first half of the second millennium and the late Western Zhou period. Its area of ..... towards the early Dynastic period (see Figure 4-â2). In the Longshan culture ...
CHANGES IN NUTRITIONAL AND OCCUPATIONAL STRESS OVER TIME: THE DIETARY IMPACT ON POPULATION HEALTH DURING THE TRANSITION FROM INCIPIENT TO INTENSIVE AGRICULTURE IN NORTHEAST CHINA by Mauricio Hernandez Submitted in partial fulfillment of the Requirements for the degree of Master of Arts in Anthropology, Hunter College, The City University of New York. 2009 Thesis Sponsor: ___________________________ _______________________________________ Date Signature Thomas L. McGovern, Professor Anthropology ___________________________ ______________________________________ Date Signature Ignasi Clemente, Assistant Professor Anthropology
I dedicate this thesis to my grandmother, Juana Concepcion, who witnessed every milestone in my life until this year, and whom I will never stop making proud.
ii
AKNOWLEDGEMENTS
Many people have been involved in the completion of this thesis in one way
or another. First, I would like to extend my gratitude to the Anthropology faculty of the Research Center for Chinese Frontier Archaeology in Jilin University, Changchun. Prof. Zhu Hong was kind enough to provide me access to all the skeletal material, the most important facet from which I developed my project. Prof. Wei Dong was my initial contact at Jilin University whom I was able to correspond with, arranging my trip and the outline of my research proposal. Once I arrived, he was able to answer all my questions, facilitated archaeological site reports and pinpointed where the collections were located. Zeng Wen also helped me to communicate my ideas to Prof. Wei and was critical during my stay at the institute.
To the Department of Anthropology at CUNY Hunter College, I take a bow. BY
placing their trust in admitting me to their program and the quality education I have received while being a student there, I would not have had the chance to develop my research interests, and this thesis would not exist today. Within the faculty, I would like to thank Prof. Thomas McGovern who provided all documentation and support for me to carry out my research in the other side of the world and who gladly accepted to become my thesis supervisor. In a big way, he is responsible for everything contained within this manuscript. Prof. Ignasi Clemente is my writing guru. His advise has been very useful while drafting my research proposal and, later, my thesis. Profs. Michael Steiper and Jessica Manser helped me figure out several statistical analyses, which allowed me to put my ideas together in a more
iii
meaningful way. The rest of the faculty and students from the program, I thank them from the bottom of my heart. By sharing three semesters of laughter and sorrows, they made me realize that we really are all in this together.
I also must express my humble gratitude to Gisselle Garcia-‐Pack, curator at
the Department of Anthropology at the American Museum of Natural History in New York City. She allowed me to access several East Asian individuals and medical school collections within the stacks that I was able to use as the control sample within my study. Also, Dr. David Hunt, curator at the Department of Anthropology at the Smithsonian National Museum of Natural History in Washington, D.C.. He was extremely helpful in allowing me access to a large and well-‐preserved Asian collection, which made up the large portion of my control sample.
Within Changchun, China, several people kept me from becoming too
overwhelmed with my data collections and welcomed me as if I was one of their own, from the very first hours after arriving. Rocio Dones, Spanish professor at Jilin University and her husband Ivan Wang were there for me from day one, as promised, and have shown me how friendship can make all the difference in an unfamiliar land. Anabel Garcia Ugrotte, also Spanish professor at Jilin University has provided countless hours of academic stimulation through deadly honest opinions and discussions in our many, sometimes hurried chats over coffee or mate at her flat, and her determination continues to leave me in awe. I truly believe she shares a nonverbal connection with some of the greatest minds, whilst her humility does not allow her to confirm this to me. To Changchun Friends, the great club of ex-‐pats and Chinese nationals in Changchun, I owe them many moments of fun, debauchery and
iv
sincerity. They really keep it all together. Also in Changchun, my friend Dr. Kenji Okazaki, whom I have been honored in interacting with, both professionally and earnestly. I hope to continue collaborating with in the future, thanks to our shared interests and values. His wife Chen Xi is an amazing cook, a very sweet individual, and has the distinction of being my very first, albeit unofficial, Mandarin professor. I hope to one day speak to her in this wonderful language.
Last but definitely not least, my best friends in the world, who through the
years have put up with my complaining and have provided loads of emotional support that I will not ever begin to pay them back for. Christian Jarquin, my longest friend knows me inside and out, he is my brother. Dalia Potosme, who I have had the pleasure of keeping in touch with, and getting to know better through the years, and Vincent Lau Chan, I could not ask for more down to earth individuals. Life would be a darker place without them. Chin-‐hsin Liu, my mentor and confidant, who is responsible for cultivating my interests in bioarchaeology and who has guided me through very tough professional moments as I have done with her; no one can break this bond. Shuyu Lin, another longtime friend who helped me a lot with the translation of several archaeological reports in order to complete my thesis. Selma Hodge, who has been a mother to me for many years, her hugs and smile never fail to show me the positive side of humanity. To my dear roommate, Lisa Konkel, who has allotted me peace in our house and whom I have shared the rowdiest and calmest of times in this crazy city of New York.
One special mention should be reserved for Prof. Ekaterina Pechenkina of
CUNY Queens College. I thank her for helping me to discover bioarchaeological
v
research in China firsthand and providing me data from which to get started in my research career. It is also because of her that I have realized the necessity and importance of working on my own.
vi
TABLE OF CONTENTS PAGE AKNOWLEDGEMENTS..........................................................................................................................iii LIST OF TABLES.......................................................................................................................................ix LIST OF FIGURES......................................................................................................................................x CHAPTER 1 – INTRODUCTION...........................................................................................................1 Research Goals..........................................................................................................................................2 Stress Indicators.......................................................................................................................................2 Paleopathology.........................................................................................................................................4 CHAPTER 2 – ARCHAEOLOGY OF THE SITES..............................................................................9 JIANGJIALIANG Geography...................................................................................................................................................9 History and Cultural Background....................................................................................................10 Contemporary Period...........................................................................................................................10 MINHE X, MINHE M AND LGS SITES Geography..................................................................................................................................................11 History and Cultural Background....................................................................................................12 Contemporary Period...........................................................................................................................13 CHAPTER 3 – MATERIALS AND METHODS................................................................................14 Background of samples........................................................................................................................14 Measurements and statistical calculations.................................................................................16 CHAPTER 4 – RESULTS........................................................................................................................19 Stature.........................................................................................................................................................20 Body Mass..................................................................................................................................................23 Correlation between stature and body mass.............................................................................24 Bilateral asymmetry..............................................................................................................................24 Paleopathology........................................................................................................................................28 Non-‐specific and Specific Response to Infection......................................................................28 Degenerative Joint Disease.................................................................................................................34 Congenital Conditions..........................................................................................................................37 Trauma........................................................................................................................................................39 Paleodemography...................................................................................................................................41
vii
CHAPTER 5 – DISCUSSION.................................................................................................................49 Diet and Disease.......................................................................................................................................50 Stature, Sexual Dimorphism and Body Mass ..............................................................................51 Indicators of Activity..............................................................................................................................54 Health Model for Chinese Populations from the Neolithic period to the Bronze Age.................................................................................................................................................58 CHAPTER 6 – CONCLUSION................................................................................................................62 REFERENCES CITED .............................................................................................................................65
viii
LIST OF TABLES PAGE
Table (4-‐1) – Mean femoral length and stature estimates for each site surveyed. ...................................................................................................................................................20 Table (4-‐2) – Femur length and estimated stature from Chinese Neolithic samples (adapted from Pechenkina et al., 2007)...................................................21 Table (4-‐3) – Mean femoral head diameter and estimated adult body mass (in kg) for populations surveyed at each site. ................................................................23 Table (4-‐4) – Bilateral asymmetry scores in the population of Jiangjialiang. .............................................................................................................................................26 Table (4-‐5) -‐ Bilateral asymmetry scores in the populations of Minhe X, Minhe M and LGS. ..................................................................................................................................26 Table (4-‐6) -‐ Bilateral asymmetry scores in the population of MXY. ..............................27 Table (4-‐7) -‐ Bilateral asymmetry scores in the Control population. .............................27 Table (4-‐8) – Incidence of pathological conditions per individual in all samples. ....................................................................................................................................................32 Table (4-‐9) – Demographic of all surveyed populations.......................................................42
ix
LIST OF FIGURES PAGE
Figure (1-‐1) – Lesions in specific and nonspecific inflammation of bone. From Suzuki (1991).................................................................................................................................6 Figure (2-‐1) – Map of Jiangjialiang site............................................................................................9
Figure (2-‐2) – Map of Minhe X, Minhe M and LGS sites...........................................................11 Figure (4-‐1) – Comparison of male and female mean statures between the Peiligang site of Jiahu and all Yangshao sites.......................................................................43 Figure (4-‐2) – Comparison of male and female mean statures between Longshan, early Dynastic sites and the Control population. ................................................44 Figure (4-‐3) – Mean body mass (in kg) for males and females at each surveyed site. ............................................................................................................................................44 Figure (4-‐4) – Linear regression for stature as a function of body mass for males and females of each surveyed population. ...................................................45 Figure (4-‐5) – Incidence of infectious disease in males and females across all surveyed sites. .....................................................................................................................45 Figure (4-‐6) – Distribution of skeletal lesions. On the skeleton displaying periostitis, the darker shade indicates more common involvement; the light-‐shaded areas are where the condition is less commonly found (after Kelley, 1989). ..........................................................................................46 Figure (4-‐7) – Incidence of degenerative joint disease on males and females at all surveyed sites. .............................................................................................................47 Figure (4-‐8) – Incidence of congenital conditions in males and females at all surveyed sites. ..............................................................................................................................47 Figure (4-‐9) – Trauma patterns between males and females across all surveyed populations. ..........................................................................................................................48 Figure (4-‐10) -‐ Paleodemography comparison between all surveyed populations. ..............................................................................................................................................48 Figure (4-‐11) – Age ranges for males and females at each surveyed site. ....................49
x
CHAPTER 1 INTRODUCTION At every segment of life, from before birth, through childhood and all the way into late adulthood, an organism may experience stress. Stress can come in the form of inadequate nutrition, infectious disease, environmental factors, reproductive needs or interpersonal conflict. Prolonged stress can leave behind evidence in the form of insults on bone, due to the inherent plasticity of this tissue. Levels of stress and disease can be discerned from skeletal remains through the analysis of bone tissue, either through simple visual examination, microscopy of cross sections, x-‐ray or chemical analysis. Such information can reveal age, sex, stature, diet, intensity of activity and even possible cause of death. However, more useful data can be obtained when assessing patterns of stress at the population level, or at least a certain segment of it, such as the very young or the very old (Goodman and Armelagos, 1989). Further comparisons of skeletal markers across geographical regions and through time can reveal how well populations adapt to varying climates, differential availability of resources and cope with disease. In the following study I present data collected from several populations in China and attempt to reveal how subsistence changes undertaken by these groups affect their health in the long term. I became interested in this region of the world because of the dearth of published archaeological and bioarchaeological studies in western journals, as well as the potential of this region of the world to yield a wealth of data on disease and the development of social systems: China is the birthplace of the oldest continuous civilization.
1
Research Goals The aim of this study is to quantify different stress markers related to development, nutrition, susceptibility to disease, and subsistence activities from the Neolithic period to the Bronze Age in China. The collections under study came from two regions of the country: Qinghai province, located in the Northwest, and Hebei province, located to the East. Together with previous research on pathological conditions, archaeological finds from the region and models for the beginnings and intensification of agriculture, this study seeks to elucidate patterns of stress caused by subsistence changes in these populations.
