(Re)Thinking the Little Ancestor: New Perspectives on the Archaeology of Infancy and Childhood 9781407308456, 9781407338293

Table of contents :
Front Cover
Title Page
Copyright
Dedication
Table of Contents
Foreword
The Osteology of Infancy and Childhood: misconceptions and potential
Subadult or subaltern? Children as serial categories
Etruscan Infants: Children’s Cemeteries at Tarquinia, Italy as indicators of an age of Transition
Thrown Out with the Bathwater or Properly Buried? Neonate and Infant Skeletons in a Settlement Context on the Dürrnberg bei Hallein, Austria
The Children in the Bog
Parenting, child loss and the cilliní of Post-Medieval Ireland
The Disposal of Dead Infants in Anglo-Saxon England from c.500-1066: an overview
Where have all the flowers gone? Bronze Age children’s burials in South East England: initial thoughts
Ble mae’r babanod? (Where are the babies?): Infant burial in Early Medieval Wales
Childhood in Roman Egypt: Bioarchaeology of the Kellis 2 Cemetery, Dakhleh Oasis, Egypt
Constituting Childhood: Identity, Conviviality and Community at Windmill Hill
The Divine Power of Childhood in Ancient Mesoamerica
‘What Must be Done’: Ideology and the Children of Sand Canyon Pueblo
Living Children

Citation preview

BAR S2271 2011

(Re)Thinking the Little Ancestor: New Perspectives on the Archaeology of Infancy and Childhood

LALLY & MOORE (Eds)

Edited by

Mike Lally Alison Moore

(RE)THINKING THE LITTLE ANCESTOR

B A R

BAR International Series 2271 2011

(Re)Thinking the Little Ancestor: New Perspectives on the Archaeology of Infancy and Childhood Edited by

Mike Lally Alison Moore

BAR International Series 2271 2011

ISBN 9781407308456 paperback ISBN 9781407338293 e-format DOI https://doi.org/10.30861/9781407308456 A catalogue record for this book is available from the British Library

BAR

PUBLISHING

Dedicated to the memory of Trevor Anderson, a dearly missed inspirational friend and Kent’s finest human osteologist.

The editors also wish to thank Dr Anthony Ward, University of Kent at Canterbury, for all his assistance in the hosting of the conference.

Contents Foreword Mike Lally and Alison Moore ................................................................................................................. ii The Osteology of Infancy and Childhood: Misconceptions and potential Mary Lewis ............................................................................................................................................. 1 Subadult or Subaltern? Children as serial categories Frederik Fahlander .............................................................................................................................. 14 Etruscan Infants: Children’s cemeteries at Tarquinia, Italy, as indicators of an age of transition Marshall J. Becker ............................................................................................................................... 24 Thrown Out with the Bathwater or Properly Buried? Neonate and infant skeletons in a settlement context on the Durrnberg bei Hallein, Austria Raimund Karl and Klaus Löcker ......................................................................................................... 37 The Children in the Bog Grete Lillehammer ............................................................................................................................... 47 Parenting, Childloss and the Cillini of post-Medieval Ireland Eileen Murphy ...................................................................................................................................... 63 The Disposal of Dead Infants in Anglo-Saxon England from c.500-1066: An overview Sally Crawford .................................................................................................................................... 75 Where Have All The Flowers Gone? Bronze Age children’s burials in south-east England: Initial thoughts Dawn McClaren ................................................................................................................................... 85 Ble Mae’r Babanod? Infant burial in early Medieval Wales Marion R. Page .................................................................................................................................. 100 Childhood in Roman Egypt: Bioarchaeology of the Kellis 2 cemetery, Dakhleh Oasis, Egypt Sandra M. Wheeler, Lana Williams, Tosha L. Dupras, Matthew Tocheri and J. E. Molto ......................................................................................................................................... 110 Constituting Childhood: Identity, conviviality and community at Windmill Hill Oliver Harris ...................................................................................................................................... 122 The Divine Power of Childhood in Ancient Mesoamerica Traci Ardren ...................................................................................................................................... 133 ‘What Must Be Done.’ Ideology and the children of Sand Canyon Pueblo Cynthia Smith Bradley ....................................................................................................................... 139 Living Children Anna Kjellstrom, Camilla Olofsson, Ulrika Stenback Lonnquist and Stig Welinder ..................................................................................................................................... 152

i

Foreword In 2005, the first major international and multi-disciplinary conference, organised by Mike Lally, was held at the University of Kent at Canterbury to explore the archaeology of infancy and childhood. One of the primary aims of the conference was to bring together wide-ranging approaches towards the study of a previously overlooked social group and contributions were sought from leading specialists and researchers in bioarchaeology, archaeology, ethnology, history and anthropology. As a result of this important conference at Kent in 2005, a new Society – The Study of Childhood in the Past (SSCIP) – emerged, devoted exclusively to promoting, exploring and understanding the social construction of infancy and childhood in the past. The formation of SSCIP led to the publication of a new journal Childhood in the Past, now in its 4th year, a truly international and multi-disciplinary publication which brings together the very best in research and researchers in its field. The papers in this collection were originally presented at the Kent conference in 2005 and have been adapted and bought up-to-date by the authors. The papers reflect the multi-disciplinary and international nature of the conference, utilising specialisms in archaeology, anthropology and ethnology with material culture and visual and literary evidence relating to children and childhood from a wide spectrum of time periods from across the world. Mary Lewis opens the proceedings with an analysis of the problems and potential inherent in the bioarchaeological study of infants. Outlining recent developments in infant skeletal analysis, the paper explores the impact of the physical and cultural environment on infants and children in relation to issues such as infanticide, malnutrition and child abuse. Lewis also identifies the problem of what constituted an infant in the past and this point is explored by Frederik Fahlander who addresses social categorizations of children. Fahlander suggests that utilising the wider distinctions inherent in Satre’s ideas regarding series and groups to childhood in the past will enable us to identify tasks, spaces and social spheres in which different sub-categories of children related to adults and to each other at various stages of development. A key feature identified in many of the papers within these proceedings is that age-related distinctions in burial between infants and older children were practiced in a wide range of diverse societies. Marshall Becker’s study of infant and child burials from 7th-6th CE Tarquinia, Etruria (Italy) infers, through the absence of the very young, a separate burial area for perinates and infants below c.5 years. He suggests that this seperation does not involve practices such as infanticide or sacrifice but represents a long Etrurian tradition relating to the life stage of infants. Cultural beliefs associated with the ‘liminal’ status of infants may also have related to differential burial treatment. The burial of infants within settlement contexts in Iron Age Austria is explored by Raimund Karl and Klaus Löcker who support the archaeology with etymological evidence to suggest that the special burial rite underlined the liminality of infants who died prior to full social acceptance. A similar sense of the liminality of infants is suggested by Grete Lillehammer who considers through folklore, mythology and etymology possible meanings behind the inclusion of infants within a bog in Iron Age Norway. The textual and contextual evidence of wetland environments, sacrifice and infanticide suggests a strong degree of liminality relating to this particular bog deposit. Whilst stillborn and unbaptised infants may have been considered liminal within the broader context of Church and society in rural post-Medieval Ireland, Eileen Murphy considers the emotional and psychological impact of child loss. Through placing cillini (infant burial grounds) within the broader context of verbal and written testimonies of those who had buried an infant Murphy argues that it is possible to gain insights into parental attitudes towards these dead infants. Another feature of infant burials identified within these proceedings is the wide range of burial treatments afforded to infants. Sally Crawford provides an overview of infant burial in pre- and postConversion Anglo-Saxon England and identifies the diverse range of normative and non-normative burial treatment afforded to infants. Crawford argues the Christian rite of baptism, whilst affording a new burial treatment, did not radically alter the Anglo-Saxon sense that infant death was different from that of older children and adults. In her examination of infant and child burial from Bronze Age contexts in south-east England, Dawn McClaren suggests that the inclusion of certain types of material culture in infant burials and the care shown in burial of these infants suggests that the very young were recognised and treated as individuals. However, in certain societies infants and young children were afforded the same burial rite as adults. Marion Page suggests that the apparent absence of infants in the cemeteries of early Medieval Wales may relate to taphonomy rather than cultural practice. Page argues that grave cuts of less than one ii

metre suggests the presence of infants within formal burial contexts. The inclusion of children of all ages, including pre-term infants, was a feature of burial in Roman Egypt explored by Sandra Wheeler et al., who utilise a range of bioarchaeological techniques, including isotopic analysis, to explore the health status, weaning patterns, diet and environmental adaptability of these infants. Trying to understand the lived experience of infants and children in the past was also an aim of the conference and four papers address different aspects of the lived experience. Oliver Harris adopts philosophical approaches and anthropological evidence to understand how children may have engaged with the world at the Neolithic site at Windmill Hill, Wiltshire, UK. Traci Ardren explores the transmission of state cultural values and ideals in the experience of childhood in ancient Mesoamerica through the political power welded by young rulers and the state-sanctioned religious sacrifice of ‘powerful’ infants. Cynthia Smith Bradley examines the lived experience of a group of, mainly female, young adolescents from atypical mortuary contexts at Sand Canyon Pueblo, south-western United States. Skeletal analyses of the remains suggest that the good level of nutrition and high percentage of peri-mortem injuries may indicate that these high-status young adolescents may have been especially targeted during an episode of ideologically based factionalism with a neighbouring village. The experience of lived childhood is explored through lost ‘milk’ teeth found at the late Medieval sites of Tibrandsholm and Västerhus, Sweden by Kellstrom et al. Through relating the bio-archaeological material to ethnographic evidence, they open a window onto patterns of weaning, diet and folk belief between the two sites. Mike Lally and Alison Moore

iii

iv

The Osteology of Infancy and Childhood: misconceptions and potential Mary Lewis, University of Reading

Introduction

Defining childhood

This paper reviews the current status children's skeletal remains in bioarchaeology. Child skeletons provide a wealth of information on their physical and social life from their growth and development, diet and age at death to the social and economic factors that expose them to trauma and disease at different stages of their brief lives. Cultural attitudes dictate where and how infants and children are buried, when they assume their gender identity, if they are exposed to physical abuse and at what age they are considered adults. The study of children and childhood in social archaeology emerged out of gender theory in the 1990s, and has gradually increased in its sophistication, moving children out of the realm of woman's work, to participating and active agents in the past, with their own social identity, material culture and influence on the physical environment around them. Children who were once invisible in the archaeological record are slowly coming into view. The primary data for the archaeology of childhood are the children themselves, and in order to progress this new discipline, it is important to examine how bioarchaeologists derive the data from which social interpretations are made and the limitations that are inherent in the methods and nature of immature skeletal material, including the impact of the burial environment on their recovery.

No matter what period we are examining, childhood is more than a biological age. It is a series of social and cultural events and experiences that make up a child’s life, where children learn about their society, gender roles and labour through play. Term Embryo Fetus Stillbirth Perinatal, perinate Neonatal, neonate Post-neonatal Infant Non-adult Child Adolescent Adult

Period First 8 weeks of intra-uterine life From 8 weeks of intra-uterine life to birth Infant born dead after 28 weeks gestation Around birth, from 24 weeks gestation to 7 postnatal days Birth to 27 postnatal days 28-346 postnatal days (1 years) Birth to 1 year 17 years

Table 1: Age terminology used to describe non-adult samples The initial dependence on their parents for nourishment and protection slowly diminishes as the child ages and becomes an independent member of society. The time at which these transitions take place varies from one culture to another, and has a bearing on the level of interaction a child has with their environment, their exposure to disease and trauma and their contribution to the economic status of their family and society. What is clear is that we cannot simply transpose our view of childhood directly onto the past.

Comparative studies of children from archaeological contexts have been complicated by the eclectic use of terminology that both describes the skeleton as a child and prescribes an age for the individual. One of the most popular terms used by osteologists to describe children is 'sub-adult'. This term is problematic as it has been used to define a specific age category within the childhood period. More fundamentally, sub-adult implies that the study of these remains is somehow less important than that of the adults (i.e. sub = below). Throughout this paper children are described as 'non-adults' encompassing all children recovered from the archaeological record up to the age of 17 years. Although the term ‘non-adult’ is also not without its critics (see Halcrow and Tayles, 2011), it has never been used to define a particular age category within the overall child group. Additional terms divide this overarching category into critical physiological periods of the child's life (Table 1) and take their lead from the work of Scheuer and Black (2000). It should be stressed, that these terms are used for ease of reference and provide a biological benchmark for sample comparison and discussion; they are not intended to describe the complex social experience of the youngest members of every society, past or present.

Bogin (1997; 1998) takes an evolutionary approach to childhood theory. Childhood, is a period in the human life-cycle not found in any other mammal, and for Bogin (1997:64) this is defined as a period of time between 3-7 years, when "the youngster is weaned from nursing but still depends on older people for feeding and protection". The child is constrained by their immature dentition, small digestive system and calorie-demanding brain, which influence the type and amounts of food they can consume. ‘Juvenility’ occurs with the eruption of the permanent dentition, and when children are able to procure and consume their own foods, as the brain and

1

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD

Birth to 8 months

8 months to 1.5 years

1.5 to 3 years

Lifts and holds up head

Begins to crawl and may stand aided by furniture Can throw without losing balance

Stands on one foot or on tip-toe Can run, skip, climb and ride a bike

Turns over unaided (7 months) Reaches towards objects

Smiling and gazing

Handles finger-foods Uses spoons and cups Becomes anxious when separated from loved ones

Dresses and undresses

Imitates others Understands people and objects still exist when they cannot be seen

Shows affection by kissing and hugging

Expresses pride, pleasure, embarrassment and shame Listens to stories Understands the future and the past

Responds to name Explores environment

Gurgles and babbles to communicate

5 years

Interacts with other children Forms simple sentences

Uses sentences to communicate feelings and needs

Has no understanding of 'male' and 'female'

Understands 'male' and 'female' through dress and over time, but not changing situations

Tells stories

Social interaction and roleplaying Asks questions about the meaning of words Understands 'male' and 'female' through time and situations: "gender consistency"

Table 2: Child development milestones from ages 1 to 5 years, Collated from Berhrman et al. (1996) and Kohlberg (1966) body growth diminished to less than 50% of total energy needs, and they undergo a cognitive shift. This period begins at the age of seven and ends with the onset of puberty (ca.10 years in girls, ca.12 years in boys). Bogin (1998) asserts that the childhood period allows a mother to wean her child and produce other offspring, by passing the energy expenditure of feeding and caring for the child onto post-reproductive members of society, such as grandparents (Key, 2000). As the body and brain slow in their growth during this period, they require less energy expenditure to feed but are protected during times of hardship (Bogin, 1998). Many would object to this purely biological view of childhood, as it ignores social theories of when a child becomes an ‘adult’ and a fully-fledged member of a society, something that is culturally defined. Hanawalt (2002) argues that in order for a child to survive, they must not only be nursed, fed and kept warm (biological survival), but also be played with and talked to (cultural survival).

but in order to examine the past life-course we need to have consistency in the raw data (the skeletal remains), and use accurate osteological assessments of age and physiological development as a marker from which to base our interpretations of the social understanding of age in the past. Contrary to the views of scholars such as Lally and Arden (2008) biological age is entirely relevant to the way in which society treats a child. It affects the child’s connection to their physical and social environment, from total dependency during infancy, to when they begin to crawl, walk, talk and communicate with the adults and children around them (Table 2). These abilities are physiologically determined and they dictate how the child interacts. In particular, the misuse of the term ‘infant’ to refer to children between the ages of 1-3 years or 1-5 years in studies that use skeletal evidence as their data, misses this point. As an infant (under one year), the child is particularly vulnerable to disease and death, and their chances of survival significantly increase after the first year. Children who die around two years of age may be reflecting inadequate weaning methods or unsanitary conditions, and those that make it to three years are talking, playing and actively mobile. By five years they are capable of contributing to the household with minor chores. To categorise this most vital developmental period into one age category ‘infant’, masks important physiological and, hence, social advancements.

Defining the child: biological versus cultural age One of the resounding issues with the definition of a ‘child’ in archaeological contexts is the use of physiological age to determine a social category (Gowland, 2001; Baxter, 2005). Physiological age is a biological reality, whereas ‘child’ is a culturally loaded term. The age at which an individual leaves the world of dependency, learning and play, and takes on roles of work and social responsibility is neither distinct nor universal. That there are three types of age category, ‘biological’, ‘chronological’ and ‘social’, is not denied,

Wiley and Pike (1998) suggested the use of developmental stages, rather than chronological age to devise child mortality rates to take into account the

2

THE OSTEOLOGY OF INFANCY AND CHILDHOOD: MISCONCEPTIONS AND POTENTIAL activity of the child (crawling, weaning, walking), which is often related to their cause of death through exposure to disease and accidental injury. Although they propose this method for use in modern communities where calendar age is rarely recorded, the application of such developmental age categories into archaeological studies has the advantage of placing the child at the centre of the study by examining the environment from their vantage point.

began to emerge, and would dominate research in nonadult remains for the next forty years. The most prolific researcher in this area at the time was Francis Johnston, who examined the growth of children from Indian Knoll in Kentucky (Johnston and Snow, 1961; Johnston, 1962; 1968). Following Johnston’s example, by the late 1970s the majority of studies that included child remains were focussed on diaphyseal length measurements to estimate growth attainment (Armelagos et al., 1972; Merchant and Ubelaker, 1977; Y'Edynak, 1976). With the increasing interest in palaeodemography, researchers began to assess the impact of under-representation of child remains on life tables (Moore et al., 1975), but only a few were interested in what this data could contribute to our understanding of perinatal and child mortality (Brothwell, 1971; Henneberg, 1977; Mulinski, 1976). In 1978, Mensforth and colleagues heralded a way forward when they examined the prevalence of anaemia and infection (i.e. porotic hyperostosis, periostitis and endocranial lesions) in 452 infants and children from the Late Woodland ossuary sample from the Libben site in Ottawa County, Ohio. For the first time, the health of children in the past was the primary focus of study. This research also demonstrated the importance of healed and active lesions in determining the precise age at which children were most at risk; the kind of detail not available when using adult evidence.

Kamp (2001) provides an excellent review of the development of childhood studies and argues that the age categories employed by osteologists are often selected and compared without reference to the society in which the children lived. In the later medieval period, the ages of 8-12 years represented a time children would begin their apprenticeships (Cunningham, 1995), children as young as 12 and 14 years could be married in ancient Egypt and Rome respectively, leaving the realm of child for that of wife and mother. Childbirth is not a common interpretation for the cause of death for older children within the burial record. Today in the UK, a child reaches adulthood by degrees. At 16 they can legally have sex, at 17 they can learn to drive, at 18 they can drink, get married, and vote, reflecting their status as full members of society. Attempts to define periods of transition in childhood in the past have been carried out by examining the burial of children and the engendered nature of their grave goods at certain ages (Rega, 1997; Gowland 2001; 2002; Stoodley 2000). While the study of childhood has come some way in elucidating a particular section of the human life-course, Gilchrist (2004) calls attention to the fact that other age categories are still neglected, among them, what it was to be an adolescent in the past.

In 1980, Buikstra and Cook summed-up child studies as being hindered by poor preservation, lack of recovery and small sample sizes, despite, they argued, many researchers becoming aware of their importance in determining the overall success of a population. Instead, there was a proliferation of papers on the lack of preservation of non-adult remains compared to adults (Gordon and Buikstra, 1981; Von Endt and Ortner, 1984; Walker et al., 1988), an assumption that still prevails today. The prevalence of stress indicators in children became more popular as researchers began to assess the impact of agriculture, colonisation and urbanisation on child health (Blakey and Armelagos, 1985; Storey, 1986; 1988; Owsley and Jantz, 1985). Goodman and Armelagos (1989) highlighted the importance of children under five as the most sensitive members of society to environmental and cultural insults, whose stress experience would impact on the overall population’s ability to rally from disease in adulthood.

Children in bioarchaeology The study of children in biological anthropology has earlier beginnings than in social archaeology, but they were no less focussed. Most studies were stimulated by an interest in fertility levels, or the information that child survival could provide on adult adaptation to their changing surroundings. These endeavours were constantly being frustrated by the perceived notion that infant and child remains could not survive the burial environment (Lewis, 2007). It was only in the 1990s that the study of non-adult skeletons began to concentrate on the information that could be provided on the growth and health of the child itself, providing information on their activities and risk of infection or injury in contrasting environments. Examination of the physical remains of children provides us with the most direct and intimate evidence for them in the past. This section outlines the development of the study of child skeletal remains in biological anthropology and palaeopathology up until the present day.

