madison: Prehistory Press. 2006 Comparing Landscapes of Transportati on: Riverine Oriented and Land Oriented Systems in the Indus Civilization and the Mughal Empire. In Space and Spatial Analysis in Archaeology. E. C. Robertson, J. D. Seibert, D. C. Fernandez and M. U. Zender, eds. Pp. 281 292. Calgary: University of Cal gary Press and University of New Mexico Press. 2006 Water Supply, Labor Requirements, and Land Ownership in Indus Floodplain Agricultural Systems. In Agriculture and Irrigation in Archaeology. Charles Stanish and Joyce Marcus, eds. Pp. 92 128. Los An geles: Cotsen Institute of Archaeology Press.

PAGE 241

241 2008 Issues in the Determination of Ancient Value Systems: The Role of Talc (Steatite) and Faience in the Indus Civilization. In Intercultural Relations between South and Southwest Asia: Studies in Commemoration of E. C. L. during Caspers (1934 1996). E. Olijdam and R. H. Spoor, eds. Pp. 145 157. Oxford: Archaeopress. Miller, H eidi 2005 A New Interpretation of the Stratigraphy at Chanhudaro. In South Asian Ar chaeology 2001. C. Jarrige and V. Lefevre, eds. Pp. 253 256. Paris: Editions Rech erche s ur les Civilizations. Miller, Laura J. 2003 Secondary Products and Urbanism in South Asia: The Evidence for Traction at Harappa. In Indus Ethnobiology: New Perspectives from the Field. Steven A. Weber and William R. Belcher, eds. Pp. 251 326. Lanham: Lexington Books. 2004 Urban Economies in Early States: The Secondary Products Revolution in the Indus Civilization. Ph.D. dissertation, Department of Anthrop ology, New York University. Miller, Daniel 1985 Ideology and the Harappan Civilization. Journal of Anthropological Archaeology 4(1):34 71. Montgomery, Janet 2002 Lead and Strontium Isotope Compositions of Human Dental Tissues as an Indicator of Anci ent Exposure and Population Dynamics. Ph.D. dissertation, Department of Archaeological Sciences, University of Bradford. Montgomery, Janet, Jane A. Evans, and Rachel E. Cooper 2007 Resolving Archaeological Populations with Sr Isotope Mixing Models. Appl ied Geochemistry 22(7):1502 1514. Montgomery, Janet, Jane A. Evans, Dominic Powle sland, and Charlotte A. Roberts 2005 Continuity Or Colonization in Anglo Saxon England? Isotope Evidence for Mobility, Subsistence Practice, and Status at West Heslerton. American Journal of Physical Anthropology 126(2):123 138. Morris, Ian 1991 The Archaeology of Ancestors: The Saxe/Goldstein Hypothesis Revisited. Cambridge Archaeological Journal 1(2):147 169. Morrison, Kathleen D. 1994 The Intensification of Product ion: Archaeological Approaches. Journal of Archaeological Method and Theory 1(2):111 159.

PAGE 242

242 Moseley, Michael E. 1975 The Maritime Foundations of Andean Civilization. Menlo Park: Cummings. Mughal, M. Rafique 1968 Harappa 1966 (Cemetery R 37). Pakistan A rchaeology 5:63 68. 1982 Recent Archaeological Research in the Cholistan Desert. In Harappan Civilization: A Contemporary Perspective. Gregory L. Possehl, ed. Pp. 85 95. Delhi: Oxford & IBH & the American Institute of Indian Studies. 1990 Further E vidence of the Early Harappan Culture in the Greater Indus Valley: 1971 90. South Asian Studies 6(1):175 199. 1992 Jhukar and the Late Harappan Cultural Mosaic of the Greater Indus Valley. In South Asian Archaeology 1989. C. Jarrige, ed. Pp. 213 222. Madison: Prehistory Press. 1994 Ancient Cities of the Indus. Lahore Museum Bulletin 7(1 2):53 59. 1994 The Harappan Nomads of Cholistan. In Living Traditions: Studies in the Ethnoarchaeolog y of South Asia. B. Allchin, ed. Pp. 53 68. New Delhi: Oxford and IBH. 1997 Ancient Cholistan: Archaeology and Architecture. Lahore: Ferozsons. Najman Y., M. Bickle, and H. Chapman 2000 Early Himalayan Exhumation: Isotopic Constraints from the Indi an Foreland Basin. Terra Nova 12:28 34. Nandan, K. R. R., B. K. D. Rao, and M. L. Singhal 1981 Exploration for Copper, Lead and Zinc in India. Bulletins of the Geological Survey of India, Series A: Economic Geology 47:1 197. Nath, Amarendra 1998 Rakh igarhi: A Harappan Metropolis in the Saraswati Drishadvati Divide. Puratattva 28:39 45. 1999 Further Excavations at Rakhigarhi. Puratattva 29:46 49. 2001 Rakhigarhi: 1999 2000. Puratattva 31:43 45. Naum, M. 2010 Re Emerging Frontiers: Postcolonial Theory and Historical Archaeology of the Borderlands. Journal of Archaeological Method and Theory 17(2):101 131. O'Leary, Marion H. 1988 Carbon Isotopes in Photosynthesis. Bioscience 38(5):328 336.

