Bioarchaeology of the Near East 1 (2007) 9781463222505

Citation preview

Bioarchaeology of the Near East

Bioarchaeology of the Near East

Volume 1 (2007) Edited by

Arkadiusz Sołtysiak

9

34 2009

Editor Arkadiusz Sołtysiak Department of Historical Anthropology, Institute of Archaeology, University of Warsaw ul. Krakowskie Przedmieście 26/28, 00-927 Warszawa, Poland email: [email protected]; phone: +48 225522845; fax: +48 225522801

Editorial Board Anagnostis Agelarakis Department of Anthropology, Adelphi University, New York, USA Andrea G. Drusini Department of Biology, University of Padova, Italy Maciej Henneberg Wood Jones Chair of Anthropological and Comparative Anatomy Medical School, University of Adelaide, Australia Theya Molleson Natural History Museum, London, UK Janusz Piontek Institute of Anthropology, Adam Mickiewicz University in Poznań, Poland Friedrich W. Rösing Institute of Human Genetics and Anthropology, Ulm University, Germany Holger Schutkowski Archaeological Sciences, School of Life Sciences, University of Bradford, UK Patricia Smith Department of Anatomy and Cell Biology, Faculty of Medicine The Hebrew University of Jerusalem, Israel Douglas H. Ubelaker National Museum of Natural History, Smithsonian Institution, Washington DC, USA Alina Wiercińska State Archaeological Museum in Warsaw, Poland

Copyright © Bioarchaeology of the Near East and the Authors

www.anthropology.uw.edu.pl Hosted by the Department of Historical Anthropology, Institute of Archaeology, University of Warsaw Printed in Poland

ISSN 1898-9403 Bioarchaeology of the Near East is published annually in one volume. The aim of the journal is to promote research on the history of human populations inhabiting the areas of Southwestern Asia (chiefly Mesopotamia, Syria, Palestine, Anatolia, Iran, and Egypt). It will publish original contributions in which methods of physical anthropology and bioarchaeology are used to answer historical questions. Three kinds of texts will be considered for publication: original papers, general review articles (especially those focussing on methodological issues) and short fieldwork reports. Papers of two first categories will be subject to peer review. We welcome contributions, which focus on the biological background of historical processes observed in past populations of the area where most ancient civilisations of the Old World emerged. This includes large-scale studies e.g. on migrations, secular trends, microevolution, temporal changes or regional differences in the quality of life, disease patterns or demographical profiles, but also local studies or diagnostic case studies of distinguished individuals. Papers using not only biological, but also archaeological and textual evidence are mostly appreciated. For more effective exchange of information the journal includes also short fieldwork reports on human remains excavated at archaeological sites located in the region of interest.

Guidelines for authors 1. General information Bioarchaeology of the Near East publishes original papers, general review articles and short fieldwork reports which conform to the journal’s profile. Occassionally other kinds of texts (reviews, comments, letters to the editor, book reviews) may be taken into consideration. Each manuscript should be prepared according to the rules specified below. Manuscripts submitted for publication or published elsewhere will not be considered. Electronic submissions by email are strongly recommended. Manuscripts and tables should be submitted as RTF files, pictures as TIFF files (no compression or LZW compression), drawings as EPS or PDF files. If manuscript contains any special characters or diacriticals, additional PDF version of whole text will be required. If electronic submission is impossible, the hardcopy and floppy/CD/DVD containing text and all illustrations can be submitted by mail. Manuscripts should be written in English, with use of one spelling style throughout the entire manuscript. Both British and American spelling will be accepted. All submitted original papers and general review articles are sent for evaluation to two referees. The fieldwork reports are not reviewed although the editors reserve the right to refuse their publication if they do not follow the guidelines or are too lengthy. Galley proofs will be sent to the corresponding author via email from the editor, as PDF file. All corrections should be clearly marked and returned within two weeks as PDF file or hardcopy. This stage is to be used only to correct errors that may have been introduced during the edition process.

(no italics, eg. Mecsas et al. 2004). If there is more than one reference of the same author, years of publication should be separated by commas, e.g. (Horden 2004, 2005). References of the same author that appeared in the same year should be indicated by succeeding Latin letters and separated by commas, e.g. (Horden 2004a, 2004b). References to publications of different authors should be separated by semicolons, eg. (Dols 1977; Horden 2005). If many references are listed, alphabetical and ascending chronological order should be kept. All cited publications must be listed in the ‘Bibliography’ section at the end of the paper. Publications not cited in the text cannot be included. The bibliography should be listed in alphabetical order under the first author’s name and sorted chronologically if necessary. More than one reference from the same author(s) in the same year must be identified by the succeeding Latin letters, placed after the year of publication. In case of more than one author, names of all authors must be included. Titles of the journals should not be abbreviated. Bibliographical entries should be formatted according to the following examples: a) papers in journals: Mecsas J., Franklin G., Kuziel W.A., Brubaker R.R., Falkow S., Mosier D.E. (2004), CCR5 Mutation and Plague Protection, Nature 427:606–608. b) papers/chapters in books: Horden P. (2005), Mediterranean Plague in the Age of Justinian [in:] “The Cambridge Companion to the Age of Justinian”, M. Maas, A.D. Lee (ed.), Cambridge: University Press, pp. 134–160. c) books: Dols M.W. (1977). The Black Death in the Middle East, Princeton: Routledge.

2. Title page

5. Tables and illustrations

The title page must contain the following information: a) the full title of the manuscript, b) full name(s) of the author(s), c) affiliation(s) of the author(s) containing full name of the institution, and its postal address, d) email address, telephone and fax number of the corresponding author (indicated with an asterisk), e) running title of maximum 50 characters incl. spaces, f) list of keywords (no more than seven, no less than three), g) abstract (up to 300 words); it should be intelligible without reference to the rest of the paper. In case of short fieldwork reports the points e) – g) should be omitted.

Tables should be accompanied by an explanatory caption above; each column should carry a separate heading. Additional explanations (such as abbreviations) should be given below the tables as footnotes. Only basic horizontal lines need to be used. Tables may be included in the text or typed on separate pages; in the second case their places in the text have to be clearly indicated. Tables must fit on the B5 page. Figures should be used only if they clarify or reduce the text. No 3D diagrams or other bizarre illustrations are allowed. Data should be presented only once in a graph or a table, not in both. Only black-and-white or greyscale readable figures which do not exceed 200×120 mm will be accepted. Required resolution is 300 DPI for greyscale images and at least 600 DPI for black-and-white drawings (although in that case vector format is recommended). All figures and tables must be referred to in the text and the references should be typed in bold. The editors reserve the right to ask the authors for additional formatting of tables and figures.

3. The text The text of an original paper should be divided into sections (e.g. Introdution, Materials, Methods, Results, Discussion, Acknowledgements, Bibliography). General review paper may include sections and sub-sections. Section and sub-section headings should be clearly distinguished from the text and indicated by larger font and bold face (e.g. 16 pts for sections, 14 pts for subsections, 12 pts for regular text). Numbering of sections is not allowed. Short fieldwork reports should not be divided into sections. Legends to all figures should be listed on a separate page at the end of the text after Bibliography section; places of the figures in the text have to be clearly indicated. The title of short fieldwork report must include the name of the site and country of its location, as well as excavation season(s) reported. The following information must be covered by the text: name and affiliation of excavation’s director, chronology of studied human remains and brief description of their archaeological context, place of storage, information about used methods and aims of research, preliminary results and assumed place of more detailed publications. In text use italics only for species, genus names, or medical terms in Latin. The metric system, in SI version, must be used for all measurements. Metric abbreviations should be expressed in standard lowercase, without fullstops. Other abbreviations should be avoided. However, if it is impossible, the abbreviation must be expanded in brackets when used for the first time. Footnotes are allowed only as comments to the table. Number the manuscript pages consecutively, beginning with the title page. The length of the original paper should not exceed 80,000 characters (incl. spaces), general review papers may contain up to 160,000 characters, short fieldwork reports up to 15,000 characters.

4. References References in text should be given in brackets in the following sequence: author’s surname, year of publication, optionally page(s) after colon without spaces, eg. (Horden 2005:135–137). In case of two authors their surnames should be separated by “&” (Franklin & Brubaker 1998:124), in case of three or more authors the surname of the first one should be followed by ‘et al.’

6. Editorial procedures After submission, the corresponding author receives an initial acknowledgement. During two weeks the editor makes a decision as to whether the paper should be rejected or sent to referees. The corresponding author receives an email with explanation of this decision as soon as possible. Preliminarily accepted paper is sent to two referees who are asked to return their comments within one month (actually it may take up more time). The editor then considers the paper in the light of the referees’ reports and makes a decision on the paper. If it is declined, the editor will email the corresponding author with extracts from the referees’ reports to help explain his decision. If the paper is accepted, the editor will ask the author to revise the paper with the help of the referees’ comments. The referees may remain anonymous or disclose their names to the author. Once the author has revised the paper, it should be sent back in a form completely comformable to above guidelines. The paper will then be edited and a final version sent back to proof. Authors will be sent the galley proofs of their paper but are expected not to make any changes to the text at this point aside from minor spelling or grammatical mistakes. If the author wishes to make any other changes, the paper may be held back until the next issue.

7. Copyright Submission of a manuscript implies that the submitted work has not been published before (except as part of a thesis or lecture note or report, or in the form of an abstract); that it is not under consideration for publication elsewhere; that its publication has been approved by all co-authors as well as by the authorities at the institute where the work has been carried out; that written permission of copyright holders is obtained by the authors for material used from other copyrighted sources, and that the manuscript or parts thereof will thus not be published elsewhere in any language without the indication of the original place of publication.

Editorial This is the first volume of the Bioarchaeology of the Near East, a new journal intended to be a bridge between biological anthropologists and the archaeologists interested in the history of human populations inhabiting the Near East. Thousands of human skeletons have been excavated during more than 150 years of archaeological research in the region, but this vast amount of data has hardly been transferred into important research results, with few exceptions, such as studies of Egyptian mummies or various attempts to define the transition to agriculture in the areas of the Fertile Crescent. Especially the research on proto-historical and historical human populations seems to be seriously underdeveloped, although potentially promising if we take into account the constantly increasing amount of both archaeological and textual evidence. This situation may be explained as the result of three independent and equally important factors. First is the poor state of preservation of human remains in the climatic conditions of the region, again with Egypt as a prominent exception. In other areas, such as Mesopotamia, Syria or Iran, the most populated places were subject to annual fluctuations in humidity and temperature, which after hundreds or thousands of years made human bones very fragile. Another important issue is the context of skeletons which are more frequently found in settlement areas than in separate cemeteries. Continuous human activity over burial places contributed to further destruction of bones. In the past, when osteological research was focused on cranial and facial metric measurements, the problem of bone preservation was more serious, but even now—in the age of biochemistry and ancient DNA studies—it frequently frustrates researchers. The second factor is the political situation in the region, and relative lack of infrastructure. Most work must be done at the excavation houses far away from the laboratories, and the export of even small samples may turn out to be problematic or impossible. There are a few cases of successful transportation of large human skeletal collections abroad—as with the Japanese Hamrin project in the early 1980s—but actually such costly and difficult operations can be attempted only occasionally, under conditions of a strong support from local authorities, significant budget allocations and enthusiasm of the archaeological team leaders. All this may be secured only if the expected results are extraordinary. Again, there is one exception—Israel with its academic infrastructure, but also with religious restrictions on excavations of human remains. And the last point, contributing to the underdevelopment of studies on Near Eastern human remains, is the lack of communication among people involved in this kind of research. Again, Egypt and to some extent Israel are exceptional, but in other areas the common way of research is just taking more or less detailed observations and measurements, which are published as appendices to archaeological reports or in journals in which we would never expect to find a paper about human remains from our region of interest. Only a limited number of authors attempted to gather these data in composing more synthetic papers, which presented considerable challenges due to the dispersion of publications and sometimes the extreme differences in methodological approaches of research. The present journal may then be treated as an ambitious attempt to offer some relief at least against the third problem, and perhaps also toward some aspects of the two other issues discussed above. The journal is primarily conceived as a tool for information exchange and, in consequence, the establishment of a forum and a community of people interested in the human population history of this very important part of the world. There is no doubt that our community of interested scholars specializing in the region is rather limited and that the Bioarchaeology of the Near East will never become as thick as general anthropological journals. However, at least

2

Editorial

in my opinion, it is worth trying, should there be a chance of bringing together and better integrating our small community hence contributing to the progress of our research. As a  community-integrating device, each volume of the journal will be divided into two parts, the first publishing regular papers, subject to the peer-review process, and the second containing short fieldwork reports, as standardized brief communications (usually of 2-3 pages) about recent studies on human remains in our region of interest. Such reports will not be strictly reviewed, and the idea is to provide the readers with up-to-date, concise information about recent activity of their colleagues. I hope that many of us will acquire a habit of concluding each fieldwork season by issuing such a report for publication in the Bioarchaeology of the Near East. If so, the project will be successful and it is hoped that such exchange of data may become a base for further collaboration, larger and/or integrated projects and finally the publication of original research papers in our journal or elsewhere. There is also room for papers on methods, even if the properties of these methods are shown not solely in Oriental populations, as in the case of two articles published in the present volume. Although we focus on research on human remains, other branches of bioarchaeology, namely archaeozoology and archaeobotany, will also be represented in the journal, but only under the condition that these papers would be directly connected to the research on past human populations. There were some dilemmas with the name of the journal, and this was the subject of many discussions between the editor and the members of the Editorial Board. The forerunner to the present journal was the annual Studies in Historical Anthropology, established by the late Professor Andrzej Wierciński in 2001 as a journal dedicated to all types of research on past human populations. However, only one volume was issued before Wierciński’s death and for four years the title was suspended. In 2005 Prof. Alina Wiercińska decided to continue publishing the journal and three thematic volumes were issued in a dramatic struggle to catch up with the synchronization between the calendar and the volume numbers. However, very quickly we realized that the publication of the next thematic volumes would have been more and more strenuous because of approaching deadlines. The third volume of the Studies in Historical Anthropology was dedicated to the research on the Pre-Pottery Neolithic site of Nemrik 9 in Iraq, and contained also some papers on the subject of human remains excavated at other Near Eastern sites. The present journal is established to follow this track. Originally we decided to give it the title Historical Anthropology of the Near East to underline that the journal would be focused to the studies on human remains from protohistorical and historical periods, between the Chalcolithic and the Islamic period. So far much more effort was put in the research on the transition from hunting-gathering to agriculture than in the studies of such important processes as urbanization or formation of super-regional empires. Therefore, it seemed suitable to support research on these later less well published periods. In such a context, ‘historical anthropology’ should be understood as methods of the biological anthropology used for answering historical questions. However, recently the term ‘historical anthropology’ has been intensively exploited by cultural anthropologists who define it in many ways (as history of anthropology, anthropology of history, ethnic history, even history of material culture) but always far away from our field of research. In order to avoid a misunderstanding of the journal’s name, it was eventually turned into Bioarchaeology of the Near East, which is perhaps a bit less precise but hopefully better understood outside our field. Perhaps even more problematic is the second part of the name, i.e. the Near East. There is general agreement that this term denotes Anatolia, the Levant and Mesopotamia, or—in other words—the Asiatic part of modern Turkey, Syria, Iraq, Lebanon, Jordan, and Israel. However,

Editorial

3

some authors also include here Armenia (or Transcaucasia in general), the Arabian Peninsula, Egypt or even Nubia (northern Sudan) and Iran. Moreover, the relation between the definitions of the Near East and of the Middle East is unclear. For that reason we will adopt the broadest possible interpretation of the Near East, and define it as the region more or less resembling the former Achaemenid Empire and limited by the Black Sea, the Caucasus Mountains, and the Caspian Sea in the north, the mountains of eastern Iran in the east, the Persian Gulf and the Arabian Sea in the south, and the Nile valley from the IVth Cataract northward as well as the eastern shore of the Mediterranean Sea and the Ægean Sea in the west. With such geographical limits perhaps more adequate would be the title Bioarchaeology of the Southwest Asia and Egypt, but again, the use of the term ‘Near East’ is better grounded in archaeological literature. To avoid any further terminological confusion, it may be taken as a rule that papers on past human populations from neighbouring regions, i.e. Greece, Sudan, Central Asia, and the Indian Subcontinent will also be accepted. In that way all of us will be able to choose our individual definition of the Near East, even if as broad as possible, but in the limits of common sense. Most authors of papers published in the journal are and will be people doing research on human remains, but—apart from the integration of our small community—the second aim of the Bioarchaeology of the Near East is to provide archaeologists, historians, and philologists interested in our region with up-to-date results of studies on Near Eastern human remains. To achieve this aim, it is very important to secure efficient channels of distribution. The journal will be printed in the traditional way in a small number of copies, but more important is the distribution through the Internet. Bioarchaeology of the Near East is an open access journal and the copyright is only to secure the integrity of the papers and their proper citation. All volumes will be available without any restriction at www.anthropology.uw.edu.pl, a site provided by the Department of Historical Anthropology, Institute of Archaeology, University of Warsaw, Poland, which is the institutional host of the journal. After issuing the first volume, it will be the duty of the editor to submit the journal to those indexing directories and services, which are visited by potential readers. Having presented the journal, I would like to thank wholeheartedly all members of its Editorial Board, who agreed to take a  part in such a  risky venture as the establishment of a new journal, and strongly supported the editor in the initial stage of the project. Should the Bioarchaeology of the Near East become a well recognized and widely quoted a journal, it would be the consequence of their vast knowledge, engagement and hard work for the common good of our field of research. Many thanks are also due to Mrs. Elżbieta Jaskulska who transformed the manuscripts into a material volume. Basic information about the Bioarchaeology of the Near East, and the guidelines for the authors may be found in the inner sides of the journal’s cover.