By the 1990s, non-adults were becoming incorporated into biocultural studies of different populations (e.g. Grauer 1993; Stuart-Macadam, 1991; Higgins and Sirianni,1995; Ribot and Roberts, 1996). These studies were encouraged by the increase in non-adult material available. Children of known-age and sex from Christ Church Spitalfields and St Bride’s Church in London became accessible (Molleson and Cox, 1993), and data began to be published on non-adults from Wharram Percy (Mays, 1995) and St Thomas’ Church in Belleville, Ontario (Saunders et al., 1993; 1995). These samples encouraged a revival of methods to estimate the sex of non-adults (De Vito and Saunders, 1990; Loth and Henneberg, 1996; Mittler and Sheridan, 1992; Molleson

Before the 1980s, studies of non-adult skeletal remains concentrated on devising ageing and sexing methods based on medical studies (e.g. Schour and Massler, 1941; Hunt and Gleiser, 1955; Boucher, 1955; 1957). In the 1960s studies on the physical growth of past populations 3

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD et al., 1998; Schutkowski, 1993). Saunders (1992) carried out a review of non-adult growth studies, outlining their advantages and limitations, particularly the issue of comparing deceased children to living healthy populations (Saunders and Hoppa, 1993). While others began to highlight the potential and extent of pathological evidence that could be derived from their study (Anderson and Carter, 1994; 1995; Lewis and Roberts, 1997). Nearly ten years after Stuart-Macadam (1988) had raised the issue, Ortner and colleagues began to address the diagnosis of rickets and scurvy (Ortner et al., 1999; Ortner and Ericksen, 1997; Ortner and Mays, 1998). By the end of the decade, dental microstructure was being used to refine ageing techniques (Huda and Bowman, 1995) and stable isotope analysis to address the age of weaning in contrasting past populations throughout the world were beginning to dominate the literature (Herring et al., 1998; Katzenberg and Pfeiffer, 1995; Wright, 1988; Dupras and Tocheri, 2007).

morphological traits indicative of sex in the non-adult skeleton (Saunders, 2000). However, the application of ancient DNA analysis in determining the sex of non-adult skeletal material holds promise for the future. The second half of this paper outlines the potential of non-adult skeletal remains to explore various episodes within a child's life as well as some of the difficulties this analysis entails. Preservation: myth or reality? One of the most commonly cited limitations in the study of children from archaeological contexts is their poor, or total lack of preservation. The absence of neonatal remains or ‘infant-sized pits’ no longer containing their ‘tiny occupants’ (Watts, 1989:377), are often cited in archaeological reports as evidence that these babies do not survive the burial environment. The high organic and low inorganic content of children’s bones (Guy et al., 1997), in addition to their small size, certainly means they are susceptible to taphonomic processes. However, in some cases non-adult remains have been shown to be better preserved than adults (Katzenberg, 2000), and in general, where adult preservation is good, children will also be well-preserved. As testament to this, large numbers of non-adult remains continue to be housed in museums and universities ready for study. In the UK, the largest include the multi-period site of St Peter’s Church, Barton-on-Humber (n=ca.1000), Poundbury Camp (n=395) and St Mary’s Spital in London, (ca.=1740).

Today, studies of infant and child skeletal remains are receiving much more attention (Lewis, 2007). The publication of texts on non-adult osteology has increased the number of researchers familiar with their identification and anatomy (Scheuer and Black, 2000; 2004; Schaefer et al., 2009; Baker et al., 2005). Children are routinely included in wide-ranging studies of health in the past (Cook and Powell, 2005; Steckel and Rose, 2002), while the analysis of children themselves from sites all over the world continues (Baker and Wright, 1999; Bennike et al., 2005; Blom et al., 2005; Buckley, 2000; Lewis, 2002a). As we refine our ageing techniques and statistical methods (Gowland and Chamberlain, 2002) our understanding of the importance of childhood diseases and their diagnosis is becoming more advanced and widely publicised (Lewis, 2002b; 2004; 2010a; Piontek and Kozlowski, 2002; Santos and Roberts, 2001; Ortner, et al., 2001; Glencross and Stuart-Macadam, 2000; Brickley and Ives, 2008). New understanding of trauma in the child has meant we can now reassess the evidence for physical abuse and occupational injury, to gain a fuller understanding of the child’s life experience in past society.

There are many factors that account for the often small numbers of infants and children from archaeological sites, both during and after excavation. These factors include the common confusion of fetal remains with animal bones (Wicker, 1998), their presence in the often neglected unstratified skeletal material (Buckberry, 2005), and even the wilful disposal of child remains (Robbins, 1977). In Scandinavia, before the 1980s, it was the practice of museums to only keep intact skulls and, as children’s skulls are made up of several fragile bones, they were generally discarded (Sellevold, 1997). The absence of infant remains from cemetery sites at different periods is probably revealing more about their status within the society, rather than their ability to 'dissolve in the ground'. Most striking in the UK is the frequent absence of these tiny remains in Anglo-Saxon cemeteries (Lucy, 1994). When children are included in the communal cemeteries, they are often found to cluster, but these clusters are not universally defined or easy to predict. For example, at St Andrews Fishergate in York, 76% of the under fives were found in the western third of the cemetery (Stroud and Kemp, 1993), while excavations at Raunds Furnells, Jarrow, Whithorn and Winchester have revealed groups of infants along the east side of the church (Crawford, 1993). At Taunton, 85% of the non-adults were located at the western end of the cemetery, while only three children were excavated from the east (Rogers, 1984). During historic periods in Ireland children were buried in separate and marginal cemeteries or ‘cílliní’ (Finlay, 2000).

In biological anthropology, we still wrestle with the issue of children in the archaeological sample representing the ‘non-survivors’ from any given population. Their pattern of growth or frequency of lesions might not reflect that of the children that went on to survive into adulthood (Saunders and Hoppa, 1993; Wood et al., 1992). The early death of these individuals provides other challenges in the study of non-adult palaeopathology. Chronic diseases need time to develop on the skeleton, but the children that enter the archaeological record have usually died in the acute stages of disease before the skeleton has had time to respond (Lewis, 2000). At the present time, studies that concentrate on non-adult material are hindered by the inability to make reliable sex estimations, due to absence of the secondary sexual characteristics evident on the adult skull and pelvis. Although sexual dimorphism has been identified in utero, there is still a disagreement about the validity of identifying 4

THE OSTEOLOGY OF INFANCY AND CHILDHOOD: MISCONCEPTIONS AND POTENTIAL Schofield and Wrigley (1979) reported that in modern Western populations the mortality rates of people under 10 years was only 2.4%, compared to 34% in preindustrialised countries. It is now common to cite this modern pre-industrial figure as the norm for child mortality in archaeological populations from many periods all over the world. That 30% of the sample should contain non-adults has become the gold standard by which under-representation is measured (Weiss, 1973). In many cemetery samples, the proportion of non-adults frequently fails to reach this target. Modern pre-industrial societies are not directly analogous with those in the past, and they are not untouched by the pressures of industrialisation that surrounds them. Factors that will affect infant and child survival, such as their weaning diet, exposure to pollution and disease, living conditions and economic status, are often dictated by the wealthier industrial countries. Therefore, it is open to debate whether 30% child mortality is the correct minimum we should expect from an archaeological sample where all the children are included.

1985). In children, most methods fail to yield an accuracy of 70%, with the degree of overlap between male and female traits. In the late 1990s there was an increase in the use of DNA sex typing for archaeological child remains. Many of these studies focussed on the sex of infants thought to be the victims of infanticide (Faerman et al., 1997; 1998; Mays and Faerman, 2001), although some studies have used DNA analyses to sex older children. In 2000, Cunha and colleagues attempted to sex the remains of two 11 year old children recovered from a Medieval convent in Portugal, although not successful for one of the children, DNA results for the second child suggested it was a boy, raising questions about the true exclusivity of the convent, where male burials were documented to be prohibited. The differential amplification of the X-specific and Y-specific sequences in ancient bone can mean males will be wrongly identified as female if the X-specific sequence is preferentially amplified over the shorter, more easily degraded Y-chromosome, and caution in the type of technique used is essential e.g. employing the Dot Blot test to confirm results (Colson et al., 1997; Saunders and Yang, 1999). In addition, the choice of bone sample can influence the success of the analysis with Faerman and colleagues (1995) reporting that DNA was better preserved in the teeth, cranial bones and cortices of long bones, than in the ribs. As yet, DNA has not been amplified for known male and female non-adult skeletal series with the expense of testing samples hindering any large scale studies, while we still struggle with problems of contamination, consistency, control, false negatives and false positives.

Tom, Dick or Harriet..? Sexing non-adult remains Sex is generally determined in the adult skeleton using morphological features of the skull and pelvis, and measurement of the long bones. In bioarchaeology we determine biological sex. This term is not to be confused with gender, a cultural construct that refers to the social significance placed upon males and females in society (Mays and Cox, 2000). As with biological age, biological sex is the reference point from which inferences about gender in past society can be made. Sex estimation is the ‘holy grail’ of non-adult osteology. As females grow and mature faster than males, accurate sex estimation of nonadults would allow for age estimates to be refined and the social treatment of males and females at birth and during childhood to be examined. Sex identification of fetuses would be used to answer questions about changing sex ratios and their predisposing factors (e.g. parental investment and economic status (Koziel and Ulijaszek, 2001), parental investment, and the fragility of male foetuses. The proportion of male to female children displaying various degrees of trauma and pathology could tell us much about their preferential treatment and lifecourse, their susceptibility to disease, handling in society through burial rites and grave inclusions, division of labour and, at what age boys and girls became accepted as adult members of a society.

Infanticide The study of infant mortality in past populations can provide us with valuable information on the success of a population to adapt to its environment and protect the most vulnerable members of its society. Estimates of neonatal versus post-neonatal mortality may help us untangle some of the causes of these deaths, although we can age our infants less accurately than in clinical studies. Perhaps the most widely explored and controversial aspect of mortality studies in past populations is the issue of infanticide. Previously, clusters of infants in noncemetery sites or hidden under the floors of houses, in addition to their complete absence from some burial grounds or, unusual male to female ratios in the adult sample was enough to convince archaeologists that infanticide was being practiced (Watts, 1989; Mays, 1993). In particular, it was commonly assumed that female infants were the victims. With the advent of DNA these preconceptions are being challenged. Smith and Kahila (1992) reported on 100 neonates from a late Roman-early Byzantine sewer in Ashkelon, Israel. DNA results revealed that of the 19 specimens that could be tested, 14 were male and only five were female. It has been suggested that perhaps illegitimate females were reared to continue in the profession of their mothers, as prostitutes of the bathhouse (Faerman et al., 1997; 1998). In 1999, Waldron and colleagues also carried out DNA analysis on infants from the Romano-British villas of

Unfortunately, sex estimations from non-adult skeletons are notoriously difficult, making associated anthropological techniques unreliable or forcing us to add greater error ranges to our results in order to account for unknown sex. Most anthropological techniques used to assign sex to non-adults focus on the size and morphological traits of the dentition (De Vito and Saunders, 1990), cranium (Molleson et al. 1998), mandible (Loth and Henneberg, 2001) and pelvis (Rissech et al., 2003). When all cranial and postcranial traits are used, sex determination in adult skeletons can provide an accuracy of 98-100% (Buikstra and Mielke, 5

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Beddingham and Bignor, once again, the remains were not exclusively female. In the light of these studies, Mays and Faerman (2001) tested the theory that females were predominantly killed in favour of males at several Romano-British sites, and again, of the 13 cases in which sex identification was successful, nine were males and four were females. These studies all suggest that, if these neonates were victims of infanticide, male preference was not practiced in Israel or in Britain in the Roman period. As male fetuses and infants are more susceptible to mortality and disease than their female peers (Stevenson et al., 2000), the higher ratios of males within an infant sample would be expected at these sites if they, in fact, represented cases of stillbirths or neonatal deaths from natural causes, due to male vulnerability at this age (Lewis, 2007).

(vitamin D deficiency) and infantile scurvy (vitamin C deficiency). The frequency and age at which these specific and non-specific metabolic lesions occur is often argued to coincide with weaning. How and why weaning ages differed in past populations is an absorbing area of study showing weaning ages to have declined from ca. 4 years in hunter-gatherer populations (Dettwyler, 1995) to just 7 months in the eighteenth century (Fildes, 1986). Measurements of nutritional stress in non-adult remains provide direct evidence on their state of health, and avoid the problems of bone remodelling and dental wear that may eliminate such signs in the adults. Advances in stable isotope analysis has begun a new era in the examination of weaning stress in past populations (Katzenberg, 2000), and it is now possible to not only identify more precisely the age at which weaning occurred, but also what type of diet the child was weaned onto. Studies of past populations have revealed a variety of practices. For example, at Wharram Percy in later medieval Yorkshire, weaning is reported to have occurred at around two years (Fuller et al., 2003; Richards et al., 2002), with children up to the age of 11 years having a diet lower in animal protein than the adults. Osteological evidence from the site indicates that enamel hypoplasia formation peaked at 1.5 years, but that child mortality, cribra orbitalia and non-specific infections peaked much later, indicating that weaning was successful in maintaining a healthy child (Lewis, 2002a). Katzenberg and Pfeiffer (1995) showed abrupt weaning at 1.5 years in children from the Methodist cemetery at Newmarket, Ontario – an age consistent with the documentary sources. Other sites have shown a more gradual introduction of weaning foods (White and Schwartz, 1994; Schurr, 1998; Schurr and Powell, 2005). Our ability to measure an increase and decline of nitrogen in the infant has shown potential for indicating children who may have died at certain points in the weaning process (Williams et al., 2002). Improvements in our non-adult ageing techniques, microscopic imaging of enamel defects and refinement of diagnostic criteria for rickets and scurvy (Ortner and Mays, 1998; Ortner et al., 1999) have enabled us to provide a much more detailed picture of the interaction between childhood nutrition, infection, mortality and environmental change in past societies (for a recent example from the Romano-British period see Lewis, 2010b).

Criticism of the anthropological evidence for infanticide in archaeological populations does not rest with sex identification and small sample sizes. Recent studies have begun to question the use of regression equations to derive age from long bone length. A number of authors have demonstrated that ageing techniques based on regression equations are statistically biased towards producing an age structure akin to that of their reference population, in the case of neonatal age estimations, usually infants who died shortly after birth (Konigsberg et al., 1997; Lucy et al., 1996). Numerous studies have since attempted to rectify this problem using Bayesian statistics. Bayesian data analysis allows us to make inferences from data using probability models for observable quantities (known-age data) and for quantities that are unknown (archaeological data), but that we wish to learn about (Gelman et al., 1995). In their reassessment of infanticide in Romano-British samples using Bayesian statistics, Gowland and Chamberlain (2002) demonstrated that infant mortality profiles actually fell within the parameters expected if fetal stillbirths and neonatal deaths were included within the burial sample. A review of infant mortality in four English medieval sites demonstrated that the use of regression equations clustered the age at death of the neonates around 38 weeks. This profile may have been interpreted as a sign of infanticide, whereas Baysian statistics applied to the same data provided a more flattened profile for infant deaths (Lewis and Gowland, 2007). At present, there is no unequivocal evidence for infanticide in British archaeological material.

Suffer the children…? Identifying child abuse Little waifs: malnutrition and disease Limited evidence of child abuse in the archaeological record, from both Europe and North America, has led many anthropologists to suggest that child abuse, in its current form, is a recent phenomenon. To date there are four cases in the archaeological literature where child abuse has been suggested (Blondiaux et al., 2002; Wheeler et al. 2007), including a 1.5 year old child from St Oswald’s Priory, Gloucester with rickets and an unhealed arm fracture (Lewis, np) and a ‘bucket-handle’ fracture to the ankle of a toddler at Poundbury Camp (Lewis, 2010b), but these cases are inconclusive. Documentary evidence would suggest otherwise. For

Since the early 1980s, combinations of osteological indicators have been used to provide indirect evidence for malnutrition in past populations. Non-specific metabolic stress is evident with the presence of dental enamel hypoplasia, radiographic Harris lines in the long bones, and porotic lesions of the skull, known as cribra orbitalia and porotic hyperostosis. These lesions can only form in the developing child, and with the exception of dental enamel hypoplasias, will remodel if the child recovers. Less commonly reported are the pathological lesions associated with specific malnutrition, including rickets 6

THE OSTEOLOGY OF INFANCY AND CHILDHOOD: MISCONCEPTIONS AND POTENTIAL example, Stone (1977) makes reference to a plea from medieval physicians against the common practice of vigorously rocking children to make them sleep when in fact they rendered the child unconscious, and in the Anglo-Saxon period there is reference to mistreatment of deaf and mute children (Kuefler, 1991) . The lack of evidence in the skeletal remains is more likely the result of our limited ability to identify trauma in non-adult skeletons, as a result of the plastic nature of their bones and non-specific nature of the lesions (Lewis, forthcoming). Today, child abuse results from a complex combination of personal, social and cultural factors, such as intergenerational transmission of violence, social stress and isolation, large family size, alcohol and drug abuse, poverty and mental illness. Urbanisation and industrialisation in the medieval and post-medieval periods resulted in displacement of families into cities, and a greater social divide between the poor and the wealthy would all have provided the circumstances in which abuse could occur, with tired and fractious parents returning from work in dirty and often lethal factories. The fact that gin drinking in the lower classes was a serious social issue in the eighteenth century is well documented (Abel, 2001). However, most post-medieval excavations are of the wealthy individuals who could afford to inter their children into family vaults, and who may have paid other people to care for their children on a daily basis. Up until the late eighteenth century, physical punishment of children was considered the norm, to the extent that ‘heaven’ was described to children as ‘a place where children are never beaten’ (Newman, 2004). In many societies children were kept away from their parents (e.g. Greco-Roman), but it is when they are living in closed contact with them that problems arise.

bones, and of course, the death of the child. Many victims of abuse will survive and injuries will be unlikely to survive in the adult skeleton (unless there is severe damage to the epiphyses). The radiographic features used by pathologists to diagnose lesions on modern cases are also invisible to us; hairline fractures and metaphyseal ‘chips’ of bone are unlikely to be recognised in archaeological material. We do have the opportunity to record the more subtle changes invisible to radiographers, such as periosteal new bone formation indicating traumatic periosteal ripping and ‘bone bruises’. With increasing interest in the diagnosis of trauma in non-adult remains, social questions about the treatment of children in different societies of the past will begin to open up to us. Conclusions The health and survival of the children from any given community is believed to represent the most demographically variable and sensitive index of bio cultural change. Patterns of infant and child mortality have been shown to have a profound effect on the crude death rates of a population and, when coupled with evidence of childhood morbidity, have become accepted as a measure of population fitness. Roth (1992) described childhood as the most sensitive portion of the human life cycle, but it has taken us a long time to fully realise this potential. Only in the last decade have studies into the growth, development and health of non-adults been used to provide insights into the impact of the physical and cultural environment on the children themselves. The osteological study of non-adults occurred independently and at a much earlier period than theoretical concepts of childhood in social archaeology, but both have now reached a level of sophistication that should encourage communication and integration of the disciplines. Only then can we develop a fuller understanding of what it was like to be a child in the past.