Stress Indicators Paleopathologists investigating disease processes in bone do not have the benefit of soft tissue analysis and laboratory tests that clinical doctors do. In fact, Wood et al. (1992) argue that skeletal data can have more than one interpretation, due to the paradoxical nature of diagnosing living conditions from dry bones, many times without cultural, dietary or archaeological context. However, Buikstra and Cook (1980) note that useful models from modern correlates can be developed to recreate conditions from the past, such as utilizing clinical cases to discern lesion patterning. These authors promote employing modern techniques such as x-‐rays, microscopy and isotope analysis to arrive at more conclusive diagnoses of disease. Also, different pathologies can elucidate different aspects of a person’s life and their patterning on the body of an individual or on a specific population demographic can itself help with identification. As Angel (1984) points out, studying pathologies
2
together, rather than in isolation, helps to better understand how disease processes work. In the present study, the presence, incidence and severity of several stress indicators were analyzed. In the following introductory sections, I give some background of each stress indicator and why it was chosen. One of the most obvious indicators of stress in adults is stature reduction, especially in a region where there is little genetic variation (Cook, 1984). Bone growth and development depends strongly on the uptake of adequate amounts of vitamins and minerals. Such nutrients are largely absent on the carbohydrate-‐rich foods that became the first cultigens, as a supplement to foraging and hunting, and in more recent periods, took over as the main food source in agricultural populations. Increase in body mass has been cited by Styne and McHenry (1993) as an indicator of improving health due to better socioeconomic conditions in modern populations. However, it can also be used to discern increased reliance on grain with high caloric content. Cohen (1977) notes that this pattern stems from humans looking for the simplest and quickest ways to cope with population growth and pressure during the Neolithic period. Several authors have analyzed bilateral asymmetry in populations, in order to discern changes in rates and levels of activity within a population and across populations over time (Auerbach and Ruff, 2005; Mintz and Fraga, 1973; Stirland, 1993). An extensive review of past studies by Kennedy (1989) points to the many patterns of occupationally related changes that can occur in bone (cortical-‐ trabecular ratio, increased surface area of muscle attachments, overall size and
3
shape changes). Such osteological changes, Stirland notes, are by definition, nonrandom and habitual, and thus can be used to reconstruct past life ways, such as those related to subsistence strategies. Wolff’s Law, proposed by anatomist Julius Wolff in 1892, states that bone will change shape due to frequency and level of pressure or force that it is subjected to, therefore, by judging levels of asymmetry, one can calculate rates and levels of activity. Paleopathology Rates of infections increased in archaeological populations after agriculture was adopted by hunter-‐gatherers in different regions of the world (Cohen, 1977). Similarly, chronic conditions exhibit gross pathological changes to bone, largely absent in hunting and gathering groups, begin to appear around this time. Most commonly appearing at this time is periostitis, an inflammatory bone lesion displaying reactive changes with irregular, fine-‐porous and spongy deposition located on the exterior of the cortex with the exclusion of underlying tissues (Suzuki, 1991). Periostitis may be caused by an adjacent soft tissue infection, bacterial transport through the bloodstream or the spread of organisms from a compound fracture, but may have other causes such as trauma, hemorrhage or skin ulcers (Aufderheide and Rodriguez-‐Martin, 1998). Due to such varied etiology, and because the morphological changes associated may appear in either an active or healed stage, it is often considered a nonspecific indicator of infection (Larsen, 1997).
4
A second nonspecific infectious condition more severe in nature is osteomyelitis. Osteomyelitis, unlike periostitis, is defined as inflammatory changes, which spread through the medullary cavity and the cortical bone (see figure 1-‐1). Its most common features are the appearance of sequestrum, involucrum, and cloacal formation due to the simultaneous process of bone destruction and bone formation (Suzuki, 1991). Sandison (1968, cited in Roberts and Manchester, 1995) notes that since bone is a single biological unit, the different terminology for these nonspecific morphological changes is arbitrary. In this study, periostitis and osteomyelitis, although caused by similar nonspecific factors are kept separate for two reasons. The first is to provide continuity of understanding and interpretation from all previous paleopathological studies of infection. The second is because this separation of severity is meaningful when analyzing prehistoric agricultural groups in comparison to recent industrialized societies in relation to diseases, which are population density-‐dependent.
5
Figure (1-‐1) – Lesions in specific and nonspecific inflammation of bone. From Suzuki (1991)
Infectious diseases that cause specific changes to bone morphology such as leprosy or treponemal diseases begin to appear along with interregional trade routes and overcrowded population centers, both of which promote close contact between many people. Robbins et al. (2009) report a case of leprosy in India at around 4,000 BP. The adult individual displays bilateral erosive lesions of the supraorbital region and glabella, including erosion/remodeling of the margin of the nasal aperture. The northern inland route of the Silk Road passed through Qinghai and Gansu provinces, which would have allowed for long distance trade networks increasing the spread of infectious diseases. Degenerative joint disease (DJD) is a multifactorial non-‐inflammatory disorder affecting articular joints. Associated skeletal changes include proliferative exophytic growths on vertebrae, a porous joint surface, or in severe cases where
6
cartilage has completely worn away, eburnation (Waldron and Rogers, 1991). The causative factors for DJD include hormones, nutrition, metabolism, infection, trauma and heredity (Larsen, 1997, 1998). Cultural differences in activity, stemming from subsistence patterns or from sexual division of labor can create differing osteoarthritis incidence between males and females, as seen in studies of living populations (Kennedy, 1989; McCarty and Koopman, 1993). Lastly, DJD incidence greatly increases with age, so populations composed of older individuals have a greater incidence of the condition. This study also notes two fairly common sacral malformations present in the collections examined: lumbosacral transitional vertebra and spina bifida oculta. Lumbosacral transitional vertebra, or LSTV is a congenital condition that carries an increased risk for disc degeneration (Aihara et al., 2005), herniated discs (Wigh et al., 1981) and may be linked to genetic factors (Tini et al., 1977). Some studies found no link between LSTV to lower back pain (Magora and Schwartz, 1978; Vergauwen et al., 1997). However, the clinical literature indicates that when combined with LSVT, lower back pain may be more severe (Tini et al., 1977). The condition may be partial or complete, evidenced by anomalies on the transverse processes. It may present itself in the more common sacralization of lumbar segments, or the less common lumbarization of sacral segments. Spina bifida oculta is the least severe form of a condition (the other, spina bifida cystica, which is incompatible with life) referring to incomplete fusion of the posterior neural arch on the sacrum (Roberts and Manchester, 1995). The condition is fairly common on modern populations and usually decreases with age, especially
7
among females (Aufderheide and Rodriguez-‐Martin, 1998). Patterns of both of these congenital conditions will be examined to establish any possible genetic links between individuals at the site. This may help determine how frequently these conditions were inherited. Trauma can be caused by cultural activities, warfare or isolated accidents. Whereas isolated trauma evident on skeletal material may not imply much, consistent patterns of trauma in a population can reveal intergroup conflict (amputations, scalping), risky food procurement (animal encounters while hunting and foraging from tall trees which can cause fractures), or other cultural practices (trepanation, foot binding). Within this study, I classify several subcategories of trauma in order to discern any patterns of activity in the surveyed populations: forelimb and hindlimb fractures, rib fractures, herniated discs, and ossified ligament/tendon or joints. Fractures occur due to impacts and low bone mass may increase their propensity. Herniated discs may occur from a weakening of the vertebral centrum via high frequency of lifting heavy loads. Although herniated discs occur in individuals suffering from tuberculosis (Pott’s disease) or Paget’s disease, other osteological patterning usually helps in their diagnosis. Lastly, traumatic injuries may cause for tendons or ligaments to spontaneously ossify. Joint fractures also lead to the fusing of bones forming part of the fracture, which reduces movement and may hinder function. The conditions to be analyzed will now be put in better perspective after an overall look of the sites surveyed for the study and the diet of the inhabitants.
8
CHAPTER 2 ARCHAEOLOGY OF THE SITES JIANGJIALIANG Figure (2-‐1) – Map of Jiangjialiang site
Geography Jiangjialiang (姜家梁) sits atop a hill east of the Xishuidi (西水地) village north of the Sanggan (桑干) river, located just east of the Nihewan Basin in Jiangjiakou (張家口) county, Hebei province. The site is positioned at an altitude of approximately 30 meters above the riverbed. It is divided into three sections to account for the altitude differences and the effects of riverine erosion processes.
9
History and Cultural Background The Nihewan basin has been studied since 1921 by French scholar E. Lecent who began collecting prehistoric zoo-‐archaeological material. In 1924, American geographer G.B. Barbour gathered geographical information in region. In 1930, French scholar P. Teilhard de Chardin hypothesized human presence in the region after his analyses of mammalian fossils. In 1957, Chinese archaeologists Wang Jiang and Jia Lanpo hypothesized that the area would yield the earliest human occupation. In 1974 a team led by Jia Lanpo and Wei Qi discovered the Paleolithic site of Syujiayao (許家窯), and in 1976 found the first human skeletal remains (nine fragments) during the excavation of the site. By 1977, another Chinese scholar, Wu Maolin found another eight human skeletal fragments. In 1978, three scholars You Yuzhu, Tang Yingjun and Li Yi discovered the site of Xiaochangliang (小長梁), some of the earliest Paleolithic remains in East Asia. Since the 1990s, Xie Fei from the Hebei Institute of Cultural Relics has discovered another seven archaeological sites in the region. Contemporary Period Two excavations were undertaken at Jiangjialiang. The 1995 excavation discovered section I of the site. Sections II and III were discovered during the 1998 excavation. By area, Jiangjialiang is the largest Neolithic site in Hebei province, with all three sections measuring 1,600 m2. The age of the site is around 6,850 +/-‐ 80 BP via C14 dating and represents the transition between two cultural periods: terminal
10
Yangshao and very early Longshan. Between both excavations, a total of nine houses and 78 tombs were discovered, all within close proximity, although it is noted that they do not seem to be contemporaneous. The Jiangjialiang cemetery is very organized, consisting of five rows of rectangular tombs facing east to west. The number of these tombs located within each row varies. MINHE X, MINHE M AND LGS SITES Figure (2-‐2) – Map of Minhe X, Minhe M and LGS sites.
Geography Minhe autonomous county (民和回族土族自治县) is located in a mountainous region on the eastern portion of Qinghai province, bordering Gansu province to the east. Most of these mountains reach 3,500 meters above sea level. The region features thick and widespread loess soil deposits. Due to the loose
11
texture of this sediment, water and soil erosion are a big problem and many gullies can be seen throughout the land. Elevation differences are significant in the region. The altitude ranges from 1,650 meters up to 4,220 meters with three main altitude levels currently utilized for agriculture. The Huang Shui (黃水) valley at 1,650-‐2,000 meters of elevation has a long history of cultivation, and is currently used for planting vegetable and fruit crops such as melon. At around 2,000-‐2,700 meters of elevation, the region frequently experiences drought conditions, so it is not an ideal environment for cultivation except for wheat, a cultivated cereal that does not require much irrigation. Areas with an altitude of 2,700-‐3,200 meters are mainly employed for growing oil-‐bearing crops.
Minhe is located between two major highlands, so the weather has several
apparent transitional features. There is sufficient light and sun throughout the year and little unevenly distributed rainfall. The temperature of the region is relatively moderate, between 7-‐8° C in the Huang Shui valley down to -‐5° C at the highest altitudes. History and Cultural Background In China, the first inhabitants of the Qinghai region, around 30,000 BP consisted of hunter-‐gatherers who utilized advanced stone and bone tools (Suzuki et al., 2005). Yangshao ruins have been discovered in the Minhe region, confirming that inhabitants had close connections with Central China over 6,000 years ago. A large number of sites (113 in total) belonging to the Majiayao culture (5,300-‐3,050 BP), which has strong local characteristics, 176 sites belonging to the Qijia culture
12
(4,000-‐3,600 BP) and other Bronze Age culture sites have also been discovered in the region. The populations studied are part of the Xindian culture, which dates to the first half of the second millennium and the late Western Zhou period. Its area of distribution is the Hehuang River Valley and the eastern part of Gansu. It originated in the late Qijia culture (due to the continuity in pottery style). The Xindian culture later expanded toward the west and became closer to the Kayue Culture, by which it was possibly absorbed (Loewe & Shaughnessy, 1999). Contemporary Period Swedish scholar Johan Gunnar Andersson carried out the first archaeological studies of the region in 1923 and 1924 and discovered several Neolithic and Bronze Age sites. The site was accidentally discovered by villagers while building a house. In the years 1978 through 1980, full-‐scale excavations were undertaken by the Qinghai Provincial Administration of Cultural Relics, uncovering 367 graves, 500 ceramic vessels and 2,690 stone, bone or bronze tools and vessels (Qinghai Provincial Administration, 2004). The collections included in this study come from three physically and culturally contemporaneous sites and are thus lumped together during analysis: the Xiaohandi (小旱地) cemetery – represented as Minhe X; the Mapai (馬排) cemetery – represented as Minhe M; and LGS. Within the Xiaohandi cemetery, only about 100 simple internments (27.5% of the site) are present the rest being disturbed, incomplete or missing.