Arkadiusz Sołtysiak Editor

Bioarchaeology of the Near East, 1:5–33 (2007)

A method for the study of activity related skeletal morphologies Theya Molleson Department of Palaeontology, Natural History Museum, London SW7 5BD, United Kingdom

Abstract: Our knowledge of the social and economic development of past societies and the people within them can be increased through an understanding of specialized activities. If an activity was time-consuming or arduous, especially when begun at an early age and the bones still growing, the bone morphology can be modified and techniques of production used in the past are recorded. The predisposing requirements for bone morphology to be distinctively modified are a restricted series of movements that are energetic and carried out for long periods probably on a daily basis and from a young age. Interpretation of an activity depends on the objective reconstruction of the biomechanical anatomy of the signs imprinted on the bones leading to a probability based on a differential diagnosis of the possibilities, in much the same way that a clinician evaluates a patient’s signs and symptoms. The significant morphology is that which differs from the normal range for the population being studied. Key words: bone morphology; craft technology; differential diagnosis; load bearing; repetitive activity; task related

The need to know The morphology of the human bones from Pre-Ceramic Neolithic levels at Nemrik, Iraq, indicates that running was of such major importance that the leg bones developed characteristic forms in relation to the stresses persistently imposed on them (Molleson 2006b). It is suggested that the method of hunting was to run down prey and catch the animals by grappling with them or by launching bolas stones at them. Stones, usually in groups of three, slung in a woven net or bolas have been widely used in the past for entangling animals. How likely this suggestion is can to some extent be checked by reference to independent sources of evidence for lifestyle of the Nemrik Neolithic populations. The detailed analysis of the teeth by Krystyna Szlachetko and Małgorzata Zadurska (2006) is the basis for further interpretation. In some jaws the shape of the coronoid process of the mandible and the extreme crushing-type occlusal wear of the teeth indicates a powerful puncture-crushing component to mastication that is associated more with task related than nutritional purposes. Bone chewing, large seed crushing, or fraying of reeds for making cord, could all be involved to a greater or lesser degree depending on the individual, season or other needs. Barley grains are among the hardest of the cereal grains and unless boiled in a broth (difficult to do before the advent of pottery) demand hard chewing to crush and moisten them before swallowing. Barley has an affinity for chromium and this element would be worth looking for in the human remains (Molleson 1995). Barium has been found at very high levels Received 6 November 2007; accepted 23 January 2008; published online 15 March 2008 on www.anthropology.uw.edu.pl

6

Theya Molleson

in the teeth from all levels at Nemrik indicating that plant foods were a significant part of the diet (Szostek et al. 2006). Low strontium-zinc ratios indicate a higher animal intake in the lowest burials (clusters aa and nn) from the site. In general animal protein indicated by zinc-calcium ratios was low, higher in older than younger individuals although this may be a cumulative effect. Agriculture as a subsistence economy seems to have been adopted during the late phase c. 9500–9000 BP and grinding stones, rubbers, mortars and pestles are part of the lithic assemblage. Thus, there is the possibility that hunting, as indicated by the lithics and bone morphology, might have had more of a social than nutritional function, as it does in so many societies. This however does not account for the very large number of animal bones, especially wild antelope, recovered from the site. Although my contention that the form of the human bones from Nemrik demonstrates that hunting could have been achieved by running down the prey cannot be fully verified, the findings from Nemrik illustrate how our knowledge of the social and economic development of past societies and the people within them can be increased through an understanding of activity related bone changes. We need to know when crafts and techniques arose and who the specialists were. Who were the craftsmen, were they men, women, children or slaves? When did caste systems emerge? Only the human bones themselves can give the answers, at least before the emergence of writing. The human remains that become available to historical anthropologists are largely reduced to bone fragments. Fortunately bone is remarkably pliable and responds to stresses and pressures exerted on it by muscular activity or weight loading. The growing skeleton is particularly responsive. Pressure imposed on bone can easily distort the form and robusticity. The muscles involved in repetitive movements carried out over a restricted range can be greatly developed and their enlarged insertions distinctively imprinted on the supporting bones. Thus a cluster or syndrome of morphological signs that are related to a specific activity can be established from changes to the usual shape of the bone in known performers and by reference to those muscles activated during the performance of specific actions. This information is surprisingly hard to assemble. Even where the activities are still being practiced in the traditional manner access to the skeleton is obviously denied. Interpretation of morphologies usually depends on the objective reconstruction of the biomechanical anatomy of the signs on the bones, signs of which the individual practioner might not be aware. Probably no two people carry out a given task in exactly the same way. In some their muscles will not develop despite the exertion, in others the muscle development might be quite disproportionate to the exertion involved; and conversely quite different tasks can require similar muscular exertion. Therefore we can never achieve certainty in interpreting bone morphology but a probability based on a differential diagnosis of the possibilities, in much the same way that a clinician evaluates a patient’s signs and symptoms. The clinical approach to assessing individual cases seems most appropriate to the needs of archaeological material at least initially. Once specialized skills had emerged the practitioners were usually relatively rare in the population and so have to be treated on an individual basis. This contrasts with the introduction of major behavioural changes that are taken up by whole sections of the population, such as hunting with spears or the bow and arrow, or a community of miners, which have been effectively documented by case control studies using comparative methods (Dutour 1986, 1993; Pálfi 1992; Bailly-Maître et al. 1996; Peterson 1997; Eshed et al. 2004; Weiss 2004). Great innovations like hunting with spears involving handed laterality are reflected in changes at the population level. The study of changes at the population level will not bring to notice individual specialists.

A method for the study of activity related skeletal morphologies

7

Nowadays probably only musicians and sportsmen in training impose such stresses on the young body that the bone morphology is actually modified. In the past the conditions of demanding and unrelenting exertion may have been more common and more evident, especially if the stresses were imposed while the bones were still growing. It is for these reasons that very occasionally we can recognise changes on excavated bones that can be attributed to specific postures and associated with particular tasks. The predisposing requirements for bone morphology to be distinctively modified are a restricted series of movements that are energetic and carried out for long periods probably on a daily basis and from a young age often producing entheses of muscle insertions. There may or may not be associated patterns of trauma; healed fractures, enthesopathies and degenerative joint osteoarthropathies are of secondary consequence (Knusel et al. 1977, but see Merbs 1983; Mafart 1996). Some degenerative joint lesions, however, have been repeatedly associated with certain activities including osteoarthrosis of the medial condyle of the left knee and myostitis ossificans of the thigh muscles (rider’s spur) in horse riders. Others, like vertebral crushing, are far too widespread to have much diagnostic value unless particular vertebrae are involved (e.g. T12 in the cereal grinders of Abu Hureyra, which risk is avoided when the quern is placed on a plinth as at Çatalhöyük). These assumptions underlie any attempt to infer cultural variation in the use of the body from physical anthropology, more specifically from the skeletal evidence provided by excavated material.

Establishing the syndrome The methodological approach has been to recognize when morphologies displayed by the bones or teeth of an individual are outside the range normally displayed by the population being studied. An attempt is then made to reconstruct the physical activity that might have led to the changes with reference to the biomechanics of the muscles and ligaments noted. The problems that would have confronted the early Neolithic people in the preparation of their newly acquired technologies and the appropriate solutions that are used by present day people are considered. This approach differs from that of Peterson (1997) and others (e.g. Eshed et al. 2004) who analyse variation between groups. It is more akin to a clinical approach where Table 1. Identification of the normal range of variation.

Normal range:

age

anomalies and variants

sex

body weight and robustness

ethnicity Pathology or abnormality:

deficiency and systemic disease genetic or congenital conditions healed trauma and degenerative change

Taphonomy:

erosion, gnawing, abrasion breakage and distortion fossilization, staining and diagenesis

8

Theya Molleson

the individual is examined and a differential diagnosis based on comparison with other possible normal and abnormal conditions. Since the same exertion can be demanded of different activities and a  given activity can be executed in different ways association of a  pattern of morphological changes with any particular task must be suggested with caution. Further one individual may have undertaken a number of strenuous tasks. The identification of activity related morphology seen in individuals is approached in five stages: exclusion of the normal, pathological and taphonomically altered; description of the morphology; differential diagnosis; case control analysis; and evaluation using independent evidence. The first need is to establish what is normal for the population or sample and is summarized in Table 1. Individuals that are exceptions to the norm can then be identified. A description of the morphological signs displayed forms the basis for the syndrome that characterizes the changes observed. Following clinical diagnostic practice several morphological signs will usually constitute the syndrome although not all will be developed equally or at all. An interpretation of the changes and reconstruction of the possible ways in which they might have come about leads, by a process of elimination, to a differential diagnosis of the most probable reason for the condition of the bones. Ideally this should be by reference to evidence based on known cases (Kennedy 1989; Blondiaux 1994). Where there are several individuals showing similar changes they can then be subjected to measurement and analysis by comparison with an equivalent normal sample—a case control study. Finally the concluding interpretation can be assessed in the light of the faunal, floral, lithic and environmental evidence (Table 2). Table 2. Steps towards the identification of task related activity.

1. The syndrome is defined from the description of the pattern of signs observed on the teeth or bones that differ from the normal. 2. A differential diagnosis is arrived at following comparison of the observed signs with those seen in other possible conditions. 3. Case controlled measurement and analysis of the identified cases contrasted with a generalized sample. 4. Independent environmental or comparative evidence used to evaluate the diagnosis.

Range of skeletal change Ken Kennedy documented an impressive range of activities that have been associated with unusual bone morphology (Kennedy 1989 and references). A large number of pathological conditions associated with particular occupations have also been documented (Mafart 1996). This overview tries to synthesize the range of activities reflected in dental modification, bone form and resulting from muscular exertion. Details of some examples are presented in the Appendix. 1. Task related dental wear. The reasons for dental modification can sometimes be established by reference to living practitioners who use their teeth as a third hand. Individuals develop their own ways of performing specific tasks but often a pattern develops that is characteristic of the task. 2. Pressure deformation. Passive pressure on a bone can deform its shape in a surprisingly short space of time especially in the young. Cranial deformation, sitting positions and load bearing can all result in characteristic changes to bone form and articular surfaces of the joints.

A method for the study of activity related skeletal morphologies

9

3. Biomechanical responses. Labour intensive activities such as running, load bearing, cereal grinding, horse riding, and acrobatics can be cautiously deduced by reconstructing the mechanical forces that must have induced the exceptional muscle and ligament imprints observed on some bones.

1. Task related dental wear Evidence for task related dental wear is seen in many sites from the Neolithic and beyond. By ‘task related’ I exclude food mastication, tobacco chewing, pipe smoking, beer drinking, labrets and tooth evulsion, but include any activity that involves using the teeth as a third hand: cord and basket makers, spinners, scribes, miners, boat builders (Molleson 2006a; Tuominen & Tuominen 1991; Heyerdahl 1971), Volga bargemen and Inuit ladies who have to chew their husband’s boots (Molleson 2004; Hansen et al. 1991). Jaw morphology can be modified by habitual activity, even normal mastication, involving jaw muscles and, conversely, underdevelopment of the masticatory muscles leading to dental crowding can be a significant problem of oral hygiene in some modern populations (Inoue 1993; Lucas 2006). Abrasions on the dental crowns can often be matched by examining the teeth of living practitioners. Cord makers, basket makers, spinners, and bark strippers have all been recognized, notably at Abu Hureyra, Mallaha, Çatalhöyük (Molleson 2006a; Bocquentin et al. 2005; Molleson et al. 1996; Ravy et al. 1996; Schultz 1977; Hoffman 1989). Using scanning electron microscope studies Michel Billard (1996) demonstrated important differences between task-related dental wear and tooth pick abrasion. There are now sufficient examples of task induced dental wear in the literature that an attempt to identify the resulting patterns seems worthwhile. The use of the mouth and teeth as a ‘third hand’ is doubtless more widespread than has been appreciated up to now and there must be many examples as yet unrecognized. The range of signs includes changes to bone form, accelerated tooth abrasion, tooth fracture and ante-mortem loss of the teeth, tooth-totooth attrition, enamel chipping, and possibly calculus deposits, hypercementosis, and periodontitis. The recognition of task-related dental modification must in each case rest on the fact that the features seen are beyond the normal range for the age of the individual compared to the rest of the group. Ideally, comparison should be made with reference to cases of documented history, which is sometimes possible with task related dental wear. In the pre-pottery Neolithic levels at Abu Hureyra, Syria, dental abrasion is characteristically severe but there are cases where the abrasion and ante-mortem tooth loss is extreme for the age of the individual. Morphological characteristics of the jawbones indicate that extraordinary forces had been applied over a period of time sufficient to modify the usual shape of the bone. In a number of cases it has been possible to interpret the morphological changes as being consistent with certain task related activities that involve the use of the teeth as tools. Cord makers identified by dental wear and enlargement of the head of the mandible; basket makers identified by grooved teeth, broad ramus, and enlargement of anterior border of the ascending ramus and a possible spinner, identified by sharp edged grooves on the anterior teeth. Distinctive jaws were identified by case study and defined by metric and statistical quantification of groups (see Appendix 1 for details of cases from Abu Hureyra, from Molleson 2006a). An initial study of the mandibles from Abu Hureyra identified a number with enormously enlarged condyles and extreme bevelled tooth wear. It was evident that these jaws had been

10

Theya Molleson

Figure 1a. Egyptian scribes prepared the reed used for writing hieroglyphs by chewing.

Figure 1b. Molar teeth of scribe Horemkenesi have uneven wear on the left side, presumably from chewing the writing reed. 21st Dynasty, Thebes (Bristol Museum H7386).

subject to stresses that were not purely to do with normal crushing chewing movements but due to abnormal forces imposed by task related activity. In a population with minimal dental disease teeth had been lost prematurely through the excessive crushing and pulling forces used. Tooth abrasion is extreme and asymmetric; even the palatal roots of the upper chewing teeth can be exposed and abraded. Sometimes anterior teeth, sometimes lateral teeth have been lost, reflecting the differing practices of individuals. The best comparison is with casts of dentitions of inhabitants of Danger Cave who chewed quids of Scirpus reeds and retted the reeds by pulling them between the teeth (Jennings 1957). The frayed reeds were then twisted together to make cord for binding rush mats. Although the ancient Egyptians had notoriously bad teeth, the healthy teeth of Horemkenesi, an Egyptian scribe, have exceptional abrasion that has resulted in mesio-distal depressions along the molar teeth presumably from the scribe’s habit of retting by chewing the reed brush he used to write hieroglyphs (Figure 1). Other jaws from Abu Hureyra had untypical abrasion of the anterior teeth and unusual morphology. In addition to wide grooves along the occlusal surfaces of upper and lower anterior teeth several traits of the mandibular body, the breadth of the ramus, and the shape of the coronoid process distinguish this group as having protruded the jaw while using the teeth

A method for the study of activity related skeletal morphologies

11

Figure 2. Sharp edged grooves on the front teeth of probable spinner from Abu Hureyra Neolithic Period (Tr.A 73.853).

to grip something. The front of the mandible is lifted to compensate for the anterior wear, which impedes occlusion, and the jaw has a tendency to rock when placed on a flat surface (Houghton 1977). It seems that reeds were held in the mouth to manipulate them in the process of weaving sieves and baskets, a worldwide practice (see Larsen 1985). Statistical analysis of the jaw dimensions of the basket makers revealed them to be morphologically distinct from the rest of the Abu Hureyra mandibles. Not all of the mandibles that showed evidence of task related dental wear were sufficiently complete for measurement but a  striking pattern did emerge that distinguished the basket makers especially from the bulk of the Neolithic sample from Abu Hureyra. All have a robust body of constant height, the mylohyoid line is parallel to the alveolar border, the ascending ramus is broad and has a strongly convex anterior border, and the coronoid process has a broad, double upper edge where the temporalis muscle is attached. This feature is also seen on jaws from Neolithic Nemrik, Iraq and Mallaha, Israel (Szlachetko & Zadurska 2006; Bocquentin et al. 2005). There are additional features especially in the gonial region that suggest that basket making at Abu Hureyra was a family based craft (Molleson 2006a). Sharp edged grooves have been noted on teeth from Neolithic levels at Abu Hureyra (Figure 2). These compare best with abrasion of teeth caused by running threads over the front teeth described by Ravy et al. (1996). In spinning the fine thread may be habitually drawn across one or more teeth eventually cutting a sharp-sided facet especially if the thread has been created from plant fibre.

2. Passive pressure remodelling 2i. Pressure modification of bones Passive pressure applied to the cranial bones of a newborn child can achieve permanent deformation in a matter of days. Manual manipulation, cradling boards or straps have all been widely used in the past (Dingwall 1931; Kiszely 1978; Molleson & Campbell 1995). Caps with a tight headband were worn until recently by young girls to ensure an æsthetically pleasing round, symmetrical head-form (Billard & Simon 1995). Similarly the carrying of loads in a basket slung from a head-

12

Figure 3a.The straps of a cassie are taken over the arms, which are pinioned to the sides.

Theya Molleson

Figure 3b. Shoulder joint of Norseman. The scapula is impacted below the neck of the humerus (Orkney 69/67A).

band or tumpline can result in a shallow depression from side to side across the frontal or parietal bones of the cranial vault. This is often easier to feel than to see and can be palpated on the living. Pronounced unciate processes on the cervical vertebrae giving head support seem to be developed in those who carry loads on the head; inferior articular processes of the thoracic vertebrae are buttressed against the adjacent superior articular processes in those who carry heavy loads on the back or slung from the shoulders (Sofaer 2000); lower back loading that puts pressure on the sacro-iliac joints extends the sacral-iliac articulations (Kennedy 1989). A deep radial fossa on the distal end of the humerus could result from the habit of carrying loads in a bag slung over the shoulder and held in place with the hand. The arm is tightly flexed at the elbow so that the head of the radius impacts above the distal epiphysis of the humerus. The ultimate in this type of deformation has been found in a medieval boatman who, it is suggested, had to pull heavy boats overland by bearing on a strap placed across his chest and over his upper arms, which were thus pinioned to his sides. Boats until recently were often dragged overland between rivers and hauled upstream in this way (Figure 3). The bones of both upper arm bones atrophied and became hollowed out where the head of the scapula had impinged on the humerus shaft (see Appendix 2i from Molleson 2004). (In rickets even body weight loading of the under-mineralized bones results in characteristic deformities e.g. angulation of the sacrum and bowing of the femora and tibiae, while obesity can result in changes to weight transfer resulting in knock knees and enlargement of the medial condyles of the tibias and asymmetry of the cortical thickness of the femora). 2ii. Pressure modification of articular surfaces a. Positions of rest. The pattern of torsion and squatting facets observed on hip and leg bones has been used to suggest a variety of sitting positions that might have been taken up by people in the Neolithic and more recent times (Molleson 2007a; see Appendix 2ii). Despite the discovery of a seated figurine by James Mellaart in the 1960s (Mellaart 1962; 1967) it is almost certain that the Neolithic people of Çatalhöyük, central Turkey, did not normally use chairs but squatted or sat directly on the ground when at rest or to undertake

A method for the study of activity related skeletal morphologies

13

Figure 3c. A Norseman from Orkney has extreme modifications to his teeth, shoulders, wrists and knees (69.67A Medieval Period).

Figure 3d. The unevenly worn teeth of the Norse bargeman who gritted his teeth from the exertion needed to haul the boat (Orkney 69/67A).