So why, if it is likely that child abuse took place in the past, and it seems from the literature that it did, do we not find any clear evidence of abuse in the skeletal record? There have been several explanations for this discrepancy. Waldron (2000) argues that children who died from this form of abuse may have been buried in clandestine burials, rather than in the general cemetery. Walker (1997) suggests that the nature of physical abuse in the past was different than that reported in the forensic literature today, and that it was older children, working as apprentices and within factories, who may have been victims of abuse, rather than newborn children. It is possible that the practice of tightly swaddling children in the seventeenth and eighteenth centuries (Stone, 1977) may alter the pattern of lesions we would expect to see, making head trauma more likely and the characteristic postcranial fractures less common (Knight, 1986). De Mause (1974) refers to swaddled children being tossed between windows as a form of adult amusement, causing ribs to be crushed inwards as they were caught. Today, the relatively small number of children suffering physical abuse has an impact on the number of cases we would realistically expect in the archaeological record. Waldron (2000) suggests we would need to examine at least 2000 child remains if we are to pick up just one who may have suffered abuse. Diagnosis also relies on the complete preservation of the skeleton, including the ribs and cranial

References Abel, E. 2001. Gin Lane: did Hogarth know about fetal alcohol syndrome? Alcohol and Alcoholism 36, 131-34. Anderson, T. and Carter, A. 1994. Periosteal reaction in a new born child from Sheppey, Kent. International Journal of Osteoarchaeology 4, 47-48. Anderson, T. and Carter, A. 1995. An unusual osteitic reaction in a young medieval child. International Journal of Osteoarchaeology 5, 192-95. Armelagos, G., Mielke, J., Owen, K., Van Gerven, D., Dewey, J. and Mahler, P. 1972. Bone growth and development in prehistoric populations from Sudanese Nubia. Journal of Human Evolution 1, 89-119. Baker, B., Dupras, T. and Tocheris, M. 2005. The Osteology of Infants and Children. Texas: Texas A.M. University Press.

7

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Baker, J. and Wright, L. 1999. Introduction - childhood nutrition and health in prehistory. American Journal of Physical Anthropology Suppl. 28, 86-7.

Buckley, H. 2000. Subadult health and disease in prehistoric Tonga, Polynesia. American Journal of Physical Anthropology 113, 481-505.

Baxter, J. 2005. The Archaeology of Childhood: children, gender and material culture. Walnut Creek: Altamira Press.

Buikstra, J. and Cook, D. 1980. Palaeopathology: an American account. Annual Review of Anthropology 9, 433-70.

Bennike, P., Lewis, M., Schutkowski, H. and Valentin, F. 2005. Comparisons of child morbidity in two contrasting medieval cemeteries from Denmark. American Journal of Physical Anthropology 127, currently available on-line.

Buikstra, J. and Mielke, J. 1985. Demography, diet, and health. In R. Gilbert and J. Mielke (eds), Analysis of Prehistoric Diets, 359-422. London: Academic Press. Colson, I., Richards, M., Bailey, J., Sykes, B. and Hedges, R. 1997. DNA analysis of seven human skeletons excavated from the Terp of Wijnaldum. Journal of Archaeological Sciences 24, 911-917.

Berhrman, R., Kliegman R., and Arvin A, eds. (1996) Nelson Textbook of Pediatrics. 15th Edition. Philadelphia: WB Saunders.

Cook, D. and Powell, M. (eds.) 2005. The Myth of Syphilis. The Natural History of Treponematosis in North Americal. Gainsville: University of Florida Press.

Blakey, M. and Armelagos, G. 1985. Deciduous enamel defects in prehistoric Americans from Dickson Mounds: prenatal and postnatal stress. American Journal of Physical Anthropology 66, 371-80.

Crawford, S. 1993 Children, death and the afterlife in Anglo-Saxon England. Anglo-Saxon Studies in Archaeology and History 6, 83-91.

Blom, D., Buikstra, J., Keng, L., Tomczak, P., Shoreman, E. and Stevens-Tuttle, D. 2005. Anemia and childhood mortality: latitudinal patterning along the coast of preColumbian Peru. American Journal of Physical Anthropology 127, 152-69.

Cunha, E., Fily, M-L., Clisson. I., Santos, A., Silva, A., Umbelino, C., Cesar, P., Corte-Real, A., Crubezy, E. and Ludes, B. 2000. Children at the convent: comparing historical data, morphology and DNA extracted from ancient tissues for sex diagnosis at Santa Clara-a-Velha (Coimbra, Portugal). Journal of Archaeological Science 27:949-52.

Blondiaux, G. Blondiaux, J. Secousse, F. Cotton, A. Danze, P-M. and Flipo, R-M. 2002. Rickets and child abuse: the case of a two year old girl from the 4th century in Lisieux (Normandy). International Journal of Osteoarchaeology 12(3), 209-215.

Cunningham, H. 1995. Children and Childhood in Western Society Since 1500. London: Longman.

Bogin, B. 1997. Evolutionary hypotheses for human childhood. Yearbook of Physical Anthropology 40, 63-89.

De Mause, L. 1974. The evolution of childhood. In L. De Mause (ed), The History of Childhood. New York, 1-73. London: The Psychohistory Press.

Bogin, B. 1998. Evolutionary and biological aspects of childhood. In C. Panter-Brick (ed), Biosocial Perspectives on Children, 10-44. Cambridge: Cambridge University Press.

De Vito, C. and Saunders, S. 1990. A discriminant function analysis of deciduous teeth to determine sex. Journal of Forensic Sciences 35, 845-58.

Boucher, B. 1955. Sex differences in the foetal sciatic notch. Journal of Forensic Medicine 2, 51-4.

Dettwyler, K. 1995. A time to wean: the hominid blueprint for the natural age of weaning in modern human populations. In P. Stuart-Macadam and K. Dettwyler (eds), Breastfeeding. Biocultural Perspectives, 39-73. New York: Aldine de Gruyter.

Boucher, B. 1957. Sex differences in the foetal pelvis. American Journal of Physical Anthropology 15, 581-600. Brickley, M. & Ives, R. 2008. The Bioarchaeology of Metabolic Bone Disease. Oxford: Academic Press. Brothwell, D. 1971. Palaeodemography. In W. Brass (ed), Biological Aspects of Demography, 111-128. London: Taylor & Francis.

Dupras, T. and Tocheri, M. 2007. Reconstructing infant weaning histories at Roman Period Kellis, Egypt using stable isotope analysis of dentition. American Journal of Physical Anthropology 134(1), 63-74.

Buckberry, J. 2005. Where have all the children gone? The preservation of infant and children's remains in the archaeological record. Paper presented at The Archaeology of Infancy and Childhood conference, 6-8 May 2005.

Faerman, M., Filon, D., Kahila, G., Greenblatt, C., Smith, P. and Oppenheim, A. 1995. Sex identification of archaeological human remains based on amplification of the X and Y amelogenin alleles. Gene 167, 327-32.

8

THE OSTEOLOGY OF INFANCY AND CHILDHOOD: MISCONCEPTIONS AND POTENTIAL Faerman, M., Kahila, G., Smith, P., Greenblatt, C., Stager, L., Filon, D. and Oppenheim, A. 1997. DNA analysis reveals the sex of infanticide victims. Nature 385, 212-3.

Grauer, A. 1993. Patterns of anemia and infection from medieval York, England. American Journal of Physical Anthropology 91, 203-13. Guy, H., Masset, C. & Baud, C. 1997. Infant taphonomy. International Journal of Osteoarchaeology 7, 221-29.

Faerman, M., Kahila Bar-Gal, G., Filon, D., Greenblatt, C., Stager, L., Oppenheim, A. and Smith, P. 1998. Determining the sex of infanticide victims from the late Roman era through ancient DNA analysis. Journal of Archaeological Science 25, 861-65.

Halcrow, S. and Tayles, N. 2011. The bioarchaeological investigation of children and childhood. In: S Agarwal & S Glencross (eds.) Social Bioarchaeology, 333-360. Chichester: Wiley-Blackwell.

Fildes, V. 1986. Breasts, Bottles and Babies: A History of Infant Feeding. Edinburgh: Edinburgh University Press.

Hanawalt, B. 2002. Medievalists and the study of childhood. Speculum 77, 440-60.

Finlay ,N. 2000. Outside of life: traditions of infant burial in Ireland from cíllín to cist. World Archaeology 31, 40722.

Henneberg, M. 1977. Proportion of dying children in palaeodemographical studies: estimation by guess or by methodical approach. Przeglad Antropologiczny 43, 107113.

Fuller, B., Richards, M. and Mays, S. 2003. Stable carbon and nitrogen isotope variations in tooth dentine serial sections from Wharram Percy. Journal of Archaeological Science 30, 1673-84.

Herring, D., Saunders, S. and Katzenberg, M. 1998. Investigating the weaning process in past populations. American Journal of Physical Anthropology 105, 425-39.

Gelman, A., Carlin, J., Stern, H. and Rubin, D. 1995. Bayesian Data Analysis. London: Chapman & Hall. Gilchrist, R. 2004. Archaeology and the life-course: a time and age for gender. In L. Meskell and R. Preucel (eds.), A Companion to Social Archaeology, 142-60. Oxford: Blackwell Publishing.

Higgins, R. and Sirianni, J. 1995. An assessment of health and mortality of nineteenth century Rochester, New York using historic records and the Highland Park skeletal collections. In A. Grauer (ed.), Bodies of Evidence: Reconstructing History through Skeletal Analysis, 121-36. New York: Wiley-Liss, Inc.

Glencross, B. and Stuart-Macadam, P. 2000. Childhood trauma in the archaeological record. International Journal of Osteoarchaeology 10, 198-209.

Huda, T. and Bowman, J. 1995. Age determination from dental microstructure in juveniles. American Journal of Physical Anthropology 97, 135-50.

Goodman, A. and Armelagos, G. 1989. Infant and childhood morbidity and mortality risks in archaeological populations. World Archaeology 21, 225-43.

Hunt, E. and Gleiser, I. 1955. The estimation of age and sex of pre-adolescent children. American Journal of Physical Anthropology 13, 79-87.

Gordon, C. and Buikstra, J. 1981. Soil pH, bone preservation, and sampling bias at mortuary sites. American Antiquity 48, 566-71.

Johnston, F. 1962. Growth of the long bones of infants and young children at Indian Knoll. American Journal of Physical Anthropology 20, 249-54.

Gowland, R. 2001. Playing Dead: implications of mortuary evidence for the social construction of childhood in Roman Britain. In G. Davies, A. Gardner and K. Lockyear (eds.), TRAC 2000. Proceedings of the Tenth Annual Theoretical Roman Archaeology Conference, University College London April 2000, 15268. Oxford: Oxbow.

Johnston, F. 1968. Growth of the skeleton in earlier peoples. In D. Brothwell (ed), The Skeletal Biology of Past Human Populations, 57-66. London: Pergamon Press. Johnston, F. and Snow, C. 1961. The reassessment of the age and sex of the Indian Knoll skeletal population: demographic and methodological aspects. American Journal of Physical Anthropology 19, 237-44.

Gowland, R. 2002. Age as an Aspect of Social Identity in Fourth to Sixth Century A.D. England: the archaeological funerary evidence. Unpublished PhD, University of Durham, Durham.

Kamp, K. 2001. Where have all the children gone? The archaeology of childhood. Journal of Archaeological Method and Theory 8, 1-34.

Gowland, R. and Chamberlain, A. 2002. A Bayesian approach to ageing perinatal skeletal material from archaeological sites: implications for the evidence for infanticide in Roman-Britain. Journal of Archaeological Science 29, 677-85.

Katzenberg, A. 2000. Stable Isotope analysis: a tool for studying past diet, demography, and life history. In A. Katzenberg and S. Saunders (eds), Biological Anthropology of the Human Skeleton, 305-27. New York: Wiley-Liss. 9

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Katzenberg, M. and Pfeiffer, S. 1995. Nitrogen isotope evidence for weaning age in a nineteenth century Canadian skeletal sample. In A. Grauer (ed.) Bodies of Evidence: Reconstructing History Through Skeletal Analysis, 221-35. New York: Wiley-Liss.

Journal of Osteoarchaeology published on-line DOI: 10.1002/oa.1229. Lewis, M. 2010b. Life and Death in a Civitas Capital: metabolic disease and trauma in the children from late Roman Dorchester. Dorset. American Journal of Physical Anthropology 142, 405-416.

Key, C. 2000. The evolution of human life history. World Archaeology 31, 329-50.

Lewis, M. and Gowland, R. 2007. Brief and precarious lives: infant mortality in contrasting sites from medieval and post-medieval England (AD 850-1859). American Journal of Physical Anthropology 134(1), 117-129.

Knight, B. 1986. The history of child abuse. Forensic Science International 30, 135-41. Kohlberg, L. 1966. A cognitive-developmental analysis of children's sex-role concepts and attitudes. In C. Maccoby (ed.), The Development of Sex Differences, 2034. Stanford: Stanford University Press.

Lewis, M. and Roberts, C. 1997. Growing pains: the interpretation of stress indicators. International Journal of Osteoarchaeology 7, 581-86.

Konigsberg L., Frankenberg, S. and Walker, R. 1997. Regress what on what? Palaeodemographic age estimation as a calibration problem. In R. Paine (ed.): Integrating Archaeological Demography: Multidisciplinary Approaches to Prehistoric Populations, 64-88. Carbondale: Center for Archaeological Investigations, Southern Illinois University.

Loth, S and Henneberg, M. 1996. Mandibular ramus flexure: a new morphologic indicator of sexual dimorphism in the human skeleton. American Journal of Physical Anthropology 99, 473-485. Loth, S. and Henneberg, M. 2001. Sexually dimorphic mandibular morphology in the first few years of life. American Journal of Physical Anthropology 115, 179186.

Koziel, S. and Ulijaszek, S. 2001. Waiting for Travers and Willard: do the rich really favor sons? American Journal of Physical Anthropology 115, 71-9.

Lucy, D., Aykroyd, R., Pollard, A. and Solheim, T. 1996. A Bayesian approach to adult human age estimation from dental observations by Johanson's age changes. Journal of Forensic Sciences 41, 189-94.

Kuefler, M. 1991. A wryed existence-attitudes towards children in Anglo-Saxon England. Journal of Social History 32, 823-34.

Lucy, S. 1994. Children in early medieval cemeteries. In J.S. Derevenski (ed.): Perspectives on Children and Childhood, 21-34. Cambridge: Archaeological Review from Cambridge.

Lally, M. and Arden, T. 2008. Little artefacts: rethinking the constitution of the archaeological infant. Childhood in the Past 1, 62-77.

Mays, S. 1993. Infanticide in Roman Britain. Antiquity 67, 883-88.

Lewis, M. 2000. Non-adult palaeopathology: current status and future potential. In M. Cox and S. Mays (eds.), Human Osteology in Archaeology and Forensic Science, 39-57. London: Greenwich Medical Media.

Mays, S. 1995. The relationship between Harris lines and other aspects of skeletal development in adults and juveniles. Journal of Archaeological Sciences 22, 511-20.

Lewis, M. 2002a. The impact of industrialisation: comparative study of child health in four sites from medieval and post-medieval England (850-1859 AD). American Journal of Physical Anthropology 119, 211-23.

Mays, S. and Cox, M. 2000. Sex determination in skeletal remains. In M. Cox and S. Mays (eds), Human Osteology in Archaeology and Forensic Science, 117-30. London: Greenwich Medical Media.

Lewis, M. 2002b. Infant and childhood leprosy: clinical and palaeopathological implications. In C. Roberts, M. Lewis and K. Manchester (eds.): The Past and Present of Leprosy, 47-53. Oxford: BAR Publishing.

Mays, S. and Faerman, M. 2001. Sex identification in some putative infanticide victims from Roman Britain using ancient DNA. Journal of Archaeological Science 28, 555-59.

Lewis, M. 2004. Endocranial lesions in non-adult skeletons: understanding their aetiology. International Journal of Osteoarchaeology 14, 82-97.

Mensforth, R., Lovejoy, O., Lallo, J. and Armelagos, J. 1978. The role of constitutional factors, diet and infectious disease in the etiology of porotic hyperostosis and periosteal reactions in prehistoric infants and children. Medical Anthropology 2, 1-59.

Lewis, M. 2007. The Bioarchaeology of Children. Perspectives from Biological and Forensic Anthropology. Cambridge: Cambridge University Press. Lewis, M. 2010a. Thalassaemia: its diagnosis and interpretation in past skeletal populations. International 10

THE OSTEOLOGY OF INFANCY AND CHILDHOOD: MISCONCEPTIONS AND POTENTIAL Merchant, V. and Ubelaker, D. 1977. Skeletal growth of the protohistoric Arikara. American Journal of Physical Anthropology 46:61-72.

Scott (eds), Invisible People and Processes, 229-47. London: Leicester University Press. Ribot, I. & Roberts, C. 1996. A study of non-specific stress indicators and skeletal growth in two Mediaeval subadult populations. Journal of Archaeological Science 23, 67-79.

Mittler, D and Sheridan S. 1992. Sex determination in subadults using auricular surface morphology: a forensic science perspective. Journal of Forensic Sciences 37, 1068-75.

Richards, M., Mays, S. & Fuller, B. 2002. Stable carbon and nitrogen isotope values of bone and teeth reflect weaning age at the medieval Wharram Percy site, Yorkshire, UK. American Journal of Physical Anthropology 119, 205-10.

Molleson, T and Cox, M. 1993. The Spitalfields Project, Volume II - The Middling Sort. York: Council for British Archaeology Research Report 86. Molleson, T., Cruse, K. & Mays, S. 1998. Some sexually dimorphic features of the human juvenile skull and their value in sex determination in immature skeletal remains. Journal of Archaeological Science 25, 719-28.

Rissech, C., Garcia, M. & Malgosa, A. 2003. Sex and age diagnosis by ischium morphometric analysis. Forensic Science International 135, 188-96. Robbins, L. 1977. The story of life revealed by the dead. In R. Blakey (ed), Biocultural Adaptation in Prehistoric America, 10-26. Athens: University of Georgia Press.

Moore, J., Swedlund, A. & Armelagos G. 1975. The use of life tables in paleodemography. In A. Swedlund (ed), Population Studies in Archaeology and Biological Anthropology. A Symposium. American Antiquity 39, 57-70.

Rogers, J. 1984. Skeletons from the lay cemetery at Taunton Priory. In P. Leach (ed), The Archaeology of Taunton Excavations and Fieldwork to 1980, 16-30. Bristol: Western Archaeological Trust Excavation Monograph 8.

Mulinski, T. 1976. The use of fetal material as a measure of stress at Grasshopper Pueblo. Paper Presented at the Society for American Archaeology, 6-8 May, St Louis.

Roth, E. 1992. Applications of demography models to paleodemography. In S. Saunders and M. Katzenberg (eds), Skeletal Biology of Past Peoples: Research Methods, 175-88. New York: Wiley-Liss.

Newman, D. 2004. Sociology: Exploring the Architecture of Everyday Life. New York: Pine Forge Press. Ortner, D. & Ericksen, M. 1997. Bone changes in the human skull probably resulting from scurvy in infancy and childhood. International Journal of Osteoarchaeology 7, 212-20.

Santos, A. & Roberts, C. 2001. A picture of tuberculosis in young Portuguese people in the early 20th century: a multidisciplinary study of the skeletal and historical evidence. American Journal of Physical Anthropology 115, 38-49.

Ortner, D. & Mays, S. 1998. Dry-bone manifestations of rickets in infancy and early childhood. International Journal of Osteoarchaeology 8, 45-55.

Saunders, S. 1992. Subadult skeletons and growth related studies. In S. Saunders and M. Katzenberg (eds), Skeletal Biology of Past Populations: Advances in Research Methods, 1-20. New York: Wiley-Liss.

Ortner, D., Kimmerle, E. & Diez, M. 1999. Probable evidence of scurvy in subadults from archaeological sites in Peru. American Journal of Physical Anthropology 108, 321-31.

Saunders, S. Herring, D. & Boyce, G. 1995. Can skeletal samples accurately represent the living populations they come from? The St. Thomas' cemetery site, Belleville, Ontario. In: A Grauer (ed). Bodies of Evidence: Reconstructing History Through Skeletal Analysis. 69-89, New York: Wiley-Liss & Sons, Inc.

Ortner, D. Butler, W. Cafarella, J. & Milligan, L. 2001. Evidence of probable scurvy in subadults from archaeological sites in North America. American Journal of Physical Anthropology 114(4), 343-351. Owsley, D. & Jantz, R. 1985. Long bone lengths and gestational age distributions of post-contact period Arikara perinatal infant skeletons. American Journal of Physical Anthropology 68, 321-8.

Saunders, S. 2000. Subadult skeletons and growth-related studies. In M. Katzenberg and S. Saunders (eds), Biological Anthropology of the Human Skeleton, 1-20. New York: Wiley-Liss.

Piontek, J. & Kozlowski, T. 2002. Frequency of cribra orbitalia in the subadult medieval population from Gruczno, Poland. International Journal of Osteoarchaeology 12, 202-208.