PAGE 243

243 O'Shea, John M. 1984 Mortuary Variability: An Archae ological Investigation. Orlando: Academic Press, Inc. Paleo Labo AMS Dating Group 2011 AMS Radiocarbon Dating. In Excavations at Farmana: District Rohtak, Haryana, India, 2006 2008. Vasant Shinde, Toshiki Osada and Manmohan Kumar, eds. Pp. 831 838. Kyot o: Indus Project, Research Institute for Humanity and Nature. Palmer M. R., and J. M. Edmond 1992 Controls Over the Strontium Isotope Composition of River Water. Geochimica e t Cosmochimica Acta 56(5):2099 2111. Pande, Kanchan, M. M. Sarin, J. R. Trivedi, S. Krishnaswami, and K. K. Sharma 1994 The Indus River System (India Pakistan): Major Ion Chemistry, Uranium and Strontium Isotopes. Chemical Geology 116(3 4):245 259. Parkes, Peter 2001 Alternative Social Structures and Foster Relations in t he Hindu Kush: Milk Kinship Allegiance in Former Mountain Kingdoms of Northern Pakistan. Comparative Studies in Society and History 43(1):pp. 4 36. Parry, Carol, and Joseph Eaton 1991 Kohl: A Lead Hazardous Eye Makeup from the Third World to the First W orld. Environmental Health Perspectives 94:121 123. Passey, Benjamin H., Todd F. Robinson, Linda K. Ayliffe, Thure E. Cerling, Matt Sponheimer, M. Denise Dearing, Beverly L. Roeder, and James R. Ehleringer 2005 Carbon Isotope Fractionation between Diet, Breath CO 2 and Bioapatite in Different Mammals. Journal of Archaeological Science 32(10):1459 1470. Pearson, Parker 1982 Mortuary Practices, Society, and Archaeology: An Ethnoarchaeological Study. In Symbolic and Structural Archaeology. Ian Hodder, ed. Pp. 99 113. Cambridge: Cambridge University Press. Peebles, Christopher S., and Susan M. Kus 1977 Some Archaeological Correlates of Ranked Societies. American Antiquity 42(3, Essays on Archaeological Problems):421 448. Pend all, Elise, and Ronald Am undson 1990 Soil/Landform Relationships Surrounding the Harappa Archaeological Site, Pakistan. Geoarchaeology 5(4):301 322.

PAGE 244

244 Peters, S. G., S. D. Ludington, G. J. Orris, D. M. Sutphin J. D. Bliss, and J. J. Rytuba 2007 Preliminary Non Fuel Mineral Re source Assessment of Afghanistan. United States Geological Survey Open File Report 2007 1214. Pin, Chri stian, and Chantal Bassin 1992 Evaluation of a Strontium Specific Extraction Chromatographic Method for Isotopic Analysis in Geological Materials. Ana lytica Chimica Acta 269(2):249 255. Pollard, A. M., M. Pellegrini, and J. A. Lee Thorp 2011 Technical Note: Some Observations on the Conversion of Dental Enamel 18 O p Values to 18 O w to Determine Human Mobility. American Journal of Physical Anthropology 145(3):499 504. Pollock, Amanda A. 2008 Change and Continuity in the Indus Valley Cultural Tradition as Manifested by Mortuary Practices. M.A. thesis, Department of Anthropology, Un iversity of Toronto. Pollock, Susan 1991 Of Priestesses, Princes and Poor Relations: The Dead in the Royal Cemetery of Ur. Cambridge Archaeological Journal 1(2):171 189. Possehl, Gregory L 1977 The End of a State and the Continuity of a Tradition. In Realm and Region in Traditional India. R. Fox, ed. Pp. 234 254. Durham: Carolina Academic Press. 1979 Pastoral Nomadism in the Indus Civilization: An Hypothesis. In South Asian Archaeology 1977. M. Taddei, ed. Pp. 537 551. Naples: Instituto Universi tario Orientale. 1980 Indus Civilization in Saurashtra. Delhi: B. R. Publishing Corporation. 1992 The Harappan Civilization in Gujarat: The Sorath and Sindhi Harappans. Eastern Anthropologist 45(1 2):117 154. 1992 The Harappan Cultural Mosaic: E cology Revisited. In South Asian Archaeology 1989. C. Jarrige, ed. Pp. 237 244. Madison: Prehistory Press. 1993 Harappan Civilization: A Contemporary Perspective. In Harappan Civilization: A Recent Perspective. Gregory L. Possehl, ed. Pp. 15 28. New D elhi: Oxford and IBH. 1997 The Transformation of the Indus Civilization. Journal of World Prehistory 11(4):425 472. 1999 Indus Age: The Beginnings. Philadelphia: University of Pennsylvania Press.