14

Theya Molleson

various stationary tasks. In the hocker squatting position a squatting facet persists where the talus pressed against the front of the tibia (Huard & Montagné 1951); and the patella has a notch where it pressed against the end of the femur (Buchet 1989; Boule 2001). Squatting in human evolutionary terms is the natural position of rest and is adopted by most infants when they start to move about. Squatting with the heels together flat on the ground is a position that many adults these days who are used to sitting on chairs find physically impossible or at best difficult to maintain for any length of time but one which is comfortable for those who have the appropriate bone morphology—and who comprise most of the world’s population. Whilst positions will have varied according to the activity, there will have been a general tendency by an individual to revert to a preferred position. Some individuals will tend to an active position, balanced on their toes from which they can get up quickly, others with a more passive temperament or lifestyle would take up a more static position. Since these habits are established and may even be enforced in childhood and are held for long periods of time each day, putting continued pressure on the limb bones over a restricted area, the shape of the affected bones and articular surfaces will be modified. It may be possible to infer preferred (or imposed) positions of rest from the morphological features of the leg bones, including long bone torsion, cross sectional shape, squatting facets, kneeling articulation and general asymmetry. Eight positions have been very tentatively identified and the expected bone changes predicted (Molleson 2007a, 2007b). Even though it will be impossible to test these predictions without reference skeletons from individuals of known habit we can at least postulate where the biomechanical forces will be and see if the expected patterns occur. The contrary effects of resting in the squatting or hocker position on the one hand and sitting cross-legged in the sartorial position on the other hand, are well known to orthopediatrics, and we should be able to identify them in skeletal material and thus infer preferred sitting positions (see Appendix 2ii from Molleson 2007a). In the sartorial or cross-legged position lateral rotation of the femur is combined with abduction and flexion exceeding 90 degrees. The range of rotation depends on the angle of anteversion of the femoral neck, which is usually quite wide in the child. In the hocker position the wide angle of anteversion is maintained or even increased when children become accustomed to sitting with their heels pressed against each other and their hips flexed. This posture causes medial rotation of the femur and accentuates the angle of anteversion as a result of the great plasticity of the young skeleton (Kapandji 1987:12). Paediatricians encourage children to sit cross-legged or even adopt the lotus position to correct the rotation of the thighs, to avoid problems in later life. The Chinese have a very low incidence of hip osteoarthrosis which has been related to a protective effect of squatting (Hoaglund et al. 1973). Males at Neolithic Çatalhöyük were more likely to develop a notch on the patella (6 of 8) than were females (1of 5). They also had large lateral or central squatting facets at the anterior distal articulation of the tibia or on the talus. If these traits are determined by posture, males at rest were most likely to squat either on their toes or with their heels flat on the ground. Females display a wider range of positions including squatting (Molleson 2007a). Their leg bones often have an asymmetrical development of features. The (comfortable) position taken up by an individual may relate to convention or to task. Body size may play a part but no trait was found to be strongly associated with stature. The pattern of squatting facets, restricted in the men, varied in the women, must reflect the differences in habitual resting positions at Çatalhöyük. Men at rest took up a preferred position, sitting on their haunches, feet together flat on the ground. This squatting position

A method for the study of activity related skeletal morphologies

15

is one of the most ergonomically efficient; it is static and can be maintained for a long time. We can infer that men at rest at Çatalhöyük were at rest not doing very much except perhaps with their hands. In contrast the variety of squatting prints on the bones of women from Çatalhöyük must reflect less uniformity in their sitting positions through individual preference or the most convenient postures to accommodate the range of activities undertaken by the women (and girls). There is no indication at Çatalhöyük that girls were expected to sit in a given position as they are in some parts of the world today (where there may be specific words to describe particular positions; Huard & Montagné 1951). Squatting facets aren’t solely indicative of positions of rest. Evidence from the prevalence of strongly developed squatting facets at the ankle and knees of the majority skeletons from the site of a Medieval lead-silver mine at Brandes, France, demonstrated that the miners had worked in cramped kneeling positions (Bailly-Maître et al. 1996). This observation was substantiated by the evidence from entheses for strongly developed muscles of the arm, hand and thigh; and was further substantiated by the analytical evidence for high levels of lead in the bones. b. Restricted repetitive movement. The saddle quern when used to grind cereal grains is both labour intensive and, without adjustments, risky. At Abu Hureyra saddle querns were found in the houses placed directly on the ground (Moore et al. 2000). The operator would have had to kneel before the quern with her toes curled under—on to the upper surface of the foot. To grind the cereal she would drive the rubber over the grains pushing off from her toes, pivoting from her hips and knees. The range of movement is constrained by the need to be positioned close to the quern and puts pressure on the hip, knee and toe joints. The bending moments imposed between the hip and knee result in bowing and buttressing of the femur. The articular surface of the head of the first metatarsal is extended on to the upper surface of the bone; it is diagnostic of the kneeling position (Ubelaker 1979). The plantar surface of the head of the second metatarsal can become flattened through being pressed on the ground (Wood Jones 1944). Sometimes the grinder overshot the saddle-quern resulting in injury to toes and back—specifically the 12th thoracic vertebra. The consequences are seen in crushing of the vertebra and traumatic osteoarthrosis of the articular surface of the first metatarso-phalangeal joint (Molleson 1994). Not surprisingly a way was found to avoid the painful accidents and evidence for these lesions has not been seen in sites where the saddle quern has been placed on a plinth, as at Çatalhöyük (Mellaart 1962). 2iii. Mobility of the joints Repeated uneven pressure on different parts of an articular surface may result in a division of the joint surface. The double occipital condyle which is observed in many of the workers from Spitalfields who were weavers seems to be associated with the repeated turning of the head from left to right and right to left to follow the shuttle (see Appendix 2iii for details). The Neolithic (c. 3000 BC) sample from Isbister, Orkney, is the only other documented sample in which a high incidence (30 of 83 condyles) of the anomaly has been recorded (Chesterman 1983:100). Here weaving is improbable. They are not associated with signs of head loading. It is possible that double condyles of the talo-calcaneus joint, also seen at Isbister, relate to mobility of the ankle in those who walk over uneven terrains and would be less prevalent in those who wear fitted shoes rather than sandals. The stability of the calcaneus is a function of the surface of the subtentaculum tali (Kapandji 1987:170). A persistent separate acromion process of the scapula, one of the last epiphyses to unite, is associated with strain on the shoulder joint in such activities as climbing rigging by young sailors (Stirland 2000).

16

Theya Molleson

3. Biomechanics and musculoskeletal stress markers Muscle attachments on bones are formed by the extension of the bone into the tendinous tissue of the muscle. They are reconstructed throughout life. In the young the bony ridges or entheses are less conspicuous and may even be represented by a depression or groove (Hamilton 1976:33; Lovejoy et al. 1976, 2002). With activity the entheses can become markedly rugose in response to the strength of the muscle action. Traumatic strain can result in tearing of the area of muscle attachment. Methods for scoring muscle impressions on bone have been developed (Hawkey & Merbs 1995; Robb 1998; Weiss 2004). Overall it seems best not to divide scoring of muscle impressions into too many stages unless the sex and age of the individual can be ascertained, since these two factors influence the degree of development of the impressions (see discussion in Mafart 1996; Lagier 1991). The effects of mechanical loading on different bones are discussed in detail by Pearson and Lieberman (2004). Their studies of the effects of exercise on the skeleton in different age groups showed that mechanical loading stimulates periosteal growth mostly prior to skeletal maturity. This is a major reason for concentrating on specialist activities that were initiated in the young. When bones function as beams the resulting form develops to resist bending, twisting and compression and the crosssectional area often reflects this. 3i. Hunting Human beings have caught, trapped and chased animal prey for millions of years. The various hunting stategies become deeply integrated into the social structure of society in ways that are characteristic of the method (Merbs 1983; Dutour 1986; Legge in Moore 2000). The morphological consequences of trapping at Abu Hureyra can be contrasted with running down animals at Nemrik (Molleson 2006b). The human bones from Nemrik are characterised by a number of features not often seen at other sites of the same general period and by the absence of a number of features that are observed elsewhere. To judge from the development of the hypotrochanteric fossa on the femora of young and old, we see what must have been the enormous development of the attachment area of the gluteus maximus muscles on the femora. The gluteal tuberosity where the gluteus maximus inserts on the femur is often long and deep. Gluteus maximus is important in running and jumping more so than in walking. It is a major extensor of the thigh especially when the thigh is flexed as in rising from a forward flexed position and in climbing and running when the thigh is thrust backwards. It would develop as a powerful muscle in those who had to run down an animal, whether it was to be taken live or killed. Further, there is a general lack of any marked development of supinator crest on the ulna. This suggests that hunting with spears or arrows was not important. In fact there is a lack of spear straighteners or arrowheads in the lithic assemblages at Nemrik, but there are bolas stones. These can be used to ensnare animals alive. The increase in bolas stones implies a sudden increase in hunting activities or in catching live animals for domestication (Mazurowski 1997:163). The fauna at Nemrik comprises some sheep, cattle and pig but mainly antelope, all of which have been herded from time to time. Horse, deer, wild cattle, boar, beaver, badger, buffalo, jackel and panther are also present. The Nemrik tibia is invariably strongly platycnemic and, where it could be observed, the proximal articulation with the fibula, which is a relatively robust bone, is well developed indicating that the fibula was a weight bearing bone, adapted to locomotion over uneven terrain. Tibias from Abu Hureyra are also platycnemic and it is possible that the side-to-side flattening of the shinbone can

A method for the study of activity related skeletal morphologies

Figure 4a. Pounding develops the arm and chest muscles (from Richards 1939).

17

Figure 4b. Skull of C18 was buried with a large pounder. (Al-’Ubaid, Iraq).

Figure 4c. The sternum and clavicle of C18 have large attachment areas for the pectoralis major muscles (2nd Dynasty, Al-’Ubaid, Iraq).

Figure 4d. The anterior teeth of C18 are very worn, from clenching the teeth while pounding (2nd Dynasty, Al’Ubaid, Iraq).

18

Theya Molleson

result from rotational stresses imposed in running over uneven terrain, especially where shoes are not worn. The side-to-side flattening of the tibia would allow large flexor muscles of the ankle that attach between the tibia and fibula, to be accommodated (Lovejoy et al. 1976). 3ii. Load bearing The Porters of Ur (see Appendix from Molleson & Hodgson 2000). Study of the extraordinary development of certain muscle imprints on some of the bones from burials in the Royal Cemetery at Ur, Iraq, has enabled us to suggest a probable occupation for some of the individuals. It was noted that several of the skeletons had extremely marked muscle attachment areas and it was suggested that these might offer clues as to the activities of the individuals during their lifetime. The muscles, which had been used so extensively, were noted and the activities, which might have led to such a change assessed. These activities were then compared with those, which were known to have taken place historically from the archaeological record so that some could be eliminated. We could then look more closely at the activities, which fitted both the changes in the skeleton and the time at which the individual had lived. Pathologies, joint osteoarthrosis or enthesopathies, were not of primary consideration, since morphological modifications that develop in response to stresses imposed on the bones during growth are considered to be more reliable indicators of habitual activity. A surprising number of the people appear to have laboured hard at very specific tasks that could be tantamount to employment or even slavery. The attendant PG1648d is one of a group of individuals whose occupation was probably similar to that of the porters shown on the Peace side of the Standard of Ur (Woolley 1934). If he used a head restraint as illustrated this would have encouraged an upright posture and effected the transfer of weight down the ventral aspect of the spine. Consequently, the sacrum is unusually straight and depressions (Schmorl’s nodes) where the weight carried had caused the intervertebral discs to protrude into the bone are on the ventral sides of the centra of the lumbar vertebrae. The clavicles have large areas for attachment of the subclavius muscles and coraco-costo-clavicular and trapezius ligaments, which must have been particularly powerful, although the bones themselves are not especially robust, but blade-like. Despite, or perhaps because of, the headband there must still have been considerable muscular tension stabilizing the sterno-clavicular joint. 3iii. Muscular exertion a. Pounding (see also Appendix from Molleson & Hodgson 2003). The woman C18 from Al-’Ubaid, Iraq, buried with a  large pounder shows major expansion of the sternum and clavicles related to massive development of the pectoralis muscles (Figure 4). It is significant that it is the bones of the arms and chest that are distinctive in pounding not the legs and feet (they are not constrained). The sternum is also expanded in the cripple from Jericho. It is the muscles used whatever the activity that can result in bone modification not the activity per se, which can be carried out in different ways. b. Rotary quern. There is potentially a  differential development of insertion areas for the teres major and t. minor muscles on left and right humeri and scapulae in populations where cultural mores demand that the wheel of the rotary quern is always turned in the direction of the sun (D. Lorimer pers. comm.). The teres spur on the lateral edge of the scapula was noted by Hrdlička (1942). The teres muscles together with the deltoid, back and leg muscles are also important in rowing a boat. In rowing the strongest muscles of the thigh, the quadriceps

A method for the study of activity related skeletal morphologies

19

and gluteus maximus, power the initial drive when the boat is moving at its slowest; the back and arm muscles are more effective in the second half of the stroke. Marked asymmetry of the shoulder muscles can develop in oarsmen who habitually row on one side of the boat. 3iv. Ligamental strain Horse riding (see Appendix from Molleson & Blondiaux 1994). In a rare documentation from cavalry records Joel Blondiaux established a number of traits, mostly on the femora, that characterize habitual horse-riders (Blondiaux 1994). These include hypertrophied ligament attachment areas around the fovea of the femur, the hypotrochanteric spur in the greater trochanter, as well as a pronounced linear aspera which supports attachment of the muscles that a rider uses to grip the back of the horse. These same traits were observed in femora from Bronze Age Kish, Iraq, thus confirming the pictorial evidence that equids were being ridden at this time. See Molleson and Blondiaux (1994) for discussion of the significance of the evidence for riding in Early Dynastic times. Subsequent work on bones from 19th century Spitalfields, London, has noted that degenerative joint changes occur on the medial condyle of the left knee: riders mount a horse from the left side. At The Royal Cemetery of Ur, one of the burials has been identified as a chariot or cart driver from the evidence for pronounced attachment areas on the knees for the cruciate ligaments—a response to twisting and turning movements centred on the knees and extreme development of the calf or ‘bouncing’ muscles of the lower leg (Molleson & Hodgson 1993). He would have needed these muscles to balance on the back of the cart or chariot, as depicted on the war side of the Standard of Ur.

Evolution of social behaviour Studies that demonstrate methods of hunting, load-bearing, food preparation and craft activities have established the value of bone morphology as a pointer to the role of the individual in past societies. During the Neolithic the development of technology seems to have been driven by the needs of food production. Some consequences of the methods of production used are recorded in the physique and demography of the populations. At Abu Hureyra, the advent of agriculture during the Neolithic about 10,000 years ago, introduced large cereal grains into the regular human diet. The need to process these seeds to prepare the ‘daily bread’ imposed long periods of energetic movement of a restricted range on the operator, who would have started to perform these tasks from a very young age. The effects of stresses imposed on the joints and bones of the skeleton can be recognised in the distinctively altered morphology of the bones and in the patterns of activity related trauma. The new cereal grains had to be dehusked by pounding, then ground on a saddle quern. Sieves and baskets to sort and to contain the grain had to be woven from prepared fibres. Subsequently pottery was manufactured to cook the food. Ropes or cords had to be made for catching, leading and tethering animals and for binding mats and baskets. All are needed and their manufacturers often had defined roles sometimes at the domestic level, sometimes as specialist craftsmen. Evidence for some of these task related morphologies is found predominantly on female skeletons. The use of the saddle quern is labour intensive, as much as five hours a day to grind a bushel (36 litres) of grain. By demonstrating that the first metatarsal is dimorphic between

20

Theya Molleson

the sexes it was possible to show that the majority of the toe bones that show the articular changes associated with kneeling are from females. Therefore, it can be inferred that most of the grain preparation at Abu Hureyra was carried out by women and girls. From the number and distribution of querns it is probable that each household prepared its own grain products. Thus there was a division of labour at the domestic level, with the females undertaking most of the grinding. From an early age daughters would become involved in the cycle of preparation, becoming identified with the female roles of the household, eventually to be buried within the dwelling which had been their domain in life (Molleson 1994; Wright 2000). Thus we can probably take the origins of gender-based roles back to the beginnings of plant production, all because the regular processing of plant products is obligatory; whereas there is circumstantial evidence from femur size that both sexes followed the hunt at Nemrik. The tendency of girls to undertake the more sedentary tasks would have resulted in a lower energy output. This together with the change in diet towards a (somewhat) higher carbohydrate intake must have predisposed them to lay down fat reserves. The dimensions of the head of the femur is a good indicator of body weight in human beings since so much of the body weight is taken on these joints with each step taken. At Çatalhöyük (Level 6) the diameter of the head of the femur is almost the same in males and females indicating that body weight at the end of growth was similar in the two sexes; yet the girls were shorter than the boys. Thus girls must have been relatively heavier than the boys (Molleson et al. 2005). At Abu Hureyra there is physical evidence for craft specialisation at a very early date. There are differences both in different parts of and during the Neolithic at this site. Basket makers and other specialists who worked fibres in the mouth were concentrated in trench A, the seed grinders in trenches B and E. Although these trenches are not fully contemporary with one another, the evidence for division of roles, specialisation and craft production so early in the Neolithic is important for our understanding of the development of the structure of society. The dental evidence for weaving and basket making is rare, presumably because the skills for these crafts were confined to a few people. Although not everyone would necessarily have used her or his teeth to manoeuvre the fibre there does seem to be conformity in the method of production over a wide, if not global, area. When the mouth is used to work the fibre mandibles have an enlarged development of the ascending ramus and coronoid process. These features are already present in the mandible of a five or six year old. The presence of the juvenile (Tr.B 73.2127) aged about five or six years at Abu Hureyra displaying many of the morphological features of the basket maker jaw raises the question as to whether the traits are familial or acquired. If acquired, the traits must be the consequence of forces imposed on the developing jaw from a very young age and five years would not be too soon for a child to become involved in the family craft. The distinctive jaw morphology, if familial, links this group of basket makers from Abu Hureyra to similar mandibles from Wadi Halfa, Nubia (Greene & Armelagos 1972), Jebel Sahaba (Anderson 1968) and Ishango (Twisselmann 1958) and their presence in northern Syria at this time even raises the possibility of a  tribe of specialist craftsmen. Once economic factors became relevant, craft specialisation might have been family based or restricted to a defined group, even caste. The increasing complexity of society sees the emergence, alongside the trade systems of local agriculturalists, of controlled labour in increasingly urbanised societies with agrarian economies, religious and military institutions dominating the control of labour forces. While many of these issues are best examined demographically, the extraordinary development of certain muscle imprints on the bones from Bronze Age Ur, Iraq, has enabled us to suggest

A method for the study of activity related skeletal morphologies

21

a probable occupation for some. It implies that their activities were followed intensively, from childhood and for long periods and with little alternative exercise, a concentration that is seen today only among sportsmen and musicians but in the past would also have included slaves and client or bonded workers. This role specialization can be taken as reflecting the level of economic activity and urbanization of the communities at Ur. To help us here we have the unique evidence from pictorial representations of people performing tasks that can be directly related to the skeletons that we have to study. The ‘Standard of Ur’ shows the nation at Peace and at War and we have examples from both among the surviving bones in the collection. Woolley noticed the continuity of activity from 5,500 years ago to the present day in the role of the porter, carrying large loads on his back, taking some of the strain from a headband, and still to be seen in the streets of Middle Eastern cities (Woolley 1934). Porterage was by many methods—the backpack with or without a headband; loads could be carried on the head or slung over the shoulder; and two porters might carry a load on a pole between their shoulders. Like the female, G48 AH, from Ur, carrying loads on the head is still the preferred method for women in many parts of the world. Men were occasionally depicted carrying loads on their heads e.g. at Khafaje (Early Dynastic III) on a wall plaque but of a decorative subject that must have occurred also at Ur (Lloyd 1984:115). That such depictions are not more frequent is likely to be for reasons of art composition—the inevitably taller figure distorts the symbolism of the pictorial narrative or must be represented as a very short person. Otherwise it would be tempting to assume that differentiation of an elite well fed and tall class from a  labouring inadequately nourished and undersized class had been established 5000 years ago. Estimates of sexual dimorphism from bone dimensions would be a more reliable way of looking at such social differences (Molleson 1996). Physically many of the skeletons reflect the importance of human effort in everyday life. There is evidence of social organization and a stratification of society from the skeletal remains. Some of the skeletons do not appear to have undertaken much physical activity at all including unsurprisingly Meskalam Dug and Puabi, whose graves were richly furnished—evidence that should not be overlooked. Sometimes even socially imposed rituals can be demonstrated. Daphne Lorimer identified the spur of the teres minor muscle on the scapula of several individuals from the medieval and earlier periods of Scotland. The attachment of this weak muscle could become hypertrophied where the left arm is required to turn in a clockwise direction, as would be the case in cultures that expect a rotary quern always to be operated in the direction of the sun’s orbit. The equivalent movement with the right hand uses the stronger, larger t. major muscle (Hrdlička 1942; Lorimer pers. comm.).

Conclusions Excavated human bones occasionally show modifications that can be interpreted as resulting from constrained activities carried out over long periods from a young age. The significant morphology is that which differs from the normal range for the population being studied. Where possible the postulated interpretation of the skeletal morphology should be checked against independent evidence from other sources, cultural and environmental including fauna and plant remains. This can point to the direction of further research. Many of the suppositions put forward in this paper need to be tested. Of course other interpretations of the bone morphologies observed are possible, that is the hazard of this type of

22

Theya Molleson

work. Perhaps all we can achieve is the identification of some remarkably energetic labourers who commenced their work at a very young age whilst the growing bones could respond to the pressures put upon them.

Acknowledgements Students and colleagues over the years have made many valuable contributions, more or less favourable, all much appreciated. Photographs of the teeth and bones were taken by Phil Crabb, Photo Studio, Natural History Museum, London, UK. Thanks too to the referees for their comments.