13
The cemeteries under study are located east of the Hetao (核桃) village. The village residents grow several cultigens such as potatoes, white carrots and white radish in one annual season. The average age-‐at-‐death for males is 60 years, and for females, 62-‐64 years. I have given a brief introduction on the study and the health parameters I sought to examine and described some of the background information of the surveyed sites. Now I will describe the methodology and materials that I used in this study, noting some of the limitations encountered.
CHAPTER 3 MATERIALS AND METHODS Background of samples For the purposes of this study, only post-‐crania were examined. Further
information regarding the paleopathology of Jiangjialiang crania can be found in Okazaki (in preparation) and Li (2004). The skeletal collections analyzed for the experimental portion of this study are curated at the Institute of Field Archaeology, Jilin University, China. The Jiangjialiang collection, from the Neolithic period of China, consists of 66 individuals from the transitional period between the Yangshao (7,000 to 5,000 BP) and Longshan cultures (5,000 to 3,000 BP). For this study, the site of Jiangjialiang is still treated as a Yangshao site, as the dietary changes that occurred during the formal Longshan period would not have been readily apparent on the skeletons. The Minhe X, Minhe M, and LGS collections occur later in time, and are associated with the Xindian culture (ca. 3,000 BP) from the Bronze Age of China.
14
Since all three occur in close temporal and spatial proximity, and due to their small sample quantity (96 individuals altogether), all three are placed together in the analyses. For information regarding craniometric studies, the study by Wang (1995) should be consulted. Lastly, the MXY collection, consisting of 40 individuals also comes from the Xindian culture, although it is not in physical proximity to the other three sites. The comparative data for this study come from two sources. First, a total of eight individuals of Chinese and Japanese descent housed at the Department of Anthropology in the American Museum of Natural History, New York City are part of an anatomical collection donated by the Cornell University Medical College in 1945. The second collection consists of over 70 individuals (all male, except for one female) of Chinese descent who worked at the Karluk Cannery in Kodiak Island, Alaska. They were collected by Aleš Hrdlička in the early 1940s and are housed at the Department of Anthropology within the National Museum of Natural History, Washington, D.C.
Analyses for all samples included gross examination of the entire skeleton to
discern and exclude development abnormalities and pathologies that may hinder measurement. Any post-‐mortem breaks were glued back together with epoxy glue after cleaning in order to be properly measured. To limit intraobserver error, all measurements were taken twice with a Carolina osteometric board (catalog #249800), or a Mitutoyo Absolute Digimatic digital caliper. Photographic record of the collections was taken with a Nikon D80 camera and a Sigma 17-‐70mm f/2.8 lens.
15
Measurements and statistical calculations Age assessments were performed according to Todd (1921) pubic symphysis protocol or the Lovejoy et al. (1985) method for the auricular surface. Infants, when possible, were aged according to humeral length or femoral length after Johnston (1962), Table 2, and/or via dental development after Ubelaker (1989), Figure 71. Limb bone measurements from sub-‐adults (individuals whose age-‐at-‐death was under 17-‐18) were omitted in this study, as their epiphyses had not yet fully fused. Pathological parameters were also not measured, as they would have skewed results. Sex assessments depended heavily on the presence and condition of the os coxae, and were determined, according to Phenice (1969) as cited in Buikstra and Ubelaker (1994). Due to curation bias, most individuals within the Minhe X, Minhe M, LGS and MXY collections were missing all carpals, metacarpals, tarsals, metatarsals, phalanges, ribs, sternums, and in some cases os coxae and sacra, so discerning sex and the incidence of many pathological conditions was only possible for a limited number of individuals. Within the Minhe X collection, several individuals’ limb bones, along with os coxae, incomplete scapulae and sacra were found comingled in a box. To utilize these comingled femora for population comparisons, I used a sex determination method devised by Iscan and Shihai (1995). The authors determined that within Mongoloid populations, the distal epiphyseal breadth of the femur is the most dimorphic part, and thus best for discriminating between the sexes (with an average success rate of 94.9% between both sexes).
16
Limb bone measuring protocols for this study are also based on Buikstra and Ubelaker (1994). They include maximum and physiological lengths, as well as maximum and minimum widths at the midshaft, or anterior-‐posterior and medio-‐ lateral parameters for all forelimb and hindlimb bones including the clavicle. Epiphyseal breaths and diameter of the femoral and humeral heads were also taken. Lastly, maximum length and breadth of all os coxae and sacra was measured. Scapulae were not included in the study due to their very fragmentary condition within the archaeological populations analyzed. All bones were oriented in standard position according to Ruff & Hayes (1983, Figure 2a). Stature, in centimeters, was calculated from the average of left and right femur formulas devised for Mongoloid femora in Trotter and Gleser (1958). When both femora of an individual were present, the left was chosen for the study, as they were generally more numerous within the sample. Average femur length for males and females, their estimated stature and sexual dimorphism were then compared to data from North China, found on Pechenkina et al., (2007), as this study also employed the stature estimate formula from Trotter and Gleser (1958). Body mass, in kilograms, was calculated by taking the average of the Ruff et al. (1991) equation for the male and female femoral heads (BM = (2.741 X FH – 54.9) X .90). Left and right sides were measured in order to test bilateral asymmetry between males and females and across all sites. All analyses could not be performed in every collection due to small samples sizes and preservation/collection biases. Following the work of Auerbach and Ruff (2005), asymmetry data were converted
17
into percentage directional asymmetries (%DA) to allow for the direct comparison of asymmetries in different skeletal elements within right and left dominant groups: %DA = ((right – left) / (average of all left and right)) * 100 For comparison percent absolute asymmetries (%AA) between sides were also calculated to find out the magnitude of random asymmetry for every dimension measured: %AA = ((maximum – minimum) / (average of all maximum and minimum)) * 100 As these analyses were performed using non-‐parametric tests on non-‐ transformed data, median asymmetry values are emphasized and mean asymmetry values are included as well, for comparison with earlier studies that employ this statistic (Auerbach and Ruff, 2005). Due to time constraints during data collection and analysis restrictions, more invasive methods such as cross-‐sectional analyses data and x-‐rays could not be taken. Also, about one third of the Jiangjialiang femora were previously sectioned for a study approximately along the 20% and 80% portions of the diaphysis, which did not allow for bilateral asymmetry comparison. Pathologies were classified into five categories based on their perceived etiologies. The first category is periostitis, which results from non-‐specific disease processes that only act upon the outer bone table. This category is divided into Type I (inactive and mild periostitis), Type II (active and moderate or severe periostitis), unspecific inflammation, osteomyelitis, which involves the inner table of bone and has characteristic sequestrum, involucrum and cloaca formation and osteitis. The second category includes bone inflammation more specific in origin, often exhibiting
18
a granulomatous lesion, such as in the case of tuberculosis, leprosy and treponemal diseases. The third category is degenerative joint disease (DJD), which derives from activity-‐related wear and tear of joints (unlike rheumatoid arthritis, an autoimmune disorder) and encompasses osteoarthritis and osteophytosis. The fourth category includes congenital malformations of the sacrum such as lumbosacral transitional vertebra and spina bifida. The fifth category includes all traumas, evident in healed or unhealed fractures, amputated limb segments and post-‐traumatic joint, tendon and/or ligament ossification.
Now after describing my methodology, I will present the results of my analysis
on the skeletal populations I studied, including the comparative data obtained from museum collections.
CHAPTER 4 RESULTS Several health markers were examined in these populations. Each pathology
elucidates a particular aspect in the livelihood of these individuals. Taken together, they allow for a more complete picture of their health profile to emerge. Below, I present the results for the data on stature, sexual dimorphism bilateral asymmetry and body mass. After, I present paleopathological data, which encompasses nonspecific and specific infections, degenerative joint disease, congenital conditions and trauma. Lastly I examine the paleodemography of the sites before making inferences on the health of populations in Northern China from the Neolithic period up to the Bronze Age.
19
Stature There is a visible difference in the stature of different populations, compared to the modern control group, and also between males and females. The control group mainly composed of factory workers of Chinese descent, and an anatomical collection, both from the 20th century displays a slightly reduced stature from the early Dynastic populations, at 165.04 cm. These modern samples, although generally shorter, are still of similar stature to those of the early Dynastic periods. Unfortunately, a sufficiently large female sample was missing for stature and sexual dimorphism comparisons. Table 4-‐1 shows that the earlier Yangshao culture site at Jiangjialiang displays the tallest statues in both males and females, with males averaging more than eleven centimeters in height over females, and more than five centimeters in height over the anatomical collections analyzed. The lowest average male statures are represented at the Xindian culture site, MXY, whereas the lowest female statures are comparable at all the Xindian culture sites, Minhe X, Minhe M, LGS and MXY. Table (4-‐1) – Mean femoral length and stature estimates for each site surveyed. a
Femur Length (mm) Site Yangshao Jiangjialiang Xindian MHX, MHM, LGS MXY Modern Control
Males [N]
Females [N]
Estimated Stature (cm) Sexual Males Females Dimorphism
457.33 +/- 16.2 [3]
405 +/- 14.12 [4]
170.81
159.40
3.45
438.57 +/- 17.38 [7] 417.38 +/- 26.1 [8]
401.25 +/- 17.69 [10] 401.43 +/- 16.08 [7]
166.72 162.10
158.58 158.62
2.50 1.08
430.85 +/- 20.37 [65]
165.04
a Stature estimates are based on the average of both, left and right femur formulas
from Mongoloid males from Trotter and Gleser (1958).