PAGE 245

245 2002 Harappans and Hunters: Economic Interaction and Specialization in Prehistoric India. In Forager Traders in South and Southeast Asia. Kathleen D. Morrison and L. L. Junker, eds. Pp. 62 76. Cambridge: Cambridge University Press. 2002 The Indus Civilization: A Contemporary Perspective. Walnut Creek: Altamira Press. Posseh l, Gregory L., and C. F. Herman 1990 The Sorath Harappan: A New Regional Manifestation of the Indus Urban Phase. In South Asian Archaeology 1987. M. Taddei, ed Pp. 295 320. Roma: Instituto Italiano per il Medio ed Estremo Oriente. Possehl, Gregor y L., and Kenneth A. R. Kennedy 1979 Hunter Gatherer/Agriculturalist Exchange in Prehistory: An Indian Example. Current Anthropology 20(3):592 593. Prabhakar, V. N. 2012 Burial Practices of the Harappans: Sanauli Excavations, A Case Study. Ph.D. dissertation, Department of Ancient Indian History, Culture, and Archaeology, Kurukshetra University. Price, T. D ouglas ., J ames H. Burton, and R. A lexander Bentley 2002 The Characterization of Biologically Available Strontium Isotope Ratios for the Study of Prehistoric Migration. Archaeometry 44(1):117 135. Price, T. Douglas, James H. Burton, Robert J. Sharer, Jane E. Buikstra, Lori E. Wright, Loa P. T raxler, and Katheri ne A. Miller 2010 Kings and Commoners at Copan: Isotopic Evidence for Origins and Movement in the Classic Maya Period. Journal of Anthropological Archaeology 29(1):15 32. Prowse, Tracy L., Henry P. Schwarcz, Peter Garnsey, Martin Knyf, Roberto Macchiare lli, and Luca Bondioli 2007 Isotopic Evidence for Age Related Immigration to Imperial Rome. American Journal of Physical Anthropology 132(4):510 519. Qu, Chang Sheng, Zong Wei Ma, Jin Yang, Yang Liu, Jun Bi, and Lei Huang 2012 Human Exposure Pathways of Heavy Metals in a Lead Zinc Mining Area, Jiangsu Province, China. PLoS ONE 7(11):e46793. Ramakri shnan, M., and R. Vaidhyanadhan 2008 Geology of India. Bangalore: Geological Society of India. Ramesh, R., and M. M. Sarin 1992 Stable Isotope Study of the Ganga (Ganges) River System. Journal of Hydrology 139(1 4):49 62.

PAGE 246

246 Rao, L. S., N. B. Sahu, Samir Diwan, and U. A. Shastry 2004 New Light on the Excavations of the Harappan Settlement at Bhirrana. Puratattva 35:60 68. Rao, Shikaripur Ranganatha 19 73 Lothal and the Indus Valley Civilization. Bombay: Asia Publishing House. 1979 Lothal: A Harappan Port Town (1955 62). Vol. 1. New Delhi: Archaeological Survey of India. 1985 Lothal: A Harappan Port Town (1955 62). Vol. 2. New Delhi: Archaeologica l Survey of India. Ratnagar, Shereen 2001 Understanding Harappa: Civilization in the Greater Indus Valley. New Delhi: Tulika. Reid, D. J., and M. C. Dean 2006 Variation in Modern Human Enamel Formation Times. Journal of Human Evolution 50(3):329 346. Renfrew, Colin 1972 The Emergence of Civilisation: The Cyclades and the Aegean in the Third Millennium B.C. London: Methuen. 1973 The Explanation of Culture Change: Models in Prehistory. London: Duckworth. 2000 Archaeogenetics: Towa rds a Population Prehistory of Europe. In Archaeogenetics: DNA and the Population Prehistory of Europe. K. Boyle, ed. Pp. 3 11. Cambridge: McDonald Institute Monographs. Reynolds, Amanda C., Jay Quade, and Julio L. Betancourt 2012 Strontium Isotopes a nd Nutrient Sourcing in a Semi Arid Woodland. Geoderma 189 190:574 584. Richards, Michael P., Paul B. Pettitt, Erik Trinkaus, Fred H. Smith, Ma ja Paunovic, and Ivor Karavanic 2000 Neanderthal Diet at Vindija and Neanderthal Predation: The Evidence from Stable Isotopes. Proceedings of the National Academy of Sciences 97(13):7663 7666. Rizvi, Uzma Z. 2007 Configuring the Space in between: Redefining the Ganeshwar Jodhpura Cultural Complex in Northeastern Rajasthan, India. Ph.D. dissertation, Department of Anthropology, University of Pennsylvania.