Appendix 1 Task related dental wear (from Molleson 2006a). The strong crushing forces used in some task related activities involving the teeth can result in a  modified mandibular form. These changes include enlargement of the head of the mandible. The human jaw joint has a structure inappropriate for stress bearing and the jaw closing muscles combine to ensure that all the muscle force exerted is effective at the teeth (Moore 1981:154). During incisal biting the mandible acts as a lever and condylar reaction forces are maximal. During powerful unilateral chewing the mandible is clearly functioning as a lever, the mandible head acting as a fulcrum, although when a food bolus is gently clamped during bilateral chewing neither head is acting as a fulcrum (Aiello & Dean 1990:100). It is evident that larger than normal condyles must be the result of abnormal forces exerted at the joint; such as the unilateral chewing required of cord makers who chew reeds first on one side then on the other in order to ret the fibres for cord making. Grinding chewing which follows the angle of the cusp surfaces can exert a force up to 750 Newtons (about body weight) with the second molars, but only 450 with the canines (Alexander 1992:136). Opinion seems to vary as to whether the greater force is on the side of chewing or on the counter-balanced side. Moore (1981:156) clearly indicates that the greater force (63:15) is on the counter-balanced side. And this is supported by the fact that those with a fracture of the condylar neck of the jaw find it easier to chew on the injured side (Dean & Pegington 1996:106). Structurally the mandible has thickened bone between the oblique line and the mylohyoid line due to the tension stresses set up when teeth are clenched. The direction in which the teeth are thrust may explain the thickened bone on the inner side of the sockets and the lower border of the body (Hamilton 1976:80). Thus a deep residual sulcus in the anterior margin of the ascending ramus is formed between the temporalis attachment on the external side and a bony bar on the inner side. Where the medial pterygoid muscles, which attach to the inner side of the gonial region, aid in chewing first on one side and then on the other the inner angles of the jaw may be turned inwards. The jaw of the basket maker has enlarged attachment areas for those muscles that act to close the jaws. The main chewing muscles are the masseters, the temporalis and the medial pterygoids. Where the protruded mandible is required to exert maximum crushing, the anterior part of the temporalis muscle is developed around the anterior border of the mandible and on the coronoid process, which develops a forward expansion and a double edge along the upper edge. This development is most often seen where the mandible is pulled forward

A method for the study of activity related skeletal morphologies

23

in order to carry out a task, including chewing leather, basket making, spinning thread. The ascending ramus is broad with large attachment areas for the masseter muscle on the lateral aspect. Together with the medial pterygoid muscle, which is attached to the inner side, the masseter produces bite. The greatest force is exerted when the teeth are in contact. The forces exerted by the basket makers must have been prolonged, as is to be expected in performing a task. The lower and upper (or superior) lateral pterygoids arise from the lateral pterygoid plate and from the greater wing of the sphenoid and insert on the neck of the mandible below the condyle. The lower lateral pterygoid draws the condyle forward in opening and in protracting the mandible; the upper lateral pterygoid closes the mandible. The lateral pterygoids do not appear to be modified in task related activity, presumably because they are not under any extra strain in pulling the mandible forward or back, the normal action. The action of the lateral pterygoids tends to pull the condyles together, with the resultant forces concentrated at the mandibular symphysis (Aiello & Dean 1990:81). Where the activity is concentrated on the anterior teeth, the chin and symphysis of the mandible seem to be raised relative to the body of the bone, the jaw is not stable when placed on a flat surface but ‘rocks’ (Houghton 1977). ‘Rocker’ jaws are seen in young individuals where the protruding action is applied but where the anterior attrition is still minimal.

Appendix 2 2i. Passive pressure remodelling Norse Age Boatman (from Molleson 2004). The remains of a Norse Age skeleton, 69/67A, from Newark Bay, Orkney, comprise the skull and post-cranial skeleton of an old adult male. At 1.749m he was above average height for an Orcadian male of the time. His teeth are extremely worn especially on the back molars. In the lower jaw the second and third molars are more worn than the first. On both humeri there is a marked depression on the inner (medial) side below the humeral neck where the head of the scapula impacted against the shaft. The bone is extremely thin or missing altogether. This seems to be a constriction atrophy that can only have been made by pressure forcing the two bones together. Both arms are affected. The constant pressure of the edge of the glenoid of the scapula on the humerus shaft has led to a depression and loss of bone in the area below the humeral neck of both arms. Until very recently, the carrying of loads in a basket suspended from the back with the holding straps taken across the chest and over, rather than under, the arms was widespread in Europe, the Near East and North Africa. Whether the degenerative changes to the man’s shoulders resulted merely from carrying a cassie wasn’t at all clear without the evidence of the dental attrition. The wear on the teeth is concentrated on the back molars, a reversal of the normal wear pattern. From this we can conclude that the teeth were subjected to something other than simple chewing, and that stress was exerted on individual molars. Tooth to tooth contact from clenching the teeth to fix the jaw during forced exertion would lead to attrition of the back teeth. This implies that some great force was required to achieve his task. Attrition can become extreme from clenching the teeth during sustained exertion, for example when using a hand screwdriver. The wear is angled because the clenching has a crushing action and there may be over-closure of the jaws causing wear on the cheek side of the lower molar teeth.

24

Theya Molleson

A possible explanation for the changes observed is that the man had hauled a boat with the hauling rope tied across the shoulders to add the weight of the body rather than just the strength of the arms. Formerly tens of thousands of burlaki were employed in dragging laden boats up the Volga River and its tributaries. The work of the burlaki or barge haulers was notoriously hard. Illustrated by Ilya Repin in the nineteenth century the haulers were linked to the boat by means of a strap that passed over the arms, which were pinioned to the sides of the body. It would take many continuous hours of intense hauling for the atrophy of the bone seen in the upper arm bones of 69/67A to develop.

2ii. Modification of articular surfaces Positions of rest and squatting (from Molleson 2007a). Thomson (1889) was one of the first to describe the changes associated with squatting. Charles (1893) amplified the earlier work adding the effects of the sartorial position—sitting cross-legged. The extreme flexion of the knee is rendered easy by a greatly increased articular area of the upper surface of the internal condyle of the femur, which is received upon the internal condylar surface of the tibia and has a very oblique plane. In the cross-legged (sartorial) sitting position lateral rotation of the femur is combined with abduction and flexion exceeding 90 degrees. The range of rotation depends on the angle of anteversion of the femoral neck, which is usually quite wide in the child. This leads to medial rotation of the femur. With time remodelling of the femoral neck leads to a more retroverted position (Kapandji 1987:12). During abduction the femoral neck impacts onto the rim of the acetabulum. In extreme flexion the antero-superior aspect of the neck comes into contact with the rim of the acetabulum and in some the neck at this point bears an impression just above the edge of the articular cartilage (Kapandji 1987:24). The sartorius muscle produces lateral rotation of the femur at the hip and flexion and medial rotation at the knee and can be seen as marked imprints on the hip and tibia. In the squatting or hocker position, the legs are flexed upon the thighs and the thighs on the trunk; the heels are generally some distance apart (about the width of the buttocks or ischial callosities) and the toes turned outwards. In this posture the knee joint is in a state of extreme flexion associated with a certain degree of rotation of the femur. A backward curve of the external articular surface of the tibia is associated with extreme flexion of the knee. The ankle joints are also in a state of extreme flexion. Squatting facets present on the anterior surface of the distal tibia and upper surface of the talus are associated with the extreme flexion of the ankle. These facets are present in the foetus (Boulle 2001). The heels and backs of the tibias support the weight of the trunk in the hocker squatting posture. From the talus the weight is transmitted to the inferior calcaneo-navicular ligament. Therefore the tibialis postici muscles will have more work to bring about support giving rise to an increase in anterio-posterior diameter—platycnemia of the tibia. Aitken (1905) doubted the direct role of the tibialis posticus in creating a platycnemic (flattened side to side) tibia. Lovejoy et al. (1976) recognized that side-to-side flattening of the tibia would allow large flexor muscles of the ankle that attach between the tibia and fibula, to be accommodated. Charles (1893) also noted differences in the hip. In squatting the head of the femur rests against the ischial portion of the acetabulum. The femoral head has a large articular area and the anterior-superior border forms a marked convexity—Poirier’s facet. Allen’s fossa, an im-

A method for the study of activity related skeletal morphologies

25

print on the anterior inferior medial part of the neck of the femur may also be present. These features, however, are not specific to squatting (Kostick 1963). Lamont (1910) described facets of the inner surface of the patella as being influenced by squatting.

2iii. Joint mobility Double occipital condyle. The dominant role of the double occipital condyle and large mastoid process in the 19th century sample from Spitalfields was revealed by statistical analysis. It can be argued that the two traits are linked; they are each both developmental and functional traits. At birth the occipital bone consists of four units, the basilar, two condylar and the squamous. The basilar unit includes the anterior parts of the condyles, the condylar units the posterior parts of the condyles. By the age of six years the four elements are united and form a single bone (Hamilton 1976:72). The occipital condyles support the skull on the spine. Changes in posture during growth are accompanied by small changes in growth and remodelling within the cranial base. These changes affect the degree of cranial base flexion and also the position of the foramen magnum (Aiello & Dean 1990:213). Rotation of the occiput in one direction is associated with linear displacement of 2–3mm in the opposite direction and lateral flexion (Kapandji 1974:182). It is possible that repeated movement of this joint at a critical time in growth served to delay synchondrosis of the two halves of the occipital condyle (just as the metopic suture will remain open when there is intracranial pressure in infancy). Large mastoid processes provide mechanical advantage for the two muscles that stabilize and rotate the head from side to side (Kapandji 1974:218ff; Aiello & Dean 1990:230). These are the sterno-cleido-occipito-mastoid, which is inserted into the superior and anterior borders of the mastoid, and longissimus capitis, which is inserted into the posterior border of the mastoid process. The significance of the double occipital condyle and the large mastoid process in the Spitalfields sample can be understood in terms of a functional response to habitual behaviour from an early age; in much the same way that the sutural ossicles respond to stress induced by artificial cranial deformation of skulls (Konigsberg et al. 1993). The majority of the occupations identified in the named sample at Spitalfields were associated with the textile industry (Molleson et al. 1993). Many of the men (and some women) were journeymen weavers or Master Weavers who would have started weaving in childhood. They would have worked 12-hour days, six days a week (Molleson et al. 1993). Inevitably their head movements would have followed the shuttle as this was thrown repeatedly from right to left and from left to right. It seems just possible that this arduous and lengthy activity influenced the growth of the cranial base.

Appendix 3 3ii. Biomechanics I Porters of Ur (see Molleson and Hodgson 2003). Attendant PG 1648d is the skeleton of a mature adult male. It includes parts of the cervical, thoracic, lumbar and sacral vertebrae; the clavicles; fragments of scapulae and pelvis. All long bones are represented and the bones of the hands and feet. Both patellae are present.

26

Theya Molleson

Figure 5. Attendant PG 1648d from Ur, Iraq. The bone morphology is consistent with the strain of carrying heavy loads on his back (from Molleson & Hodgson 2000).

The vertebral bodies of C6–C7 and T1 are large; the unciate processes are not pronounced and the neural arches are rather gracile; C6 is slightly crushed. In the lower back T12–L5 are present and fairly complete. They show marked degenerative changes: degenerative changes

A method for the study of activity related skeletal morphologies

27

are present in the bodies and there are additional arthritic changes along the proximal ventral margins of T12–L4 indicative of disc damage. There are arthritic changes to the proximal joints of L3. The sacrum is very straight. These changes to the spine could be the result of carrying loads persistently from adolescence. The clavicles are robust and have especially deep grooves for insertion of the subclavius muscles. The impression on the conoid tubercle and trapezoid line for the clavicular ligament and for the coraco-costo-clavicular ligament is also marked. If a heavy load is carried on the back, it tends to pull the scapula away from the clavicle. The trapezoid ligament stabilizes any strain on the acromioclavicular joint and the costo-clavicular ligament similarly stabilizes the sterno-clavicular joint. The humerus is robust, with well-developed deltoid impressions. The two ulnas, two radii, and the carpals have a slight lipping of the articular surface margins. The metacarpals of the right hand are complete. The head of the IInd metacarpal is deflected towards the palm. The bone is robust with particularly marked insertions for the 2nd palmar interosseus and the 1st dorsal interosseus muscles, so that there is a ridge between the two on the palmar surface. On the IIIrd metacarpal the 2nd and 3rd dorsal interosseus insertions are also marked. The head of MCI (the base of the thumb) is flattened and the proximal phalanx has a marginal tubercle for the adductor pollicis muscle. This combination of features suggests that the hand was used for gripping—with the fingers splayed— perhaps the straps holding the pack to relieve the weight on the back. The femora are robust and each has a strong torsion of the head relative to the distal condyles, so that it is anterior to the knee. This form, in bearing a heavy load, would ensure that weight transfer was vertical from hip to knee if the knees were slightly bent. The insertions for the medial and lateral heads of the gastrocnemius muscle are pronounced on the femora implying well developed calf muscles used in flexing the knees when walking. The pattern of ligament and muscle development observed on the bones of PG 1648d would be those of one who had to carry loads on his back. We have a pictorial representation of such a porter on the “Peace” side of the Standard of Ur where there is depicted a man carrying on his back a great bundle which, like a Stambul porter, he supports by a band across his forehead (Woolley 1934:273). It is therefore possible that PG1648d could have been one such porter.

3iii. Biomechanics II Pounding (from Molleson & Hodgson 2003). Skeleton C18 comes from al-’Ubaid, about four miles from Ur, Iraq. Woolley noted in his field notebook that the bones of C18 were fairly well preserved; that there was a large spindle whorl of green stone (later identified as pottery), and behind the head a rough stone pounder almost a cube (BM Ms). The relatively complete skeleton is probably that of a robust mature adult female. The skull is unusually well preserved and most of the teeth are present. The pattern of wear is strange; it is much heavier on the upper teeth, especially on the lingual side of the anteriors; the dentine is more worn than the enamel so that the surface of the tooth is scooped out, and an incisor and two molars are chipped. Strong development of the insertion areas for the masseter, temporalis, buccinator and anterior belly of the digastric muscles give the jaw a masculine appearance. These muscles are involved in closing and opening the jaws. The styloid is long and robust. The hyoid bone with greater horn and ossified hyoid cartilage are preserved. Muscles between the hyoid and styloid are involved in stabilising the jaw in a fixed position when the weight of the body is required to exercise force on an outside object.

28

Theya Molleson

Most of the vertebrae are present and reveal a spine with unusually large lower thoracic vertebrae and rather narrow lumbars; the first two vertebrae of the sacrum are cleft. The clavicles both have a remarkably broad area for the attachment of the clavicular part of pectoralis major muscle and the sternum is unusually wide, presumably to support a large sterno-costal part of this muscle. The muscle insertion areas on the humeri for this muscle are well developed. As a whole the pectoralis major takes an active part in the movements of adduction and medial rotation of the humerus, but the activity is only marked if resistance is to be overcome—as when using a pounder? On both ulnae the supinator crest is pronounced and extends to merge with the interosseus border. The supinator muscle rotates the radius, so that objects, which may be heavy, can be picked up with the forearm pronated; the more powerful supinators can lift an object against gravity. The femora have strongly developed linea aspera for the adductor muscles; and there is a clear Allen’s fossa on the superior border of the femoral neck. The tibia does not have a squatting facet. Both the first metatarsals have extended articular surfaces associated with kneeling. So we have a picture of this strong woman, energetically and with teeth clenched, working away with the heavy pounder that was found in her grave.

3iv. Biomechanics III Horse riding at Kish (from Molleson & Blondiaux 1994). Fifteen femora from Early Dynastic and later levels at Kish, Iraq were examined for traits that have been associated with habitual riding. No attempt was made to distinguish between the characteristics of horse and other equid riders. Four of the femora from Kish have a strongly developed linea aspera indicative of strong adductor muscles. Two of these also have pronounced areas of insertion of all three gluteal muscles, but especially gluteus minimus and g. medius on the greater trochanter; and one has a distinct spicule in the trochanteric fossa. The trochanteric spicule is an enthesis at the insertion on the medial aspect of the greater trochanter of the obturator internus, a muscle important in the lateral rotation of the thigh, a position taken up with force during riding (Blondiaux 1994). The obturator interna and gemelli muscles are placed under strain when the thigh is flexed and rotated laterally as in sitting on the back of an equid. The adductor longus muscle, which inserts in the mid-shaft region of the linea aspera, adducts the thigh at the hip, and ensures that the rider maintains his seat on the animal’s back. Riders often sustain an injury to the adductor longus (‘rider’s strain’). Riding always throws great strain on the adductor muscles, at no time more than when a horse refuses a jump or stumbles, or throws the rider out of the saddle. The rider makes a reflex gripping action, the adductors are put under sudden additional load and the weakest of these is likely to give way. Three conditions follow: traumatic periostitis, chiefly at the attachment of the tendon of the adductor longus to the ischial ramus; tendon-muscle strain of the adductor longus; and rupture of the muscle-belly of adductor longus. These strains would be frequent among equestrians who had to ride, without stirrups, a mettlesome animal. Two of the femora from Kish have well marked iliac impressions (Poirier’s facets) on the anterior aspect of the femoral neck. Poirier’s facet develops where the femur abuts against the rim of the acetabulum (iliac impression) as also in sitting cross-legged. Whilst any one of these features might result from a variety of activities, the association of the three, trochanteric spicule, gluteal and adductor development, would seem to result from habitual

A method for the study of activity related skeletal morphologies

29

riding. Only one of the Kish femora, from a tall male, displays all three traits. It also has a marked development of the adductor tubercle for insertion of the tendon of the adductor magnus muscle, which may be important in medial rotation of the thigh. The area of attachment of the medial head of the gastrocnemius muscle on the popliteal surface is pronounced. This muscle flexes the knee.