20
Table (4-‐2) – Femur length and estimated stature from Chinese Neolithic samples (adapted from Pechenkina et al., 2007). a
Femur Length (mm) Site Peiligang Jiahu Yangshao Beiliu Jiangzhai I Jiangzhai II Shijia Guanjia Xipo Longshan Meishan b Mengzhuang Dynastic Kangjia Western Zhou
Males [N]
Females [N]
Estimated Stature (cm) Sexual Males Females Dimorphism
470.31 +/- 17.4 [32]
428.09 +/- 33.8 [21]
173.69
164.49
2.72
448.00 +/- 27.3 [4] 455.87 +/- 19.4 [4] 450.73 +/- 17.9 [7] 447.71 +/- 26 [14] 440.60 +/- 10.9 [5] 448.98 +/- 24.5 [11]
408.84 +/- 20.1 [4] 427.85 +/- 19.9 [5] 416.9 +/- 11.3 [6] 410.08 +/-16.5 [10] 413.44 +/- 20.7 [9] 398.65 +/- 23.3 [5]
168.89 170.58 169.46 169.26 167.3 169.1
160.5 164.53 162.26 161.11 161.46 158.28
2.55 1.81 2.17 2.47 1.78 3.31
441.8 +/- 15.1 [5] 438.9 +/- 29.2 [9]
389.6 +/- 31 [3] 418.5 +/- 15 [4]
167.56 166.93
156.33 162.55
3.47 1.33
433.4 +/- 18.8 [8] 437.0 +/- 18.3 [8]
366.8 +/- 14.8 [4] 397.1 +/- 24.6 [6]
165.76 166.52
155.74 157.94
3.12 2.64
a Stature Estimates are based on the average of both, left and right femur formulas
from Mongoloid males from Trotter and Gleser (1958). b Based on Henan Institute (2003): 546 By comparing data from this study to data in Table 4-‐2 from eleven populations, expanding a wider time period in North China, I was able to obtain a better consensus of inter-‐site stature changes over time. The Early Neolithic site of Jiahu dated to the Peiligang culture (7,500-‐6,900 BP) displays a wide disparity between male and female stature, which could not be attributed to insufficient sampling, as it contained measurements for over 53 individuals (see Figure 4-‐1). Yangshao culture sites, occurring later in the sequence (7,000-‐5,000 BP), are much less sexually dimorphic than Jiahu. The population at Jiangjialiang exhibits one
21
of the most sexual dimorphism of all sites (similar to Xipo); it also displays the tallest males in the Yangshao culture with an average stature of 170.81 cm. The smallest sexual dimorphism in the Yangshao period occurs at the site of Guanjia with 1.78 cm. At the Yangshao culture sites, average male stature ranges from 170.81 cm to 167.3 cm and females statures range from 164.53 cm to 158.28 cm. At the Longshan and Xindian cultures there is an overall reduction of stature in both males and females across all sites, more apparent as time progressed towards the early Dynastic period (see Figure 4-‐2). In the Longshan culture (5,000-‐ 4,000 BP), male stature range from 167.56 cm to 166.93 cm and female stature, from 162.55 cm to 156.33 cm. In the slightly later Xindian culture (ca. 3,000 BP) sites analyzed for this study, average male statures ranged from 166.72 cm to 162.10 cm and females statures ranged from 158.62 cm to 158.58 cm, with the shortest males at the site of MXY. By the early Dynastic periods, stature seems to make a small rebound in males, although for females, the reduction trend persists until the Western Zhou period.
Sexual dimorphism is reduced to some degree in the Longshan culture
(except at the site of Meishan), with the Xindian culture experiencing the least amount of dimorphism between all periods. Stature differences between males and females at the sites of Minhe X, Minhe M, LGS and MXY only vary between 3.48 cm to 8.15 cm, unlike in earlier Yangshao culture periods, where differences per site were as high as 11.41 cm at Jiangjialiang. For females in the early Dynastic period, there seems to be an increase in sexual dimorphism compared to their male counterparts; the implications for which will be evaluated in the discussion.
22
Body Mass Body mass was determined from the femoral head diameter using the average from the male and female formulas from Ruff et al. (1991). Data on Table (4-‐3) indicate that at the Jiangjialiang site, male body mass was the highest of all six populations surveyed, including the modern anatomical samples, at 70.55 kg, and females were also the heaviest, at 59.02 kg. All four Xindian populations also experienced a reduction in body mass similar to that in stature. The site of MXY had the least average body mass for both males and females, with 61.2 kg and 49.06 kg respectively (see Figure 4-‐3). Table (4-‐3) – Mean femoral head diameter and estimated adult body mass (in kg) for populations surveyed at each site. Jiangjialiang MHX, MHM, LGS MXY Control
Femoral Head Diameter (mm) Males [N] Females [N] 47.76 +/- 2.88 [18] 42.8 +/- 2.39 [19] 44.98 +/- 2.13 [11] 40.74 +/- 1.12 [14] 43.74 +/- 3.07 [12] 38.52 +/- 1.81 [12] 44.99 +/- 2.16 [64]
Estimated Body Mass (kg) Males Females 70.55 59.02 64.09 54.23 61.20 49.06 64.11
The control population had a slightly higher average body mass than individuals from the Xindian culture, at 64.11 kg. Such results indicate a slight rebound in body weight after the beginning of the Dynastic period, although it would be useful to have more samples from the 2,000 years that separate these periods to form more solid conclusions.
23
Correlation between stature and body mass
An ordinary least squares (OLS) regression plot of stature relative to body
mass is reported in Figure (4-‐4) in order to determine the quantitative relationship between both metric variables dealing with levels of sexual dimorphism. The plot displays a high positive correlation (.916), which indicates both variables are inherently connected. Males have a wide spread with those from Jiangjialiang clearly being the tallest and largest among all samples. Males from the Xindian culture and the control group are located near the middle of the graph, with the MXY males closer to the female cluster. The three female groups surveyed cluster closer together than to any of the male groups, displaying the lowest average body mass and stature form all samples. From the graph, it can again be noted that the Jiangjialiang population was on average the tallest, but had the highest degree of sexual dimorphism, whereas the MXY population was on average the shortest, and had the least amount of sexual dimorphism. Bilateral asymmetry Degrees of bilateral asymmetry shifted little from population to population, although there were slight to moderate levels between males and females on certain measurement parameters. Male humeral midshaft diameter asymmetry drastically increases from the Jiangjialiang population to the Minhe X, Minhe M, and LGS populations from a mean percentage of directional asymmetry (%DA) of 3.67 to a mean %DA of 11.63, indicating more single-‐side-‐focused activities at the Xindian population sites (see
24
Tables 4-‐4 and 4-‐5 respectively). For females however, mean %DA decreases between both periods. At Jiangjialiang, the median %DA is 4.69, whereas at the later Minhe X, Minhe M and LGS sites, it decreases to only 0.63, indicating activities that require the use of both arms equally. The MXY site, dated to the same time period does not show a similar median %DA of the humeral midshaft in males (see Table 4-‐ 6). Instead, the male value (1.14) and the female value (0.77) indicate that females and males may have been performing more similar activities than at the Minhe X, Minhe M, and LGS sites. Males from the control sample display a moderate median %DA with 2.68, and a median %AA of 3.74 (see Table 4-‐7). Their radial midshaft median %DA and median %AA are also moderate, at 1.83 and 4.77 respectively, which may have had to do with their work at the cannery. Tibial midshaft diameters also show moderate changes in median %DA and median %AA in both males and females. At the Jiangjialiang site, median %AA for males and females is 2.97 and 2.30 respectively. At the MXY site, median %DA is 2.26 for males and 1.88 for females, but the median %AA for males is 2.88 and 3.57 for females, indicating that females have a higher percentage of absolute asymmetry of their tibia midshaft diameters. Lastly, males from the control population also display a moderate median %AA of 3.18.
25
Table
(4‐4)
–
Bilateral
asymmetry
scores
in
the
population
of
Jiangjialiang.
Measure Humeral maximum length Humeral distal epicondylar breadth Humeral head diameter Humeral average 50% diaphyseal diameter Radial maximum length Femoral AP head diameter Tibial maximum length Tibial average 50% diaphyseal diameter
Median %DA (mean %DA) Total Males Females 0.71 (0.56) -0.24 (0.00) 1.79 (1.41) -1.37 (-1.50) -1.03 (-0.82) -1.37 (1.04) 0.06 (0.32) -0.62 (0.22) 1.11 (0.47) 3.97 (3.75) 3.67 (3.63) 4.19 (3.88) 0.22 (0.28 -0.21 (0.05) 0.43 (0.43) -0.31 (-0.32) -0.63 (-0.32) 1.22 (-0.31) 0.00 (-0.11) -0.47 (-0.10) 0.15 (-0.12) 0.18 (0.57) 0.29 (0.71) 0.07 (0.41)
Median %AA (mean %AA) Total Males Females 1.26 (1.22) 1.18 (1.10) 1.79 (1.41) 1.37 (1.96) 1.17 (1.04) 1.37 (3.38) 1.74 (1.65) 1.86 (1.69) 1.74 (1.58) 4.32 (4.81) 4.91 (5.12) 4.19 (4.49) 0.53 (0.71) 0.52 (0.57) 0.76 (0.80) 1.30 (1.48) 0.76 (0.88) 2.05 (2.49) 0.54 (0.63) 0.54 (0.64) 0.46 (0.62) 2.57 (3.47) 2.97 (4.07) 2.30 (2.79)
Table
(4‐5)
‐
Bilateral
asymmetry
scores
in
the
populations
of
Minhe
X,
Minhe
M
and
LGS.
Measure
Median %DA (mean %DA) Males Females -1.02 (-1.24) 3.49 (4.57) 11.63 (11.85) 0.63 (2.15) -0.30 (-0.12) -0.62 (-0.68) 0.12 (0.25) -0.40 (-0.43) 0.29 (0.39) -1.10 (-1.06) 0.01 (-0.06) -0.27 (-1.00) 0.30 (0.34) -0.29 (-0.05) -1.15 (-1.19) 2.24 (2.61)
Total Humeral head diameter Humeral average 50% diaphyseal diameter Femoral maximum length Femoral AP head diameter Tibial maximum length Tibial condylar breadth
26
Median %AA (mean %AA) Males Females 1.02 (1.28) 5.13 (3.49) 11.63 (11.85) 1.76 (2.89) 0.53 (0.69) 0.68 (0.71) 0.43 (0.70) 0.98 (1.41) 0.46 (0.96) 1.22 (1.76) 0.74 (1.36) 0.54 (1.36) 0.88 (1.36) 1.54 (1.65) 1.16 (1.36) 2.24 (2.24)
Total
Table
(4‐6)
‐
Bilateral
asymmetry
scores
in
the
population
of
MXY.
Measure Humeral maximum length Humeral distal epicondylar breadth Humeral head diameter Humeral average 50% diaphyseal diameter Radial maximum length Femoral maximum length Femoral epicondylar breadth Femoral AP head diameter Tibial maximum length Tibial condylar breadth Tibial average 50% diaphyseal diameter
Median %DA (mean %DA) Total Males Females 1.45 (1.33) -0.40 (-0.87) 2.21 (2.24) 4.19 (4.19) 0.81 (0.94) 0.95 (1.14) 1.14 (1.10) 0.77 (1.17) 1.12 (0.52) 0.36 (-0.11) 0.36 (-0.08) 0.19 (-0.13) 0.00 (0.51) -0.34 (0.17) 0.74 (0.74) -0.14 (-0.08) -0.60 (-0.82) 0.75 (0.77) -0.30 (-0.30) -0.36 (-0.36) -0.30 (-0.27) -0.38 (-0.38) 2.15 (1.45) 2.26 (1.90) 1.88 (0.93)
Median %AA (mean %AA) Total Males Females 1.45 (1.33) 0.94 (1.59) 2.21 (3.04) 4.19 (4.19) 2.00 (2.27) 2.27 (2.52) 2.24 (2.06) 3.01 (2.87) 1.12 (1.12) 0.63 (0.90) 1.43 (1.27) 0.50 (0.63) 1.05 (1.06) 1.02 (1.18) 1.11 (0.99) 0.75 (1.52) 0.74 (1.81) 0.84 (1.18) 0.72 (0.64) 0.36 (0.36) 0.91 (0.76) 0.77 (0.77) 3.39 (4.26) 2.88 (4.46) 3.57 (4.04)
Table
(4‐7)
‐
Bilateral
asymmetry
scores
in
the
Control
population.
Measure Total Humeral maximum length Humeral distal epicondylar breadth Humeral head diameter Humeral average 50% diaphyseal diameter Radial maximum length Radial average 50% diaphyseal diameter Femoral maximum length Femoral epicondylar breadth Femoral AP head diameter Femoral average 50% diaphyseal diameter Tibial maximum length Tibial condylar breadth Tibial average 50% diaphyseal diameter
Median %DA (mean %DA) Males Females 0.66 (0.70) 1.25 (0.96) 1.20 (1.29) 2.68 (2.75) 0.85 (0.84) 1.83 (2.41) -0.23 (-0.27) 0.65 (0.53) 0.29 (0.21) 0.26 (0.54) -0.07 (-0.12) 0.00 (0.35) 1.24 (0.36)
27
Total
Median %AA (mean %AA) Males Females 0.66 (0.97) 1.65 (1.79) 1.90 (3.09) 3.74 (4.81) 0.85 (1.04) 4.77 (6.04) 0.58 (0.67) 0.65 (1.01) 1.08 (1.37) 2.02 (2.20) 0.43 (0.53) 1.37 (1.63) 3.18 (3.55)
Paleopathology Paleopathological conditions examined in this study were divided into five categories, each representing a related etiology. Every marker points to a different aspect on an individual’s life, and can be used to determine not only incidence, but propensity to infectious disease, nutritional stress, strenuous labor, intergroup conflict or other chronic health conditions that has left their mark on bone. Non-‐specific and Specific Response to Infection Periostitis is noted to be more common in agricultural populations than in hunter gatherers, reaching high incidence after growing group numbers promoted an intensification of this subsistence strategy (Cohen and Armelagos, 1984; Larsen, 1984). At the site of Jiangjialiang, 35 bones in 17 individuals (ten male and seven female) display the woven striated bone typical of the condition (see Figure 4-‐5). There is an unusually high percentage of type I periostitis in males (39.13%), as well as, but less so, for females (18.18%).