PAGE 247

247 Robbins Schug, Gwen, Kelsey Gray, V. Mushr if Tripathy, and A. R. Sankhyan 2012 A Peaceful Realm? Trauma and Social Differentiation at Harappa. International Journal of Paleopathology 2(2 3):136 147. Roscoe, Robert J., Janie L. Gittleman, James A. Deddens, Martin R. Pe tersen, and William E. Halperin 1999 Blood Lead Levels among Children of Lead Exposed Workers: A Meta Analysis. American Journal of Industrial Medicine 36(4):475 481. Rozanski, K., L. Arag us Aragus, and R. Gonfiantini 1993 Isotopic Patterns in Modern Global Precipitation. In Climate Change in Continental Isotopic Records. P. K. Swart, K. L. Lohmann, J. McKenzie and S. Savin, eds. Pp. 1 36. Washington: American Geophysical Union. Sahni, Rai Bahadur Daya Ram 1921 Excavations at Harappa. In Annual Progress Report of the Superintendent, Hindu and Buddhist Monuments, Northern Circle, for the Year Ending 31st March 1921. Pp. 8 26. Lahore: Archaeological Survey of India. 1926 Exploration and Research, Northern Circle, Punjab, Harappa. In Annual Report of the Archaeological Survey of India, 1923 24. J. H. Marshall, ed. Pp. 52 54. Calcutta: Government of India Press. 1927 Exploration and Research, Northern Circle, Harappa. In Annual Report of the Archaeological Survey of India, 1924 25. J. F. Blakiston, ed. Pp. 73 80. Calcutta: Government of India Press. Sai ni, H. S., and S. A. I. Mujtaba 2012 Depositional History and Palaeoclimatic Variations at the Northeastern Fringe of Thar Desert, Haryana Plains, India. Quaternary International 250:37 48. Sana Ullah, K. B. M. 1940 The Sources, Composition, and Technique of Copper and its Alloys. In Excavations at Harappa. M. S. Vats, ed. Pp. 378 382. New Delhi: Government of India. Sastri, K. N. 1965 New Light on the Indus Civilization. Vol. 2. Delhi: Atma Ram and Sons. Saxe, Arthur A. 1970 Social Dimensions of Mortuary Practices in a Mesolithic Population from Wadi Halfa, Sudan. Ph.D. dissertation, Department of Anthropology, University of Michigan.

PAGE 248

248 1971 Social Dimensions of Mortuary Practices in a Mesolithic Population from Wadi Halfa, Sudan. In Approaches to the Social Dimensions of Mortuary Practices. James A. Brown, ed. Pp. 39 57. Washington: Society for American Archaeology. Scarborough, Vernon L. 2003 The Flow of Power: Ancient Water Systems and Landscapes. Santa Fe: School of American Research Press. Schoeninger, M. J., M. J. Kohn, W. John, and J. W. Va lley 2000 Tooth Oxygen Isotope Ratios as Paleoclimate Monitors in Arid Ecosystems. In Biogeochemical Approaches to Paleodietary Analysis in Archaeology. Stanley H. Ambrose and M. A. Katzenberg, eds. Pp. 117 140. New York, NY: Plenum Publishers. Scholl, Martha A., James B. Shanley, Jan Pa ul Zegarra, and Tyler B. Coplen 2009 The Stable Isotope Amount Effect: New Insights from NEXRAD Echo Tops, Luquillo Mountains, Puerto Rico. Water Resources Research 45(12): W12407. Schroeder, Hannes, Tamsin C. O'Connell, Jan e A. Evans, Kristrina A. Shuler, and Robert E. M. Hedges 2009 Trans Atlantic Slavery: Isotopic Evidence for Forced Migration to Barbados. American Journal of Physical Anthropology 139(4):547 557. Schuyler, Robert L. 1968 The U se of Historic Analogs in Archaeology. American Antiquity 33(3):390 392. Schwarcz, Henry P. 2002 Some Biochemical Aspects of Carbon Isotopic Paleodiet Studies. In Biogeochemical Approaches to Paleodietary Analysis. Stanley H. Ambrose and M. Anne Katzenberg, eds. Pp. 189 209. Ne w York: Kluwer Academic. Schwarcz, Henry P., L. Gibbs, and M. Knyf 1993 Oxygen Isotope Analysis of Bones from Snake Hill as Indicator of Place of Origin. In Snake Hill Burial, an Investigation of a Military Cemetery from the War of 1812. S. Pfeiffer and R. Williamson, eds. Pp. 263 268. Toronto: Dundurn Press. Sengupta, Saikat, and A. Sarkar 2006 Stable Isotope Evidence of Dual (Arabian Sea and Bay of Bengal) Vapour Sources in Monsoonal Precipitation Over North India. Earth and Planetary Science Letter s 250(3 4):511 521. Shaffer, J. G. 1992 The Indus Valley, Baluchistan and Helmand Traditions: Neolithic through Bronze Age. In Chronologies in Old World Archaeology, Volume 1. R. Ehrich, ed. Pp. 441 464. Chicago: University of Chicago Press.