Bibliography In these days of internet access no attempt has been made to provide a complete bibliography and what follows is frankly idiosyncratic and overloaded with my own publications simply because I can most easily quote at length from them. Aiello L., Dean C. (1990), An introduction to human evolutionary anatomy, London: Academic Press. Aitken D.M. (1905), A note on the variation of the tibia and astragalus, Journal of Anatomy 39:489-491. Alexander R.M. (1992), The human machine, London: Natural History Museum. Anderson J.E. (1968), The late Paleolithic skeletal remains from Nubia [in:] “The prehistory of Nubia”, vol. II, F. Wendorf (ed.), Dallas: Southern Methodist University Press. Bailly-Maître M.-C., Simonel B., Barré N., Boulle, E.-L. (1996), Travail et milieu. Incidences sur une population au Moyen Âge [in:] “L’identité des populations archéologiques. XVIe Rencontres Internationales d’Archéologie et d’Histoire d’Antibes”, Sophia Antipolis: CNRS, pp. 211-243. Billard M. (1996), L’abrasion artificielle des dents comme aspect revelateur d’habitudes culturelles: observation comparée de deux sites du Chalcolithiques — Bronze ancien (Drôme et Ardèche) [in:] “L’identité des populations archéologiques. XVIe recontre internationales d’archéologie et d’Histoire d’Antibes”, Sophia Antipolis: CNRS, pp. 329-343. Billard M., Simon C. (1995), L’os révélateur d’habitude culturelle, Dossiers d’Archéologie 208:22-33. Blondiaux J. (1994), A propos de la dame d’Hochfelden et de la practique cavalière: discussion autour des sites fonctionnels fémoraux [in:] “La femme pendant le Moyen Âge et l’Époque Moderne”, L. Buchet (ed.), Paris: CNRS, pp. 97-110. Bocquentin F., Sellier P., Murail P. (2005), Abrasion dentaire et travail spécialisé dans la population natoufienne de Mallaha (Israël), Comptes rendus Palevol 4:351-357. Boule E.-L. (2001), Evolution of two human skeletal markers of the squatting position: a diachronic study from antiquity to the modern age, American Journal of Physical Anthropology 115:50-56. Buchet L. (1989), L’usage de la station accroupie dans les sociétés antiques et médiévales de Gaulle [in:] “Homme et Milieu, approches paléoanthropologiques”, L. Buchet (ed.), Paris: CNRS, pp. 113-120. Charles R.H. (1893), The influence of function, as exemplified in the morphology of the lower extremity of the Punjabi, Journal of Anatomy and Physiology 28:1-18. Chesterman J. (1983), The human remains [in:] J.W. Hedges Isbister, “A chambered tomb in Orkney”, Oxford: BAR British Series 115. Dean C., Pegington J. (1996), Core anatomy for students, London: Saunders. Dingwall E.J. (1931), Artificial cranial deformation, London: Bale. Dutour O. (1986), Enthesopathies (lesions of muscular insertions) as indicators of the activities of Neolithic Saharan populations, American Journal of Physical Anthropology 71(2):221-224. Dutour O. (1993), Les marqueurs d’activités sur l’os humain fossile. Une tracéologie paléoanthropologique? [in:] “Traces et fonction: les gestes retrouvés. Actes du colloque international

30

Theya Molleson

de Liège 8–9–10 décembre 1990”, P.C. Anderson, S. Beyries, M. Otte, H. Plisson (ed.), Liège: Editions ERAUL, pp. 59-66. Eshed V., Gopher A.V., Galili E., Hershkovitz I. (2004), Musculoskeletal stress markers in Natufian hunter-gatherers and Neolithic farmers in the Levant: the upper limb, American Journal of Physical Anthropology 123:303-315. Greene L.D., Armelagos G.H. (1972), The Wadi Halfa Mesolithic population, “Department of Anthropology, University of Massachusetts, Research Report No. 11”, Amherts. Hamilton W.J. (1976), Textbook of human anatomy, 2nd edition, London: Macmillan. Hansen J.P.H., Meldgaard J., Nordquist J. (1991), The Greenland Mummies, London: British Museum. Hawkey D.E., Merbs C.F. (1995) Activity-induced musculoskeletal stress markers (MSM) and subsistence strategy changes among ancient Hudson Bay Eskimos, International Journal of Osteoarcheology 5:324-338. Heyerdahl T. (1971), The RA expeditions, London: George Allen and Unwin. Hoaglund F.T., Yau A., Wong W.L. (1973), Osteoarthritis of the hip and other joints in southern Chinese in Hongkong: incidence and related factors, Journal of Bone and Joint Surgery. American Volume (Boston) 55:545-557. Hoffman M.P., Cutress T.W., Crooks M.C. (1989), Some epidemiological and scanning electron microscopic features of crazing of the dental enamel of Polynesians, New Zealand Dental Journal 85:86-90. Hole F. (1994), Environmental stability and urban origins [in:] “Chiefdoms and early states in the Near East. Monographs in World Archaeology” 18, G. Stein, M.S. Rothman (ed.), Madison: Prehistory Press, pp. 121-151. Houghton P. (1977) Rocker jaws, American Journal of Physical Anthropology 47:365-370. Hrdlička A. (1942), The scapula: visual observations, American Journal of Physical Anthropology 29:73-94, 287-310, 363-415. Huard P., Montagné M. (1951), Le squelette humain et l’attitude accroupie, Bulletin de la Societè des Etudes Indochinoises 26:401-426. Inoue N. (1993), Collapse of dentition in Japan [in:] “Culture of food and oral health in Maori”, N. Inoue (ed.), Tokyo: Therapeia publishing co., pp. 67-77. Jennings J.D. (1957) Danger Cave, “University of Utah Anthropological Papers” 27, Salt Lake City, pp. 224-227. Kapandji I.A. (1974, 1982, 1987) The physiology of the joints, vols. 1–3. Edinburgh: Churchill Livingstone. Kennedy K.A.R. (1989) Skeletal markers of occupational stress [in:] “Reconstruction of life from the skeleton”, M.Y. Iscan, K.A.R. Kennedy (ed.), New York: Alan Liss, pp. 129-161. Kiszely I. (1978), The origins of artificial cranial deformation in Eurasia, “BAR International Series” 50, Oxford. Knusel C.J., Goggel S., Lucy D. (1997), Comparative degenerative joint disease of the vertebral column in the medieval monastic cemetery of the Gilbertine Priory of St Andrew, Fishergate, York, England, American Journal of Physical Anthropology 103:481-495. Kostick E.L. (1963), Facets and imprints on the upper and lower extremities of femora from western Nigerian population, Journal of Anatomy 97:393-402. Lagier R. (1991), Entheses normales et enthésopathies [in:] “Pathologie des insertions et enthésopathies”, L. Simon, Ch. Hérisson, J. Rodineau (ed.), Paris: Mason.

A method for the study of activity related skeletal morphologies

31

Lamont J.C. (1910), Note on the influence of posture on the facets of the patella, Journal of Anatomy 44:149-150. Larsen C. (1985), Dental modifications and tool-use in the western Great Basin, American Journal of Physical Anthropology 67:393-402. Lloyd S. (1984), The archaeology of Mesopotamia, London: Thames and Hudson. Lovejoy O., Burstein A.H., Heiple K.G. (1976), The biomechanical analysis of bone strength: a method and its application to platycnemia, American Journal of Physical Anthropology 44:489-505. Lovejoy C.O., Meindl R.S., Ohman J.C., Heiple K.G., White T.D. (2002), The Maka femur and its bearing on the antiquity of human walking: applying contemporary concepts of morphogenesis to the human fossil record, American Journal of Physical Anthropology 119:97-133. Lucas P. (2006), Facial dwarfing and dental crowding in relation to diet, International Congress Series 1296:74-82. Lukacs J.R., Hemphill B.E. (1990), Traumatic injuries of prehistoric teeth: new evidence from Baluchistan and Punjab Provinces, Pakistan, Anthropologischer Anzeiger 48:351-363. Mafart B.-Y. (1996), Quelques particularité anthropologiques et palépathologiques [in:] “L’identité des populations archeologiques. XVIe rencontre internationale d’archéologie et d’histoire d’Antibes”, Sophia Antipolis: CNRS, pp. 345-350. Mazurowski R.F. (1997) Nemrik 9: ground and pecked stone industry in the Pre-Pottery Neolithic of Northern Iraq, Warsaw: Instytut Archeologii UW. Mellaart J. (1962), Excavations at Catal Hüyük, Anatolian Studies 12:41-65. Mellaart J. (1967), Çatal Hüyük, London: Thames and Hudson. Merbs C.F. (1983), Patterns of activity-induced pathology in a Canadian Inuit population, “Papers of the National Museum of Man Mercury Series” 119, Ottawa: Archaeological Survey of Canada. Molleson T. (1994), The eloquent bones of Abu Hureyra, Scientific American 271(2):70-75. Molleson T. (1995), The importance of porridge [in:] “Nature and Culture. Colloque de Liège”, M. Otte (ed.), Liège: Eraul, pp. 481-488. Molleson T. (1996), Skeletal evidence for identity and role in the Neolithic [in:] “L’identité des populations archéologiques. XVIe recontre internationale d’archéologie et d’Histoire d’Antibes”, CNRS: Sophia Antipolis, pp. 345-350. Molleson T. (2004), A Norse Age boatman from Newark Bay [in:] “Orkney Archaeology Trust, Papers and Pictures in honour of Daphne Home Lorimer MBE”, http://www.orkneydigs. org.uk/dhl/papers/tm/index.html. Molleson T. (2006a), The third hand: Neolithic basket makers of Abu Hureyra [in:] “Current Trends in Dental Morphology Research. Proceedings of the 13th International Symposium on Dental Morphology”, E. Żądzińska (ed.), Łódź, pp. 233-243. Molleson T. (2006b), Hunters of Nemrik, Studies in Historical Anthropology 3(2003): 5-18. Molleson T. (2007a), Times of stress at Çatalhöyük [in:] “Faces from the Past: Diachronic Patterns in the Biology of Human Populations from the Eastern Mediterranean. Papers in honour of Pat Smith”, M. Faerman, L.K. Horwitz, T. Kahana, U. Zilberman (ed.), “BAR International Series” 1603, Oxford: Archaeopress, pp. 140-150. Molleson T. (2007b), Bones of work at the origins of labour [in:] “Archaeology and women. Ancient and modern issues”, S. Hamilton, R.D. Whitehouse, K.I. Wright (ed.), Walnut Creek, pp. 185-198. Molleson T., Hodgson D. (1993), A cart driver from Ur, Archaeozoologia 6:93-106. Molleson T., Cox M., Waldron A.H., Whittaker D.K. (1993), The people of Spitalfields: The Middling Sort, “CBA Research Report” 86. Molleson T., Blondiaux J. (1994), Riders’ bones from Kish, Cambridge Archaeological Journal 4:312-316.

32

Theya Molleson

Molleson T.I., Campbell S. (1995), Deformed skulls at Tell Arpachiyah: the social context [in:] “The Archaeology of Death in the Ancient Near East”, S. Campbell, A. Green (ed.), Oxbow Monograph 51, pp. 45-55. Molleson T., Boz B., Nudd K., Alpagut B. (1996), Dietary indications in the dentitions from Çatalhöyük, Arkeometri Sonuçlari Toplantisi 11:141-150. Molleson T., Hodgson D. (2000), The Porters of Ur, Isimu 3:101-117. Molleson T.I., Hodgson D. (2003), The human remains from Woolley’s excavations at Ur, Iraq 65:91-129. Molleson T., Andrews P, Boz B. (2005), Chapter 12: Reconstruction of the Neolithic people of Çatalhöyük [in:] “Inhabiting Çatalhöyük: reports from the 1995-1999 seasons. Çatalhöyük research project” vol. 4, I. Hodder (ed.), McDonald Institute and British Institute of Archaeology at Ankara, pp. 279-300. Moore A.M.T., Hillman G.C., Legge A.L. (2000), Village on the Euphrates. From foraging to farming, Oxford University Press. Moore W.J. (1981), The mammalian skull, Cambridge University Press. Pálfi G. (1992), Maladies, environnement et activités: traces sur l’os humain ancien. Exemple de la série anthropologique de Solliès-Toucas (IIIe à IVe siècle A.D., Var, France), Préhistoire et anthropologie méditérannéennes 1:61-72. Pearson O.M., Lieberman D.E. (2004), The aging of Wolff’s “Law”: ontogeny and responses to mechanical loading in cortical bone, Yearbook of Physical Anthropology 47:63-99. Peterson J. (1997), Tracking activity patterns through skeletal remains: a case study from Jordan and Palestine [in:] “The Prehistory of Jordan II. Perspectives from 1997. Studies in Near Eastern Production, Subsistence and Environment” vol. 4, H.G.K. Gebel, Z. Kafafi, G.O. Rollefson (ed.), Berlin: ex oriente, pp. 475-492. Ravy E., Clère J., Puech P.-F. (1996), Traces d’activités humaines sur des dents du Chalcolithique Ardéchois, L’Anthropologie 100:574-588. Robb J.E. (1998), The interpretation of skeletal muscle sites: a statistical approach, International Journal of Osteoarchaeology 8:363-377. Schulz P.D. (1977), Task activity and anterior tooth grooving in prehistoric California Indians, American Journal of Physical Anthropology 46:87-92. Sofaer J.D. (2000), Sex differences in activity-related osseous change in the spine and the gendered division of labour at Ensay and Warram Percy, UK, American Journal of Physical Anthropology 111(3):333-354. Stirland A. (2000), Raising the Dead. The skeleton crew of Henry VIII’s great ship, Mary Rose. London: John Wiley. Szlachetko K., Zadurska M. (2006), Human teeth from Nemrik, Studies in Historical Anthropology 3(2003):31-92. Szostek K., Głąb H., Kaczanowski K. (2006), An analysis of the content of macro- and microelements in the teeth of an early Neolithic population from Nemrik (Iraq), Studies in Historical Anthropology 3(2003):19-30. Thomson A. (1889), The influence of posture on the form of the articular surfaces of the tibia and astragalus in the different races of man and the higher apes, Journal of Anatomy 23:616-639. Tuominen M., Tuominen R. (1991), Tooth surface loss among people exposed to cement and stone dust in the work environment in Tanzania, Community Dental Health 8:233-238. Twisselmann F. (1958), Les ossements humains du site mésolithique d’Ishango, Exploration du parc national Albert Mission Jean de Heinzelin de Braucourt 5:3-123.

A method for the study of activity related skeletal morphologies

33

Ubelaker D.H. (1979), Skeletal evidence for kneeling in Prehistoric Ecuador, American Journal of Physical Anthropology 51:679-686. Weiss E. (2004), Understanding muscle markers: lower limb, American Journal of Physical Anthropology 125:232-238. Wood Jones F. (1944), Structure and function as seen in the foot, London: Baillière. Woolley C.L. (1934), Ur excavations. The Royal Cemetery, Joint Expedition of the British Museum and of the University of Pennsylvania to Mesopotamia. Wright K. (2000), The social origins of cooking and dining in early villages of western Asia, Proceedings of the Prehistoric Society 66:89-121.

Bioarchaeology of the Near East, 1:35–57 (2007)

The practice of cremation in the Roman-era cemetery at Kenchreai, Greece The perspective from archeology and forensic science

Douglas H. Ubelaker*1, Joseph L. Rife2 1

Department of Anthropology; National Museum of Natural History, MRC 112 Smithsonian Institution; Washington, D.C. 20560; email: [email protected] (corresponding author) 2 Classics Department; Macalester College 1600 Grand Avenue; St. Paul, MN 55105–1899

Abstract: Since 2002 the Kenchreai Cemetery Project has explored subterranean chamber tombs of Roman date in the main cemetery of the ancient port of Kenchreai, on the eastern coast of the Isthmus of Corinth, Greece. Analysis of the human remains recovered from three tombs has furnished evidence for cremation as well as inhumation. The cremated remains represent both adults and immature individuals. Forensic analysis indicates that the original event of cremation reached high temperatures over a long duration, and that only a fraction of the cremated remains were transferred to the tombs. Ancient mortuary sites represent valuable repositories of information regarding not just the burial customs of past societies, but also biocultural information about the people represented and their attitudes toward life and death. Such information is augmented through a thoughtful bioarcheological approach in which knowledge gleaned from skeletal analysis is integrated with archeological interpretation following meticulous excavation. This report exemplifies this sort of analysis, focusing on the interpretation of cremated remains from chamber tombs of Roman date in the main cemetery at Kenchreai in southern Greece. Key words: Greece; cremation; bone; Roman; Kenchreai

1. Kenchreai and the cemetery on the Koutsongila Ridge Kenchreai was located on the eastern coast of the Isthmus of Corinth, which connected central Greece with the Peloponnesos, or the southern Greek peninsula (Figure 1). The cemetery under discussion, the most important burial ground serving the port–town, was located on Koutsongila, a  low coastal ridge immediately north of the ancient harbor (Figure 2). Kenchreai was the eastern port of Corinth throughout classical antiquity, but the harbor was especially active and the associated settlement was particularly prosperous during the Roman Empire, between roughly the first and seventh centuries AD. Since 2002, Joseph L. Rife has directed the Kenchreai Cemetery Project (KCP), an interdisciplinary program of excavation, conservation and research sponsored by Macalester College (St. Paul, Minnesota) and worked under the auspices of the American School of Classical Studies in Athens and with the permission of the Hellenic Ministry of Culture (Rife et al. 2007). KCP has explored all the known cemeteries at Kenchreai but concentrated on the major site on the Received 23 April 2007; accepted 22 August 2007; published online 15 March 2008 on www.anthropology.uw.edu.pl

36

Douglas H. Ubelaker, Joseph L. Rife

Figure 1. Greece (insert) and the Isthmus of Corinth (J.L. Rife).

Koutsongila Ridge. The site has been selectively studied by Greek and American archeologists since 1906, but KCP has conducted the first comprehensive investigation of the area, responding in part to rampant looting that has continued for decades. Intensive survey and excavation by KCP over five summers has documented 30 subterranean chamber tombs and 28 individual cist graves that had been cut into the bedrock along the seaward slope of the ridge and extending northward into adjacent areas. The most impressive context for burial in the cemetery on Koutsongila was the chamber tombs (Rife et al. 2007). These were large structures cut into the bedrock that were used over a long period (see below) and contained numerous individuals (Rife et al. 2007). Each tomb (Figure 3) has a stairway leading from ground level down into a rectangular chamber area, which measures on average approximately 3.7 m long by 3.3 m wide by 2.5 m high. Certain tomb entrances were enclosed at the surface by a rectangular structure with a monumental

Practice of cremation in the Roman-era cemetery at Kenchreai, Greece

37

Figure 2. Cemetery on the Koutsongila Ridge, 1 m contour interval (C. Mundigler, J.L. Rife, D. Edwards, M.C. Nelson).

38

Douglas H. Ubelaker, Joseph L. Rife

Figure 3. Tomb 13, views of south (top left) and west walls (top right) of Tomb 13 and plan at level of niches (above) and loculi (below) (J.L. Rife).

façade, which marked the site of the tomb, protected the entrance, and displayed an epitaph. Those epitaphs that survive reveal that the tombs were built and first used by parents and their children, followed by their descendents and freedpersons (Rife et al. 2007). The tomb

Practice of cremation in the Roman-era cemetery at Kenchreai, Greece

39

interiors were finished in white or painted plaster and furnished with benches and altars. Each tomb accommodated two different mortuary procedures. The primary burial of bodies most frequently occurred in long, narrow compartments (loculi) cut into the lower part of the chamber walls, though in rare cases cists were cut directly into the chamber floor. In addition, urns with cremated remains were deposited in small niches cut into the upper part of the chamber walls (Rife et al. 2007). The evidence of monumental architecture, wall-painting, and rich funerary artifacts shows that the tombs were owned and used by members of the local elite class. They buried their dead in the chambers over a span of four to eight generations between roughly the middle or late first and the middle or late third centuries AD. Apparently the loculi were subsequently used for further burial during the Late Roman period, or the fifth to sixth or seventh centuries, but these individuals are most likely unrelated to the original owners of the tombs (Rife et al. 2007). Investigation of the burials within the tombs has revealed complex procedures for burial. During the Early Roman phase of use, multiple individuals were placed inside the chambers either as intact corpses or as cremated remains. Bodies were placed in the loculi and covered with a thin layer of soil, after which the compartments were sealed with large limestone slabs or terracotta tiles that could be removed for the addition of other individuals. Cremated bone and tooth fragments without artifacts were placed in heavy, cylindrical urns, which were then deposited in the open niches. Burial during the Late Roman phase of use involved the placement of bodies on top of the loculus covers and possibly the disturbance of the skeletal remains of those previously deposited (Rife et al. 2007). Thus so-called disturbed remains associated with loculi likely represent an aspect of normal mortuary practice for the period. Since antiquity, diverse environmental and anthropogenic factors have altered the preservation of the human remains in the tombs on the Koutsongila Ridge (Ubelaker and Rife, in press). Small animals, especially mice, rats, and snails, have burrowed into the tombs and displaced bones on a small scale. Water transporting fine sediment seems to have continuously entered the tombs through the porous, calcareous bedrock into which the tombs had been cut and directly through the entrances. This moisture and soil stimulated the growth of plant roots, especially from pine trees at the surface. These taphonomic conditions left a clayey crust on some remains and contributed to their general degradation and displacement. The main cause of the disturbance of the tombs and their contents has been illicit digging. Looting on Koutsongila probably began in ancient times, but it has intensified since the 1960s. Looters apparently searching for intact pottery and precious jewelry have tunneled into the tombs and disturbed the contents of loculi, cists, and niches. This clandestine activity, possibly coupled with seismic factors, has relocated human remains from their primary state of deposition in the niches and loculi to scattered, commingled deposits across the tomb floors, and it has mixed the human remains preserved within single loculi.