Saunders, S. & Hoppa, R. 1993. Growth deficit in survivors and non-survivors: biological mortality bias in subadult skeletal samples. Yearbook of Physical Anthropology 36, 127-51.

Rega, E. 1997. Age, gender and biological reality in the early Bronze Age cemetery at Morkin. In J. Moore and E. 11

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Saunders, S. & Yang, D. 1999. Sex determination: XX or XY from the human skeleton. In S. Fairgrieve (ed), Forensic Osteological Analysis: a book of case studies, 36-59. Springfield: Charles C. Thomas.

Stone, L. 1977. The Family, Sex and Marriage in England 1500-1800. London: Weidenfeld and Nicolson. Stoodley, N. 2000. From the cradle to the grave: age organisation and the early Anglo-Saxon burial rite. World Archaeology 31, 456-72.

Saunders, S., Hoppa, R. & Southern, R. 1993. Diaphyseal growth in a nineteenth century skeletal sample of subadults from St Thomas' Church, Belleville. International Journal of Osteoarchaeology 3, 265-281.

Storey, R. 1986. Perinatal mortality at pre-Columbian Teotihuacán. American Journal of Physical Anthropology 69, 541-8.

Schaefer, M., Black, S. & Scheuer J. 2009. Juvenile Osteology. London: Academic Press.

Storey, R. 1988. Prenatal enamel defects in Teotihuacán and Copan. American Journal of Physical Anthropology 75:275-6.

Scheuer, L. & Black, S. 2000. Developmental Juvenile Osteology. London: Academic Press.

Stroud, G. & Kemp, R. (eds). 1993. Cemetery of St. Andrew Fishergate. London: Current British Archaeology.

Scheuer, L. & Black, S. 2004. The Juvenile Skeleton. London: Elsevier.

Stuart-Macadam, P. 1988. Rickets as an interpretative tool. Journal of Paleopathology 2, 33-42.

Schofield, R. & Wrigley, E. 1979. Infant and child mortality in England in the late Tudor and early Stuart period. In C. Webster (ed), Health, Medicine and Mortality in the Sixteenth Century, 61-95. Cambridge: Cambridge University Press. Schour, I. Massler, M. 1941. The development of the human dentition. The Journal of the American Dental Association 28, 1153-60.

Stuart-Macadam, P. 1991. Anemia in Roman Britain: Poundbury Camp. In H. Bush and M. Zvelebil (eds), Health in Past Societies: Biocultural Interpretations of Human Skeletal Remains in Archaeological Contexts, 101-13. Oxford: British Archaeological Research International Series.

Schurr, M. 1998. Using stable nitrogen isotopes to study weaning behaviour in past populations. World Archaeology 30, 327-42.

Von Endt, D. & Ortner, D. 1984. Experimental effects of bone size and temperature on bone diagenesis. Journal of Archaeological Science 11, 247-53.

Schurr, M. Powell, M. 2005. The role of changing childhood diets in the prehistoric evolution of food production: an isotopic assessment. American Journal of Physical Anthropology 126, 278-94.

Waldron, T Taylor, G. & Rudling, D. 1999. Sexing of Romano-British baby burials from the Beddingham and Bignor villas. Sussex Archaeological Collections 137, 7179.

Schutkowski, H. 1993. Sex determination of infant and juvenile skeletons: 1. morphological features. American Journal of Physical Anthropology 90, 199-205.

Waldron, T. 2000. Hidden or overlooked? Where are the disadvantaged in the skeletal record? In J. Hurbert (ed), Madness, Disability and Social Exclusion, 29-45. London: Routledge.

Sellevold, B. 1997. Children's skeletons and graves in Scandinavian archaeology. In G. De Boe and F. Verhaeghe (eds), Death and Burial in Medieval Europe, 15-25. Brugge: IAP Rapporten 2.

Walker, P. 1997. Skeletal evidence for child abuse: A physical anthropological perspective. Journal of Forensic Sciences 42, 196-207. Walker, P., Johnson, J. & Lambert, P. 1988. Age and sex biases in the preservation of human skeletal remains. American Journal of Physical Anthropology 76, 183-8.

Smith, P. & Kahila, G. 1992. Identification of infanticide in archaeological sites: a case study from the late Romanearly Byzantine Periods at Ashkelon, Israel. Journal of Archaeological Sciences 19, 667-75.

Watts, D. 1989. Infant burials and Romano-British Christianity. Archaeological Journal 146, 372-83.

Steckel, R. & Rose, J. 2002. The Backbone of History: Health and Nutrition in the Western Hemisphere. Cambridge: Cambridge University Press.

Weiss, K. 1973. Demographic Models for Anthropology. Society for American Archaeology, Memoir 27: New York.

Stevenson, D, Verter J, Fanaroff A, Oh W, Ehrenkranz R, Shankaran S, Donovan E, Wright L, Lemons J, Tyson J, Korones S, Bauer C, Stoll B, and Papile L-A 2000. Sex differences in outcomes of very low birthweight infants: the newborn male disadvantage. Archives of Diseases in Childhood Fetal and Neonatal Edition 83,182-185.

Wheeler, S. Beauchesne, P. Wiliams, L. & Molto, J. 2007. Fractured childhood: a case of probable child abuse from the Kellis 2 cemetery, Dakhleh Oasis, Egypt.

12

THE OSTEOLOGY OF INFANCY AND CHILDHOOD: MISCONCEPTIONS AND POTENTIAL Poster presented at the 72nd Annual Meeting of the Society of American Archaeologists. Austin, Texas.

Williams, J., White, C. & Longstaffe, F. 2002. Trophic level and macronutrient shift effects associated with the weaning process in the Maya Post-classic. American Journal of Physical Anthropology 34, 165.

White, C. & Schwartz, H. 1994. Temporal trends in stable isotopes for Nubian mummy tissues. American Journal of Physical Anthropology 93, 165-87.

Wood, J. Milner, G. Harpending, H. & Weiss, K. 1992. The osteological paradox. problems of inferring prehistoric health from skeletal samples. Current Anthropology 33(4), 343-370.

Wicker, N. 1998. Selective female infanticide as partial explanation for the dearth of women in Viking Age Scandinavia. In G. Halstall (ed), Violence and Society in the Early Medieval West, 205-21. Woodbridge: Boydell Press.

Wright, L. 1998. Stable carbon and oxygen isotopes in human tooth enamel: identifying breastfeeding and weaning in prehistory. American Journal of Physical Anthropology 106, 1-18.

Wiley, A. & Pike I. 1998. An alternative method for assessing early mortality in contemporary populations. American Journal of Physical Anthropology 107, 315-30.

Y'Edynak, G. 1976. Long bone growth in Western Eskimo and Aleut skeletons. American Journal of Physical Anthropology 45, 569-74.

13

Subadult or subaltern? Children as serial categories Fredrik Fahlander Department of Archaeology and Classical Studies, Stockholm University variability in contemporary societies. Of this follows that is difficult for us to find any child or concept of childhood to define in static terms. Rather, children’s experiences are many and they are likely to have been treated very differently in diverse social and material situations. Any category of age also needs to be analysed in relation to other social variables such as sex, class and ethnicity, etc (Heywood 2001, 4f, cf. Jordanova 1990, Alexandre-Bidon and Lett 1999, 138). We cannot thus aspire to find the situation or the world-view of children in the past or present, their stories and life-courses are many and heterogeneous.

For some time now, children have been a neglected category in archaeological analysis and our narratives of the past. Despite the fact that we know that there always has been children around, they are seldom included in our social models and fictions, and the children’s overall impact on the archaeological record are often ignored. To speak with the terminology of Gyatri Spivak, we may say that children are a subaltern group of prehistory. They do not have a voice in the contemporary discourse (Fahlander 2003, 22-4). The later decades have, however, displayed a growing interest in the archaeology of children and childhood. Nowadays, one can find a great variety of approaches of e.g. how to find and visualize children, what it can mean to be a child in different times, and how their presence and agencies may have formed the archaeological record (e.g. Gräslund 1973, Lillehammer 1986; Beaumont 1994; Sofaer-Derevenski 1997; 2000b, Moore and Scott 1997; Johnsen and Welinder 1995; Andersson 1999, Scott 1999, Alt and Kemkes-Grottenthaler 2002, Gustavsson and Lundin 2004, Baxter 2005). These and many other studies have substantially contributed to a richer understanding of past societies. Still, the research on children in archaeology is hampered by the fact that the object of study is very diffuse and fluid. It has proven difficult to establish a general threshold when a child is turning adult in a social sense. Even if that was feasible, would it be unfortunate to treat all individuals under a certain age, like, for instance, puberty, as one category: there are too great differences in e.g., corporeality and social abilities among pre-pubertal individuals. To make progress, we thus need to find ways of defining and differentiate the vague category of children into more relevant social dittos. In this paper I explore Jean-Paul Sartre’s concept of serial collectivity as a means to establish a range of series of differently empowered children based on their corporeal, mental and social abilities.

So, the image of the ‘child’ is dissolving and slipping through our fingers. We cannot acknowledge social ambiguity and diversity and at the same time continue to hold on to a vague distinction between children, adults and elderly. That would simplify social complexity and variation too far. From a social point of view, however, we can approach the subject by discussing the corporeality of individual bodies rather than of social subjects. If we can discern some more or less general corporeal and mental stages in children’s development that potentially may be socially significant. From such a basis, we may continue and discuss their potential social roles and identities at a given time and place. To do this, we first need to penetrate the problematic issues of social identity and social aspects of the body. Social identity, corporeality and social categorisation The issue of social identity is perhaps one of the most complex concepts in social analysis which far transcends, or crosses, simple notions of sex, gender, status and ethnicity. Many archaeologists have pointed out the inconsistencies in viewing social subjects as homogeneous, equivalent and socially able (e.g. Berggren 2000, Gero 2000, Fahlander 2003). Instead we rather find that individuals in most societies differ in many respects when it comes to power, abilities and knowledgeability, etc. As a starting point, however, the social subject is perhaps best viewed as a corporeal being, or rather, an embodied being with body and mind in conjunction. The literature on the body and the corporeality of the body has grown vast in recent decades and embraces various strands of thought (e.g. Davis 1997; Welton 1998; Burkitt 1999, Meskell 1999, Hamilakis et al. 2002, Rautman 2002, Sweeney and Hodder 2002). Some projects depart from the phenomenology of Merleau-Ponty, stressing that all sensory input is transmitted through the physical body and hence is affected by its corporeal constitution (e.g. Bigwood 1991). Others follow strands within poststructuralism, such as Foucault’s theories of the disciplined body, the body as imprisoned by the soul, exposed to (and exposing) subjectivation and power (e.g.

Who are the children? It is nowadays generally recognized that the category of children and the concept of childhood are culturally constructed concepts (e.g. Sofaer Derevenski 2000, Baxter 2005). Indeed, the social position and status of an individual of young age are likely to vary a lot in (pre)history and there can only be few, if any, general aspects of children and childhood valid for all time periods. Nonetheless, are there many ideas of what it is to be a child, or rather, what it ought to be? In contemporary discourse we find ideas and images of certain ‘natural’ properties that children supposedly share. Children are sometimes regarded naïve and innocent, physically and mentally vulnerable, or on the other hand, evil or savage. We can, however, quite easily reject the any universal property of children only by looking at the great 14

SUBADULT OR SUBALTERN? CHILDREN AS SERIAL CATEGORIES time and place. It is perhaps obvious that our present day ideas and experiences of the different aspects of social identity and social categorisation do not need to have relevance in our analysis of the past. Many such ’natural’ aspects and categorisations have been refuted by social scientists. Haraway (1991) has pointed out that many ’objects of knowledge’, such as race, class and homosexuality, have rather short histories. Homi Bhabha (1994, 1) has, from a post-colonial perspective, stressed that the focus on sex and ethnicity are rather narrowminded and that other aspects probably are more important in certain social situations (e.g. generation, institutional affinity, geopolitical place and sexual preferences). It seems that whenever we try to define an object of study in detail, we find that any category becomes fluid and heterogeneous (Butler 1993, 168). Indeed, it would be pretty simplistic or even self-fulfilling to stick with our present-day categories in our analysis of the past. Social subjects are multifaceted and changeable actants with varying means of agency and exposed to different subjectivation processes related to performative acts and appearance. We need to recognise the fact that individual subjects are not alike, with identical properties and (dis)advantages. We also need to dismiss the idea of the ‘sameness’ of a person, as bodies go through corporeal alterations over time, through childhood, maturity, old age and death (Turner 1996, 30; Gilchrist 1999, 54-78; Stoodley 2000). Which of the varying social and corporeal attributes that lay behind social categorisation in various pasts is certainly a prime objective for social analysis in archaeology. The outline and applicability of such a corporeal approach is, however, not evident. The critical discussion about social identity and social categories leaves us rather helpless, since few authors provide any new methods or ways to discuss fluid entities such as ’children’. We can, however, perhaps somewhat surprisingly, find one promising perspective in Jean-Paul Sartre’s notion of serial collectives.

Foucault 1980; Butler 1997). In archaeology, the corporeal dimension of the body can be discussed in two general ways. One facet is that our corporeality is often related to the way we are able to act in the world. For instance, Joanna Brück (1998, 28) has argued that pregnant women or disabled individuals may (but do not necessarily) have different ways of negotiating with monuments like the Dorset cursus. In a similar way, the general abilities and means of children of different stages will, by and large, affect their agencies in other ways than adults. To acknowledge such variability in agency and ability is crucial for discussing categories of children because of their different corporeal constitution in relation to adults. The other facet of corporeal characteristics is found in their potential as imperatives for social subjectivation and categorization. The body can be seen as a material node subjected to social and ideological processes. The body is on constant display and thus works as a social actant like any other material object. Corporeal attributes are generally active social signifiers that work consciously or unconsciously in the social process of arranging and subjectivating individuals into social categories or groups. For instance, phenotypic aspects of individual corporeality such as sex, age, skin colour, etc. are (today) conventional bases for the construction of social categories and identity (Moore 1994, 13; Fahlander 2001, 78ff). We may also consider less attentive, physical differences in weight and length, as well as aesthetic aspects such as ‘beautiful’ and ‘ugly’ and normal and deviant body forms as potentially important social factors (cf. Higate 1998, 191f) We can add several other, less striking, phenotypic characteristics (hair colour, facial attributes, etc.) that may or may not be socially significant in a given case. Of course, many corporeal traits can be purposely hidden or rearranged for different purposes (e.g. bodily manipulations and mutilations). There are also corporeal aspects which are invisible ‘inside the body’ (e.g. menopause, genetic abnormalities, deafness, diseases, etc.). Such internal variations may, despite their invisibility, nevertheless be exposed by their effects or lack of effects.

Series, groups and materialities Sartre has produced some interesting thoughts on how individuals often act ‘together as solitudes’. Especially his concept of series is helpful in analysing diffuse collectives based on individual agency (1991, 256-69; 1977). Sartre has frequently, but perhaps somewhat unfairly, been labelled an idealist or subjectivist, advocating an individual-oriented perspective. This simplified view of the works of Sartre emanates from his more popular writings, but in his later works we find a more complex and interesting perspective. In his magnum opus, Critique of Dialectical Reason, (1991) he seeks to merge existential philosophy with Marxist theory. He aims to sketch an image of a social totality, the structural elements, without excluding individual intentions and perspectives. The multiplicity of sociality is not simply described as constituted by atomised and autonomous subjects, but as a totalising process, whereas the social multiplicity is interiorised. This process is not that unlike Giddens’ notion of the duality of structure or Bourdieu’s concept of habitus (Fahlander 2003). Sartre argues that a

From anthropology we have ‘learnt’ that kin, sex, age and ethnicity are the prime variables around which social identity and categorisations is based, but when discussing aspects of prehistoric social structures, we probably need to attend to a wider array of corporeal aspects than the usual. The variability, ambiguity, complexity and heterogeneity of the traditional social and biological categories have been pointed out by a great number of post-structuralists and third-wave feminists (e.g. Irigaray 1985, Butler 1993; Moore 1994). Not only gender as a concept is dissolving as a social construction without essence, but also the duality of biological sex has been questioned (e.g. Laqeur 1990; Foucault 1984, 98; Nordbladh and Yates 1990; Moore 1994, 23; Žižek 2005, 59-86). This critique is not only valid for gender and sex, but applies to age and age-related phases as well. We cannot assume that there is any quintessence to the concept of childhood, teens or old age that is valid for all 15

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD abilities and material resources, statistics remind us that women in general suffer from structural inequality in relation to the average man. Young’s solution to the problem is to see ‘women’ and the notion of ‘gender’ as series rather than as social categories. This particular type of series is not defined by biological sex or ‘femininity’, but rather by the unequal situations that many women face and by the practices that they perform in a patriarchal social order. It would, for instance, not be surprising to find many women in the queue waiting for the bus while their husbands have the family car at their disposal. In this particular case, we find a direct relationship between a superstructure (patriarchal ideology) with individual agency and materialities.

larger social collective (a practico-inert ensemble) “makes of everyone both a polyvalent isolation (with millions of facets) and an integrated member” (Sartre 1991, 257; 1977). Sartre argues that many socially constituted collectives are better seen as series (the inert effect of individual activities) than as social categories or groups. To be a member of a group, in Sartre’s terms, one has to enter consciously, as in a fraternity, and join in a common cause (Sartre uses the example of the Resistance to the German occupation, 1940-44). Most of what are normally considered social groups or categories do not qualify for this definition. Instead, many social collectives are better seen as combinations of momentary series, constituted by the common circumstances of situated individuals (1991, 258).

It is important to recognise that the series are not necessarily defined by the intentions of the subjects, but by their actions. The people in the queue have acted in different ways and of various reasons to situate themselves at that same particular point in time and space. Marion Young’s argument is thus partly sustained by Butler’s definition of gender as ‘stylized repetition of acts’ (1990b, 140). Sartre’s concept of series seems hence very fit to describe social categories as the result of both individual bodily constitution and motives, as far as social subjection/subjectivation processes and other structural force are concerned. To make Sartre’s concept of serial action even more useful, we may add a notion of serial categories. Such categories are defined from a number of repetitive serial actions, instead of single, momentary series, as in the bus-queue example (Fahlander 2003, 34). Most individuals participate in a number of serial collectives during their day-to-day activities. After the bus queue, some individuals may join another type of series at their workplace and so forth. All the serial memberships that a person is involved in do not necessarily constitute a part of social identity. The line of profession can mean very much to some, while others base their identity on other grounds, but very few define themselves as ‘public transporters’. We may thus speak of individual, momentary, serial situations in favour of subject positions or status-roles, and serial categories, instead of social categories (Fahlander 2003; 2007). The interesting aspect of Sartre’s concept of series is that it allows for some degree of heterogeneity and fluidity, but is still operational. As previously hinted, the serial perspective suits the fuzzy category of children and juveniles as hand in glove. Children in different stages of corporeal and social development certainly form various serial collectives according to the possibilities and constraint of their bodily constitutions. The agency of a two year old is likely to differ (but not in all respects) from the one of a seven year old and they may thus be found in different series of day to day life. The kinds of series we can expect to find certainly vary by ideologies and material circumstances, but the physical and social development of a child does, however, follow certain general courses. Some of these aspects are nearly universal, while others depend more on the situation of the whole society (sufficient food, no major conflicts, etc.) and social norms and ideologies. To be able to construct relevant series based on stages of development

As a simple but illustrative example, Sartre discusses a group of people waiting for a bus. These individuals may be of different sex and age, have different body postures, belonging to different fraternities, religions, etc. and are not integrated as a group in the strict sense but are united by their intention of riding in the same bus. The individuals in the queue would probably neither recognise themselves as a group, nor do they need to share a common social origin (habitus) or individual, discriminating attributes. This particular series is defined by a ‘fluid homogeneity’. Sartre’s example may seem naive and his distinctions less meaningful, but this perspective may prove to be very helpful in finding interesting clusters in the social plurality. It is interesting to note that the serial perspective does not necessarily depart from what the agents think of the materialities that surround them (like the city, the bus, etc.). Individuals may in many situations act and think as solitaries, but they nevertheless reproduce patterns according to general, social and material conditions. Of course, individuals understand materials in various ways; some they find ugly, others are practical, some we like, some we don’t, and in everyday life we do not think about most of them in particular. For instance, it is obvious that some people do not like buses and seek other ways of transportation; these we probably found in other series such as cyclists and car commuters. Sartre mentions, in addition to serial praxis, the existence of serial feelings and serial thoughts: hence, “a series is a mode of being for individuals both in relation to one another and in relation to their common being and this mode of being transform all their structures” (1991, 266). Marion Young (1997) has successfully employed Sartre’s serial perspective in her struggle to define women as a subjectivated category (cf. Fahlander 2006, see also Johannesen 2004 on seriality and ethnicity). Feminist theory is troubled in defining a valid category of women since the individual situation of women of different social contexts is varied and many do not share the same goals or problems (Mohanty 1984, Butler 1990a, 324f, Braidotti 1991, 158; Moore 1994, 10f). The differing experiences of women make them more of a heterogeneous collective or, in Braidotti’s terms, a collective singularity (Braidotti 1991, 132). But still, despite the differences in power, physical and mental 16

SUBADULT OR SUBALTERN? CHILDREN AS SERIAL CATEGORIES it is in place to discuss some corporeal, mental and social variables related to age with potential of being socially significant.