Furthermore, four cases in females and one in males is considered type II (severe). The tibia is the most affected bone, encompassing 20 of 35 cases, with the fibula
28
making up an extra 10 of those cases. Demographically, periostitis occurs early at the site, with 13 of the 17 cases present on individuals between 20 and 39 years old.
Other affected bones include the femur (3), and even more rare, the os coxa (1). Most bones (except for the healed rib) exhibiting healed fractures were associated with periostitis. A female around 40 to 54 years old displays a healed compound fracture of the humeral shaft with extensive bone remodeling that could be considered osteomyelitis, with remains of a cloaca on the midshaft.
However, the lesion is completely healed and no sequestrum or involucrum are apparent, only inactive woven bone. Also an important observation to be made is that in more than three cases, fractures are also associated with periostitis of the limb bone, usually involving the adjacent bone element, as in the case of healed
29
distal fibulae. This shows one of the many ways that periostitis could develop and the importance to document patterns of occurrence.
At the combined sites of Minhe X, Minhe M and LGS, 12 bones in three individuals display periostitis. One of the three cases is type II periostitis (see Table 4-‐8). Two of the three cases could not be aged or sexed, so determining the demographic profile at this site was not possible. At the MXY site, six bones in four individuals exhibit periostitis. Three or possibly all four cases are males between 20 to 39 years of age (since in the last case it could only be ascertained that the individuals was young). None of the individual cases is severe enough to be considered type II. Within the control group, there are 10 bones in four males affected by periostitis. Here, three of the four cases are severe enough to be considered type II. The demographic incidence is less focused on a particular age group. Here two males are between 20 and 39, another is between 40 and 54, and the fourth is over
30
55 years old. Within this population of cannery workers, two individuals suffer from acute chronic osteomyelitis, encompassing several skeletal elements, and displaying severe bone remodeling (see Figure 4-‐6). The first individual, a male between 44 and 48 years old, displays bone inflammation on right ulna, left fibula diaphysis, left proximal tibia diaphysis, and both clavicles. The left os coxa and humerus display severe bone remodeling and active woven bone lesions are apparent on inner table with sequestrum and involucrum present. The second case of an individual around 38 to 45 years old is more severe. Both humeri display active woven bone lesions with bone inflammation present. Right humeral head has withered, losing its shape, and sequestrum and involucrum appear throughout both diaphyses. Osteological changes extend distal to the proximal ulna and radius on the left arm. Both scapulae show spread of infection, especially the right one near the articulation with the withered humeral head. This individual may have been capable of only 10-‐15 degrees of anterio-‐posterior rotation on right arm. Sternum also displays woven bone lesions with the left first rib fused onto the manubrium after the cartilage ossified and there is swelling on the body of sternum. The diaphyses of the left and right clavicles show reactive lesions (woven bone comprises entire left clavicle). The left femur displays moderate osteomyelitic changes on shaft and linea aspera. The left tibia also displays inflammation on the proximal diaphysis and woven bone.
31
Table
(4‐8)
–
Incidence
of
pathological
conditions
per
individual
in
all
samples.
Lesion
JJL
Females
MHX/MHM/
LGS
Males
MHX/MHM/
LGS
Females
MXY
Males
MXY
Females
Control
Males
%
[N]
%
[N]
%
[N]
%
[N]
%
[N]
%
[N]
%
[N]
Type
I
periostitis
39.13
[9]
18.18
[4]
20
[2]
0
[0]
36.36
[4]
0
[0]
1.54
[1]
Type
II
periostitis
4.35
[1]
13.64
[3]
10
[1]
7.14
[1]
0
[0]
0
[0]
4.62
[3]
Osteitis
0
[0]
4.55
[1]
0
[0]
14.29
[2]
9.09
[1]
8.33
[1]
3.08
[2]
Osteomyelitis
0
[0]
0
[0]
0
[0]
0
[0]
0
[0]
0
[0]
0
[0]
0
[0]
0
[0]
0
[0]
0
[0]
0
[0]
3.08
[2]
0
[0]
Osteophytosis
47.83
[11]
22.73
[5]
20
[2]
14.29
[2]
18.18
[2]
0
[0]
1.54
[1]
Osteoarthritis
8.70
[2]
4.55
[1]
0
[0]
14.29
[2]
9.09
[1]
0
[0]
4.62
[3]
Spina
bifida
Sacralized
L4
and/or
L5
13.04
[3]
4.55
[1]
40
[4]
0
[0]
9.09
[1]
25
[3]
4.62
[3]
13.04
[3]
4.55
[1]
0
[0]
14.29
[2]
18.18
[2]
8.33
[1]
13.85
[9]
Forelimb
fracture
4.35
[1]
4.55
[1]
0
[0]
7.14
[1]
9.09
[1]
0
[0]
3.08
[2]
Hindlimb
fracture
8.70
[2]
0
[0]
0
[0]
0
[0]
9.09
[1]
0
[0]
3.08
[2]
Rib
Fracture
8.70
[2]
0
[0]
0
[0]
0
[0]
0
[0]
0
[0]
0
[0]
Herniated
disc
Ossified
ligament/
tendon
8.70
[2]
9.09
[2]
0
[0]
0
[0]
0
[0]
0
[0]
0
[0]
8.70
[2]
0
[0]
20
[2]
21.43
[3]
9.09
[1]
0
[0]
3.08
[2]
0
[0]
9.09
[2]
0
[0]
0
[0]
0
[0]
0
[0]
3.08
[2]
Specific
Infection
Ossified
joint
JJL
Males
32
Both cases of osteomyelitis and the third possible case on the Jiangjialiang
humerus differ from the periostitis cases in the patterns of bone that are affected, so they must occur via a different etiology (see Figure 4-‐6). Also, all three cases seem to stem from infections cause by traumatic bone fractures, whereas only two cases of periostitis are associated with fractures. Seven cases of non-‐specific inflammation are noted within the collections, which do not qualify as periostitis but rather osteitis because of the involvement of inner cortical bone.
These cases are uncommon, occurring only once in a Jiangjialiang female, twice in females from the Minhe X, Minhe M and LGS sites; twice at the MXY site affecting a male and a female; and twice in males within the control group. All samples were also analyzed to assess any possible specific inflammatory changes stemming from treponemal disease, tuberculosis and leprosy but no evidence was found. A possible case of Brodie’s abscess was found in the MXY sample. The pathology is described as a one-‐centimeter foramen roughly of elliptical shape located on the posterior medial supracondylar region of distal left femur. The cortical bone appears normal and the bone itself does not appear swollen.
33
The edges are smooth and the lesion is deep enough to reach the medullary cavity, although no trabecular bone is discernible. However, radiographs are needed in order to be certain of this diagnosis. Lastly, it should be mentioned that no changes in morphology resulting from specific infections were noted at any of the sites, although such diseases may have been prevalent in the region due to trade. Degenerative Joint Disease Arthrititic lesions resulting from DJD can occur between synovial joints on limbs or apophyseal joints on vertebra (osteoarthritis). It can also manifest itself as bony outgrowths on the centra of vertebral segments (osteophytosis). At the Jiangjialiang site, there was a high incidence of osteophytosis among males (affecting 11 out of 34 individuals) whereas only one female was affected.
34
At the Minhe X, Minhe M, and LGS sites, the incidence of osteophytosis was lower for males, occurring only on two individuals, and also twice on females. At the MXY site, two males and no females were affected. In the control group, only one male showed osteophytotic lesions. Incidence of osteoarthritis was low throughout the sites, only appearing in two males and one female at Jiangjialiang, two females from the Minhe X, Minhe M and LGS sites, one male at the MXY site and three males in the control group.
35
A possible case of ankylosing spondylitis or diffuse idiopathic skeletal hyperostosis (DISH) was found within the comingled remains from the Minhe X site. Three fused thoracic and/or lumbar vertebra fragments, along with a possible fourth fragment of sacrum display anterior fusion between vertebral bodies without involvement of the centrum.
The tissue that appears to have ossified, linking up the vertebrae is the longitudinal ligament of the spine, which may point to DISH (Verlaan et al., 2007). However, due to the lack of osteophytosis of the spinal segments, the individuals suffering from this condition may not have been old, pointing to ankylosing spondylitis, which primarily affects individuals over 50 years of age (Rogers et al, 1985).
36
Congenital Conditions Data from only two congenital conditions, spina bifida and lumbosacral transitional vertebra (LSTV) are analyzed in this study. Most other non-‐metric traits occur on the cranium, especially in the dentition, both of which were not examined. Sacralization of the fifth lumbar vertebra is evident in three males and one female from the Jiangjialiang site. At the Minhe X, Minhe M and LGS sites, no males and only two females exhibit the condition. At the MXY site, the incidence is also quite low: only two males and one female exhibit L5 sacralization.
In the control group however, males exhibit a high incidence of this condition (13.85% in 67 individuals). One case of the less common lumbarization of a sacral
37
vertebra was found in the LGS collection, but was not included in the study since it was part of the comingled remains to which no sex could be attributed.
All cases of spina bifida analyzed are comprised of either incomplete or complete spina bifida oculta, which are common in modern populations. The less common condition spina bifida cystica is fatal, and therefore not observed in adults. At Jiangjialiang, three males and one female exhibit the condition, whereas at the Minhe X, Minhe M and LGS sites, four males and no females exhibit it. The other Xindian culture site MXY, has one male and three females who display the condition, whereas in the control group, three males display spina bifida.
38
Trauma Several types of trauma can be discerned in Jiangjialiang males, but have low incidence. Healed fractures occur only seldom in the forelimb (4.35%), hindlimb (8.7%) and ribs (8.7%) within males, whereas only 4.35% of females exhibited forelimb fractures. Herniated discs occur at a slightly higher incidence in females (9.09%) than in males (8.70%) at Jiangjialiang.
Some cases of Schmorls nodes, which are evidence of heavy loads, were also noted at the Jiangjialiang site.
39
Furthermore, Jiangjialiang males had an 8.70% incidence of ossified tendons/ligaments, whereas females have a 9.09% incidence of ossified joint segments.
The only apparent pathologies found within the three sites of Minhe X, Minhe
M and LGS are relatively high incidences of ossified tendons and/or ligaments, with males displaying the condition 20% of the time and females 21.43% of the time. In addition, females also display forelimb fractures 7.14% of the time. At the MXY site, males display similar incidences of forelimb (evidenced by an amputation halfway down the forearm) and hindlimb fractures, as well as ossified tendons/ligaments (9.09%).
40
No pathologies can be discerned on the females of these populations. Within the males of the control group, there are fairly low incidences of forelimb and hindlimb fractures, as well as ossified tendons/ligaments and joints (all at 3.08%). Paleodemography
Regarding the paleodemography of all sites, it is useful to build a life table for
comparisons between sites, such as Table 4-‐9. For the purpose of this study it is assumed that any sampling biases for sub-‐adults are sufficiently small to be omitted. At Jiangjialiang, four sub-‐adults (three males and one of indeterminate sex) have an age-‐at-‐death around 10-‐15 years. At the combined sites of Minhe X, Minhe M and LGS, 17 sub-‐adults were found, 14 whose age could be discerned. Six had an approximate age-‐of-‐death from around 6 months to 3.5 years. Another four had age-‐ at-‐death estimates between 6 and 10 years old and the last four were between 11 and 18 years old when they died. Three individuals at the MXY site has age-‐at-‐death estimates of 4.5 through 10 years, and another two were approximately between 13 to 15 years old when they died.