2. Cremation in antiquity The presence of cremated human remains in association with unburned human remains within the tombs in the cemetery on Koutsongila at Kenchreai represents a mortuary custom known in many areas of the world. As shown by Prothero (2001), the practice of cremation was widespread in ancient times outside Europe and the Mediterranean Basin. For example,

40

Douglas H. Ubelaker, Joseph L. Rife

in North America, cremation is well known among Native Americans from the Southwestern United States, especially in Arizona (Merbs 1967) and in New Mexico (Toulouse 1944), where it increased over time in the prehistoric Mimbres area (Creel 1989). Burned bone and evidence for cremation have been reported at early sites in Australia (Bowler et al. 1970) and Africa (Cain 2005; Sillen & Hoering 1993). Researchers have also studied cremated remains from British sites (Williams 2004) dating to the Early Bronze Age (Brown et al. 1995; Bond 1996) and the Roman Empire (McKinley 2001; Williams 2004; Faber et al., in press). Cremation is also reported at archeological sites in France and Luxembourg (Van Doorselaer 1967; Pajot 1987; Toupet 1987; Polfer 1996, 2001), Norway (Oestigaard 1999), Sweden (Kaliff 1998), Denmark and Germany (Pearson 2000), Holland and Luxembourg (Musgrave 2005), the Czech Republic (Holnerova 1978), and among ancient hunters and fishers of northern Eurasia (Häusler 1968), the peoples of India (Van Gennep 1960), Nepal (Oestigaard 2000), Bali (Downes 1999), and other cultures of Asia (Irion 1968). Metcalf and Huntington (2006) describe practices in groups in Borneo and Bali in which remains are accumulated over long periods of time, cremated and then the ashes and fragments are collected and then stored in wooden containers (Borneo) or dispersed at sea (Bali). Cremation was a common mode of mortuary treatment in Greece from prehistory through the Roman era (Kurtz & Boardman 1971; Morris 1987). Irion (1968) observes that cremation was widespread in ancient Europe after about 1500 BC and entered Greece from the north around 1000 BC. After that point, cremation occurs simultaneously with inhumation, though with variable frequency, down through the Classical period, or until the late fourth century BC (Robinson 1942; Childe 1945; Kurtz & Boardman 1971; Popham et al. 1979; Garland 1985; Morris 1987; Musgrave 1990; Liston 1993; Little & Papadopoulos 2004; Musgrave 2005). During the Hellenistic era, or the late fourth to first centuries BC, cremation was widely practiced in Greece, and it was the standard custom in chamber tombs belonging to the royalty and elites of Macedonia (Prag et al. 1984; Musgrave 1990; Miller 1993), though as noted by Kurtz and Boardman, the transition toward cremation was gradual. Burial by cremation increased in frequency during the Early Roman period, when it occurred with high frequency in many Greek communities but particularly those with ties to Roman Italy, such as the colonies at Patras and Corinth (Blegen et al. 1964; Walbank 2005). This is not surprising, because cremation was the predominant mortuary procedure in the western provinces during this time (Audin 1960; Toynbee 1971). Garland (1985:78) notes that cremation of the very young (infants and young children) was rare in ancient Greece. Papathanasiou (2001) described the possible cremation of two children (1.5 and 3.5 years of age) in addition to primary inhumations from the Neolithic Alepotrypa Cave in southern Greece. The evidence suggests that neonates and infants were sometimes cremated in the Early Bronze Age and Early Iron Age (Branigan 1998). Note also that calcined animal bones (sheep, goat, fox, and cattle) were reported from the Sanctuary of Apollo Hylates at Kourion in Cyprus dating to 2500 years ago (Davis 1996; Buitron-Oliver 1996) and also from sites in Greece (Forstenpointner 2003). Some evidence from prehistoric Greek sites also indicates that human remains can be burned through exposure to the ritual use of fire, and thus may not represent intentional body cremation practices. Gallou (2005) provides a useful summary of such examples, noting that even calcination can result from such accidental exposure.

Practice of cremation in the Roman-era cemetery at Kenchreai, Greece

41

3. Reconstructing Greek cremation practices While it is often difficult to reconstruct the full procedure of cremation from the cremated remains alone, the archaeological study of cremated bone has clearly demonstrated that practices varied considerably over time and from place to place. Cremation in ancient Greece was hardly a  consistent or standardized custom. Research on Greek cremation has focused on evidence from Iron Age, Classical and Hellenistic sites (see references above). No large-scale study has attempted to reconstruct the procedure of cremation in a cemetery of Roman date in Greece, even though it was a widespread practice in the Greek world at this time. In his discussion of the human remains recovered during excavations at the Classical city of Olynthus in northern Greece (fifth to fourth century BC), Robinson (1942) provides some perspective on the likely procedures employed for cremation in the area. Robinson noted that although evidence for cremation occurred at a variety of archeological sites in Greece, methodology and frequency varied considerably. For example, Robinson indicates “no cremations were reported from Corinth” (1942:145), while about 94% of burials at the site of Vroulia on Rhodes (eighth to fourth century) involved cremations. At Olynthus, about 25% of the cremations recovered were thought to have originated from children, but infants were not included. Robinson (1942) notes that in some areas of Greece, the expense of cremation, particularly the cost of wood for fuel, may have been a factor influencing decisions regarding cremation versus inhumation. However, Robinson argues this was probably not the case in Olynthus, where wood presumably was relatively plentiful. Other possible motivations for cremation include the transport home of individuals who had died abroad, general attitudes toward the significance of death, the conferring of public honors, and the expression of wealth and status (Irion 1968). The final decision likely was an individual or family one. The procedure, as interpreted by Robinson (1942), initially involved washing and dressing the body, usually by family members. After lying in state, the body was carried to the crematory on a bier or coffin and placed on the pyre. The pyre was constructed within a trench, and, if the cremation represented the primary burial site, the residue following cremation was covered with soil. Robinson suggests that in constructing the pyre, straw or hay was first placed on the floor of the trench to serve as kindling. Wood was then stacked in layers upon the kindling. Robinson notes that the wood at Olynthus likely originated from deciduous trees, but that a variety of types of wood could have been utilized. The pyre also could be constructed on the ground surface, especially for those intended for secondary deposition elsewhere. Such pyres probably had better air flow during the burning. Trench cremations, especially deep ones, often had secondary vent–type structures, which enhanced the circulation of air (Kurtz & Boardman 1971). Kurtz and Boardman (1971) note that trench cremations frequently are thought to be primary, meaning that the remains of the individual were left in the trench and not transported elsewhere. Secondary cremains, in which remains are transported to another location, frequently require the use of ash urns. A similar method has been reconstructed for the cremation of the Macedonian royalty in the Late Classical period, or the late fourth century BC (Prag et al. 1984). Musgrave (Prag et al. 1984) has proposed that the central individual who was cremated and deposited in Tomb II at Vergina, identified by some as King Philip II, was burned in an oven. The body was enclosed in a brick box, which was packed with fuel and heated up to 900oC for as long as 45 to 90 minutes. The bones show the degree of warping, fracturing, and shrinkage typical of exposure to such intense heat over this interval, but the skeleton is relatively intact, Musgrave argues, because it was burned in the container rather than on an open pyre. While the extended application of heat to the skeleton

42

Douglas H. Ubelaker, Joseph L. Rife

would have been typical of Greek cremations, the testimony of ancient art and literature indicate that the use of a pyre, not a brick box or oven, was the conventional practice. Ethnographic analogy helps to draw a plausible picture of Greek methods at the usual site of the pyre. Oestigaard (2000) notes that in Nepal, as the pyre cremation advances, those in charge used long sticks to reorganize the burning wood and to break apart the burning human remains. Such intervention during the cremation was likely needed to ensure that ample fuel was present, that proper airflow was maintained, and that the heat was directed toward the human remains.

4. The evidence for cremation at Kenchreai As noted above, the excavation of tombs on the Koutsongila Ridge at Kenchreai has revealed that cremation was practiced in addition to inhumation. Although cremated remains were found primarily in niches, small quantities were also found dispersed throughout deposits on the chamber floors and rarely in loculi. The past activities of people inside the tombs, including occasional squatters and frequent looters, have apparently displaced cremated bones from the niches, where they were originally deposited, to the floors. This movement is understandable if we imagine, for instance, that residents removed the sturdy cinerary urns from the niches for practical re-use outside the tombs but first dumped out the contents on the floor, or that looters sifted through the cremated fragments in the niches in search of artifacts. During such activity, it seems that a small quantity of cremated fragments was sometimes inadvertently moved from the niches or swept from the floors into the fronts of the loculi. Table 1 summarizes the amount of human remains in three tombs, nos. 13, 14, and 22. These tombs were selected for this discussion because their excavation and analysis are complete, and they provide good evidence for interpreting cremation and inhumation in the cemetery. Of the 12 tombs so far explored on Koutsongila by KCP and earlier archaeologists, these three provide the most comprehensive view of cremation at Kenchreai. Two other tombs at the site (nos. 4 and 9), excavated in 1990 and 1969, respectively, have produced very small quantities of cremated bone (35 g and 152 g, respectively). These have not been included here because they were only partially excavated and all the remains were not recovered. A third tomb (no. 10), has produced large quantities of cremated bone, but it is still under study. As KCP expands into a large–scale excavation in 2007–2009, it is hoped that further investigation will uncover more evidence for mortuary practices, in particular cremation, at Roman Kenchreai. Table 1. Comparison of the weights of burned remains versus the total weights of samples of human remains recovered from three tombs in the cemetery on Koutsongila.

Tomb

Total Weight (g)

Total Weight Burned (g)

% Burned

13

12,245

2,949

24

14

34,235

2,471

7

22

12,486

221

2

Practice of cremation in the Roman-era cemetery at Kenchreai, Greece

43

As shown in Table 1, the total weight of human remains recovered from the three tombs varies dramatically, despite careful and exhaustive excavation and complete recovery of all materials within the tombs. At the extremes, Tomb 13 yielded 12,245 g and Tomb 22 yielded 12,486 g, while Tomb 14 produced over two and half times as much bone, or 34,235 g. This disparity of representation reflects not only the different numbers of individuals deposited within each tomb during the Roman era, but also the differential destruction of human remains by natural and cultural processes. Over the centuries, the slow influx of sandy to gravelly sediments along with numerous cobbles, scattered structural debris, and occasional boulders through the opened tomb entrances, together with the periodic collapse of calcareous bedrock from the ceilings and walls of the chambers, have crushed and pulverized many smaller bone fragments. The violent displacement of bones by looters, who have churned through deposits in niches and loculi to find valuables, has also led to the destruction of already fragmentary remains. It is, however, highly improbable that interlopers in the tombs after antiquity have collected and removed large quantities of bone or introduced bone that was originally deposited in other chamber tombs. Apart from the possible extraction of select bones during the Late Roman to Byzantine era (Ubelaker & Rife in press), it seems that no bones have left or entered the tombs since ancient times. The representation of individuals also varied considerably between tombs. While the relatively large deposits within Tomb 14 represented at least 52 individuals, the smaller deposits within Tombs 13 and 22 represented at least 19 and 13 individuals, respectively. Thus, at least 84 individuals are represented in the three tombs, with a mean of at least 28 individuals per tomb. The minimum number of individuals within each tomb correlates generally with the amount of human material recovered, but it should be noted that the calculation of the minimum number of individuals (MNI) has involved certain assumptions. In the cases of Tombs 13 and 14, the calculation assumes that there was no mixing of bones from separate individuals between loculi or between niches, that the movement of cremated bone fragments from the niches to the loculi was minimal and easily recognized, and that material recovered from the tomb floor originated from both the niches and loculi. The calculation of MNI in Tomb 22 required the combined study of bone from all contexts, because both burned and unburned bone was found only in commingled deposits on the chamber floor, below the niches, and above the looted cists that were built into the floor. Separate tombs also displayed different frequencies of burned material within the bone assemblage. Of the 58,966 g of all human remains recovered from the three tombs, only 5,641 g, or 9.6%, revealed clear evidence of burning. Of the 5,641 g of burned bone, most originated from Tomb 13, even though Tomb 14 yielded more human remains overall. The cremated fragments ranged in color from gray to white, reflecting the results of extreme heat over a long duration (Figure 4). Many of the calcined (white) fragments reveal transverse fracture patterns, warping and twisting suggestive of burning in the flesh (Figure 5). Some of these fragments (Figure 6) display extreme alterations in bone morphology suggesting prolonged exposure to high temperatures. Fragmentation, especially of the cremated material, limited the inventory of skeletal elements and the interpretation of sex and age at death. Within the total sample of burned bone from the three tombs, all different regions of the skeleton were represented. The estimated ages at death among the cremated individuals ranged from young child to adult. At least four subadults were present, two in Tomb 14, a two to three year old child in Tomb 22 and another child of undetermined age from Tomb 22. No cremated infant remains were detected. Standard methods were employed for the estimation of age at death (Ubelaker 1999), as appropriate.

44

Douglas H. Ubelaker, Joseph L. Rife

Figure 4. Calcined bone fragments from the chamber of Tomb 13 showing warping and transverse fracture patterns indicative of in-flesh cremation (J.L. Rife).

Figure 5. Calcined cortical bone fragment from Loculus V in Tomb 13 (locus T13–040) showing warping and transverse fracture patterns indicative of in-flesh cremation (A.J. Suehle).

Figure 6. A calcined left radius from Loculus IV in Tomb 13 (locus T13–037) displaying severe alterations in morphology indicative of prolonged exposure to high temperatures (A.J. Suehle).

Practice of cremation in the Roman-era cemetery at Kenchreai, Greece

45

5. Approaches to the analysis and interpretation of cremation In the past cremated bone from archeological sites was discarded because of its presumed lack of analytical potential. But in recent decades, the experience brought by physical anthropologists (Cook 1999), modern forensic work, commercial cremations, experimentation and broader concepts of research design have led to a growing literature on what information can be extracted from such samples (e.g., Musgrave 1990, 2005; Liston 1993, 2007). A variety of authors have addressed methodological approaches to the analysis and interpretation of cremation in different contexts (Gejvall 1969; Lisowski 1968; Kurzawski et al. 1987; Masset 1987; Mayne Correia & Beattie 2002; McKinley & Roberts 1993; McKinley 2000, 2001; Thompson 2004). Together these sources reveal the broad range of information that can be gleaned from the study of cremation when the proper methods are applied.

5.A. Perspective from forensic science The practice of modern forensic science furnishes considerable insight into the analysis of cremation. Even commercial cremations present situations requiring forensic investigation, which in turn yield valuable information on the cremation process. For example, in Tucson and Pima County, Arizona in 2002–2003, 0.8% of requests for cremations became medical examiner cases after appropriate review (Nelson & Winston 2006). Burning cases also result from criminal attempts to destroy forensic evidence (Fanton et al. 2006), from suicide (Shkrum & Johnston 1992), and from a variety of disasters such as aircraft accidents, bombing and explosive use, and earthquakes (Sledzik & Rodriguez 2002). Many of these forensic scenarios present challenges for evidence recovery (Dirkmaat 2002) and analysis. Heat related changes to the human remains frequently limit analysis and shift the focus toward those aspects that survive (Owsley et al. 1995; Ubelaker & Scammell 1992; Ubelaker et al. 1995; Houck et al. 1996; Ubelaker 1999). In a typical house fire, ceiling temperatures increase to 500oC in about 10 minutes, climb to between 700 and 800oC in about 55 minutes, and subsequently diminish (Richards 1977), providing an opportunity for the extensive heat alteration to human victims. Forensic experience has yielded considerable information on both the effects of heat on the human body and the limitations to analysis. For fleshed remains, extreme heat produces the classic pugilistic pose in which post-mortem muscle contracture causes joint flexure. According to Saukko and Knight (2004), the muscle in such conditions is shortened by “dehydration and protein denaturation,” causing the bulkier flexors to contract more that the extensors. As soft tissue is reduced by heat exposure, bone is increasingly exposed, creating a  contrast between heat-related changes in the exposed bone and in the bone still protected by soft tissue. The analysis of burned bone recovered in a forensic context involves such diverse approaches as simple reconstruction of fragments to form larger segments or complete bones (Grèvin et al. 1998), scanning electron microscopy (Carr et al. 1986; Bush et al. 2006), and elemental analysis (Brooks et al. 2006), including trace element analysis (Herrmann & Grupe 1988). Trace element data may, however, reveal environmental factors, and it may be limited due to “volatilization and crystal modification” (Grupe & Hummel 1991). In spite of the fragmentation that is frequently associated with skeletal exposure to extreme temperatures, evidence of pre-existing trauma can survive (Pope & Smith 2004), especially sharp force trauma (Her-

46

Douglas H. Ubelaker, Joseph L. Rife

rmann & Bennett 1999), including chop marks (De Gruchy & Rogers 2002). Although DNA recovery can be challenging in cases involving burned bone, techniques are available to increase both the purity and yield of DNA in such cases (Ye et al. 2004).

5.B. Perspective from modern cremation practice The growing popularity of modern cremation as a method of disposal of the dead gives an additional perspective on the effect of high temperature on human remains. As noted by Wells (1960), crematoria use gas jets to reach temperatures of 820 to 980o C, or 1500 to 1600o F (Eckert et al. 1988). These regulated temperatures and burning times can be correlated with changes in bones and teeth. The alterations to bones reveal both the sequence of soft tissue reduction and the body position in relation to the heat source (Wells 1960). Reporting on the practice of cremation in the United Kingdom, McKinley (1989) has observed that modern retorts (firing chambers) reach temperatures as low as 500oC and as high as 1000o C. McKinley (1994) also reports the weights of remains resulting from the cremations of 15 individuals in the United Kingdom. Male weights ranged from 1735 g to 3001 g with a mean of 2284 g. The female range was 1227 g to 2216 g, with a mean of 1616 g. Warren and Maples (1997) have presented a study of 100 individuals who had been commercially cremated. In particular, they compare the pre–cremation body weight with the weight of the recovered cremated material. For adults, the weights of the latter ranged from 876 g to 3,784 g, with a mean of 2,430 g. They found that all individual weights above 2750 g originated from males and those below 1887 g originated from females. The resulting cremated matter represented 3.5% of the body weight in adults, 2.5% in children, and about 1% in fetuses. Murad (1998) summarized unpublished research by Alexander Sonek in 1992 on the weights of the cremated remains of 150 individuals. Although these materials were not separated by sex, the weight of remains from individuals ranged from 892 g to 4,056 g, with a mean of 2,348 g. Murad (1998) also noted that, though the heat source in retorts is shut down at about 1600o F (889o C), temperatures within the retort can climb as high as 2300oF (1278oC). Those analyzing materials from commercial cremations face the challenge of extreme skeletal fragmentation but recognize that non-skeletal inclusions, such as dental restorations and surgical materials, can facilitate identification. One such analysis even reported evidence of arteriosclerosis that had survived in cremated material (Warren et al. 1999). The nature of the equipment used in commercial cremations also contributes to the qualities of the resulting human materials (Warren & Schultz 2002). DeHaan and Nurbakhsh (2001) make the important point that the body itself can represent a fuel source in the process of incineration. In fact, the body can become a primary heat source if an initial outside heat source is present along with a “wick” and adequate oxygen. Bass and Jantz (2004) have also addressed the topic of cremation weights. In a regional study of commercial cremations in East Tennessee with retort temperatures documented between 1600oF and 1800oF, cremation time varied between two and three hours, and several hours were required for the retort to cool down before removal of the cremations. Male cremations weighed between 1,865 g and 5,379 g, with a mean of 3,380 g. Weights of females ranged from 1,050 g to 4,000 g, with a mean of 2,350 g.