There are a number of ways in which we can discuss individual age. Bodily aspects can be defined as, i.e. corporeal age, which refers to bodily statue, bodily skills, level of growth, sexual maturity or generally the appearance of age. The bodily aspects are not simple visual; we can also assume that the ability to move increases by age as well as development of fine motor coordination and general bodily skills. Most children are hampered by their corporeal constitution which decreases their possibilities to reach, climb, swim, and affect the speed in which they do things (of course, this can be an advantage in some situations, for instance, when work needs to be done in small and narrow places). Children are also generally limited in bodily knowledge, that is, how to do things properly, and certain tasks need a level of fine motor skills that younger children do not normally possess. Linked to this biological category, we may also add mental age, which include IQ and EQ in a general sense, but more specifically the level of social capabilities, rate of communicative abilities, and problem-solving skills. In psychology, some general steps in mental development are recognized, for instance, the mirror-stage (Lacan 1977; cf. Alcorn 1994) or Freud's (1962) five stages of psychological development (oral, anal, phallic, latency, genital). These distinctions may be helpful in describing the child’s psychological development, but their social implications are not necessarily that important. A more interesting concern, however, is the development of a ‘theory of mind’; a socio-psychological threshold that most children pass at the age of four (Cox 1991, 177ff). To possess a theory of mind means that one can differ between real circumstances and how other people may react to delusions. In short, to understand that other people may have different beliefs than yourself. A popular test to determine if one has acquired a theory of mind is the Sally-Ann test (Cox 1991, 167ff). The child is presented to two dolls, Sally and Anne. While all three are present in a room a small bag of candy is placed under a pillow. Then one of the dolls, Sally, ‘leaves’ the room and the candy is moved to the pocket of the remaining doll Anne. The child is now asked where Sally, when she enters the room again, would believe the candy is hidden. Children under the age of four generally answers ‘in Ann’s pocket’, while children above four correctly answers ‘under the pillow’ often with the addition; ‘but they are not’, delivered with a conspiratorial smile.

What’s my age again? Corporeal and social aspects of childhood In archaeological analysis we normally employ osteological and/or dental evidence for age estimation in chronological years. We should, however, rather use the terms osteological age and dental age, as such age estimates normally are calculated in reference to a given normality and are thus not always sensitive enough to account for individual variation. There are also a number of methodological problems in case of incomplete or highly degraded skeletons and sex determinations are generally difficult to establish before the individual have reached sexual maturity (Manchester 1989, 11, Sheuer and Black 2000, 12). The main problem is, however, that estimations of chronological age from osteological and dental evidence may differ from phenotypic age and indeed from social age. Archaeologists are in general not interested in the long and incomprehensible tables of measurements that sometimes are offered by the osteologists but are normally satisfied by the single column of summarized age (Table 1). Detailed osteological reports are, however, very valuable in social analysis because we relate and compare actual measurements before that are recalculated according to sex and other bell curves. We need more data than general estimations of statue, sex and gender to be able to discuss social aspects of embodiment and the social significance of bodily appearance. Another implication is found in the age-groups of osteologists and odontologists whose thresholds are based upon normal development of certain bones and teeth. They are thus not always sensitive enough for social thresholds that may occur within or in between two osteologically visible traits.

Fetus Infant Infans 1 Infans II Juvenilis Adultus Maturus Senilis

Martin & Saller 1957 – – 0 – 6/7 years 6/7 – 14/15 years 14/15 – 18-20 years 18/20 – 35/40 years 35/40 – 50/60 years –

Sjøvold 1978

Sellevold et al 1984

Arcini 1999

– 0 – 1 year

– –

0–7 years 5 – 14 years 10 – 24 years

0 – 6 years 6 – 12/14 years 12/14 – 17/19 years

0? 0–9 months 1–6 years 7 – 14 years 15 – 19 years

18 – 44 years

20 – 35 years

20 – 39 years

35 – 64 years

35 – 55 years

40 – 59 years

50 – 79 years

>60 years

>60 years

Other similar studies suggest that children under the age of three have difficulties in masking their feelings, a social skill that is gradually developed and normally handled at the age of six. This is, of course, connected to the ability to understand and read other peoples feelings which normally begins at the age of two, but is generally more developed at the age of six-seven (Cox 1991, 187, 191). Another crucial parameter is the development of language. This is a process that accelerates rapidly at the age of c.30 months and culminates at the age of six-seven when the child normally has acquired basic communicative and grammatical knowledge. To conquer a language does not simply mean that one can

Table 1. Different categories of age according to a number of osteological manuals. (After Gustafsson and Lundin 2004:81). 17

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD child or juvenile in contrast to adults. By analysing the importance of these variables in a given case we can actually find several different categories of children instead of only one fuzzy and diffuse category. Based upon the aspects discussed above we can quite easily discern at least four different series that ought to have social significance (Table 3). Although, these are important steps in any child’s corporeal and social development, we cannot assume that they are always significant in any given case. In order to validate their respective social importance we need to discuss them in relation to archaeological data. The series may, however, prove helpful to us as a starting point when we arrange and construct our fictions and models of the past.

communicate better with others. The grammatical skills also give rise to a greater ability to understand the surrounding world and to grasp an idea of self in future stages. This self-awareness is often both a pain and a blessing, but carry important implications both for the individual itself as to its primary others. The mental age is, of course, very much tied to what we may call social age. To possess language, self awareness and theory of mind is crucial in social life. It is important to recognise that other people can have other beliefs than oneself in order to cooperate and negotiate with others. It may also to be useful to take advantage of a situation and even manipulate others (as well as avoiding being manipulated and abused). A theory of mind is also socially important as it enables the child to make sense of stories and thus be able to comprehend cultural knowledge (i.e. ethnic narratives). Social age also concerns some aspects that lie outside the corporeal and mental characteristics. A few such practical examples are the level of self-sufficiency and ability to manage without help, passing the stage of weaning, and the rate of mobility. Communication is often of essential social importance. To function as a group in everyday life one need to be able to express taboos, rights and duties, negotiate and express emotions and plan collaborative work to some extent. These skills are normally something that takes time to learn and interiorize.

Series I (Suggested chronological age-span: c.0-2years) Low rate of communication Need of care Minimal mobility Minimal agency Low self-awareness Series II (Suggested chronological age-span: c.2-4 years) Communicative (language and grammar) Beginning to cooperate (play) with others Limited mobility Low social skills (no theory of mind) Self-awareness Need of guidance and help (but normally weaned)

To summarise so far, children are able to participate socially in various ways according to their specific corporeal and mental aspects as well as by acquiring physical and social skills. In most societies an individual can reach different socially highlighted thresholds, i.e. achieved age. Such thresholds can be reached by passing tests or by completing certain tasks or perhaps ‘automatically’ by reaching sexual maturity (or menopause), marriage, deaths of significant others, adoption and divorce, etc. Common for all these examples is that they tend to vary between individuals, and between societies, and we cannot assume any natural or logical stages that separate individuals from one phase of the life-cycle from another. We can, however, point out some general aspects related to chronological age that may have social importance (Table 2). Corporeal Low stature Ability to move

Mental Mirror stage Verbalism

Fine motor coordination Strength Sexual maturity

Theory of mind Knowledge IQ/EQ Dispositions

Series III (Suggested chronological age-span: c.4-6/7 years) Communicative (language and grammar) Developed fine motor coordination Increased mobility Theory of mind Greater strength Developed social skills and knowledge Series IV (Suggested chronological age-span: c.7-10/12 years) Fully communicative Basic social skills and cultural knowledge Full mobility Ability of fine motor coordination ‘Fit for work’

Social Social thresholds Tacit ‘bodily’ knowledge Cultural knowledge

Table 3. Examples of series based on corporeal, mental and social properties of young children. The suggested chronological time-spans of the series in Table 3 are attached with some hesitation; perhaps we would be better off without any reference to chronological age at all? Our prime focus of interest is aspects of corporeality, such as appearance and socially significant abilities. Reference to age does, however, serve a purpose of clarifying the level of the properties of the series. But when it comes to actual case studies, archaeologists are generally confined to estimates of osteological age. The suggested series are thus only

Ability to propagate Self-sufficiency Trustworthiness

Table 2. Suggested corporeal, mental and social variables related to the development of children. Table 2 outlines some corporeal, mental and social aspects that may be important properties in the ways different people categorise an individual as a kind of 18

SUBADULT OR SUBALTERN? CHILDREN AS SERIAL CATEGORIES relationships between bodily attributes and elements of the grave. What kind of bodies (i.e. not individuals) are buried with certain kinds of material, at certain places and in certain constructions? Such corporeal relationships can in turn hint at social categorisation and subjectivation of the living population. In this manner, we may end up with more relevant and even unexpected series of individuals and stages of the life-course than by the ‘normal’ way of procedure.

tentative and needs to be apprehended in a polythetic manner since children do not develop physically and mentally simultaneously and at a constant speed. Just to pick one example, a socially important variable is weaning. Here we have little means to establish a breaking point. Children can be breast-fed up to quite an advanced age or can manage without almost from birth (cf. Alexandre-Bidon and Lett 1999, 33). We therefore need to view the variables of the series like fibres of a thread. A thread is made up of separate fibres of different lengths wrapped over each other, and each of them does not necessarily extend all through the thread. Of course, we probably need to adjust our series according to varying social and material conditions. There are probably (but not necessarily) differences in which variables that prove socially and practically significant between, lets say, a hunter-gatherer group and life in a medieval village.

It is most peculiar how negligent burials of children have been treated in archaeology, when it is quite evident that studies of children’s burials have great potential. A common feature in many periods and places is that children generally are treated differently than the older population (Watts 1989; Lee 1994; Molin 1999). Generally, the burials of small children are not as ‘rich’ and elaborated as the graves of adults (but there are of course many exceptions).1 They are often buried at liminal places or sometimes even in profane domestic contexts. When adults are inhumated, children are often cremated or vice versa, etc (Rash 1991, 31, AlexandreBidon and Lett 1999, 30, Morris 2000, 287; HoubyNielsen 2000; Svedin 2005). These instances, the seemingly subaltern status of the dead children, are generally associated with an ambiguous identity in life (e.g. Jonsson 1999). Indeed, the young child is in a social sense often displaced, or lost, somewhere in between a ‘proper’ gender, sex, identity, citizenship, wild-domestic, animal-human, etc. Instead of simply dismiss the treatment of children as deviant and anomalies in ‘normal ritual practice’ we may instead explore these relations in order to better understand larger issues of e.g. social identity, agency, gender, corporeality, social categorisation (subjectivation) and ideology.

Series in action? There is unfortunately no room here for an extensive case-study that would have put the serial perspective in practice (but see Fahlander 2003, 87-120). I assume, however, that the reader can very well imagine a number of ways in which it can be applied to e.g. material, iconographic, and written evidence (cf. Scott 1999; Baxter 2005). I will nonetheless finish off by discussing the possibilities of a serial perspective in analysing burial evidence. Interpreting graves and grave interments are perhaps one of the most complicated and ambiguous analytical problems in archaeology. The main problem in burial analysis is to determine in which sense the construction of the grave and its interments are related to the buried individual (Fahlander 2003). To simply follow the Saxe-Binford approach and state that the burial reflects the buried individuals status or social persona is clearly not sufficient (or not even very likely). The graves of children highlights this problematic when we find tools and weapons in children’s graves that they cannot have been using during life. Buried children are often not assumed to have high social status or importance of their own. Instead, such elaborated burials are assumed to reflect the status of their families (e.g. Pader 1982, 57; Baxter 2005, 94). One example is a child grave of the Neolithic site at Ajvide, Sweden where a newborn child was buried with a large harpoon which it never could have been able to use in life. The excavators interpret this instant as evidence for a social structure based on inherited prestige (Burenhult 1997, 62, but see Strassburg 2000:200). It could, of course, also simply represent a burial gift of one of the participating adults with little or no relation to the dead child. In singular cases, we can never reach a satisfactory interpretation. We need some kind of patterning or structure to make more generalising statements. The problem is how to establish such patterning without making simple correlations between grave and individual. One way of bypassing that tricky question is by employing a corporeal perspective as discussed above. Instead of searching for patterns related to age and sex, we can relate attributes of the grave to various aspects of osteology and thus establish corporeal

To exemplify such an approach, we may briefly consider one case where different bodies (young – old, male – female, human – animal) seem to have required different burial practices. At the Mesolithic burial complex of Skateholm in southern Sweden (Larsson 2000; 2004; cf. Strassburg 2000; Nilsson-Stutz 2003, Fahlander 2008) the individually buried children under the age of five are placed marginally at the borders of the two main burial areas. The same spatial aspects are also valid for the eight individually buried dogs of the site. For some reason the dead bodies of children and dogs been separated from the adults in a similar manner. This spatial arrangement is quite obvious, but there are nonetheless interesting differences between the younger and the older phase of Skateholm. At the younger burial area of Skateholm I, we 1 One striking example is the huge Bronze Age burial cairn of Kivik in southern Sweden. Due to its exceptional size and monumental appearance it has even since the 19th century been believed to be the grave of a powerful chief. The recent performed osteologically analysis and C14 determinations do however, prove that theory wrong. It turned out that the chamber of the cairn hosted four to five different individuals, all in their early teens except for one adult. The carbon determinations also revile that the individuals were deposited in at least three different occasions (c. 1400-1200, 1200-1000 and 900-800 BCE). Interestingly is the adult individual the only one buried during the last phase and hence has little to do with the construction of the monument (Goldhahn 2005, 224-54).

19

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD traditional social categories but rather tried to explore an archaeology that would embrace and, to a greater extent, use traces of social complexity and heterogeneity of different periods and places. From such a perspective, analysis of children and related practices can prove to be a rewarding platform. Not only to increase our knowledge of ‘the small ones’, but such studies can also help us to understand larger issues of e.g. gender, corporeality and ideology.

find the children and the dogs in the southern, western and eastern borders. At the slightly older burial area of Skateholm II, however, we find the dogs at the eastern and western borders while the children limit the area in the north and the south (Fahlander 2011). There can be little doubt that the marginal placement of children and dogs are significant and meaningful. Generally, such marginal placement of burials been interpreted as an indication of low status, but that hypothesis makes little sense for Skateholm. Several of the children’s graves are, on the contrary, rich in interments and of high energy expenditure and the ‘richest’ grave of the Skateholm site was actually one of the dogs! The dead bodies of small children and dogs are clearly out of the ordinary and seem to have needed special requirements during the burial. Of course, this does not necessarily mean that children and dogs were exceptional (or alike) in life. But nonetheless these patterns between children, adults and non-humans (dogs) do provide a very interesting platform in discussing how social identity and status may be related to burial evidence. The example of bodily metaphors at Skateholm does indeed suggest that there are great possibilities in detailed studies of children’s graves; especially if we employ a serial perspective based on corporealities and social practice.

References Alcorn Jr, M. W. 1994. The subject of discourse: Reading Lacan through (and beyond) poststructuralist context. In M. Bracher (ed.) Lacanian Theory of Discourse: Subject, structure, and society. New York: New York Univ. Press, cop, 19-45. Alexandre-Bidon, D. and Lett, D. 1999. Children in the Middle Ages: Fifth-fifteenth centuries. Translation by J. Gladding. Notre Dame, Ind.: University of Notre Dame Press. Alt, K. W. and Kemkes-Grottenthaler, A. (eds.) 2002. Kinderwelten: Anthropologie, Geschichte, Kulturvergleich. Köln: Böhlau Verlag.

Summary Andersson, K. 1999. De dolda könen – om barn, ungdomar, vuxna och åldringar i romartidens Uppland och Västermanland. In K. Andersson, A. Lagerlöf & A.

It has been argued that simply discussing ‘children’ as a single category won’t get us very far in social analysis of the past; we need to open up the category into socially significant sub-categories (series). In this paper, I have suggested that Jean-Paul Sartre’s distinction between series and groups is helpful in discerning such subcategories of children. Children in different stages of corporeal, mental and social development are likely to be involved in different tasks, spaces and social spheres not only in relation to adults, but also to each other. I have proposed four such series of pre-pubertal individuals here, but, of course, such schemes can as well be extended in detail or range according to available data and questions.

Åkerlund (eds.) Forskaren i fält – en vänbok till Kristina Lamm. Stockholm: Riksantikvarieämbetet, 77-86. Arcini, C. 1999. Health and Disease in early Lund: Osteo-pathologic studies of 3,305 individuals buried in the first cemetery area of Lund 990-1536. Lund: Lunds Universitet Baxter, J. E. 2005. The Archaeology of Childhood: Children, gender, and material culture. Walnut Creek, CA: AltaMira Press.

It is probably true that children and juveniles seldom are the great producers, leaders or innovators of prehistoric societies, but they do make up a large part of the populations and must thus directly or indirectly been involved in the formation of the archaeological record. The fact that the category is diffuse and heterogeneous is no arguments for them being less interesting to study and discuss. On the contrary, it rather seems that extensive and detailed analyses of children and other age-related aspects have a great potential in archaeology. There is a large empirical material consisting of texts, iconography and burial data ready at hand and there is a whole world of children to discover by discussing the constitution and social effects of different series of children. The brief example of the Mesolithic site of Skateholm indicates that social categorisation and identity are issues that can be very different from what we normally assume them to be. Research on children and social aspects of early age would surely benefit if we did not strictly employed

Beaumont, L. A. 1994. Constructing a Methodology for the Interpretation of Childhood Age in Classical Athenian Iconography. Archaeological Review from Cambridge 13(2), 81–96. Berggren, K. 2000. The knowledge-able agent? On the paradoxes of power. In C. Holtorf and H. Karlsson (eds.) Philosophy and Archaeological Practice. Perspectives for the 21st Century. Göteborg: Bricoleur Press, 39-46. Bhabha, H. K. 1994. The Location of Culture. London: Routledge. Bigwood, C. 1991. Renaturalizing the Body (with the Help of Merleau-Ponty). Hypatia 6(3) (Fall 1991), 54-73. Braidotti, R. 1991. Patterns of Dissonance. A Study of Women in Contemporary Philosophy. Cambridge: Polity Press. 20

SUBADULT OR SUBALTERN? CHILDREN AS SERIAL CATEGORIES barndom, In: Fredrik Fahlander (ed.), Stockholm studies in archaeology 54, Stockholm: Univ., 181-196.

Brück, J. 1998. In the footsteps of the ancestors: a review of Christopher Tilley’s A Phenomenology of Landscape: Places, paths and monuments. Archaeological Review from Cambridge 15, 23-36.

Foucault, M. 1984. The Birth of the Clinic. An archaeology of medical perception. Transl. A. M. Sheridan Smith. New York: Vintage Book.

Burenhult, G. 1997. Gravarnas vittnesbörd. In G. Burenhult (ed.) Ajvide och den moderna arkeologin, ed: Göran Burenhult. Stockholm: Natur & kultur, 52-70. Burkitt, I. 1999. Bodies of Thought. London: Sage.

Foucault, M. 1980. Power, Knowledge: Selected interviews and other writings 1972-1977. Ed. and transl. by Colin Gordon, New York: Pantheon cop.

Butler, J. 1993. Bodies that Matter. On the discoursive limits of “sex”. New York: Routledge.

Freud, S. 1962. Three Essays on the Theory of Sexuality. London: Hogarth Press.