PAGE 249

249 Shaffer, J im G., and Diane A. Lichtenstein 1989 Ethnicity and Change in the Indus Valley Cultural Tradition. In Old Problems and New Perspectives in the Archaeology of South Asia. Jonathan Mark Kenoyer, ed. Pp. 117 126. Madison: Wisconsin Archaeological Reports. 1999 Migration, Philology and South Asian Archaeology. In Aryan and Non Aryan in South Asia. J. Bronkhorst and M. Deshpande, eds. Pp. 239 260. Cambridge: Harvard Oriental Series. Sharma, A. K. 1999 The Departed Harappans of Kalibangan. Delhi: Sundeep Prakashan. Sharma, D. V., K. C Nauriyal, and V. N. Prabhakar 2005 Sanauli: A Late Harappan Burial Site in the Yamuna Hindon Doab. Puratattva 34:35 44. 2007 Excavations at Sanauli 2005 06: A Harappan Necropolis in the Upper Ganga Yamuna Doab Puratattva 36:166 183. Shinde, Vasant 2011 Harappan Cemetery at Farmana. In Excavations at Farmana: District Rohtak, Haryana, India, 2006 2008. Vasant Shinde, Toshiki Osada and Manmohan Kumar, eds. Pp. 550 673. Kyoto: Indus Project, Research Institute for Humanity and Nature. 2011 Site and its Environs. In Excavations at Farmana: District Rohtak, Haryana, India, 2006 2008. Vasant Shinde, Toshiki Osada and Manmohan Kumar, eds. Pp. 5 19. Kyoto: Indus Project, Research Institute for Humanity and Natu re. 2011 Structural Remains in the Settlement Area. In Excavations at Farmana: District Rohtak, Haryana, India, 2006 2008. Vasant Shinde, Toshiki Osada and Manmohan Kumar, eds. Pp. 41 167. Kyoto: Indus Project, Research Institute for Humanity and Nature. Shinde, Vasant, Toshiki Osada, and Manmohan Kumar eds. 2011 Excavations at Farmana: District Rohtak, Haryana, India 2006 2008. Kyoto: Indus Project, Research Institute for Humanity and Nature. 2011 Excavations at Girawad: District Rohtak, H aryana, India, 2006. Kyoto: Indus Project, Research Institute for Humanity and Nature.

PAGE 250

250 Shinde, Vasant, Toshiki Osada, M. M. Sharma, Akinori Uesugi, T. Uno, H. Maemoku, P. Shirvalkar, S. S. Deshpande, A. Kulkarni, A. Sarkar, A. Reddy, V. Rao, Vivek Dan gi, and Manmohan Kumar 2008 Exploration in the Ghaggar Basin and Excavations at Girawad, Farmana (Rohtak District) and Mitathal (Bhiwani District), Haryana, India. In Occasional Paper 3: Linguistics, Archaeology and the Human Past. Toshiki Osada and Akin ori Uesugi, eds. Pp. 77 158. Kyoto: Indus Project, Research Institute for Humanity and Nature. Shinde, Vasant, Toshiki Osada, Aki nori Uesugi, and Manmohan Kumar 2008 A Report on Excavations at Farmana 2007 08. In Occasional Paper 6: Linguistics, Archaeo logy and the Human Past. Toshiki Osada and Akinori Uesugi, eds. Pp. 1 116. Kyoto: Indus Project, Research Institute for Humanity and Nature. Sillen, Andrew 1986 Biogenic and Diagenetic Sr/Ca in Plio Pleistocene Fossils of the Omo Shungura Formation. Paleobiology 12(3):311 323. 1989 Diagenesis of the Inorganic Phase of Cortical Bone. In The Chemistry of Prehistoric Human Bone. T. Douglas Price, ed. Pp. 211 229. Cambridge: Cambridge University Press. Sillen, Andrew, Grant Hall, Stephen Ri chardson, and Richard Armstrong 1998 87 Sr/ 86 Sr Ratios in Modern and Fossil Food Webs of the Sterkfontein Valley: Implications for Early Hominid Habitat Preference. Geochimica e t Cosmochimica Acta 62(14):2463 2473. Silverman, Helaine 2002 Introduction: The Spac e and Place of Death. Archeological Papers of the American Anthropological Association 11(1):1 11. Singh, Upinder 2008 A History of Ancient and Early Medieval India: From the Stone Age to the 12th Century. Upper Saddle River: Pearson Education. Smith, G. 1997 Political Units of the Mature Period Harappan Civilization. M.A. thesis, Department of Anthropology, University of Pittsburgh. Sonawane, V. H. 1992 Fresh Light on the Specialized Crafts of the Harappans in Gujarat. The Eastern Anthropologist 4 5(1 2):155 172. Sponheime r, Matt, and Julia A. Lee Thorp 1999 Oxygen Isotopes in Enamel Carbonate and their Ecological Significance. Journal of Archaeological Science 26(6):723 728.