Practice of cremation in the Roman-era cemetery at Kenchreai, Greece

47

6. Experimentation Considerable research and experimentation has generated useful data on heat-induced alterations in bone (Thompson 2005). Experiments by Baby (1954), which were later supported by Binford (1963) and Thurman and Willmore (1981), have suggested that remains cremated as dry bones could be distinguished from those cremated with fresh and/or flesh covered bones. According to Baby (1954), the former exhibited superficial surface checking, longitudinal splitting, and lack of warping, while the latter were characterized by warping with transverse fractures. Van Vark (1970) conducted controlled experiments with an electric oven in which bones were measured and examined after successive increases of 100oC. Van Vark noted that the bones became increasingly brittle but no shrinkage was recorded until about 700o. Between 700 and 800o, the bones became white with extensive shrinkage and loss of histological features. Between 800o and 1500o no additional changes were noted. Van Vark also introduced statistical procedures to evaluate sex and age at death from calcined remains (Van Vark 1970, 1974, 1975). It is important to note that many publications discuss the typical color change with burning from black to gray to white, but substances in the environment can cause variations of this pattern (Dunlop 1978). Shipman et al. (1984) conducted additional experiments using sheep and goat mandibles in a test kiln to document heat related changes in bone, dentin and enamel. Their study recorded changes in coloration, microscopic morphology, size, and crystal structure. They made the key point that the duration of the heat exposure is important, because the temperature of the heat source is usually distinct from the bone temperature. Although it has been common knowledge that cremation leads to bone fragmentation, Chandler (1987) noted that tooth size and the ability to recognize individual tooth types are affected by cremation. Christensen (2002) added that osteoporotic bone is more likely to fragment than normal bone. Experiments with electron miscroscopy reported by Wilson and Massey (1987) have shown that structural changes in tooth dentin begin at about 600oC and at 800oC enamel rod structure becomes altered. Additional experimentation by Spennemann and Colley (1989) noted that the surface texture of bone changes with exposure to temperatures of 500oC and above. They also made the important observations that staining results from materials in the environment, fragmentation is augmented during recovery, and increased heat produces increasing whiteness in bone but coloration exhibits great variation. Holland (1989) added that exposure to low temperature (less than 800oC) produces minimal shrinkage of bone. David (1990) conducted a series of experiments testing hypotheses on the factors in patterns of burning in bone. A controlled bushfire produced no calcinations but brown to black color changes and some fractures. A campfire using eucalyptus wood reached a temperature of 840oC in one hour and five minutes and produced bone fractures, some gray–black coloration, but mostly bone fragments with a white, calcined appearance. Some bones removed from the campfire after 25 minutes displayed fractures and a color range of brown, black, gray and white/blue. The study documented that the duration of the fire, not just the temperature, was a key factor in bone alteration. Similar experiments by Stiner et al. (1995) emphasized that calcination was linked to exposure to live coals. Post-mortem taphonomic issues complicate interpretations of fragment size. Certainly conditions of cremation contribute to fragmentation, but taphonomic factors such as mortuary treatment and even techniques of excavation and processing also lead to increased fragmentation (McKinley 1994).

48

Douglas H. Ubelaker, Joseph L. Rife

Key experiments by Holden et al. (1995) documented that in cremations, the combustion of the organic component of bone occurred up to 400oC. Beginning at 600oC, bone mineral displayed recrystallisation. Beginning at 1600oC, the bone mineral melted. This research shows how temperature is related to basic morphological transformations of cremated bone. In a review of the literature, Mayne Correia (1997) relates several important points concerning cremation. Calcination with a white color represents a complete loss of organic material and fusion of bone salts. The extent of shrinkage depends on the type of bone, the temperature and the amount of bone mineral present. Weight (stress) on the bones during cremation can produce plastic deformation at high temperatures. Denser bones and those protected by soft tissue will demonstrate better preservation and less fragmentation. The literature presents conflicting evidence on the ability to distinguish bones burned with or without flesh attached. Mayne Correia noted the similarity of bone to ceramics in the nature of heat reaction: dehydration is followed by pyrolysis of organic material, followed by loss of carbonates and crystal conversion, followed by fusion. Quatrehomme et al. (1998) failed to find a correlation between temperature and patterns revealed through examination with the scanning electron microscope. Using transmission electron microscopy, Koon et al. (2003) detected changes in collagen fibrils following the roasting of sheep humeri. Hurlbut (2000) noted that bone exposed to minimal heat is difficult to distinguish from unburned bone. Whyte (2001) emphasized the nature of cracks and fissures, as well as the warping noted by many others, in distinguishing bones burned in the flesh versus those burned as dry bones. Walker and Miller (2005) noted that oxygen availability, duration and temperature of the heat all affect resulting bone color. Experimentation has also focused on the correlations between heat and the preservation of DNA, dental pulp and albumin. Tsuchimochi et al. (2002) found that DNA amplification and typing of dental pulp was successful up to a temperature of 300oC but not at higher temperatures. Duffy et al. (1991) described a procedure to facilitate sex chromatin counts from tooth pulp cells but noted that the extent and duration of heat as well as the extent of flesh present affect it. Duffy et al. (1991) also noted the difference between fire temperature and temperature of the specimen. Cattaneo et al. (1994) added that human albumin survives cremation up to temperatures of 300oC but no higher. Bone integrity and exposure to water affect protein survival in such cases (Cattaneo et al. 1995). Although Sanjantila et al. (1991) reported successful DNA typing on all 26 samples from 10 fire victims exhibiting extensive charring, Cattaneo et al. (1999) reported that no mitochondrial DNA samples could be amplified in bone experiments with temperatures reaching 800 to 1200oC. In attempting to determine human versus non-human status in fragments of burned bone using standard microscopy, quantitative microscopy, albumin analysis, and DNA analysis, quantitative microscopy emerged as the method of choice. In a discussion of human cremation in an outside wood fire, McKinley (1989) noted that up to seven or eight hours would be required for calcination, and fuel would need to be added during the process. The center of the fire would be much hotter than the periphery. McKinley (1989) also argued that, in ancient times, bodies were most likely placed on top of the pyre to promote oxygen access, and stirring would be necessary during the process.

7. Interpretation of the cremated remains from Kenchreai This forensic and experimental research on cremation can shed light on the interpretation of the cremated remains from the Roman-era cemetery on Koutsongila at Kenchreai, especially

Practice of cremation in the Roman-era cemetery at Kenchreai, Greece

49

in regard to the number of individuals represented, the duration and temperature of the original cremation event, and the specific character of the mortuary procedure.

7.A. Weights A relatively large quantity of cremated remains (2,949 g) was recovered from several depositional contexts representing the contents of separate niches in Tomb 13. The remains were highly fragmentary, and therefore morphological analysis could detect the presence of only one adult older than 13 years of age. According to the estimate by Warren and Maples (1997) that the cremated remains of one adult weigh on average 2,430 g, the total weight of cremated bone from Tomb 13 could represent just over one individual. However, it is known on archaeological grounds that these combined bone fragments originated in several niches containing the burials of multiple individuals. Therefore, it is reasonable to conclude that only a fraction of all cremated remains was recovered from the burial site for deposition in the tomb. Tomb 14 presents a similar scenario for interpretation. The total sample of cremated remains from this tomb weighs 2,471 g, which falls within the estimated range of weights for the cremated remains from one adult (Warren & Maples 1997; Bass & Jantz 2004). But analysis of this sample detected the presence of at least two subadults and one adult-size individual. Moreover, as in Tomb 13, the cremated remains recovered from Tomb 14 represent multiple individuals buried in separate niches. To be sure, the weights of subadult cremated remains are less than those of adults. Nonetheless, the analysis of combined data and consideration of the archaeological context together demonstrate that the cremated remains from Tomb 14 comprise but a fraction of the total amount that originally existed at the primary cremation site(s). In contrast to Tombs 13 and 14, Tomb 22 contained very few bone fragments with signs of burning (121 g). According to the standard estimate of the weight of cremated remains from one adult, this quantity would constitute less than 5% of one skeleton. However, examination of the sample of burned bone from Tomb 22 has shown that elements from at least two subadults and at least one adult were buried here. The gradual filling of the chamber with sediment has undoubtedly crushed numerous small fragments. But this process cannot account for such a significant disparity between the preserved weight of the cremated bone sample in Tomb 22 and the expected total weight of cremated bone from at least, and probably many more than, three subadult and adult bodies. The explanation must be mortuary procedure. As in the case of Tombs 13 and 14, it appears that mourners at Roman Kenchreai collected only a small percentage of all cremated remains from the pyre at the primary event of cremation. Note that Liston (2007) reached similar conclusions of minimal representation in her study of secondary cremation burials dating to the Early Iron Age at Kavousi Vronda, Crete. In other cases, however, such as the “rich Athenian lady” buried during the Early Iron Age in the area of the Athenian Agora, mourners recovered a large percentage of the skeleton from the pyre (Little & Papadopoulos 2004).

7.B. Condition of individuals at the time of cremation A majority of the cremated fragments recovered from the tombs on Koutsongila exhibit the patterns of calcination, fragmentation, fractures, fissures, transverse fracturing and warping

50

Douglas H. Ubelaker, Joseph L. Rife

(Figures 4–6) consistent with having been burned as fresh bone with the flesh still attached, as opposed to having been burned dry, without flesh. Of course, not all burned fragments show such diagnostic qualities; some displayed an intact form but distinct discoloration. The experimental studies cited above indicate clearly that such variation is normal, even in cremations that reached relatively high temperatures. Research reported by Baby (1954), Binford (1963), Thurman and Willmore (1981), and Ubelaker (1999), among others, together indicate that the overall pattern displayed by most burned fragments in the three tombs at Kenchreai is consistent with the burning of fleshed remains.

7.C. Burning temperatures and duration The overwhelming evidence of calcined bones within the tombs shows that, during the initial event of cremation, pyre temperatures reached at least 700oC at the level of the body. The research discussed above collectively suggests that mourners employed a pyre constructed of available wood, perhaps with kindling. The body was likely placed on top of the pyre, but certainly it would have shifted toward the base of the pyre as the fire progressed. Alternative approaches may have involved placement of the individual on a burial bed surrounded by wood fuel or use of a  properly oxygenated trench structure. To achieve the extent of calcination observed in the fragments from the tombs, the individuals associated with the event of cremation probably added fuel during the burning process and may have promoted body reduction by manual prodding with long sticks or similar instruments. The greatest heat changes likely occurred when the burning had advanced so far that the human remains were in contact with the hot coals at the base of the exhausted pyre. Some bone fragments recovered from the tombs, such as a radius recovered from Tomb 13 (Figure 6), display not only calcination and other evidence of exposure to high temperatures for a long period but also extreme morphological disfigurement. This condition appears consistent with the “melting” effect described in the literature (e.g. Holden et al. 1995), in which temperatures reaching 1600oC transform bone salts. Such conditions attest to the care and attention given to the cremation event, the extent and quality of the fuel, as well as the long duration, which must have lasted several hours or even days.

8. Summary Careful excavation and analysis of three tombs in the cemetery on the Koutsongila Ridge at Kenchreai in Greece have provided important evidence for the role of cremation in the mortuary practices in the community during the early to middle Roman periods, or the middle to late first to middle or late third century AD. Local residents chose to dispose of their dead using both inhumation and cremation. Methods and conclusions gleaned from the rapidly growing body of published studies in forensics and experimental research, coupled with archeological interpretation, have contributed to the understanding of cremation at Roman Kenchreai. This study suggests that cremation pyres were constructed from a substantial amount of wood and were closely attended during the procedure of cremation. This likely involved the addition of fuel to prolong the fire and the rearrangement of the human remains during the process which resulted in extreme reduction. The extreme heat-related alterations evident in

Practice of cremation in the Roman-era cemetery at Kenchreai, Greece

51

certain bone fragments point to long contact with very hot coals. In addition, it appears that only a relatively small percentage of the cremated remains were recovered and transferred to the niches in the tombs. It is uncertain why mourners did not recover more remains from the site of cremation. Perhaps they thought that only a selection of bone fragments was sufficient to represent the deceased at the burial site. Moreover, the recovery of minute pieces from a large heap of carbonized wood must have been a difficult task. Despite a complex history of use in ancient times and extensive disturbance since antiquity, the cemetery on Koutsongila is furnishing much information about the people buried there and the community they represent. The evidence for cremation at Kenchreai attests to a common mortuary procedure during the Early Roman period in Greece, one in which great care was devoted to the dead. As the exploration of this rich site expands in coming years, it is hoped that new evidence will illuminate this important chapter in the history of Greek burial and society.

Bibliography Audin A. (1960), Inhumation et incinération, Latomus 19:312-322, 518-532. Baby R. (1954), Hopewell Cremation Practices, The Ohio Historical Society Papers in Archaeology 1:1-17. Bass W.M., Jantz R.L. (2004), Cremations Weights in East Tennessee, Journal of Forensic Sciences 49:900-904. Binford L.R. (1963), An Analysis of Cremations from Three Michigan Sites, Wisconsin Archaeologist 44:98-110. Blegen C.W., Palmer H., Young R.S. (1964). Corinth XIII: The North Cemetery, Princeton: American School of Classical Studies at Athens. Bond J.M. (1996), Burnt Offerings: Animal Bone in Anglo-Saxon Cremations, World Archaeology 28:76-88. Bowler J.M., Jones R., Allen H., Thorne A.G. (1970), Pleistocene Human Remains from Australia: a Living Site and Human Cremation from Lake Mungo, Western New South Wales, World Archaeology 2:39-60. Branigan K. (1998), The Nearness of You: Proximity and Distance in Early Minoan Funerary Landscapes [in:] “Cemetery and Society in the Aegean Bronze Age”, K. Branigan (ed.), Sheffield: Sheffield Academic Press, pp. 13-26. Brooks T., Bodkin T., Potts G., Smullen S. (2006), Elemental Analysis of Human Cremains Using ICP-OES to Classify Legitimate and Contaminated Cremains, Journal of Forensic Sciences 51:967-973. Brown K.A., O’ Donoghue K., Brown T.A. (1995), DNA in Cremated Bones from an Early Bronze Age Cemetery Cairn, International Journal of Osteoarchaeology 5:181-187. Buitron-Oliver D. (1996), The Sanctuary of Apollo Hylates at Kourion: Excavations in the Archaic Precinct, Jonsered, Sweeden: Paul Åströms förlag. Bush M.A., Bush P.J., Miller R.G. (2006), Detection and Classification of Composite Resins in Incinerated Teeth for Forensic Purposes, Journal of Forensic Sciences 51:636-642. Cain C.R. (2005), Using Burned Animal Bone to Look at Middle Stone Age Occupation and Behaviour, Journal of Archaeological Science 32:873-884. Carr R.F., Barsley R.E., Davenport W.D. (1986), Postmortem Examination of Incinerated Teeth with the Scanning Electron Microscope, Journal of Forensic Sciences 31:307-311.

52

Douglas H. Ubelaker, Joseph L. Rife

Cattaneo C., Gelsthorpe K., Sokol R.J., Phillips P. (1994), Immunological Detection of Albumin in Ancient Human Cremations Using ELISA and Monoclonal Antibodies, Journal of Archaeological Science 21:565-571. Cattaneo C., Gelsthorpe K., Phillips P., Sokol R.J. (1995), Differential Survival of Albumin in Ancient Bone, Journal of Archaeological Science 22: 271-276. Cattaneo C., Di Martino S., Scali S., Craig O.E., Grandi M., Sokol R.J. (1999), Determining the Human Origin of Fragments of Burnt Bone: a Comparative Study of Histological, Immunological and DNA Techniques, Forensic Science International 102:181-191. Chandler N.P. (1987), Cremated Teeth, Archaeology Today August:41-45. Childe V. (1945), Directional Changes in Funerary Practices During 50,00 Years, Man 45:13-19. Christensen A.M. (2002), Experiments in the Combustibility of the Human Body, Journal of Forensic Sciences 47:466-470. Cook D.C. (1999), Physical Anthropology in the Field: Recognizing Cremation, Defleshing, Exposure and Secondary Burial [in:] “The Practical Impact of Science on Near Eastern and Aegean Archaeology”, S. Pike, S. Gitin (ed.), London: Archetype Books, pp. 43-46. Creel D. (1989), a Primary Cremation at the NAN Ranch Ruin, with Comparative Data on Other Cremations in the Mimbres Area, New Mexico, Journal of Field Archaeology 16:309-329. David B. (1990), How was This Bone Burnt? [in:] “Problem Solving in Taphonomy: Archaeological and Paleontological Studies from Europe, Africa and Oceania”, S. Solomon, I. Davidson, D. Watson (ed.), Tempus 2, Queensland: University of Queensland, pp. 63-79. Davis J.M.S. (1996), Animal Sacrifices [in:] “The Sanctuary of Apollo Hylates at Kourion: Excavations in the Archaic Precinct”, D. Buitron-Oliver (ed.), Jonsered, Sweeden: Paul Åströms förlag, pp. 181-182. DeHaan J.D., Nurbakhsh S. (2001), Sustained Combustion of an Animal Carcass and its Implications for the Consumption of Human Bodies in Fires, Journal of Forensic Sciences 46:1076-1081. Dirkmaat D. (2002), Recovery and Interpretation of the Fatal Fire Victim: The Role of Forensic Anthropology [in:] “Advances in Forensic Taphonomy: Method, Theory, and Archaeological Perspectives”, W.D. Haglund, M.H. Sorg (ed.), Washington DC: CRC Press, pp. 451-472. Downes J. (1999), Cremation: a Spectacle and a Journey [in:] “The Loved Body’s Corruptions: Archaeological Contributions to the Study of Human Mortality”, J. Downes, T. Pollard (ed.), Great Britain: Cruithne Press, pp. 19-29. Duffy J.B., Waterfield J.D. and Skinner M.F. (1991), Isolation of Tooth Pulp Cells for Sex Chromatin Studies in Experimental Dehydrated and Cremated Remains, Forensic Science International 49:127-141. Dunlop J. (1978), Traffic Light Discoloration in Cremated Bones, Medicine, Science, and the Law 18:163-173. Eckert W., James S., Katchis S. (1988), Investigation of Cremations and Severely Burned Bodies, American Journal of Forensic Medicine and Pathology 9(3):188-200. Faber A., Fasold P., Struck M., Witteyer M. (ed.) (in press), Inhumations in the Roman Empire from the 1st until the end of the 3rd century A.D., Schriften des Archäologischen Museums Frankfurt 21, Frankfurt: Archäologisches Museum Fanton L., Jdeed K., Tilhet-Coartet S., Malicier D. (2006), Criminal Burning, Forensic Science International 158:87-93. Forstenpointner G. (2003), Promethean Legacy: Investigations into the Ritual Procedure of ‘Olympian’ Sacrifice [in:] “Zooarchaeology in Greece: Recent Advances”, E. Kotjabopoulou, Y. Hamilakis, P. Halstead, C. Gamble, P. Elefanti (ed.), London: The British School at Athens 203-213.