Butler, J. 1990a. Gender trouble, feminist theory, and psychoanalytic discourse, In L J. Nicholson (ed.) Feminism/postmodernism. New York: Routledge, 324340.

Gero, J. 2000. Troubled Travels in Agency and Feminism. In M-A Dobres and J. E. Robb (eds.) Agency in Archaeology. London & New York: Routledge, 34-39. Gilchrist, R. 1999. Gender and Archaeology: Contesting the past. London: Routledge.

Butler, J. 1990b. Gender Trouble, Feminism and the Subversion of Identity. New York: Routledge.

Goldhahn, Joakim 2005. Från Sagaholm till Bredarör hällbildsstudier 2000-2004. GOTARC series C, No 62, Göteborg: Arkeologiska institutionen.

Butler, J. 1997. The Physic Life of Power. Theories in subjection. Stanford: Stanford University Press. Cox, M. V. 1991. The Child’s Point of View: The development of cognition and language. London: Harvester Wheatsheaf

Gräslund, A-S. 1973. Barn i Birka. In Tor, 1972-1973, 161–179. Gustafsson, A. & Lundin, I. 2005. Och så var det barnen. Barn- och ungdomsgravar i Svarteborg och Kville Socknar. In P. Claesson and B-A Munkeberg (eds.) Projekt Gläborg – Rabbalshede. Bygden innanför fjordarna: 3 Gravar och ritualer. Kulturhistoriska dokumentationer nr 15. Uddevalla: Bohusläns museum, 71-106.

Davis, K. 1997. Embody-ing Theory: Beyond Modernist and Postmodernist Readings of the Body. In K. Davis (ed.) Embodied Practices: Feminist perspectives on the body. London: Sage, 1-23. Fahlander, F. 2001. Archaeology as Science-Fiction. A Microarchaeology of the Unknown. Gotarc serie C, no 43, Göteborg: Institutionen för arkeologi.

Hamilakis, Y., Pluciennik, M. and Tarlow, S. 2002. Thinking through the Body: Archaeologies of Corporeality. New York: Kluwer Academic/Plenum.

Fahlander, F. 2003. The Materiality of Serial Practice. A Microarchaeology of Burial. Gotarc serie B, no 23, Göteborg: Institutionen för arkeologi.

Haraway, D. J. 1991. Simians, Cyborgs, and Women: The reinvention of nature. London: Free Association Books.

Fahlander, F. 2006. Kön & Gender som seriekollektiv: sociala aspekter på korporealitet & handling, Det arkeologiske kjønn, Ed: I. Fuglestvedt and L. Skogstrand, OAS (7), Oslo: University, 27-42.

Heywood, C. 2001. A History of Childhood: Children and childhood in the West from medieval to modern times. Oxford: Polity Press.

Fahlander, F. 2007. Third space encounters. Hybridity, mimicry and interstitial practice, EncountersMaterialities-Confrontations. Archaeologies of Social Space and Interaction. In P. Cornell and F. Fahlander (eds.) Newcastle: Cambridge Scholars Press, 15-43.

Higate, P. 1998. The body resists. Everyday clerking and unmilitary practice. In S. Nettleton and J. Watson (eds.) The Body in Everyday Life. London & New York: Routledge, 180-98. Houby-Nielsen, S. 2000. Child burials in ancient Athens, In J. Sofaer Deverenski (ed.) Children and Material Culture. London: Routledge, 151-66.

Fahlander, F. 2008. A piece of the Mesolithic. Horizontal stratigraphy and bodily manipulations at Skateholm, The Materiality of Death. Bodies, burials, beliefs. In Fredrik Fahlander and Terje Oestigaard (eds), BAR International Series 1768, Oxford: BAR Publishing, 29-45.

Irigaray, L. 1985. Speculum of the Other Woman. Ithaca, N.Y.: Cornell University Press.

Fahlander, F. 2011. Bland barn och hundar på Skateholm, Spåren av de små. Arkeologiska perspektiv på barn och

James, A., Jenks, C. and Prout, A. 1998. Theorizing Childhood. Cambridge: Polity Press. 21

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Mohanty, C. T. 1984. Under Western Eyes: Feminist Scholarship and Colonial Discourses. Boundary 2 12(3), (Spring-Autumn, 1984), 333-358.

Johannesen, J. M. 2004. Operational Ethnicity. Serial Practice and Materiality. In F. Fahlander and T. Oestigaard (eds.) Material Culture and Other Things. Gotarc series C, No 61, Göteborg, 161-184.

Molin, A. 1999. Barngravfältet i Hulje. In A. Kaliff (ed.) Olika perspektiv på en arkeologisk undersökning i västra Östergötland. Skrifter Nr. 28. Stockholm: Riksantikvarieämbetet, 42-56.

Johnsen, B. and Welinder, S. (eds.) 1995. Arkeologi om barn. Uppsala: Societas archaeologica Upsaliensis. Jonsson, K. 1999. Bland barnaföderskor, spädbarn och ”vuxna barn” – social och religiös kontroll speglad i gravmaterialet från Västerhus. META No 4, 12-35.

Moore, H. 1994. A Passion for Difference: Essays in anthropology and gender. Cambridge: Polity Press.

Jordanova, L. 1990. New Worlds for children in the eigthteenth century. Problems of historical interpretation. History of the Human Sciences 3, 69-83.

Morris, I. 2000. Archaeology as Cultural History. Words and things in Iron Age Greece. Oxford: Blackwell Publishers.

Lacan, J. 1977. Écrits. A selection. Trans: Alan Sheridan, London: Tavistock Publications.

Nilsson Stutz, L. 2003. Embodied Rituals & Ritualized Bodies: Tracing ritual practices in Late Mesolithic burials. Stockholm: Almqvist & Wiksell International.

Laqueur, T. 1990. Making Sex: Body and Gender from the Greeks to Freud. Cambridge, Mass.: Harvard University Press.

Nordblad, J. & Yates, T. 1990. This perfect body, this virgin text: Between sex and gender in archaeology. In I. Bapty and T. Yates (eds.) Archaeology after Structuralism: Post-structuralism and the Practice of Archaeology. London: Routledge, 222-237.

Larsson, L. 2000. Cemeteries and mortuary practice in the Late Mesolithic of southern Scandinavia. In V. Lang (ed.) De temporibus antiquissimis ad honorem Lembit Jaanits. Tallinn: Teaduste Akadeemia Kirjastus, 81-102.

Rasch, M. 1991. Öland - en regional översikt, In A. Lagerlöf (ed.) Gravmaterialet som källa för kunskap om människans livsvillkor, religiösa och sociala värderingar. Stockholm: RAÄ, 28-36.

Larsson, L. 2004. The Mesolithic period in Southern Scandinavia: with special reference to burials and cemeteries. In A. Saville (ed.) Mesolithic Scotland and its Neighbors: The early Holocene prehistory of Scotland, its British and Irish context, and some Northern European perspectives. Edinburgh: Society of Antiquaries of Scotland.

Rautman A. (ed.) 2000. Reading the Body: Representations and remains in the archaeological record. Philadelphia: University of Pennsylvania Press. Sartre, J-P. 1991. Critique of Dialectical Reason. London: Verso.

Lee, K. A. 1994. Attitudes and Prejudices Towards Infanticide: Carthage, Rome and Today. Archaeological Review from Cambridge 13(2), 65–79.

Sartre, J-P. 1977. Life/Situations: Essays Written and Spoken. New York: Pantheon Books. Scheuer, L. and Black, S. 2000. Development and ageing of the juvenile skeleton, In M. Cox and S. Mays (eds.) Osteology in Archaeology and Forensic Science. London: GMM, 9-21.

Lillehammer, G. 1986. Barna i Nordens forhistorie. Drøft metodegrunnlaget og kildernes bærekraft. KAN No2, 3– 21. Lillehammer, G. 1989. A Child is Born. The Child’s World in an Archaeological Perspective. Norwegian Archaeological Review 22(2), 89-105.

Scott, E. 1999. The Archaeology of Infancy and Infant Death. Oxford: BAR Publishing: BAR International Series 819.

Manchester, K. 1989. Burial Archaeology: the state of the art past, present and future, In C. A. Roberts, F. Lee and J. Bintliff (eds.) Burial Archaeology Current Research, Methods and Developments. Oxford: BAR 211, 5-14.

Sellevold, B. J. 1984. Prehistoric man in Denmark: a study in physical anthropology. In B. J. Sellevold, U. L. Hansen and J. B. Jørgensen (eds.) Iron Age Man in Denmark: Vol. 3. København: Munksgaard.

Martin, R. and Saller, K. 1957-1966. Lehrbuch der Anthropologie: in systematischer Darstellung mit besonderer Berücksichtigung der anthropologischen Methoden. Stuttgart.

Sjøvold, T. 1977. Non-Metrical Divergence between Skeletal Populations: The theoretical foundation and biological importance of C. A. B. Smith's mean measure of divergence. Solna, Univ.: Osteological research laboratory.

Meskell, L. 1999. The Archaeology of Social Life. Age, Sex, Class et cetera in Ancient Egypt. Oxford: Blackwell.

Sofaer-Derevenski, J. 1997. Engendering children, engendering archaeology. In J. Moore and E. Scott (eds.) 22

SUBADULT OR SUBALTERN? CHILDREN AS SERIAL CATEGORIES Invisible People and Processes Writing Gender and Childhood into European Archaeology. London: Leicester University Press, 192-202. Sofaer Deverenski, J. 2000a. Material culture shock: confronting expectations in the material culture of children, In J. Sofaer Deverenski (ed.) Children and Material Culture. London: Routledge, 3-16. Sofaer Deverenski, J. (ed.) 2000b. Children and Material Culture. London: Routledge. Stoodley, N. 2000. From the cradle to the grave: age organization in the early Anglo-Saxon burial rite. World Archaeology 131(3), 456-72. Strassburg, J. 2000. Shamanic Shadows. One Hundred Generations of Undead Subversions in Southern Scandinavia 7,000–4,000 BC. Surte: Department of Archaeology, Stockholm University. Svedin, M. 2005. The Significance of the children in life and in death. Styrmansberget on Gotland, Sweden. A place for Communications, fertility, gender and kin relations during the Roman Iron Age and the Migration period. In T. Hjørungdal (ed,) Gender Locales and Local Genders in Archaeology. BAR International Series. Sweeney, S. & Hodder, I. (eds.) 2002. The Body. New York: Cambridge University Press. Turner, B. 1996. The Body and Society: Explorations in social theory. London: Sage. Watts, D. J. 1989. Infant burials and Romano-British Christianity. The Archaeological Journal 146, 372–383. Welton, D. 1998. Introduction: Situating the Body. In D. Welton (ed.) Body and Flesh: A Philosophical Reader. Oxford: Blackwell Publishers lnc, 1-8. Wheelan, M. 1995. Gender and Historical Archaeology: Eastern Dakota Patterns in the 19th Century. Historical Archaeology 25, 17-32. Young, I. M. 1997. Intersecting Voices: Dilemmas of Gender, Political Philosophy and Policy. Princeton, NJ: Princeton University Press. Žižek, S. 2005. Connections of the Freudian Field to Philosophy and Popular Culture. In R. Butler and S. Stephens (eds.) Interrogating the Real. London & New York: Continuum, 59-86.

23

Etruscan Infants: Children’s Cemeteries at Tarquinia, Italy as indicators of an age of Transition Marshall Joseph Becker for groups living in Rome's western empire. High rates could be inferred without elaborate "studies," based solely on what is known of morbidity patterns in 21st century third world nations. Given all these data, archaeologists working in Etruria should indeed wonder where all the bones of these children have gone.

Introduction Careful analysis of human skeletal populations in central Italy, as well as at many archaeological sites throughout the world, often leads physical anthropologists to ask, "Where have all the children gone?" Perinatal and young children's remains may be absent from among the recovered skeletons from many ancient sites. As recently as 25 years ago I found that infant remains were notably absent from many Italian skeletal collections. Angel (1969; 1971) as well as some more recent scholars (Bello 2004) believed that the fragile bones of infants simply decayed past recognition, or were lost due to skeletal recovery techniques then common in Etruria and elsewhere. I now believe that in much of Italy the small bones of perinatals are no less well preserved than those of adults, and are as likely to survive as any others (see also Jackes 1993, Herring et al. 1994, 60; Bourbou 2000). The absence of perinatal remains may reflect the use of a separate burial location for the bodies of "people" at this stage of life. In Etruria, as in several other culture areas of the Italian peninsula, the bodies of perinatals, infants, and even children up to the age of five years were interred in contexts removed from the formal cemeteries that were used for burials of adults (Becker 1995a; b).

The extent of child mortality is so high in traditional societies, especially in urban areas, that disposal of the tiny bodies is a significant task for the poor. Finding even a simple place to "bury," or dispose of, a small corpse can be a serious concern. Placement in construction fills, abandoned wells (cf. Angel 1945, 311-312, fig. 12; Little 1999), and other available locations often resolved the problem in antiquity. Simple "burial" techniques continued to be employed in colonial America (Burnston 1982) and in many parts of the modern world. The disposal of great numbers of these stillborns, plus those infants who died soon after birth, may have been a principal "birthing" role for the male members of the society (cf. Hanson 1994), meriting a study of its own. Despite the considerable numbers of small bodies involved, only recently have scholars in the Mediterranean area become interested in this fundamental aspect of life in antiquity (Becker 1983; 1994a and b; 1995a and b; 2006a; cf. Bassett 1995).

Studies of perinatal skeletons are essential to the discussion of "infant" mortality rates. The absence of these remains from any "cemetery" population, a commonly found distortion in expected percentages often called "the osteometric paradox," is notable for several reasons (cf. Konigsberg and Frankenberg 1992). Biological studies of non-industrial societies reveal normal infant mortality rates of 50% or greater, with perinatal mortality accounting for most of these figures. Even in modern industrialized societies "the incidence of preterm birth is increasing and remains the major cause of perinatal morbidity and mortality" (Bloomfield et al. 2003:606). The last months of a pregnancy, the process of parturition, and various stresses during the months after birth provide a frightening trio of challenges that lead to what is commonly called "infant mortality." For clarity in the use of terms "perinatal" deaths (perinatal mortality) includes all aborted or stillborn children from within the third trimester of pregnancy, to those who survive through the first month (28 days) after a full-term live birth. The designation of "infant mortality" will be used only for those deaths that take place from age 29 days to one year. Perinatal and infant mortality may be combined together with deaths of children up to the age of weaning under the more general term "child mortality." Russell (1958, 25) estimated child mortality in Rome during the Late Antique period at 17 to 26%, and suggested a considerably higher rate for young adults. Durand (1960, 371-2) estimated first year mortality rates of 30 to 40%

Before focusing on the cultural significance of child mortality patterns as they apply at Etruscan Tarquinia, and how they are evaluated through archaeology and physical anthropology, a brief note is offered on other aspects of human morbidity patterns. By far the highest mortality peak in any pre-industrial society appears around the time of birth (perinatal). For those newborns that live for a year or more in a traditional society, the next major challenge to their survival comes at weaning. Human morbidity curves tend to follow a similar path, with a small but significant "bump" or rise in the curve that reflects the culture's rules regarding when a child should be weaned (Fuller et al. 2006, Clayton et al. 2006). Regardless of the food provided to a child during the weaning process, the adjustment to a less than optimal diet is reflected in increased mortality. This event, or process, is seen in the archaeological record in the greater numbers of deaths that reflect this transition. Study of the mortality curve for a specific population reveals the age at which that society specifies that weaning should take place. For example, no children below the age of 2.5 years are in the population described by Dawes and Magilton (1980) from Medieval York, England. This suggests that among the numerous skeletons from this site there are none representing perinatal and infants up to the age of weaning. We may infer that they were buried in a "separate cemetery." Deaths at the age of weaning in industrial or complex societies may not reveal a strong 24

ETRUSCAN INFANTS: CHILDREN’S CEMETERIES AT TARQUINIA, ITALY AS INDICATORS OF AN AGE OF TRANSITION noted comments on the subject from ancient Greece. The earlier is from Aristotle's Historia Animalium (1965) (Book vii). Aristotle refers to infants in Athens of the 4th century BCE, noting that “most are carried off before the seventh day; that is why they give the child its name then, as by that time they have more confidence in its survival." Hundreds of years later the Greek moralist Plutarch (CE ca. 46-ca.120) observed, in his Numa (Book 12) (1914, 347-348), that there was no formal mourning for children who die under three years; an age that I suggest relates to the weaning process. Miles (1984) notes that this ancient "cut-off" age was later changed to one year, a change that may suggest that the age of weaning may have been reduced through time, or that survival rates may have improved.

rise because a standard age may not exist for the various groups that converge in a city. In complex societies weanling deaths may be spread over a wide range of ages that reflect several cultural patterns. Improvements in nutrition available to infants being weaned also may be a factor in some urban contexts. Since "life expectancy" is the average age of all people who are born (alive) into the society, high infant mortality lowers the average. Thus what we call "life expectancy" actually is a strong reflection of perinatal mortality. A low "life expectancy" indicates a high death rate among children. "Life expectancy" can vary depending on whether an infant is calculated as "a live birth" if it survives for a few hours, or even days after birth. In premodern or traditional societies a person surviving to age 25 could be expected to live to age 65 to 75 and longer with almost the same probability as people in modern society. Data on the people in ancient Tarquinia and elsewhere, who can be documented as living to more than 75 years, and some to 90 or more, is important to our general understanding of population dynamics.

The rich collection of surviving Roman documents provides a great deal about children, but commonly research on the Roman "child" pertains to those between the ages of five and 15 years (see Rawson 1991a, 4). The fetal stage of development is rarely noted, and then only in relationship to matters of abortion (Rawson 1991b, 9). Newborn children in Rome commonly are noted as having been subject to sale or fatal exposure, but their status was subject to the laws that covered matters of household order. In the later Roman periods, deliberate exposure of newborns had become a crime (see Eyben 1991). An excellent review of what is known about the Roman birthing process is provided by Rawson (1991b, 10-13). Of particular interest to this study of infant mortality rates is the ritual of lustratio that took place eight days after birth for a girl and nine days for a boy. The difference of a single day may reflect a Roman awareness of the greater biological fragility of males. The lustratio ritual involved conferring a name on the child and, as Rawson notes (1991b, 14), "thus the infants 'existed' in a formal sense."

In the reconstruction of an ancient society archaeological finds of clusters of infant burials rarely have been contextualized using basic information on expected morbidity patterns. Finds of specialized infant cemeteries, including Phoenician tophets, often had been interpreted as evidence for infanticide (e.g. Greene 2003, following Mays 1993). Gowland and Chamberlain (2002) addressed this question by reassessing the data relating to 396 child burials from 19 Romano-British sites, using a specific and outstanding approach to aging the remains of these perinatals. They found that the age distributions for these children similar to natural mortality profiles, and conclude that these deaths cannot be attributed to anything but natural causes (see Becker, in press B). Their conclusion is important when addressing any example of ancient demography, and particularly those that relate to finds of clustered infant burials. Myths relating to supposedly extensive practice of infanticide, in ancient Greece as well as in ancient Rome, are continually discussed using only texts as sources of "evidence." The "archaeological evidence," in the form of skeletal remains, points to a very different, but less dramatic interpretation of the data (Becker 1995b). Separate modes of interment for remains resulting from perinatal deaths are, as we shall see, extremely common (see Engels 1980; Pomeroy 1997) and may explain what some archaeologists interpret as evidence for infant sacrifice.

Information on perinatal mortality rates appears but rarely in the ancient texts and is similarly rare in the modern ethnographic literature (cf. Godden and Godden 1966). Reports of finds of perinatal burials have similarly been absent from the archaeological literature due to the limited attention directed to this subject, and sometimes because there are no burials of children to report. The ethnographic literature also includes but a few brief commentaries on the customs involved in disposing of the bodies on children who die at birth, or up to and around the age of weaning. While a stillbirth or death of any infant may be a personal tragedy felt most intensively by the parents, it may not be a matter commonly recorded by a male ethnographer. The limited ethnographic record does provide us with evidence for burial rituals involving perinatals that cannot be inferred solely from an archaeological record.