41
Table (4-‐9) – Demographic of all surveyed populations. Age 18-29 30-39 40-49 50+
JJL M N % 7 (23) 14 (47) 8 (27) 1 (3)
JJL F N % 8 (32) 7 (28) 6 (24) 4 (16)
MHX/MHM/LGS M N % 10 (59) 6 (35) 1 (6) 0 (0)
MHX/MHM/LGS F N % 3 (13) 13 (57) 6 (26) 1 (4)
MXY M N % 1 (9) 8 (73) 1 (9) 1 (9)
MXY F N % 3 (23) 9 (69) 1 (8) 0 (0)
Control M N % 4 (6) 25 (40) 26 (41) 8 (13)
Adult age-‐at-‐death spreads display some differences between the sites. At
Jiangjialiang, 47% of males died between 30 to 39 years and only 3% reached late adulthood (50+ years). Female age spreads at Jiangjialiang are more evenly distributed. Although they tend to accumulate earlier (60% died between the ages of 18 and 39 years), 16% of women still manage to live over 50 years (see Figure 4-‐ 11). At the combined sites of Minhe X, Minhe M and LGS, 59% of males did not live past 29 years of age, whereas 57% of females lived until 30-‐39 years old, another 26% lived up to 49 years and even still, 4% lived past 50 years of age (see Table 4-‐ 9). It should be noted that substantial amounts of demographic information from these three sites are missing due to post-‐collection comingling (38 adults could not be aged with certainty and 89 skeletal elements were found together in a box). At the MXY site, 73% of males die around 30-‐39 years of age, with another 9% living up to 49 years and a further 9% living past 50 years. For females, 69% die between 30-‐ 39 years of age, and another 8% lived up to 49 years, but none were found to live past 50 years. Around 81% of males from the control group died between 30 and 49 years of age and 13% lived past 50 years.
42
The health of populations in North China during the Bronze Age appears to
have worsened from the Late Neolithic period, in different ways. Female and male susceptibility to disease as well as their activity patterns are seen to differ, even across contemporaneous populations. Now I will discuss these results within the wider context of health during the period when agricultural intensification occurs in the region and present a health model for these populations. Figure (4-‐1) – Comparison of male and female mean statures between the Peiligang site of Jiahu and all Yangshao sites.
43
Figure (4-‐2) – Comparison of male and female mean statures between Longshan, early Dynastic sites and the Control population.
Figure (4-‐3) – Mean body mass (in kg) for males and females at each surveyed site.
Figure (4-‐4) – Linear regression for stature as a function of body mass for males and females of each surveyed population.
44
Slope: y= 0.6067 + 126.44 Coefficient of Correlation = .916 Figure (4-‐5) – Incidence of infectious disease in males and females across all surveyed sites.
45
Figure
(4‐6)
–
Distribution
of
skeletal
lesions.
On
the
skeleton
displaying
periostitis,
the
darker
shade
indicates
more
common
involvement;
the
light‐shaded
areas
are
where
the
condition
is
less
commonly
found
(after
Kelley,
1989).
Periostitis
Osteomyelitis
case
#1
Osteomyelitis
case
#2
Image
credit:
www.sciencequiz.net
46
Figure (4-‐7) – Incidence of degenerative joint disease on males and females at all surveyed sites.
Figure (4-‐8) – Incidence of congenital conditions in males and females at all surveyed sites.
47
Figure (4-‐9) – Trauma patterns between males and females across all surveyed populations.
Figure (4-‐10) -‐ Paleodemography comparison between all surveyed populations.
48
Figure (4-‐11) – Age ranges for males and females at each surveyed site.
CHAPTER 5 DISCUSSION In this chapter, I first reconstruct the population health at the Late Yangshao/Early Longshan site of Jiangjialiang and the Xindian sites of Minhe X, Minhe M, LGS and MXY in order to examine the role that diet plays in the propensity to nutritional stress and disease. I then follow with a discussion of how stature, sexual dimorphism and body mass data from these populations correlate to subsistence and levels of nonspecific response to infection at each site. Indicators of activity are then assessed in each population to determine possible stratification of labor between males and females. Some attention is given to the history of specific
49
infections in the region and the frequency of congenital sacral malformations in order to discern relatedness. Finally, I review changes in stress marker incidence overtime and present an overall model of health for each site and across time in North China. Diet and Disease Diet has a very close relation with susceptibility to disease (Kelley, 1989). Cohen (1991) notes increased infection rates in populations that adopted agriculture (and domestication). This trend skyrockets as cultivation is later intensified as nutrient-‐poor starchy foods that support larger populations become the staple diet of large populations. Disease reduces uptake of nutrients, which in turn propagates the condition and brings on added stress (Roosevelt, 1984).
In China, experimentation with millet, a small-‐seeded cereal crop may have
occurred as early as 12,000 BP (Shi, 1998, 2001, cited in Pechenkina et al, 2005). However, more formalized millet agriculture, evidenced by agricultural tools and carbonized grain, probably began around 8,500-‐7,000 BP (Hu et al., 2008). Based on the spread of Sino-‐Tibetan language, Bellwood (2009) proposes that cultural contact played a role in the adoption of agriculture in the region. Animal domestication also had its origins in North China around the early Neolithic, with osteological evidence at Cishan, in Hebei province around 8,000 BP. This may have occurred as a passive reaction to meat scarcity from increasing cultivation (Jing & Flad, 2002). This evidence of animal domestication and early cultivation suggests that individuals living at the late Yangshao site of Jiangjialiang would have had both subsistence
50
strategies available to them. Isotopic studies on the Jiangjialiang population based on nine trace elements have yielded a wealth of dietary information. Within the site, inhabitants consumed a diet primarily consisting of non-‐grain vegetables (Strontium levels were higher than normal) with some evidence for millet agriculture (Wang, 1995; Okazaki, 2008). Wang (1995) notes that food distribution at the site was uneven and all samples showed varying degrees of dietary activity. He mentions a key difference in the diet of the 17 individuals he tested. One group had a significantly higher proportion of meat consumption than the other. Wang (1995) also mentions that the population at Jiangjialiang consumed relatively low levels of protein when compared to later Bronze Age period sites from Gansu province. In contrast, although no isotopic studies were undertaken at the Minhe sites, archaeological evidence points to intensive cultivation of grains (mainly millet) supplemented by some meat (Qinghai Provincial Administration, 2004; Wei, personal communication). The wide range of nutritional profiles seen here may play a role in the different levels of stress markers examined below. As grain consumption increased through time and became the main food source for these populations, the impact of less nutrients on individual and population health will be quantified. Stature, Sexual Dimorphism and Body Mass
Three stress indicators (Stature, Sexual Dimorphism and Body Mass) are a
valuable tool for assessing nutritional difference in sex and long-‐term health changes in a population. At Jiangjialiang, both the average male stature (170.81cm)
51
and the average female stature (159.40cm) scored the highest within all surveyed populations analyzed in this study and second only to the Peiligang culture site at Jiahu (from Pechenkina et al., 2007). Further comparison with Longshan sites and Xindian culture sites examined in the study shows a steady stature reduction. The average males at Longshan sites measure 167.25cm whereas the average males and Xindian sites, later in time, measure 164.41cm. Female stature at Longshan sites does not seem to change much, measuring on average 159.44cm. However, there is a small stature decrease at later Xindian sites, where MXY females measure 158.6cm. These data suggest a slight to moderate stature decrease overtime affecting males much more than females.
Sexual dimorphism at Jiangjialiang was also the highest in all populations
surveyed (3.45), and only second highest to Meishan (3.47), a Longshan culture population from the Pechenkina et al., (2007). In the later-‐occurring Xindian culture, such dimorphism drops to 2.50 at the Minhe X, Minhe M, and LGS sites, and even lower at the MXY site (1.08). Benfer (1984) noted a similar drop in sexual dimorphism for the population at the pre-‐ceramic village of Paloma, Peru. This drastic drop in sexual dimorphism levels over time may stem from similar diet being consumed by males and females in the later periods.
Changes in estimated body mass can also be noted between males and females
over time. Jiangjialiang males and females once again score the highest of the populations, at 70.55kg and 59.02kg respectively. An additional interesting pattern between the sexes emerges from the analysis of these data. Female body mass at these sites always hovers around 10 kg less when compared to males.
52
Body mass data and sexual dimorphism data seem to disagree at this point.
One possible reason is that female anatomy is generally more compact than male anatomy within a homogeneous population and changes across populations will display the similar values. However, sexual dimorphism values decrease across time, because female stature does not decrease at the level that male stature does as a consequence of nutritional stress. Ortner (1998) attributes similar observations to differences in hormones between both sexes that allow females to have better natural buffers against stress, especially since they are adapted for childbirth. Even if both sexes maintain the same body mass dimorphism, both experience nutritional stress overtime, evidenced in their decreasing statures.
The different subsistence strategies between the Yangshao culture site and the
Xindian culture sites not only played a mayor role in stature and body mass changes, but also determined how susceptible each population would be to infectious diseases. Cohen (1991) notes that rates of infection increased as farming was adopted an intensified, although infection is traditionally with increasing population density and sedentism. Levels of Type I periostitis among males and females were highest at Jiangjialiang (39.13% and 18.18% respectively). Jiangjialiang females also suffered the highest levels of Type II periostitis out of all populations (13.64%). When comparing males and females across time however, another pattern emerges. Incidence of periostitis in females decreases drastically from the Neolithic period through to the Bronze Age. For males, however, incidence of periostitis decreases only slightly at the end of the Neolithic period before once again increasing during the Bronze Age; evidenced by males from the MXY site at 36.36%. Osteitis incidence
53
differed from this trend, increasing in both groups over time and possibly suggesting a different etiology from periostitis. Data on infectious disease indicate that although males and females may have been suffering from similar nutritional stress, males were more susceptible to disease, supporting Ortner’s (1998) model. Indicators of Activity
Indicators of activity have been used historically to argue that agricultural
populations worked harder and longer than their hunter-‐gatherer predecessors (Lee and DeVore 1968, cited in Cohen, 1991). However, more recent work has not only dispelled these myths (Larsen, 1984), but has also pointed out that are these indicators only elucidate physical demand peaks but not work length. In any case, occupationally–related paleopathology is, as defined by Stirland (1991), nonrandom and habitual, so skeletal changes directly relate to the activities performed while the individual was alive. In what follows, I investigate the intensity of labor between early agricultural populations and those practicing more intensive agriculture via levels and patterns of DJD, bilateral asymmetry and trauma. My examination also aims to determine any possible differences in labor loads between males and females.
Males and females at Jiangjialiang exhibit moderate levels of osteoarthritis
(8.7% and 4.55% respectively) and very high levels of osteophytosis (47.83% and 22.73% respectively). At the sites of Minhe X, Minhe M and LGS, levels of osteoarthritis increase for females but overall incidence of osteophytosis decreases. Discerning osteophytosis incidence in Xindian culture sites presented a challenge, as
54
most vertebra, ribs and os coxae were not curated along with limb bones (also making sex and age estimations difficult). Lastly, at the MXY site, males show low levels of DJD and no females exhibit these conditions. The decreasing levels of DJD agree with data comparing hunter-‐gatherers to agriculturalists and may signify that this trend continued from early agriculture through to intensive agriculture.
In the agricultural populations surveyed, osteoarthritis was observed mostly
on the elbow joint but at very low levels. Inouye et al., (2001) note that osteoarthritis is most often seen in East Asians at the medial femoral condyle and the medial tibial plateau of the knee joint and is more common in hunter-‐gatherers at the elbow and knee joints than agriculturalists. Findings within this study also agree with their data, suggesting that these joints were being utilized with higher frequencies in earlier periods. DJD was found mostly at the elbow joint, with one case of distal ulna DJD at the MXY site. Expanding the scope of the current study to pre-‐agricultural populations will produce more results for better data comparisons, fortifying the hypothesis that DJD decreased from hunter-‐gatherer populations to practitioners of early agriculture and intensive agriculture.