Practice of cremation in the Roman-era cemetery at Kenchreai, Greece

53

Gallou C. (2005), The Mycenaean Cult of the Dead, BAR International Series 1372, Oxford: Archaeopress. Garland R. (1985). The Greek Way of Death, Ithaca: Cornell University Press. Gejvall N. (1969), Cremations [in:] “Science in Archaeology.” 2nd Edition, D. Brothwell, E. Higgs (ed.), New York: Prager. Grévin G., Bailet P., Quatrehomme G., Ollier A. (1998), Anatomical Reconstruction of Fragments of Burned Human Bones: a Necessary means for Forensic Identification, Forensic Science International 96:129-134. De Gruchy S., Rogers T.L. (2002), Identifying Chop Marks on Cremated Bone: a Preliminary Study, Journal of Forensic Sciences 47:933-936. Grupe G., Hummel S. (1991), Trace Element Studies on Experimentally Cremated Bone I. Alteration of the Chemical Composition at High Temperatures, Journal of Archaeological Science 18:177-186. Haüsler A. (1968), Burial Customs of the Ancient Hunters and Fishers of Northern Eurasia, Arctic Anthropology 5:62-67. Herrmann B., Grupe G. (1988), Trace Element Content in Prehistoric Cremated Human Remains [in:] “Trace Elements in Environmental History: Proceedings of the Symposium from June 24th to 26th, 1987, at Gottingen”, G. Grupe, B. Herrmann (ed.), Germany: Springer-Verlag, pp. 91-101. Herrmann N.P., Bennett J.L. (1999), The Differentiation of Traumatic and Heat-Related Fractures in Burned Bone, Journal of Forensic Sciences 44:461-469. Holden J.L., Phakey P.P., Clement J.G. (1995), Scanning Electron Microscope Observations of Heat-Treated Human Bone, Forensic Science International 74:29-45. Holland T.D. (1989), Use of Cranial Base in the Identification of Fire Victims, Journal of Forensic Sciences 34:458-460. Holnerova E. (1978), Cremation Graves in the Bronze Age, Anthropologie 16(2):127-130. Houck M.M., Ubelaker D.H., Owsley D.H., Craig E., Grant W., Fram R., Woltanski T., Sandness K. (1996), The Role of Forensic Anthropology in the Recovery and Analysis of Branch Davidian Compound Victims: Assessing the Accuracy of Age Estimations, Journal of Forensic Sciences 41:796-801. Hurlbut S.A. (2000), The Taphonomy of Cannibalism: a Review of Anthropogenic Bone Modification in the American Southwest, International Journal of Osteoarchaeology 10:4-26. Irion P.E. (1968), Cremation, Philadelphia: Fortress Press. Kaliff a (1998), Grave Structures and Altars: Archaeological Traces of Bronze Age Eschatological Conceptions, European Journal of Archaeology, 1(2):177-198. Koon H.E.C., Nicholson R.A., Collins M.J. (2003), a Practical Approach to the Identification of Low Temperature Heated Bone Using TEM, Journal of Archaeological Science 30:1393-1399. Kurtz D.C., Boardman J. (1971), Greek Burial Customs. Ithaca, New York: Cornell University Press. Kurzawski V., Bouville C., Totoyan C. (1987), Fouille d’un ensenble des sépultures àcremation à Martigues (Bouches-du-Rhône) [in:] “Anthropologie physique et archéologie: méthodes d’étude des sépultures. Actes du colloque de Toulouse, 4, 5 et 6 novembre 1982”, H. Duday, C. Masse (ed.), Paris: Presses du CNRS, pp. 67-72. Lisowski F. (1968), The Investigation of Human Cremations, Anthropologie und Humangenetik 4:76-83. Liston M.A. (1993), The Human Skeletal Remains from Kavousi, Crete: a Bioarchaeological Analysis, Dissertation for the Doctor of Philosophy Degree, Knoxville: University of Tennessee. Liston M.A. (2007), Secondary Cremation Burials at Kavousi Vronda, Crete: Symbolic Representation in Mortuary Practice, Hesperia 76:57-71.

54

Douglas H. Ubelaker, Joseph L. Rife

Little L.M., Papadopoulos J.K. (2004), The “Rich Athenian Lady” was Pregnant: The Anthropology of a Geometric Tomb Reconsidered, Hesperia 73:7-38. Masset C. (1987), Le “Recrutment” d‘un ensemble funéraire [in:] “Anthropologie physique et archéologie: méthodes d’étude des sépultures. Actes du colloque de Toulouse, 4, 5 et 6 novembre 1982”, H. Duday, C. Masse (ed.), Paris: Presses du CNRS, pp. 111-134. Mayne Correia P. (1997), Fire Modification of Bone: a Review of the Literature [in:] “Forensic Taphonomy: The Postmortem Fate of Human Remains”, W.D. Haglund, M.H. Sorg (ed.), Boca Raton: CRC Press, pp. 275-293. Mayne Correia P., Beattie O. (2002), a Critical Look at Methods for Recovering, Evaluating, and Interpreting Cremated Human Remains [in:] “Advances in Forensic Taphonomy: Method, Theory, and Archaeological Perspectives”, W.D. Haglund, M.H. Sorg (ed.), Washington DC: CRC Press, pp. 435-450. McKinley J.I. (1989), Cremations: Expectations, Methodologies, and Reality [in:] “Burial Archaeology: Current Research, Methods, and Developments”, C.A. Roberts, F. Lee, J. Bintliff (ed.), BAR British Series 211, Oxford: Archaeopress, pp. 65-76. McKinley J.I., Roberts C. (1993), Excavation and Post-Excavation Treatment of Cremated and Inhumed Human Remains, Institute of Field Archaeologists Technical Paper Number 13. McKinley J.I. (1994), Bone Fragment Size in British Cremation Burials and its Implications for Pyre Technology and Ritual, Journal of Archaeological Science 21:339-342. McKinley J.I. (2000), The Analysis of Cremated Bone [in:] “Human Osteology in Archaeology and Forensic Science”, M. Cox, S. Mays (ed.), London: Greenwich Medical Media, pp. 403-421. McKinley, J.I. (2001), Phoenix Rising: Aspects of Cremation in Roman Britain [in:] “Burial, Society and Context in the Roman World”, J. Pearce, M. Millett, M. Struck (ed.), Oxford: Oxbow, pp. 38-44. Merbs M. (1967), Cremated Remains from Point of Pines, Arizona: a New Approach, American Antiquity 32:498-506. Metcalf P., Huntington R. (2006), Celebrations of Death: The Anthropology of Mortuary Ritual, New York: Cambridge University Press. Miller S.G. (1993), The Tomb of Lyson and Kallikles: a  Painted Macedonian Tomb, Mainz: Philip von Zabern. Morris I. (1987), Burial and Ancient Society: The Rise of the Greek City-State, Cambridge: Cambridge University Press. Murad T. (1998), The Growing Popularity of Cremation Versus Inhumation: Some Forensic Implications [in:] “Forensic Osteology: Advances in the Identification of Human Remains” 2nd Edition, K. Reichs (ed.), Springfield: Charles C. Thomas, pp. 86-105. Musgrave J.H. (1990), Dust and Damn’d Oblivion: a Study of Cremation in Ancient Greece, Annual of the British School at Athens 85:271-299. Musgrave J.H. (2005), Appendix A: An Anthropological Assessment of the Inhumations and Cremations from the Early Iron Age Cemetery at Torone [in:] J.K. Papadopoulos et al., “The Early Iron Age Cemetery at Torone”, vol. 1, Los Angeles: Cotsen Institute of Archaeology, pp. 243-315. Nelson C.L., Winston D.C. (2006), Detection of Medical Examiner Cases from Review of Cremation Requests, American Journal of Forensic Medicine and Pathology 27:103-105. Oestigaard T. (2000), The Deceased’s Life Cycle Rituals in Nepal: Present Cremation Burials for the Interpretations of the Past, BAR International Series 853, Oxford: Archaeopress. Owsley D.W., Ubelaker D.H., Houck M.M., Sandness K.L., Grant W.E., Craig E.A., Woltanski T.J., Peerwani N. (1995), The Role of Forensic Anthropology in the Recovery and Analy-

Practice of cremation in the Roman-era cemetery at Kenchreai, Greece

55

sis of Branch Davidian Compound Victims: Techniques of Analysis, International Journal of Anthropology 40:341-348. Pajot B. (1987), Le rite de l’incineration sur la nécropole du Frau de Cazals (Tarn-et-Garonne) [in:] “Anthropologie physique et archéologie: méthodes d’étude des sépultures. Actes du colloque de Toulouse, 4, 5 et 6 novembre 1982”, H. Duday, C. Masse (ed.), Paris: Presses du CNRS, pp. 379-388. Papathanasiou A. (2001), a  Bioarchaeological Analysis of Neolithic Alepotrypa Cave, Greece, BAR International Series 961, Oxford: J. and E. Hedges Ltd. Pearson M.P. (2000), The Archaeology of Death and Burial, College Station: Texas A&M University Press. Polfer, M. (1996), Das gallorömische Brandgräberfeld und der dazugehörige Verbrennungsplatz von Septfontaines-Dëckt (Luxemburg), Luxembourg: Musée National d’Histoire et d’Art. Polfer, M. (2001), Reconstructing Funerary Rituals: The Evidence of Ustrina and Related Archaeological Structures [in:] “Burial, Society and Context in the Roman World”, J. Pearce, M. Millett, M. Struck (ed.), Oxford: Oxbow, pp. 30-37. Pope E.J., Smith O.C. (2004), Identification of Traumatic Injury in Burned Cranial Bone: An Experimental Approach, Journal of Forensic Sciences 49:431-440. Popham M.R., Sackett L.H., Themelis P.G., Boardman, J. (1979), Lefkandhi I: The Iron Age, London: British School of Archaeology at Athens. Prag A.J.N.W., Musgrave J.H., Neave J.A.H. (1984), The Skull from Tomb II at Vergina: King Philip II of Macedon, Journal of Hellenic Studies 104:60-78. Prothero S. (2001). Purified by Fire, Berkeley: University of California Press. Quatrehomme G., Bolla M., Muller M., Rocca J.P., Grevin G., Bailet P., Ollier A. (1998), Experimental Single Controlled Study of Burned Bones: Contribution of Scanning Electron Microscopy, Journal of Forensic Sciences 43:417-422. Richards N.F. (1977), Fire Investigations- Destruction of Corpses, Medicine, Science, and the Law 17:79-82. Rife J.L., Morison M.M., Barbet A., Dunn R.K., Ubelaker D.H., Monier F. (2007), Life and Death at a Port in Roman Greece: The Kenchreai Cemetery Project 2002-2006, Hesperia 76:143-181. Robinson D.M. (1942), Excavations at Olynthus. Part XI, Necrolynthia: a Study in Greek Burial Customs and Anthropology, Baltimore: The Johns Hopkins Press. Sanjantila A., Strom M., Budowle B., Karhunen P., Peltonen L. (1991), The Polymerase Chain Reaction and Post-mortem Forensic Identity Testing: Application of Amplified D1S80 and HLADQ Loci to the Identification of Fire Victims, Forensic Science International 51:23-34. Saukko P., Knight B. (2004), Knight’s Forensic Pathology 3rd Edition, London: Arnold. Shipman P., Foster G., Schoeninger M. (1984), Burnt Bones and Teeth: an Experimental Study of Color, Morphology, Crystal Structure and Shrinkage, Journal of Archaeological Science 11:307-325. Shkrum M.J., Johnston K.A. (1992), Fire and Suicide: a Three-Year Study of Self-Immolation Deaths, Journal of Forensic Sciences 37:208-221. Sillen A., Hoering T. (1993), Chemical Characterization of Burnt Bones from Swartkrans [in:] “Swartkrans: a  Cave’s Chronicle of Early Man”, C.K. Brain (ed.), Transvaal Museum Monograph 8, Pretoria: Transvaal Museum, pp. 243-249. Sledzik P.S., Rodriguez III W.C. (2002), Damnum Fatale: The Taphonomic Fate of Human Remains in Mass Disasters [in:] “Advances in Forensic Taphonomy: Method, Theory, and Archaeological Perspectives”, W.D. Haglund, M.H. Sorg (ed.), Washington DC: CRC Press, pp. 321-330.

56

Douglas H. Ubelaker, Joseph L. Rife

Spennemann D.H.R., Colley S.M. (1989), Fire in a  Pit: The Effects of Burning on Faunal Remains, Archaeozoologia 3:51-64. Stiner M.C., Kuhn S.L., Weiner S., Bar-Yosef O. (1995), Differential Burning, Recrystallization, and Fragmentation of Archaeological Bone, Journal of Archaeological Science 22:223-237. Thompson T.J.U. (2004), Recent Advances in the Study of Burned Bone and their Implications for Forensic Anthropology, Forensic Science International 146S:S203-S205. Thompson T.J.U. (2005), Heat-Induced Dimensional Changes in Bone and Their Consequences for Forensic Anthropology, Journal of Forensic Sciences 50:1008-1015. Thurman M.D., Willmore L.J. (1981), a Replicative Cremation Experiment, North American Archaeologist 2:275-283. Toulouse J.H. (1944), Cremation among the Indians of New Mexico, American Antiquity 10:65-74. Toupet C. (1987), Un exemple d’ apport de l’analyse des vestiges fugaces en milieu funéraire: la cas de la nécropole à incineration de Longuesse (Val-d’Oise) [in:] “Anthropologie physique et archéologie: méthodes d’étude des sépultures. Actes du colloque de Toulouse, 4, 5 et 6 novembre 1982”, H. Duday, C. Masse (ed.), Paris: Presses du CNRS, pp. 62-65. Toynbee J.M.C. (1971), Death and Burial in the Roman World, Ithaca: Cornell University Press. Tsuchimochi T., Iwasa M., Maeno Y., Koyama H., Inoue H., Isobe I., Matoba R., Yokoi M., Nagao M. (2002), Chelating Resin-Based Extraction of DNA from Dental Pulp and Sex Determination from Incinerated Teeth with Y-Chromosomal Alphoid Repeat and Short Tandem Repeats, American Journal of Forensic Medicine and Pathology 23:268-271. Ubelaker D.H., Scammell H. (1992), Bones, a Forensic Detective’s Casebook, New York: HarperCollins. Ubelaker D.H., Owsley D.H., Houck M.M., Craig E., Grant W., Woltanski T., Fram R., Sandness K., Peerwani N. (1995), The Role of Forensic Anthropology in the Recovery and Analysis of Branch Davidian Compound Victims: Recovery Procedures and Characteristics of the Victims, Journal of Forensic Sciences 40:335-340. Ubelaker D.H. (1999). Human Skeletal Remains, Excavation, Analysis, Interpretation, 3rd Edition. Manuals on Archaeology 2, Washington: Taraxacum. Ubelaker D.H., Rife J.L. (in press), Approaches to Commingling Issues in Archaeological Samples: a Case Study from Roman Period Tombs in Greece, [in:] “Recovery, Analysis and Identification of Commingled Human Remains”, B. Adams, J. Byrd (ed.), Totowa, N.J.: Humana. Van Doorselaer A. (1967), Les Nécropoles d’époque romaine en Gaule septentrionale, Brugge: De Tempel. Van Gennep A. (1960), The Rites of Passage. Chicago: The University of Chicago Press. Van Vark G.N. (1970), Some Statistical Procedures for the Investigation of Prehistoric Human Skeletal Material, Groningen: Rijksuniversiteit te Groningen. Van Vark G.N. (1974), The Investigation of Human Cremated Skeletal Material by Multivariate Statistical Methods I. Methodology, Ossa 1:63-95. Van Vark G.N. (1975), The Investigation of Human Cremated Skeletal Material by Multivariate Statistical Methods II. Measures. Ossa 2:47-68. Walbank M.E.H. (2005), Unquiet Graves: Burial Practices of the Roman Corinthians [in:] “Urban Religion in Roman Corinth: Interdisciplinary Approaches”, D.N. Schowalter, S.J. Friesen (ed.), Harvard Theological Studies 53, Cambridge, Mass.: Harvard University Press, pp. 249-280. Walker P.L., Miller K.P. (2005), Time, Temperature, and Oxygen Availability: An Experimental Study of the Effects of Environmental Conditions on the Color and Organic Content of Cremated Bone, American Journal of Physical Anthropology 40S:222.

Practice of cremation in the Roman-era cemetery at Kenchreai, Greece

57

Warren M.W., Maples W.R. (1997), The Anthropometry of Contemporary Commercial Cremation, Journal of Forensic Sciences 42:417-423. Warren M.W., Falsetti A.B., Hamilton W.F., Levine L.J. (1999), Evidence of Arteriosclerosis in Cremated Remains, American Journal of Forensic Medicine and Pathology 20:277-280. Warren M.W., Schultz J.J. (2002), Post-Cremation Taphonomy and Artifact Preservation, Journal of Forensic Sciences 47:656-659. Wells C. (1960), a Study of Cremation, Antiquity 34:29-37. Whyte T.R. (2001), Distinguishing Remains of Human Cremations from Burned Animal Bones, Journal of Field Archaeology 28:437-448. Williams H. (2004), Potted Histories: Cremation, Ceramics and Social Memory in Early Roman Britain, Oxford Journal of Archaeology 23:417-427. Wilson D., Massey W. (1987), Scanning Electron Microscopy of Incinerated Teeth, American Journal of Forensic Medicine and Pathology 8(1):32-38. Ye J., Ji A., Parra E.J., Zheng X., Jian C., Zhao X., Hu L., Tu Z. (2004), a Simple and Efficient Method for Extracting DNA from Old and Burned Bone, Journal of Forensic Sciences 49:754-759.

Bioarchaeology of the Near East, 1:59–72 (2007)

Short Fieldwork Reports Nag el-Qarmila, Aswan (Egypt), season 2007 Mindy C. Pitre*1, Maria Carmela Gatto2, Serena Giuliani3 Department of Anthropology; 13–15 HM Tory Building University of Alberta; T6G 2H4; Edmonton, Alberta Canada email: [email protected] (corresponding author) 2 Loc. La Ficorella, 18; 00015 Monterotondo (RM); Italy 3 Via Gino Capponi, 96; 00179 Roma; Italy 1

Site surveys carried out between 2005 and 2006 in the Aswan-Kom Ombo region of Egypt by the British Museum and the University of Rome “La Sapienza” revealed the presence of two cemeteries and a settlement in the locality of Nag el-Qarmila, north of Kubbaniya (Gatto & Giuliani 2006–2007, 2007). The 2007 field season at the site of Nag el-Qarmila, directed by Dr. Maria Carmela Gatto, took place between January 24 and March 13. The purpose of this excavation was to rescue a Pan-Grave cemetery and two Predynastic/A-Group sites, all of which are in danger of being destroyed by building activities in the area. In addition to rescuing the sites, the overall research goals of the project were to define the extent of interaction between Egyptians and Nubians in the region over time and to determine the location of the political boundary that existed between them. This report includes a summary of the excavation of the cemeteries and a discussion of the findings of the preliminary in-field osteological analysis performed by Pitre. Pan-Grave Cemetery (WK11). At the Pan-Grave cemetery, designated WK11 and located at 24°14'23.16" N latitude and 32°51'41.95" E longitude, surveys and excavations led by Giuliani revealed 30 potential burials as well as several potholes and offering places in an area approximately 20×30 m. Many diagnostic Pan-Grave ceramics, together with Egyptian pottery were collected, dating the site to the late XII–early XIII Dynasty (Late Middle Kingdom–Early Second Intermediate Period). A surface collection was carried out of the entire Pan-Grave cemetery. Following this, a long central trench measuring 7×14 m was excavated; this area was selected in order to include the largest stone tumuli. Excavations revealed the presence of three main burials (9, 17 and 22) and two smaller ones (21 and 36), as well as several potholes and offering places (Figure 1). Unfortunately, the Pan-Grave cemetery had been extensively and most likely repeatedly plundered in the past. Although no complete burials were discovered, a large quantity of fragmented and poorly preserved bone and several funerary offerings such as ostrich-shell beads, animal bones, painted horn, mother-of-pearl plaque beads, and water-worn pebbles were recovered. Little organic material in the form of leather, matting, or linen was found. One of the main characteristics of the WK11 cemetery is its remarkable stone tumuli, one of which is up to 7 m in diameter. Each tumulus is composed of two concentric rows of stone slabs that surround a deep pit which is cut into the bedrock. The stone rings are fixed

60

Short Fieldwork Reports

Figure 1. Location of burials in the 7×14 m excavation area of WK11 (plan by R. Zineldeen).

in a prepared soil composed of mud and cobbles, at the top of which sheep/goat horn, some still bearing traces of red paint, is found. It was likely this visible superstructure that reminded Petrie of a frying pan, a characteristic that he later used to name the burials and the PanGrave culture itself. The Pan-Grave culture remains a mystery even though their cemeteries have been found at several sites in Egypt and their pottery has been found throughout Egypt and Sudan. When compared to the discoveries at Hierakonpolis (Friedman 2001), the Nag el-Qarmila site revealed a wider distribution of burial pits in the excavation area, larger stone tumuli, and a different pottery assemblage, demonstrating the existence of regional variation in Pan-Grave cemeteries and shedding new light on intra-cultural dynamics. Following the excavation, Pitre conducted a preliminary osteological analysis of the PanGrave material, which is currently stored in Kom Ombo. This examination involved a fragment count, a  basic description of the remains including sex and age, and a  preliminary investigation for pathological conditions. Since no intact burials were found, three minimum number of individual (MNI) estimates were calculated; an MNI of the material collected on surface, in the 7×14 m excavation area, and for a combination of the two. With regards to the preservation of the skeletal material, few skeletal elements were intact, and those that did preserve were very brittle. Fragments ranged in size from