Ethnographic data: Texts old and new Infant mortality rates remain high in modern developing nations as well as in contemporary tribal societies. Ethnographic data on perinatal burial customs among living cultures can provide important insights into ancient behaviours and cultural beliefs. Miles's (1984) study of high infant mortality rates in antiquity, following Hooper's study (1975), quotes the two most frequently

One of the more extensively documented ethnographic accounts relating to perinatal mortuary activities derives from modern India. Kelly (1975), in her study of gender ratios and female infanticide in two Indian states, elicited data on women's concepts of a "child." She demonstrated that during the period of British administered India 25

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD families did not register the birth of a child until it had survived its first year. Thus expectations of high infant mortality rates were reflected in the registration procedure, and probably in the patterns of burial of infants who died before they were registered. In the early 20th century in the area of Dacca, now Pakistan, a description of cremation practices (Godden and Godden 1966, 149-150) includes only a brief line concerning the disposal of "babies and little children." Ideally "adults" were cremated on the banks of a river and the burned remains consigned to the river. However, little infant bodies were not burned but were wrapped in red cloth and placed on little flower decorated rafts that were floated on the river. Poorer people in Dacca at the beginning of the 20th century, and those not wishing to make a trip to the river, simple buried these tiny corpses, but the preferred locations, if any, for these infant inhumations are not noted. Fig. 1. Map of central Italy with locations of six archaeological sites: A. Cazzanello; B. Ficana, on Monte Cugno; C. Narce; D. Osteria del’Osa (ancient Gabii); E. Poggio Gramignano; F. Tarquinia (From Becker 2007a; Fig 14.1. Drawing by R. J. Robertson).

Bari Hooper's study (1975, 375, from Still 1931), using data from an historical context, employs the London Bills of Mortality for 1762-71 for direct evidence for mortality patterns. These data indicate that, in the urban environment of London, some 50% of the children [all live births?] died by the age of two years, and that by the age of five years this had grown to 67% (cf. Steel 1988, 150-155). This corresponds closely with Rousseau's anecdotal description, in Émile (1762) for urban France, that of all the children born, scarcely half reach adolescence. This is not a purely urban phenomenon limited to 18th century France. High urban infant mortality rates contrast sharply with the rate in the countryside surrounding Philadelphia in the mid-18th century. At that time meat, cheese and grain exports from that port sustained the British Empire. Rural women in the Philadelphia area, who consumed huge quantities of high protein products, commonly had ten or more viable children. These children provided valuable additions to the agrarian work force. Infant mortality rates appear to have been extremely low in rural Pennsylvania during the latter half of the 18th century; but these data need closer scrutiny (Becker, in press A).

Archaeological evidence for high infant mortality: Children’s burials at Tarquinia The archaeological evidence used to reconstruct all aspects of life in an ancient society often derives largely from excavations of the cemeteries in which members of that culture interred its dead. Specific pottery assemblages buried with the dead, for example, are assumed to reflect the kinds of ceramics that had been used by the living. Some scholars suggest that Etruscan tombs and cemeteries reflect the organization of the associated cities. The bones of the dead also provide information useful in reconstructing the lives of these people, as well as their demographics (Becker 1986; 1995a; 2005b). While health and nutrition within a society may be inferred from visual as well as chemical studies of the bones, patterns of child mortality - a basic "fact of life" - often go unstated due to lack of skeletal evidence. High rates of child mortality were noted in the archaeological literature more than three decades ago (Angel 1971, 93), but continue to be overlooked in discussions of cemetery populations and child burials. Sub-adults in general are often ignored by archaeologists who may assume that their small skeletons will not or cannot be found.

All of these findings indicate that a complete skeletal record for any city, town or other community would include very large numbers of perinatals, and that the total subadult population (16.5 years and younger) commonly would exceed 50% of the total. Commonly what is described as a "normal" or expected population lacks most if not all of the remains of perinatals (cf. Becker Ms. A). The question "Where have all the children gone?" thus becomes a matter relevant to all cemetery sites, not just the many associated with ancient Tarquinia (Fig. 1). Philip Perkins (pers. comm. 2004) suggests that this question might be asked in a different way. Perkins asks "what does one have to do, or to be, in order to be buried in a cemetery? Posing the question in this way not only helps us to understand why perinatals may not appear at "expected" rates in a cemetery, but also provides for a broader archaeological concern involving outcasts and other people from a community, and how they may be buried.

Archaeologists involved in the excavation of human skeletal remains from any site should consider documentation of culture-specific mortuary patterns. Special attention needs to be directed toward locating the bones of sub-adults, here defined as ages 16.5 years and younger. Despite the high rate of infant mortality in preindustrial societies, relatively few sub-adults die between the ages of weaning and 16.5 years; commonly some 10 to 15% of the population above the age of weaning. Generally the age group from 5.5 to 16.5 years is well represented in cemeteries, even when the younger (below 26

ETRUSCAN INFANTS: CHILDREN’S CEMETERIES AT TARQUINIA, ITALY AS INDICATORS OF AN AGE OF TRANSITION age 5.5 years) and more numerous skeletons often are "missing." This is exactly the case at Tarquinia, where two separate sets of evaluations have reached the same conclusions (cf. Becker 2006b). At Tarquinia the various skeletal evaluators who published prior to 1987, when a focused research project began, provided information on a total of only 62 individuals from the tombs numbered up to 6101 (Mallegni et al. 1980). Of the 62, 56 were adults (31 M, 20 F, cf. Becker 2005b). Six were found to be subadults (age range 11.5 to 16.5 years; see Table 1). The percentage of sub-adults in this small sample of 62 is 9.7. Note should be made that in this sample the youngest identified by age had died at 11.5 years. Of the 198 individuals identified by Becker (Ms. C) a total of 19 subadults (age range 5.5 to 16.5 years) were identified. The percentage of subadults in this larger sample (3.2 times as large as the population published by Mallegni et al. in 1980) is 9.6%; a figure that must be noted as remarkably in agreement with the recent findings. More notable at Tarquinia is the complete absence of perinatals as well as all children up to and including to the age of five years. The 9.6% subadult population of the Tarquinians identified to date - with an age range from 5.5 to 16.5 years - appears to be a perfectly normal (expected) distribution for that group (Table 2). These bodies at Tarquinia represent a population clearly between the age of weaning - which possibly took place at age two to three years - and the age of young adults (16.6 - 21 years), when mortality rates tend to increase, often at a level higher than that associated with weaning (cf. Becker 1986; Vargiu and Becker 2005).

Tomb No. 6179B 6191 6192 6203 6204 6205 6225B 6251 6254G 6255B 6262C 6264H 6268D 6272E 6276D2 6316B 6330A 6330D 6355

Age 13 Juv. 16 16.5 11.5 Juv.

Sex F F? ? M? M? M? F? F? ? ? M? ? ? F ? ? ? F? ?

Cre/Inh I C I C C C I I I I I I I I I I I I I

*Suspected age range 12-15 years. **Tomb 6330 is the only known example of a tomb at Tarquinia with more than one sub-adult located amongst the adults. These were not studied.

Table 2. Children aged 16.5 years and younger from tombs in Tarquinia excavated between 1982 and c. 1995 (N=19; extracted from Becker Ms C: Table 2). Skeletal populations recovered from the extensive Etruscan cemeteries uniformly lack any evidence for the remains of infants. Two "children" had been reported from the 19th century excavations at an adult cemetery near Tarquinia, but both appear to date from the Villanovan period (BCE ca. 900-800 and after, and the ages of neither child has been verified by studies of the associated bones. The evidence for any children's burials in Etruria during the Villanovan period remains poorly known. Bartoloni (1989, 31) discussed child burials of the Villanovan period and noted that they commonly were inhumed within amphorae but she did not specify locations. Use of amphorae for child burials is noted for other cultures of the same period (cf. Ridgeway 1992). The use of amphorae for the interment of perinatals and small children provides archaeologists with an easily identified context from which tiny skeletons may be recovered. More rudimentary inhumations of these tiny bodies may be more difficult to identify. Extensive as well as careful excavations in settlement areas of the Archaic period in central Italy (Jarva 1981a & b) have yielded numbers of infant burials (Becker 1996), commonly in the vicinity of houses. Prayon's study of early Etruscan grave and house architecture (1975, 149181) provides considerable evidence for the burial of perinatals within residential areas. Without the skeletal remains precise study is impossible, but a careful review of locational data should provide useful information (Fig. 2).

If in normal contexts half of all people born died before reaching adulthood, then where are their remains? Data on infant and juvenile skeletons rarely appear in published excavation reports that include any data on human bones, or they are not noted as under-represented. If we lack direct archaeological evidence for high infant mortality rates, what can archaeologists do to locate the remains of sub-adults? Consideration of cases where we do have direct skeletal evidence for high infant mortality rates is a first step (e.g. Herring et al. 1994).

Tomb No. 3855 5070E 5511D 5672A 5859C 6020A

Age 11 10 Sub-adult* 16? Adolescent 13 14.5 16? 8 12 15 10 15 15.5 11 8 9** 5.5** Child?

Cre/Inh I C I I I C

Table 1. Children aged 16.5 years and younger at Tarquinia identified from excavations prior to 1981 (N=6). Extracted from Becker Ms. C.

27

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD

Fig. 2. Schematic drawing of locations of possible children’s burials from excavations within the urban area of ancient Tarquinia. Locations and letter designations are Chiaramonte Treré (1987, Table XXX). The squares represent neonates while circles represent suspected sub-adults of undetermined age except for ‘C,’ a child aged seven-eight years (Becker 2007a, 289). Many, if not all, predate the 7th century BCE (cf. Becker 1996). Drawing by R. J. Robertson. linked with marriage - marked the beginning of a second stage during which the individual became biologically Extensive excavations within the actual ancient city of able to reproduce new members of the society. Marriage Tarquinia were conducted under the direction of Prof. reflects the appropriate social and economic Maria Bonghi Jovino (1987; see also Chiaramonte-Treré arrangements made for the children. Death marked the 1999). The discovery of a few neonates and young exiting of the individual from the society. Each of these children interred in rude and undated graves within the life-transitions was marked by rites of passage. Of note city proper may reflect pre-6th-7th century BCE burials here is the discovery that "birth" rituals, on closer (Chiaramonte-Trere 1987; 1995; 1999; Bonghi Jovino examination, take place at a period several days after and Chiaramonte-Trere 1987). Alternatively, these few parturition. These rituals welcome a new member into the burials might indicate an impoverished urbanizing society. Now cultural as well as physical anthropologists perinatal mortuary pattern at later Tarquinia, similar to - and some archaeologists - recognize much greater that known from Rome and Lazio. This evidence suggests complexity in various aspects of these passages and see prior to the early 7th century, suggrundaria burials or the how much they differ from culture to culture. inhumation of perinatals beneath the "eaves" of the house, were used at Tarquinia as well as at nearby Ficana Formal rites welcoming a new member into a society, (Becker 1996), where children above the age of 5.5 years such as by baptism or circumcision, vary greatly. Often were buried in "adult" cemeteries. The development of they reflect the difficulties of the transition to "life," or to separate cemeteries for perinatals at Tarquinia developed becoming a person with a high probability of survival. after the 7th century BCE, with the rise of the Etruscan The process of being born is but one event along the long cities. road from conception to personhood. An infant is not a person until some point after birth at which the members When does life begin? of the culture recognize the probability of the infant’s survival. This view, commonly held in cultures During the first half of the 20th century many cultural throughout the world, reflects the simple reality that anthropologists saw one goal of ethnographic study as morbidity rates after birth are at their highest during the describing life within a specific culture, from birth to first year of life. The specific way in which a society death. This "life cycle" approach to descriptive views these several life stages may define the way a ethnography commonly views each "life" as passing culture actually views itself. The focus of this paper is on through three major stages, beginning at birth with the the ways in which a culture sees the transition of the entry of an individual into the society. Puberty - often 28

ETRUSCAN INFANTS: CHILDREN’S CEMETERIES AT TARQUINIA, ITALY AS INDICATORS OF AN AGE OF TRANSITION for the existence of separate perinatal burial areas. In each case a population of 200 or more adults provides a

newly born, from being provisional members of the culture to being actual members, as reflected in the archaeological evidence. The recent identification of two specialized infant cemeteries in Etruria (Soren 1999a and 1999b; Becker 1994c; 2005a) provides direct evidence for separate burial programs for children below 5.5 years of age. These findings led to greater discussion of the many similar unpublished examples noted verbally by several Italian archaeologists. At Poggio Gramignano in Lugano in Teverina, Soren et al. (1999) identified 50 small children, including 44 perinatals, five infants aged at onefive months, and one aged two-three years; this last may be a weanling death. Many were buried in amphorae. The Cazzanello population of 12 children includes 11 perinatals and one weanling. Intensive studies of the bones from newly discovered chamber tombs at Tarquinia and elsewhere (Becker 2002), dating from after the 6th century BCE, are notable for the complete absence of skeletal remains of children below the age of 5.5 years. We now know that they were buried in separate cemeteries. The absence of these infants and young children from main cemeteries indicates that the Tarquinians, as other South Etruscans, treated infants under the age of 5.4 years in the same way that they treated stillborn children. These fundamental mortuary traditions at Tarquinia and elsewhere in South Etruria continued for 1,000 years, into the Late Roman and early Antique periods, as indicated by these two specialized children’s cemeteries at Cazzanello and Poggio Gramignano.

Fig. 3. An infant burial area of the 6th-7th century, located within the abandoned 4th CE bathhouse at Cazzanello (see Becker 2007a, 290). These graves are adjacent to two cisterns built onto the eastern apse. Burials of adults of similar and later dates are found in other areas. Drawing by R. J. Robertson. significant sample and evidence for good skeletal recovery. The possibility that perinatal and infant skeletons were missed appears improbable. The ages of the youngest children found in these cemeteries provides us with evidence for an age of transition to "adult" status. Ideally the recovery of a perinatal cemetery from the same culture would provide confirmation for the existence of a specific mortality pattern within that culture as well as confirming the age of the ritual transformation. The example of burials from Cazzanello although offering only ten perinatals - provides a small sample, but one which indicates that separate burial locations were a cultural tradition that has endured in that area for two thousand years. This continuity is demonstrated by 21st century Italian cemeteries in which perinatals, as well as foreigners, are buried in specialized areas distinct from the typical inhabitant of the community.

The Cazzanello cemetery, near Tarquinia: An Etruscan perinatal cemetery The evidence from the Cazzanello cemetery and from numerous other central Italian burial areas that include only infants (Fig. 3) provides evidence for the age at which children become adults. The Cazzanello cemetery provides direct archaeological evidence for the use of separate cemeteries for perinatals, a situation suggested by the absence of certain age groups from among the bones recovered from the several principal cemeteries at Tarquinia and elsewhere. Archaeologists involved in cemetery excavations at Tarquinia had not previously noted the absence of perinatal burials; a situation that would have offered clues of use in answering two important questions. First, we must ask when did members of a specific culture believe that life "began?" Second, we must apply the archaeological evidence to consider what the specific mortuary patterns tell us about the way in which these people define themselves, and how did they relate the members of their group to the "cosmos."

Examination of data relating to the age of this transition, and the separate placement of infant burials as distinct from other members of society in different cultures, reveals significant regional variations and provides markers for delineating specific regions. These mortuary data should complement the evidence for cultural boundaries that is derived from regional pottery styles and other more commonly used means for recognizing boundaries between ancient peoples.

Studies of skeletons from several extensive excavations in cemeteries have found only the remains of adults, plus children above the age of weaning. The two primary examples noted are the cemeteries from Etruscan Tarquinia and Punic Marsala (Becker 2005a; in press B), with the absence of perinatals providing indirect evidence

Data on selective placement of the dead by age or sex help clarify problems relating to the age structure of a population. There are a number of challenges distorting our ability to use the dead to reconstruct the lives of the living population. These include mortuary customs that 29

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD distort the skeletal record, such as cremation, selective cremation, and intra-cultural differentials in mortuary customs for perinatals or other members of the society. In combination with social class differences in the mortuary rites, these cultural rules provide an understanding of population dynamics. Perinatal burials at Tarquinia: Etruscan religious beliefs

Inferences

Confirmation of the Etruscan burial thesis: Parallel data from Etruscan Blera In the small sample of skeletons recovered and evaluated from the Etruscan city of Blera (see Table 3), the seven children ages 5.4-15 years form 15.5 % of the available population (Becker 2004). This percentage appears high, but it is well within the range of expected mortality figures for this age group. These percentages for mortality rates for children in the range of ca. 5.0 to 16.0 years of age are similar to those inferred from the Isola Sacra (Imperial Roman) cemetery populations. At Blera, as at Tarquinia, no children under the age of five years were interred in the adult cemetery (table 4).

on

Every culture has its own complex set of beliefs and rituals surrounding birth, including a specific mode of treatment for stillbirths and those who die soon after birth. The amount of time that a neonate remains in this category is a cultural belief that can be identified by archaeological research and subsequent skeletal analysis. Interesting cultural variations in these beliefs enable us to use archaeology to define cultural boundaries or borders, as they existed in the past. The evidence indicates that Etruscans buried perinatals in specific areas associated with water sources, as distinct from the cemetery areas used for all other members of the community (Becker 1997a; 2007b). This placement differs considerably from that of the Archaic period Romans, who buried perinatals outside their houses in a pattern identified as "suggrundaria," or "sub-grundaria" (beneath the eaves; see Becker 1994a; 1994b; 1996a). A seven-year program of study (1987-1995) focussing on the skeletal material recovered after 1981 from several of the cemetery areas of Tarquinia (Becker 1990; 1993; 2000) provides a significant skeletal database from a major South Etruscan city. During this study more than 200 individuals were identified among the bones retrieved from approximately 77 tombs of all sizes (Becker Ms. C). Two points of interest emerge from the data from Tarquinia. First, perinatals are not represented at all. Second, sub-adults (children 5.5 to 16.5 years of age) are represented in normal numbers, as an expected percentage of the total population (Becker 2005b). These percentages apply equally to cremations as well as to inhumations. Only four of the 23 cremations excavated at Tarquinia after 1981, or 17% of the individuals evaluated, are the remains of sub-adults, with the youngest dying at age 13 years. These four include a cluster of three small tombs (T. 6203-6205) each holding the cremation burial of a male, and all three being between 13 and 16 years of age, may reflect another interesting aspect of mortuary ritual at Tarquinia. Perkins (pers. comm. 2004) also notes that these burials at Tarquinia predate the point at which the Romans came to dominate, but that Etruscan perinatal mortuary programs do not appear to be influenced by the Roman conquest. While the boundaries and traditions may not yet be clear, these observations regarding a largely ignored part of daily life among the Etruscans and help us to formulate questions that require new considerations in the process of conducting excavations in this and other parts of Italy.

Tomb

No of People

Adults

Children

1 3

13 11

13 8

0 3

4

11

8

3

7 9 N=5

8 2 N=45

7 2 N=38 (84%)

1 0 N=7 (16%)

Locus of child 2 4 10 9 10B 14 4 -

Age of child

Sex of child

7.5 5.6 5.4 15 8 7.5 15? Range 5.4-15

? F M F ? F? ? F=3; M=1; ?=3

Table 3. Children aged 16.5 years and younger identified in five tombs at the Casacce Necropolis at Blera (Viterbo) excavated by G. Barbieri (data extracted from Becker 2004). Number of People

Tarquinia Blera

Age in Years 0-5 0 0

Percentage 5.5-16.6 25 7

9.6% 15.5%

Table 4: Sub-adult populations at Blera compared with those from Tarquinia (from Becker 2004). Discussion: The bones of children Sophisticated techniques have been used for the evaluation of cemetery populations for more than 20 years (e.g. Orton & Hodson 1981, Chapman et al. 1981), but the presence of infants, or their absence, has been consistently ignored. The absence of infant remains had previously been explained by a number of theories of which the principal are listed here: 1. 2. 3.

30

Decay Theory: fragile remains are consumed by acid soil or other destructive soil conditions. Differential Recovery Theory: Excavation techniques miss small graves. Differential Mortuary Processes: The use of cremation or other treatments of infants lead their bones to be “absent” from the archaeological record.

ETRUSCAN INFANTS: CHILDREN’S CEMETERIES AT TARQUINIA, ITALY AS INDICATORS OF AN AGE OF TRANSITION identifies them (cf. Becker 1997b). Rotroff (1998; 1999) relates the Athenian rite of amphidrómia that took place on the fifth or seventh day after birth, as marking a significant transition to becoming an official member of a family. The amphidrómia in Athens involved carrying a viable child at age five or seven days, around the hearth and bestowing a name at that time. This formal ritual announced the official entry of the infant into the family. Rotroff suggests that children who died before this passage may not have been granted a formal burial, and the archaeological evidence confirms her insight.