Bilateral asymmetry is another way to judge levels of activity in populations.
The parameters that changed the most within the populations surveyed were the diameter of the humeral midshaft and of the tibial midshaft. Mean %DA for humeral midshaft diameter in males increases up to 11.63% at Xindian culture sites, although females experience an overall decrease over time. Asymmetry of the tibial midshaft diameter in males decreases from Jiangjialiang to the Xindian culture sites, although the opposite occurs in females. Judging from the data, males become more
55
involved in activities that require the use of one arm over the other, such as raising a plow or other agricultural implements. Females however, have an overall decrease in asymmetry, suggesting they took up activities that required less use of one arm over the other, such as cooking or the use of a grinding stone to process grain. One surprising result is the increased asymmetry in the tibial midshaft in females over time, as no previous studies of occupationally-‐related osteological changes chronicle this occurrence. Kennedy’s (1989) review of past literature of skeletal markers of occupational stress only yielded tibial pathologies related to nutritional deficiencies (platycnemia) and to squatting (squatting facets, retroversion of the tibial head, rounding of the lateral tibial condyle), so further research is needed to examine this phenomenon.
Trauma patterns in the form of fracture and ossified soft tissue were also
examined in order to discern activity. Jiangjialiang males exhibit several types of trauma (forelimb, hindlimb and rib fractures, along with herniated disks and ossified tissues) although occurring at low levels. Females only display low levels of herniated disks and ossified joints, suggesting that even if both sexes were doing heavy lifting of grain for storage or construction materials, males may have been doing a wider range of activities that promoted different fracture patterns. Females at Minhe X, Minhe M and LGS display three cases of ossified ligaments/tendons, suggesting increased peak intensity of labor (more pronounced loading in less time). Lastly, forelimb fractures are seen at very low levels at Xindian culture sites, mostly in males. However, this finding may not be meaningful, as the incidence does not change from Jiangjialiang. Activity at the MXY site with more intensive
56
agriculture seems to have lessened the incidence of different fracture patterns. However, new types of repetitive activity, such as grinding of cereal grains, could have increased the likelihood for ossified soft tissue, at least in females.
Although no cases of specific infections were found, it is important to note the
prevalence of these conditions in the region. Historical descriptions of possible leprosy cases in China start around 2,500 BP. An account tells of a Confucian, Pai-‐ Niu, suffering from a disease characterized by nasal destruction, eyebrow loss, crippled broken legs, anesthetic mucosa and hoarseness (McLeod and Yates, 1981). Suzuki et al. (2005) examined skeletons from the Kayue culture (2,500 BC to 1,850 BP) and found two cases showing irregular sclerotic hyperostosis with swelling on distal femoral and tibial diaphyses, suggestive of treponemal disease. The authors cite Hackett’s map of endemic syphilis (1967), noting that the most northeastern region of prevalence around 9,000 BP corresponds to Qinghai province.
These data indicate that with the development of long-‐distance trade routes,
diseases producing specific responses were likely present in China around the time of the Xindian culture. From gross examination of skeletal material from the sites in Qinghai and Hebei provinces, no pathologies characteristic of treponemal disease or leprosy could be discerned. However, there were high incidences of periostitis in all populations, which signifies low tolerance for infectious disease in the region.
Both congenital conditions examined, LSVT and spina bifida, could be used to
detect relatedness in the population, which may help in determining the level of endogamous pairings. In total, four individuals in the Jiangjialiang population and another four from the Xindian culture combined sites exhibit the condition, which is
57
not statistically significant. However, the prevalence of LSVT is moderate (13.43%) in the control population, which may suggest that several Karluk cannery workers may have been related. Health Model for Chinese Populations from the Neolithic period to the Bronze Age
The subsistence strategies at Yangshao site of Jiangjialiang emphasize a diet of
non-‐grain vegetables along with moderate consumption of meat by a certain segment of the population. In contrast, later populations belonging to the Xindian culture had a diet mainly consisting of grain supplemented by small amounts of meat. Comparison of nutritional stress markers from both time periods indicates that males and females experienced lower quality of health into the Bronze Age, evidenced by decreased stature and body mass. The health impact on females was less significant than in males, who continually experienced higher levels of non-‐ specific infection, suggesting a higher susceptibility to disease. A decrease in sexual dimorphism also points to males having suffered a more profound health impact than females.
Occupational markers of stress at the Xindian culture sites indicate lower
levels of labor intensity from a decrease in the incidence of degenerative joint disease in both sexes when compared to the Jiangjialiang population. Bilateral asymmetry data also point to different activities at the Xindian culture sites. Male humeral shafts become more asymmetric, suggesting activities that favor one side of the body over the other, such as planting grain and land management. Females on the other hand, become less asymmetric, suggesting that they performed activities
58
requiring the use of both sides of the body equally such as tending to animals, craft production or food processing.
Patterns of traumatic lesions also indicate that males at Jiangjialiang were
involved in a wider scope of activities (perhaps related to food procurement or settlement construction) than females, as they display different fracture patterns at low levels. This parameter changes at the later Xindian sites, suggesting more focused activities. Females seem to experience higher levels of joint, and ligament/tendon ossification at the Xindian sites, which may point to higher levels of focused activity – which, coupled by decreased upper limb asymmetry lends further support to hypotheses for food processing activities.
The results of this study strongly suggest that as populations in North China
shifted their subsistence strategies from vegetable cultivation in the Neolithic period to intensive grain cultivation during the Bronze Age, levels of nutrition decreased (impacting males more severely) while activities between males and females became more stratified and focused. Pechenkina et al., (2002) examined developmental stress markers in the late Neolithic period and proposed a similar model of health for these populations. They blame subsistence changes occurring during the late Yangshao period for the general decline in health evident in the later Longshan period persisting into the Western Zhou Dynasty.
An important question to ask is: what prompted populations to settle and
begin to grow their own food? Also, if the consequences of this subsistence shift were decreased nutrition and higher infection rates, why did populations intensify agriculture at a later time? Cohen (2009) posits that the adoption of agriculture
59
seems more of a matter of necessity rather than of invention and choice. In reviewing his earlier work (Cohen 1977, 1991), he notes that that there is an overall increase in population in all areas of the world as time passes. Such steady population growth, coupled with dwindling resources created an imbalance, causing population stress and eventually prompting the exploitation of different sources. Cereal grains such as millet and sorghum (first cultigens in East Asia) provided a high ratio of caloric value to labor output, helping mitigate stress from population growth. However, these grains were low in vitamins and minerals compared to other vegetable foods.
Cohen argues that it was population growth and not climatic change (itself
only a regional phenomenon) the main factor that prompted incipient cultivation. However, a possible reason why different areas of the world developed agriculture at different times may have been due to favorable regional environments for planting (such as longer rainy seasons or more suitable land). Following this argument, with the notion that climatic events occur regionally, one could make a case for the later intensification of agriculture occurring in North China near the end of the Yangshao period, after the fifth millennium, and continuing into the third millennium BP. A study by An et al. (2005) into the rapid environmental change in the Western Loess Plateau of China verifies the increased aridity of the region at this time. They propose a reduction of settlement expansion due to the decreased numbers of archaeological sites in the region caused by reduced agricultural productivity.
Pechenkina et al. (2002) indeed propose that the subsistence changes that
60
impacted population health were first triggered by environmental deterioration during that period. The sudden food scarcity that resulted thus prompted agricultural populations to increase their production output of the highly caloric and weather-‐tolerant millet grain, which had been domesticated earlier during the Neolithic period. As the main staple in the diets of Longshan and later periods, cereal grains negatively impacted population health in the region, as all the data seems to suggest. Lastly, due to its high caloric properties and dependability as a crop, millet continued to support growing populations in China, albeit having low nutrient levels and a tendency to exacerbate bad health.
From the results of this study, a working model for the health impacts caused
by an increased reliance in cereal grains over time was presented. The evidence from the various nutritional and occupational stress parameters serves to reconstruct the possible physical activities and subsistence strategies of the populations surveyed. Next, I summarize the main points of this study and how they relate to the long-‐term consequences of agricultural intensification. Finally, I will place the findings of this study within the wider context of bioarchaeological research in East Asia and discuss its significance within current and past research into the role that diet plays in individual and population health at the end of the Neolithic period.
61
CHAPTER 6 CONCLUSION This study was designed to reveal possible connections between stress and the change in subsistence strategies occurring at the end of the Neolithic period through the Bronze Age in Northern and Eastern China. Three types of stress markers were studied: nutritional stress, occupational stress and paleopathology. Nutritional stress markers show an overall stature decrease for males and females from the Neolithic period through the Bronze Age in these regions of China. This finding was also confirmed by similar body mass decreases in males and females over time. Furthermore, it became apparent that male health has been more greatly impacted during this period because of a decrease in stature and sexual dimorphism, and greater incidence of infectious disease at the Xindian culture sites. These data suggest that males were more susceptible to nutritional stress. Occupational stress markers point to less severe activities being performed by individuals at Xindian culture sites. This finding implies that intensive grain agriculture was less arduous than a more mixed system that included vegetable cultigens and greater meat consumption. In addition, there was a sexual stratification of labor during the Bronze Age, apparent from the increased asymmetry in the males humeral midshaft and a decrease of the same parameter in females. Males became involved with activities that emphasized the use of one size of their body, whereas females began practicing activities where both sides of the body were utilized equally. Degenerative joint disease patterns also point to more
62
focused activities, such as repetitive lifting, pounding or carrying heavy objects, in both sexes over time, as incidence of different traumas in earlier populations disappear in later populations. The transition to, and later intensification of agriculture worldwide were complex and risky undertakings. This study verifies the theory proposed by Pechenkina et al. (2002) that a general health decline in North and West China at the end of the Neolithic period was caused by a marginalization of diet, due to a change in subsistence strategies. The intensification of grain production was probably a direct result of regional climatic changes at the end of the fifth millennium BP. The study also expands on Cohen’s hypothesis (1977) that population pressure around the world would have prompted hunter-‐gatherers to settle gradually and take up agriculture (discounting climate as the primary global factor). Cohen suggests that eventual regional climate degradation would have been the primary motive for populations to begin intensive agricultural production. To further understand the multifaceted role of subsistence changes in this region, I propose that additional populations in North and West China be examined to expand the scope of this research. Such comparisons would provide additional dietary data and a wider sample of male and female measurements from which more confident results could be obtained. Lastly, similar analyses should be carried in other regions of the world where agriculture was prevalent at the time and climatic data is available, in order for meaningful comparisons to be made. A limitation of my current study is the number of research projects and articles published in foreign languages and lesser-‐known journals that could not be used to
63
strengthen my proposed health model. I hope that my research can be of use to Chinese scholars as well as western scholars when attempting to answer questions regarding the shifting patterns of nutrition and occupation in East Asian populations and results in more cross-‐cultural collaborations. One day, the results of this study may further help elucidate other aspects of the intensification of agriculture and the consequences it had on population health across the world.