PAGE 251

251 2001 The Oxygen Isotope Composition of Mammalian Enamel Carbonate from Morea Estate, South Africa. Oecologia 126(2):153 157. Stacul, Giorgio 1992 Further Evidence for" the Inner Asia Complex" from Swat. In South Asian Archaeology Studies. Gregory L. Po ssehl, ed. Pp. 111 122. New Delhi: Oxford & IBH Publishing Co. 1994 Neolithic Inner Asian Traditions in Northern Indo Pakistani Valleys. In South Asian Archaeology 1993. Asko Parpola and P. Koskikallio, eds. Pp. 707 714. Helsinki: Suomalainen Tiedeaka temia. Stein, Aurel 1931 An Archaeological Tour in Gedrosia. New Delhi: Archaeological Survey of India. 1937 Archaeological Reconnaissances in Northwestern India and Southeastern Iran. London: Macmillan. Stein, Gil J. 2002 From Passive Periphery to Active Agents: Emerging Perspectives in the Archaeology of Interregional Interaction. American Anthropologist 104(3):903 916. Strathern, Marilyn 1996 Cutting the Network. The Journal of the Royal Anthropological Institute 2(3):517 535. Sugiyama, Shinji 2011 Phytolith Analysis at the Farmana Site of the Indus Civilization. In Excavations at Farmana: District Rohtak, Haryana, India, 2006 2008. Vasant Shinde, Toshiki Osada and Manmohan Kumar, eds. Pp. 826 830. Kyoto: Indus Project, Research Institu te for Humanity and Nature. Thomas, K. D., and J. R. Knox 1994 Routes of Passage: Later Prehistoric Settlement and Exploitation of a Frontier Region in Northwestern Pakistan. Bulletin of the Institute of Archaeology, University College London 31:89 104. Thussu, J. L. 1995 Quaternary Stratigraphy and Sedimentation of the Indo Gangetic Plains, Haryana. Journal of the Geological Society of India 46:533 544. Tieszen, L. L., and T. Fagre 1993 Effect of Diet Quality and Composition on the Isotopic Compos ition of Respiratory CO 2 Bone Collagen, Bioapatite, and Soft Tissues. In Prehistoric Human Bone: Archaeology at the Molecular Level. J. B. Lambert and G. Grupe, eds. Pp. 121 155. New York: Springer Verlag.

PAGE 252

252 Tie szen, Larry L. 1991 Natural Variations in th e Carbon Isotope Values of Plants: Implications for Archaeology, Ecology, and Paleoecology. Journal of Archaeological Science 18(3):227 248. Tosi, Maurizio 1993 The Harappan Civilization Beyond the Indian Subcontinent. In Harappan Civilization: A Recent Perspective. Gregory L. Possehl, ed. Pp. 365 378. New Delhi: Oxford and IBH. Touzeau, Alexandra, Janne Blichert Toft, Romain Amiot, Franois Fourel, Franois Martineau, Jenefer Cockitt, Keith Hall, Jean Pierre Flan drois, and Christophe Lcuyer in press Egyptian Mummies Record Increasing Aridity in the Nile Valley from 5500 to 1500 yr before Present. Earth and Planetary Science Letters. Trickett, Mark A., Paul Budd, J anet Montgomery, and Jane Evans 2003 An Assessment of Solubility Profiling as a Decon tamination Procedure for the 87 Sr/ 86 Sr Analysis of Archaeological Human Skeletal Tissue. Applied Geochemistry 18(5):653 658. Trigger, Bruce G. 2006 A History of Archaeological Thought. 2nd ed. Cambridge: Cambridge University Press. Tripathi, Jayant K., Barbara Bock, V. Rajamani, and A. Eisenhauer 2004 Ca and Sr Dynamics in the Indo Gangetic Plains: Different Sources and Mobilization Processes in Northwestern India. Current Science 87(10):1453 1458. Tripathi, Jayant K., Bar bara Bock, and V. Rajamani 2013 Nd and Sr Isotope Characteristics of Quaternary Indo Gangetic Plain Sediments: Source Distinctiveness in Different Geographic Regions and its Geological Significance. Chemical Geology 344:12 22. Tripathi, Jayant K., P. Ghazanfari V. Rajamani, and S K. Tandon 2007 Geochemistry of Sediments of the Ganges Alluvial Plains: Evidence of Large Scale Sediment Recycling. Quaternary International 159(1):119 130. Tripa thi, Jayant K., and V. Rajamani 1999 Geochemistry of the Loessic Sediments on Delhi Rid ge, Eastern Thar Desert, Rajasthan: Implications for Exogenic Processes. Chemical Geology 155(3 4):265 278. Trivedi, Pramod Kumar 2009 Excavations at Tarkhanewala Dera and Chak 86, 2003 2004. Delhi: Archaeological Survey of India.