64
REFERENCES CITED Aihara, T., Takahashi, K., Ogasawara, A., Itadera, E., Ono, Y., Moriya, H., 2005. Intervertebral disc degeneration associated with lumbosacral transitional vertebrae: A clinical and anatomical study. J. Bone and Joint Surgery -‐ British Volume, 87-‐B (5) 687-‐691. An, C.-‐B., Lingyu T., Barton, L., Chen, F.-‐H., 2005. Climate change and cultural response around 4000 cal yr B.P. in the western part of Chinese Loess Plateau. Quarter. Res. 347–352. Angel, J.L., 1984. Health as a crucial factor in the changes from hunting to developed farming in the eastern Mediterranean. In: Cohen, M.N. and Armelagos, G.J. (Eds.), Paleopathology at the Origins of Agriculture. Academic Press, Orlando, pp. 51-‐73. Auerbach, B.M., Ruff, C.B., 2006. Limb bone bilateral asymmetry: variability and commonality among modern humans. J. Hum. Evol. 50 (2) 203-‐218. Aufderheide, A.C., Rodriguez-‐Martin, C., 1998. The Cambridge Encyclopedia of Human Paleopathology. Cambridge University Press, Cambridge. Benfer, R.A., Jr., 1984. The challenges and rewards of sedentism: the preceramic village of Paloma, Peru. In: Cohen, M.N. and Armelagos, G.J. (Eds.), Paleopathology at the Origins of Agriculture. Academic Press, Orlando, pp. 531-‐558. Bellwood, P., 2009. The Dispersals of Established Food-‐Producing Populations. Current Anthropology 50 (5) 621-‐626. Buikstra, J.E., Cook, D.C., 1980. Paleopathology: An American account. Ann. Rev. Anthropol. 9 433-‐470. Buikstra, J.E., Ubelaker, D.H., 1994. Standards for Data Collection from Human Skeletal Remains. Arkansas Archaeological Survey Research Series No. 44, Fayetteville. Cook, D.C., 1984. Subsistence and health in the Lower Illinois Valley: osteological evidence. In: Cohen, M.N. and Armelagos, G.J. (Eds.), Paleopathology at the Origins of Agriculture. Academic Press, Orlando, pp. 237–269. Cohen, M.N., 1977. The Food Crisis in Prehistory. Yale University Press, New Haven.
65
Cohen, M.N., 1991. Paleopathology and the interpretation of economic change in prehistory. In: Lamberg-‐Karlovsky, C.C. (Ed.), Archaeological Thought in America. Cambridge University Press, Cambridge, pp. 117-‐132. Cohen, M.N., 2009. Introduction: Rethinking the origins of agriculture. Current Anthropol. 50 (5) 591-‐595. Goodman, A., Armelagos, G.J., 1989. Infant and childhood morbidity and mortality risks in archaeological populations. World Archaeo. 21 (2) 225-‐243. Hackett C.J., 1967. The human treponematoses. In: Brothwell D. and Sandison A.T. (Eds.), Diseases in Antiquity. Charles C. Thomas, Springfield, pp. 152–169. Hu, Y., Wang, S., Luan, F., Wang, C., Richards, M.P., 2008. Stable isotope analysis of humans from Xiaojingshan site: implications for understanding the origin of millet agriculture in China. J. Archaeo. Sci. 35 2960–2965. Inoue, K., Hukuda, S., Fardellon, P., Yang, Z.Q., Nakai, M., Katayama, K., Ushiyama, T., Saruhashi, Y., Huang, J., Mayeda, A., Catteddu, I., Obry, C., 2001. Prevalence of large-‐joint osteoarthritis in Asian and Caucasian skeletal populations. Rheumatology 40 70-‐73. Iscan, M.Y., Shihai, D., 1995. Sexual dimorphism in the Chinese femur. Forensic Sci. Inter. 74 79-‐87. Jing, Y., Flad, R.K., 2002. Pig domestication in ancient China. Antiquity 76 724-‐732. Johnston, F.E., 1962. Growth of the long bones of infants and young children at Indian Knoll. Am. J. Phys. Anthropol. 20 249-‐254. Kelley, M.A., 1989. Infectious disease. In: Iscan, M.Y., Kennedy, K.A.R., (Eds.), Reconstruction of Life from the Skeleton. Wiley-‐Liss, Hoboken, pp. 191-‐200. Kennedy, K.A.R., 1989. Skeletal markers of occupational stress. In: Iscan, M.Y. and Kennedy, K.A.R. (Eds.), Reconstruction of Life from the Skeleton. Wiley-‐Liss, Hoboken, pp. 129-‐160. Larsen, C.S., 1984. Health and disease in prehistoric Georgia: The transition to agriculture. In: Cohen, M.N. and Armelagos, G.J. (Eds.), Paleopathology at the Origins of Agriculture. Academic Press, Orlando, pp. 367-‐392. Larsen, C.S., 1997. Bioarchaeology: Interpreting behavior from the human skeleton. Cambridge University Press, Cambridge. Larsen, C.S., 1998. Gender, health, and activity in foragers and farmers in the American southeast: Implications for social organization in the Georgia bight.
66
In: Grauer, A.L., and Stuart-‐Macadam, P. (Eds.), Sex and Gender in Paleopathological Perspective. Cambridge University Press, Cambridge, pp. 165-‐187. Li F., 2004. A Research of the Neolithic Human Skeletons from Jiangjialiang, Yangyuan in Hebei Province, China. Unpublished. M.A. Thesis. Loewe, M., Shaughnessy, E.L., 1999. The Cambridge history of ancient China: from the origins of civilization to 221 B.C. Cambridge University Press, Cambridge. Lovejoy, C.O., Meindl, R.S., Pryzbeck, T.R., Mensforth, R.P., 1985. Chronological metamorphosis of the auricular surface of ilium: A new method for determination of adult skeletal age at death. Am. J. Phys. Anthropol. 68 15–28. Magora, A., Schwartz, A., 1978. Relation between the low back pain syndrome and x-‐ ray findings. 2. Transitional vertebra (mainly sacralization). Scand. J. Rehabil. Med. 10 (3) 135-‐145. McCarty, D.J., Koopman, W.J., 1993. Arthritis and Allied Conditions: A Textbook of Rheumatology, 12th Edition. Lippincott Williams & Wilkins, Philadelphia. McLeod, K.C.D., Yates, R.D.S., 1981. Forms of Ch’in Law: An annotated translation of the Feng-‐chen shih. Harvard Journal of Asiatic Studies 41 (1) 111-‐163. Mintz, G., Fraga, A., 1973. Severe osteoarthritis of the elbow in foundry workers. Archives Of Environmental Health 27 (2) 78-‐80. Okazaki, K., Wei, D., Zhu, H., 2008. Oral and paleohealth in millet agriculturalists of the Neolithic and the Warring States-‐Qin dynasty, northern China. Paper presented at the 62nd annual meeting of the Anthropological Society of Nippon. Ortner, D., 1998. Male-‐female immune reactivity and its implications for interpreting evidence in human skeletal paleopathology. In: Grauer, A.L., and Stuart-‐Macadam, P. (Eds.), Sex and Gender in Paleopathological Perspective. Cambridge University Press, Cambridge, pp. 79-‐92. Pechenkina, E.A., Benfer, R.A., Jr., Wang Z., 2002. Diet and health changes at the end of the Chinese neolithic: The Yangshao/Longshan transition in Shaanxi province. Am. J. Phys. Anthropol. 117 15–36. Pechenkina, E.A., Ambrose, S.H., Ma X., Benfer, R.A., Jr., 2005. Reconstructing northern Chinese neolithic subsistence practices by isotopic analysis. J. Archaeo. Sci. 32 1176-‐1189.
67
Pechenkina, E.A., Benfer, R.A., Jr., Ma, X., 2007. Diet and health in the neolithic of the Wei and Middle Yellow river basins, northern China. In: Cohen, M.N., and Crane-‐Kramer, G. (Eds.), Ancient Health Skeletal Indicators of Agricultural and Economic Intensification. University Press of Florida, Gainesville, pp. 255-‐272. Phenice, T.W., 1969. A newly developed visual method of sexing of the 0s pubis. Am. J. Phys. Anthropol. 30 297-‐302. Qinghai Provincial Administration of Cultural Relics, and College for Cultural Heritage and Museology, Northwest University, 2004. Hetaozhuang of Minhe County. Science Press, Beijing. Robbins, G., Tripathy, V.M., Misra, V.N., Mohanty, R.K., Shinde, V.S., Gray, K.M., Schug, M.D., 2009. Ancient skeletal evidence for leprosy in India (2000 B.C.). PLoS ONE 4 (5) e5669 1-‐8. Roberts C., Manchester, K., 2005. The Archaeology of Disease. Cornell University Press, Ithaca. Rogers, J., Watt, J., Dieppe, P., 1985. Paleopathology of spinal osteophytosis vertebral ankylosis, ankylosing spondylitis and vertebral hyperostosis. Ann. Rheum. Dis. 44 113–120. Roosevelt, A.C., 1984. Population, health, and the evolution of subsistence: Conclusions from the conference. In: Cohen, M.N. and Armelagos, G.J. (Eds.), Paleopathology at the Origins of Agriculture. Academic Press, Orlando, pp. 559-‐84. Ruff, C.B., Hayes, W.C., 1983. Cross-‐sectional geometry of Pecos Pueblo femora and tibiae-‐A biomechanical investigation. I. Method and general patterns of variation. Am. J. Phys. Anthropol. 60 359-‐381. Ruff, C.B., Scott, W.W., Liu, A. Y.-‐C., 1991. Articular and diaphyseal remodeling of the proximal femur with changes in body mass in adults. Am. J. Phys. Anthropol. 86 397-‐413. Stirland, A.J., 1993. Asymmetry and activity-‐related change in the male humerus. Int. J. Osteoarchaeol. 3 105-‐113. Stirland, A.J., 1991. Diagnosis of occupationally related paleopathology: can it be done? In: Ortner, D.J., and Aufderheide, A.C. (Eds.), Human Paleopathology: Current Syntheses and Future Options. Smithsonian Institution Press, Washington, pp. 40-‐47.
68
Styne, D.M., McHenry, H., 1993. The evolution of stature in humans. Horm. Res. 39 (Suppl. 3) 3-‐6. Suzuki, T., 1991. Paleopathological study on infectious diseases in Japan. In: Ortner, D.J. and Aufderheide, A.C. (Eds.), Human Paleopathology: Current Syntheses and Future Options. Smithsonian Institution Press, Washington, pp. 128–139. Suzuki T., Matsushita, T., Han, K., 2005. On the possible case of treponematosis from the Bronze Age in China. Anthropological Science 113 253–258. Tini, P.G., Wieser, C., Zinn, W.M., 1977. The transitional vertebra of the lumbosacral spine: Its radiological classification. Rheumatology, 16 180-‐185. Todd, T.W., 1921. Age changes in the pubic bone. I: The male white pubis. Am. J. Phys. Anthropol. 3 285-‐334. Trotter M., Gleser, G.C., 1958. A re-‐evaluation of estimation of stature based on measurements of stature taken during life and long bones after death. Am. J. Phys. Anthropol. 16 79–123. Ubelaker, D.H., 1989. The estimation of age at death from immature human bone. In: Iscan M.Y. (Ed.), Age markers in the human skeleton. Charles C. Thomas, Springfield, pp. 55–70. Vergauwen, S., Parizel, P.M., van Breusegem, L., Van Goethem, J.W., Nackaerts, Y., Van den Hauwe, L., De Schepper A.M. (1997) Distribution and incidence of degenerative spine changes in patients with a lumbo-‐sacral transitional vertebra. Eur. Spine J. 6 168-‐172. Verlaan, J. J., Oner, F.C., Maat, G.J.R., 2007. Diffuse idiopathic skeletal hyperostosis in ancient clergymen. Eur. Spine J. 16 1129–1135. Waldron, T., Rogers, T., 1991. Inter-‐observer variation in coding osteoarthritis in human skeletal remains. Inter. J. Osteoarchaeol. 1 (1) 49-‐56. Wang, M.H., 1995. A Research on Ancient Human Bones from Hetaozhuang Cemetery of Xindian Culture in Minhe county Qinghai province. Unpublished. M.A. Thesis. Wigh, R.E., Anthony, H.F., Jr.,1981. Transitional Lumbosacral Discs: Probability of Herniation. Spine 6 (2)168-‐171. Wolff, J., 1892. Das Gesetr der Transformation der Knochen. Hirchwild, Berlin. (Translated as The Law of Bone Remodeling.) Springer, Berlin.
69
Wood, J.W., Milner, G.R., Harpending, H.C., Weiss K.M., 1992. The osteological paradox: problems of inferring prehistoric health from skeletal samples. Current Anthropology 33 343–370.
70