PAGE 253

253 Ucko, Peter J. 1969 Ethnography and Archaeological Interpretation of Funerary Remains. World Archaeology 1(2):262 280. Uesugi, Akinori 2011 Pottery from the Cemetery. In Excavations at Farmana: District Rohtak, Haryana, India, 2006 2008. Vasant Shinde, Toshiki Osada and Ma nmohan Kumar, eds. Pp. 674 800. Kyoto: Indus Project, Research Institute for Humanity and Nature. 2011 Pottery from the Settlement Area. In Excavations at Farmana: District Rohtak, Haryana, India, 2006 2008. Vasant Shinde, Toshiki Osada and Manmohan Ku mar, eds. Pp. 168 368. Kyoto: Indus Project, Research Institute for Humanity and Nature. Valentine, Benjamin, Georg e D. Kamenov, and John Krigbaum 2008 Reconstructing Neolithic Groups in Sarawak, Malaysia through Lead and Strontium Isotope Analysis. Journal of Archaeological Science 35(6):1463 1473. van der Merwe, Nikolaas J., and Ernesto Medina 1989 Photosynthesis and 13 C / 12 C Ratios in Amaz onian Rain Forests. Geochimica e t Cosmochimica Acta 53(5):1091 1094. 1991 The Canopy Effect, Carbon Isotope Ratios and Foodwebs in Amazonia. Journal of Archaeological Science 18(3):249 259. Vats, M. S. 1933 Excavations at Harappa. In Annual Report of the Archaeological Survey of India, 1928 29. H. Hargreaves, ed. Pp. 76 83. Delhi: Manager of Publications. 1935 Excavations at Harappa. In Annual Report of the Archaeological Survey of India, 1929 30. Sahni, Rai Bahadur Daya Ram, ed. Pp. 21 31. Delhi: Manager of Publications. 1936 Excavations at Harappa. In Annual Report of the Archaeological Survey of India, for the Y ears1930 31, 1931 32, 1932 33, and 1933 34. C. L. Fabri, ed. Pp. 72 90. Delhi: Manager of Publications. 1940 Excavations at Harappa. Delhi: Government of India Press. Vidale, Massimo 1989 Specialized Producers and Urban Elites: On the Role of Craft Industries in Mature Harappan Urban Contexts. In Old Problems and New Perspectives in the Archaeology of South Asia. Jonathan Mark Kenoyer, ed. Pp. 171 182. Madison: Department of Anthropolo gy, University of Wisconsin, Madison. 2000 The Archaeology of Indus Craf ts: Indus Craftspeople and Why W e Study Them. Rome: IsIAO.

PAGE 254

254 Vidale, Massimo 1992 Produzione Artigianale P rotostorica: Ethnoarchacologia e Archeologia. Pa dua: Universita degli Stud i di P adova. Vidale, Ma ssimo, and Heather M. L Miller 2000 On the Development of Indus Technical Virtuosity and its Relation to Social Structure. In South Asian Archaeology 1997. M. Taddei, ed. Pp. 115 132. Rome: IsIAO. Wanzke, H. 1984 Axis Systems and Orientation at Mohenjo Daro. In Interim Reports, Volume 2: Reports on Field Work Carried Out at Mohenjo Daro, Pakistan 1983 84 by the IsMEO Aachen University Mission. Michael Jansen and Gunter Urban, eds. Pp. 23 32. Aachen: Forschungsprojekt "Mohenjo D aro". Wasserman, R. H., F W. Lengemann, and C. L. Comar 1958 Comparative Metabolism of Calcium and Strontium in Lactation. Journal of Dairy Science 41(6):812 821. Weber, Steven A. 1999 Seeds of Urbanism: Palaeoethnobotany and the Indus Civilization. Antiquity 73:813 826. 2003 Archaeobotany at Harappa: Indications for Change. In Indus Ethnobiology.New Perspectives from the Field. Steven A. Weber and William R. Belcher, eds. Pp. 175 198. Lanham: Lexington Books. Weber, Steven A., Tim Barela, and Heather Lehman 2010 Ecological Continuity: An Explanation for Agricultural Diversity in the Indus Civilization and Beyond. Man and Environment 35(1):62 75. Weber, Steven A. and William R. Belcher, eds. 2003 Indus Ethnobiology: New Perspectives from the Field. Lanham: Lexington Books. Weber, Steven A., Ar unima Kashyap, and Laura Mounce 2011 Archaeobotany at Farmana: New Insights into Harappan Plant use Strategies. In Excavations at Farmana: District Rohtak, Haryana, India, 2006 2008. Vasant Shinde, Toshiki Osada and Manmohan Kumar, eds. Pp. 808 825. Kyoto: Indus Project, Research Institute for Humanity and Nature. Weeks, Lloyd R. 2004 Early Metallurgy of the Persian Gulf: Technology, Trade, and the Bronze Age World. Boston: Brill Academic Publish ers.

PAGE 255

255 Weiss Krejci, Estella 2011 The Formation of Mortuary Deposits: Implications for Understanding Mortuary Behavior of Past Populations. In Social Bioarchaeology. Sabrina C. Agarwal and Bonnie A. Glencross, eds. Pp. 68 106. Chichester: Wiley Blackwell. Wenke, Robert J. 2009 The Ancient Egyptian State: The Origins of Egyptian Culture (C. 8000 2000 B.C.). Cambridge: Cambridge University Press. Weyhenmeyer, Constanze E., Stephen J. Burns, H. Niklaus Waber, Werner Aeschbach Hertig, Rolf Kipfer, H. Hugo Loosli, and Albert Matter 2000 Cool Glacial Temperatures and Changes in Moisture Source Recorded in Oman Groundwaters. Science 287(5454):842 845. Weyhenmeyer, Constanze E., Stephen J. Burns, H. Niklaus Waber, Phil G. Macumber, and Albert Matter 2002 Isotope Study of Moisture Sources, Recharge Areas, and Groundwater Flow Paths within the Eastern Batinah Coastal Plain, Sultanate of Oman. Water Resources Research 38(10):1184. Wheeler, R. E. Mortimer 1947 Harappa 1946: The Defences and C emetery R 37. Ancient India 3:58 130. 1968 The Indus Civilization. 3rd ed. Cambridge: Cambridge University Press. Whipkey, C. E., R. C. Capo, O. A. Chadwick, and B. W. Stewart 2000 The Importance of Sea Spray to the Cation Budget of a Coastal Hawaiian Soil: A Strontium Isotope Approach. Chemical Geology 168(1 2):37 48. White, Christine D., Rebecca Storey, Fred J. Longstaffe, and Michael W. Spe nce 2004 Immigration, Assimilation, and Status in the Ancient City of Teotihuacan: Stable Isotopic E vidence from Tlajinga 33. Latin American Antiquity 15(2):176 198. White, Christine, Fred J. L ongstaffe, and Kimberley R. Law 2004 Exploring the Effects of Environment, Physiology and Diet on Oxygen Isotope Ratios in Ancient Nubian Bones and Teeth. Journ al of Archaeological Science 31(2):233 250. Willey, Gordon R., and Philip Phillips 1958 Method and Theory in American Archaeology. Chicago: University of Chicago Press. Wright, Rita 1991 Patterns of Technology and the Organization of Production at Harappa. In Harappa Excavations 1986 1990: A Multidisciplinary Approach to Third Millennium Urbanism. Richard H. Meadow, ed. Pp. 71 88. Madison: Prehistory Press.

PAGE 256

256 2010 The Ancient Indus: U rbanism, Economy, and Society. Cambridge: Cambridge University Press. 2013 Commodoties and Things: The Kulli in Context. In Connections and Complexity: New Approaches to the Archaeology of South Asia. Shinu Abraham, Praveena Gullapalli, Teresa P. Rac zek and Uzma Z. Rizvi, eds. Pp. 47 62. Walnut Creek: Left Coast Press. Wright, Rita P., R. Bryson, and J. Schuldenrein 2008 Water Supply and History: Harappa and the Beas Regional Survey. Antiquity 82:37 48. Wright, Lori E. 2005 Identifying Immigran ts to Tikal, Guatemala: Defining Local Variability in Strontium Isotope Ratios of Human Tooth Enamel. Journal of Archaeological Science 32(4):555 566. 2012 Immigration to Tikal, Guatemala: Evidence from Stable Strontium and Oxygen Isotopes. Journal of Anthropological Archaeology 31(3):334 352. Wright, Lori E., and Henry P. Schwarcz 1998 Stable Carbon and Oxygen Isotopes in Human Tooth Enamel: Identifying Breastfeeding and Weaning in Prehistory. American Journal of Physical Anthropology 106(1):1 18. Zahran, Sammy, Mark A. S. Laidlaw, Shawn P. McElmurry, Gabriel M. Filippelli, and Mark Taylor 2013 Linking Source and Effect: Resuspended Soil Lead, Air Lead, and Children's Blood Lead Levels in Detroit, Michigan. Environmental Science & Technology 47(6) :2839 2845.

PAGE 257

257 BIOGRAPHICAL SKETCH Benjamin Valentine was born in Denton, Texas to a military family. At a young age s travels, sparking an interest in human cultural variation that has underscored (Benjamin earned his B.S. in telecommunication from the University of Florida in 2003), Benjamin eventually returned to his intell ectual passi ons, matriculating to the University of Florida Department of Anthropology in 2005. He earned his M.A. in 2007 and his Ph.D. in 2013. He looks forward to continuing his international research and raising his two children as his family raised him ever curio us to know more about a world bigger than ourselves and the many ways we might fit into it.