I propose a fourth scenario, involving the use of separate cemeteries for infants. Separate infant cemeteries (burial areas) recently have been identified among several of the many cultures archaeologically known from the region that now is modern Italy. The best known are the tophets used by Carthaginians, which have been erroneously interpreted as reflecting sacrifice, rather than special concern for these tiny bodies (see Becker In review A). I agree with Philip Perkins (pers.comm. 2004) that tophets are "remarkable" in that they "'monumentalize' infant mortality in a way that other Mediterranean societies did not."

Traditions relating to sub-floor or near-house burials are documented from Iron Age and Roman Italy (see Jarva 1981a & b; Becker 1996). Perinatals found buried beneath or around houses, and the discovery of perinatal cemeteries, are rarely reported in the literature (but see Soren 1999a and b). Various Italian excavators with whom I have spoken noted that sub-floor burials are rarely noted in published reports. Infant sub-floor burials are known from many Greek excavations, but noted infrequently in the literature. John Younger (pers. comm. 2003) described the discovery of six perinatal burials beneath the floor of a Middle Helladic house in the southern Pelonponessus, but the data remain unpublished. It may be that the burial of bodies of Greek infants below house floors enabled their spirits or souls to be reborn into the women of the same family (Becker 1996; cf. Morris 1992), and the same may have been the case with Roman suggrundaria burials. Ethnographic analogies as well as the ancient literary records may help to clarify the cultural "meaning" of this burial practise. The evidence for sub-floor perinatal burials in Greece benefits from careful culling of the literature (cf. Nevett 1999; Becker 2001; Lagia 2003; see also Becker 1999).

The recognition of culture-specific rules for the placement of infant bodies, often linked with the use of specific types of pottery or other burial containers, is now seen as likely (cf. Becker 2005a). Many cemeteries specifically used for perinatals and infants are known from archaeological contexts and ethnographic observations in various parts of Italy as well as from other parts of the Roman Empire. These specialized cemeteries range in date from the Iron Age to the 20th century, although the preponderance date from the "Late Roman" period. Mortuary customs in general tend to reflect the ancient (or "traditional") local patterns, with the possible exception of elite burials. The elite may include grave goods and elements of form or arrangement that reflect political domination or influence, and such elaborate tombs generally are better documented by archaeology than the tombs of infants in which cultural traditions are less likely to be influenced by foreign influences. Mortuary rituals at Tarquinia and in other cultural areas of Italy during the period from ca. BCE 700 to ca. CE 600 involved the disposal of infant bodies using locations distinct from those that were applied to the remains of adults. Specific details differed from culture to culture throughout the Italian peninsula and on the major islands nearby. In some cases those born dead were treated in the same fashion as those who were born live but who died within days after birth. Both categories may also have been interred in the same way as all children who died before their second year of life, or before weaning (Becker 2007a). In fact, a number of post-natal developmental stages may be recognized in the archaeological, as well as the ancient literary records (Becker 1994d). These ancient stages reflect, in some ways, the recognition by modern physicians of fetal trimesters as separate developmental stages.

Philip Perkins (pers. comm. 2004) indicates that in the Roman region "conventional wisdom has it that juveniles are under represented and were not buried in adult cemeteries." He notes that this belief supports various interpretations of how Roman males become adult, or full citizens, and that this idea "is implicitly accepted for other reconstructions of Etruscan societal development based upon mortuary data." The evidence from Tarquinia indicates that the percentage of the population that died between 5.5 and 16.5 years of age is approximately 10%, or normal for any pre-industrial people (Becker Ms. C). Nearly identical percentages have been independently discovered by physical anthropologists working with the skeletons from Tarquinia (Mallegni et al. 1980). Similar percentages are found in the sample from Blera, another South Etruscan site (Becker 2004), and from Rome (Calza 1940, 268-269). These examples of direct evidence suggest that historians and others studying the social histories of Rome and Etruria might wish to consider how the biological evidence relates to their interpretations of the written records.

The study of infant burial practices provides another and more reliable means by which cultures may be differentiated within Italy. The same approach may be used within various parts of Magna Graecia, particularly to differentiate between the cemeteries used by Greek colonists and those of the native peoples. The use of fine ceramics, bronze and iron tools, and a host of other material goods that may be purchased or traded widely may not be useful in delineating cultural borders. Trade in some types of goods renders difficult the ability to define cultural or ethnic borders, as Lomas (1996)

Each traditional culture commonly employs some rite of passage to delineate the passage of a newly born child to "life," or to the point where it has a higher chance of survival, although these transitions may be difficult to 31

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD "Constructions of Childhood in the Ancient World" (Hood Museum, November 2003). My sincere thanks are due Professor Jeremy Rutter and Professor Ada Cohen for their invitation to contribute to that program. Special thanks are due Dr. Maria Cataldi and Dr. Gabriella Barbieri for their continued concern with the recovery and evaluation of human and other skeletal remains from their several excavations, and to Dr. Philip Perkins (The Open University) for his many helpful suggestions and careful reading of an earlier version of this manuscript. My sincere thanks also are due Prof. R. Ross Holloway, Dr. Elena Lattanzi, A. Callucci (Matera Museum), Dr. Pedar W. Foss, Dr. Helle Horsnaes, Dr. Stephen D. Jones, Dr. John K. Papadopoulos, Dr. Leonart Rutgers, Prof. Susan Rotroff, Prof. Jeremy Sabloff, and Dr. Ann M. E. Haentjens. Thanks also are due to several helpful colleagues who provided me with reprints and those who assisted me in processing these data, including Dott.ssa Marina Voluntè and Prof. B. Chiarelli, and in Cremona to Dr. Daniele Generali and his many kind associates. My thanks also are due F. P. and M. E. Gillon for their significant contributions to this effort, and to an anonymous BAR reader who provided extremely useful editorial advice. This manuscript was prepared while the author was a Consulting Scholar in the Mediterranean Section of the University Museum of the University of Pennsylvania. My thanks are due Dr. Donald White, then Curator of that section, and Dr. Richard Leventhal, Williams Director of the Museum, for their support and encouragement of this research. Thanks also are due D. B and E. Tyler. The ideas expressed in this paper, as well as any errors of presentation or interpretation, are solely the responsibility of the author.

recognize in the archaeological record (Becker 1986). When we have excavated all or a significant portion of a cemetery, and have a complete absence of perinatal remains within the known burial areas, we may infer that these small bodies were treated in some other way. I believe that the "presence" of separate infant cemeteries at such sites is one of the more probable explanations for the absence of perinatals in the principal burial area. Even when perinatal graves are identified, the general absence of grave goods does not encourage the publication of this scanty information. The importance of these simple data in the reconstruction of the entire culture is now recognized, and more publications of these data should soon be available. Conclusions 1. Infant "cemeteries" at Tarquinia may be inferred from the evidence for separate burial areas for perinatals, such as those found elsewhere in Etruria. At Tarquinia the absence of remains representing children below the age of 5.5 years from the principal cemeteries suggests that a separate means of disposal were used for the remains of very young children. These infant cemeteries at Tarquinia may be located some distance from the principal burial areas used by the population, and I believe they will be found in association with local water sources or springs. 2. The burials now identified from tombs at Tarquinia indicate that approximately 10% of the population died between the ages of 5.5 and 16.5 years of age. This percentage is consistent with rates found elsewhere in Etruria, as well as with our expectations of mortality profiles from pre-industrial societies. All children in this age range at ancient Tarquinia were afforded a place of burial within the cemeteries used for adults in this population.

References Angel, J. L. 1945. Skeletal material from Attica. Hesperia 14: 279-363. Angel, J. L. 1969. Paleodemography and evolution. American Journal of Physical Anthropology 31: 343-353.

3. High infant mortality rates were normal in antiquity. Cultural rules that cluster perinatals and small children in specialized cemeteries should not be taken as indicators of human sacrifice or infanticide. Specialized infant cemeteries should be recognized as providing important insights into each culture's views towards life and death as a part of a normal cycle.

Angel, J. L. 1971. The People of Lerna; Analysis of a prehistoric Aegean population. Princeton, NJ: American School of Classical Studies at Athens. Aristotle (384-322 BCE). Historia Animalium Book VII. Translated by A. L. Peck, 1965. Cambridge, Mass.: Harvard University Press.

4. Earlier I erroneously suggested that specialized cemeteries for children were introduced to Italy as a Christian custom (Becker 1995a, 2005a). The evidence now available suggests that the use of special cemeteries for children is a pre-Christian trait observable in Etruria as early as the 7th century BCE. Disposal of perinatals outside the traditional cemeteries is suggested for Tarquinia and probably other Etruscan cities, and is a trait known from some other regions in Italy.

Bartoloni, G. 1989. La Cultura Villanoviana: all'inizio della storia Etrusca. Roma: La Nuova Italia Scientifica. Bassett, S. (ed.) 1995. Death in Towns: Urban Responses to the Dying and the Dead, 100-1600. Leicester: Leicester University Press. Becker, M. J. 1983. Children's Burials in Puglia from the Iron Age to the IInd Century A.D: Cultural Continuities. Studi di Antichità (Lecce) 4: 261-284.

Acknowledgements: Many thanks are due Michael Lally for his kind invitation to contribute to this volume. Portions of this paper were presented at the Dartmouth College symposium 32

ETRUSCAN INFANTS: CHILDREN’S CEMETERIES AT TARQUINIA, ITALY AS INDICATORS OF AN AGE OF TRANSITION Becker, M. J. 1999. Human Skeletons from the Greek Emporium of Pithekoussai on Ischia (NA): Culture Contact and Biological Change in Italy after the 8th Century BC. In R. H. Tykot, J. Morter and J. E. Robb (eds.) Social Dynamics of the Prehistoric Central Mediterannean. London: Accordia Research Centre, University of London, 217-229.

Becker, M. J. 1986. Mandibular Symphysis (Medial Suture) Closure in Modern Homo sapiens: Preliminary Evidence from Archaeological Populations. American Journal of Physical Anthropology 69: 499-501. Becker, M. J. 1990. Etruscan Social Classes in the VI Century B.C.: Evidence from Recently Excavated Cremations and Inhumations in the Area of Tarquinia. In, H. Heres and Max Kunze (eds.), Die Welt der Etrusker. Berlin: Akademie-Verlag, 23-35.

Becker, M. J. 2000. Reconstructing the Lives of South Etruscan Women. In A. E. Rautman (ed.), Reading the Body: Representations and remains in the archaeological record. Philadelphia: Univ. of Pennsylvania Press, 55-67.

Becker, M. J. 1994a. Suggrundaria and Roman Villas: A Note on an Infant's Burial at Satricum, Lazio, Italy. Old World Archaeology Newsletter 17 (2): 7-10.

Becker, M. J. 2001. Review of House and Society in the Ancient Greek World, by Lisa C. Nevett (1999). American Journal of Archaeology 105 (3): 552-554.

Becker, M. J. 1994b. A Note on Children's Burials in Central Italy: The 1978 Excavations at a Late Roman Sepulchral Building at the Fourth Milestone along the Via Appia (proprietà Lugari) Near Rome. Archaeological News 19: 15-17.

Becker, M. J. 2002. Etruscan Tombs at Tarquinia: Heterarchy as indicated by Human Skeletal Remains. In N. Negroni Catacchio (ed.) Paesaggi d'Acque - Ricerche a Scavi. Preistoria e Protostoria in Etruria. Milano: Onlus, for the Centro Studi di Preistoria e Archeologia. Vol. II, 687-708.

Becker, M. J. 1994c. "An Infants' Cemetery, Possibly from the Late Antique Period, as an Indication of Changes in Mortuary Practices from Roman to Christian Times: An Evaluation of Children's Skeletons at the Villa Romana a Cazzanello (near Tarquinia). Excavated by a Team from the Institute for the Study of Cultural Exchange of the University of Tokyo." Revised report submitted 5 Sept. 1994 to Prof. M. Aoyagi.

Becker, M. J. 2004. Etruscan Skeletons of the Hellenistic period from the Casacce necropolis at Blera (Viterbo, Italy) excavated in 1982. Appendix II, pages 175-189. In Blera (Viterbo). Località Casacce – Necropoli rupestre di età ellenistica, by Gabriella Barbieri. Notizie degli scavi di Antichità XIII-XIV (2002-2003), 89-190.

Becker, M. J. 1994d. "When Does Life Begin? Age Evaluation of Perinatal and Infant Skeletons as a Basis for the Recognition of Cultural Change in Italy during the Late Roman to Early Christian Period." Paper delivered at the Colloquium "Science in Archaeology," Harvard University (Cambridge, MA: 15 October). Robert Tykot, organizer.

Becker, M. J. 2005a. The Cazzanello Perinatal Cemetery near Tarquinia: Continuities of Etruscan mortuary practices into the Late Antique Period and Beyond. Studi Etruschi LXX (2004), 255-267. Becker, M. J. 2005b. Etruscan Women from Tombs at Tarquinia: Skeletal evidence for tomb use. Analecta Romana Instituti Danici 31, 21-36.

Becker, M. J. 1995a. Human Skeletal Remains from the Pre-Colonial Greek Emporium of Pithekoussai on Ischia (NA): Culture Contact in Italy from the Early VIII to the II Century B.C. In Neil Christie (ed.) Settlement and Economy in Italy 1500 BC to AD 1500. Leicester, UK. Oxbow Monograph 41, 273-281.

Becker, M. J. 2006a. The Archaeology of Infancy and Childhood: Integrating and Expanding Research into the Past. [Review essay]. American Journal of Archaeology 110: 655-658.

Becker, M. J. 1995b. Infanticide, Child Sacrifice and Infant Mortality Rates: Direct Archaeological Evidence as Interpreted by Human Skeletal Analysis. Old World Archaeology Newsletter 18 (2): 24-31.

Becker, M. J. 2006b. Etruscan Women at Tarquinia: Skeletal Evidence for Tomb Use, in C. C. Mattusch, A. A. Donohue and A Brauer (eds.) Common Ground: Archaeology, Art, Science, and Humanities. Proceedings of the XVIth International Congress of Classical Archaeology. Boston, August 23-26, 2003. Oxford: Oxbow Books, 292-294.

Becker, M. J. 1996. Human Skeletal Remains recovered from the Ficana excavations. In, J. R. Brandt (ed.), Scavi di Ficana, 2 (I): Il Periodo Protostorico e Arcaico. Rome: Istituto Poligrafico e Zecco dello Stato, 453-479.

Becker, M. J. 2007a. Childhood among the Etruscans: Mortuary Programs at Tarquinia as Indicators of the Transition to Adult Status, in A. Cohen and J. B. Rutter (eds.) Constructions of Childhood in Ancient Greece and Italy. Hesperia, Supplement 41, 281-292.

Becker, M. J. 1997a. Archaeological Evidence for When Life Begins. Anthropology Newsletter 38 (5): 13-14. Becker, M. J. 1997b. Perinatal and Other Burial Patterns in Lucania: Evidence from Collina San Salvatore at Timmari, Italy During the 7th-6th Centuries BC. Old World Archaeology Newsletter 20 (3): 1-12.

Becker, M. J. 2007b. [II.4.2]. The anthropological analysis of the human skeletal remains: Tombs 6008, 6093 and 6092. Suspected Tomb, in H. Horsnaes “The 33

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Chiaramonte Treré, C. 1987. Altri dati dagli scavi alla civita sugli aspetti culturali e rituali. In, M. Bonghi Jovino and C.Chiaramonte Treré (eds.). Tarquinia: Ricerche, Scavi e Prospettive. Atti del Convegno "La Lombardia per gli Etruschi." Milano: Edizione ET, for "L'Erma" di Bretschneider, 79-89.

Excavations” in B. Tang (ed.) Hellenistic and Roman Pontecagnano: The Danish Excavations in Proprietà Avallone 1986-1990. Naples: Centre Jean Bérard, 50-52. Becker, M. J. In press A. Quaker Farmsteads in the Delaware Valley: Research at the Taylor Burying Ground Site, Chester County, PA (36CH117): A "Farm" Cemetery of the 18th and 19th centuries. Bulletin of the Archaeological Society of New Jersey 64 [2011].

Chiaramonte Treré, C. 1995. Seppellimenti in Abitato: Il caso di Tarquinia. In N. Negroni Catacchio (ed.) Preistoria e Protostoria in Etruria, Volume 1: Tipologie delle Necropoli e Rituali di Deposizione. Milano: Edizioni Et, 241-248.

Becker, M. J. In press B. Tophets as Perinatal Cemeteries among Punic Colonists: The evidence from Lilybaeum (Marsala), Sicily. Sicilia Archeologica.

Chiaramonte Treré, C. (ed.) 1999. Tarquinia. Scavi sistemmatici nell'abitato. Campagne 1982-88. I Materiali. Roma: "L'Erma" di Bretschneider.

Becker, M. J. Ms. A. Suggrundaria: Burial Practices Relating to Children and Infants in the Roman World. Manuscript in circulation.

Chapman, R. Kinnes, I. and Randsborg, K. (eds.). 1981. The Archaeology of Death (New Directions in Archaeology). New York: Cambridge University Press.

Becker, M. J. Ms. B. Tophets and Cemeteries in the Punic World: The evidence from a large skeletal sample from Lilibaeum (ancient Marsala) Sicily.

Clayton, F., Sealey, J. and Pfeiffer, S. 2006. Weaning Age among Foragers at Matjes River Rock Shelter, South Africa from Stable Nitrogen and Carbon Isotope Analysis. American Journal of Physical Anthropology 129 (2): 311-317.

Becker, M. J. Ms. C. An Analysis of Human Skeletal Remains Recovered from Excavations at Tarquinia between 1981 and 1995, together with a summary of previous research. A master list (October 2002) updating an earlier report of 1997. Manuscript prepared for M. Cataldi, Soprint. Archeologica per Etruria Meridionale.

Dawes, J. D. and Magilton, J. R. 1980. The Cemetery of St Helen's-on-the-Walls, Aldwark. Fascicule 1 of the Archaeology of York, Volume 12 (The Medieval Cemeteries). York Archaeological Trust.

Bello, S. M. 2004. Taphonomy, selective preservation and robusticity in human skeletal samples: The osteometric paradox. [Abstract]. American Journal of Physical Anthropology. Supplement 38, 61-62.

Durand, J. D. 1960. Mortality Estimates from Roman Tombstone Inscriptions. American Journal of Sociology 65 (4): 365-373.

Bloomfield, F. H., Oliver, M. H. and Hawkins, P. 2003. A Periconceptual Nutritional Origin for Noninfectious Preterm Birth. Science 300: 606.

Engels, D. 1980. The Problem of Female Infanticide in the Greco-Roman World. Classical Philology 75: 112120.

Bonghi Jovino, M. 1987. Gli Scavi nel'Abitato di Tarquinia e la scoperta dei "Bronzi" in un preliminare inquadramento. In, M. Bonghi Jovino and C. Chiaramonte Treré (eds.) Tarquinia: Ricerche, Scavi e Prospettive. Atti del Convegno "La Lombardia per gli Etruschi". Milano: Edizione ET, 59-77.

Eyben, E. 1991. Fathers and Sons. In B. Rawson (ed.) Marriage, Divorce and Children in Ancient Rome. Oxford: Clarendon Press for the Humanities Research Centre, Canberra, 114-143.

Bonghi Jovino, M. and Chiaramonte Treré, C. (eds.) 1987. Tarquinia: Ricerche, Scavi e Prospettive. Atti del Convegno "La Lombardia per gli Etruschi". Milano: Edizione ET, for "L'Erma" di Bretschneider.

Fuller, B. T., Fuller, J. L., Harris, D.A. and Hedges, R. E. M. 2006. Detection of Breastfeeding and Weaning in Modern Human Infants with Carbon and Nitrogen Stable Isotope Rations. American Journal of Physical Anthropology 129 (2), 279-293.

Bourbou, C. 2000. Infant Mortality: The Complexity of It All! Eulimene II, 187-202.

Godden, J. and Godden, R. 1966. Two under the Indian Sun. New York: Alfred A. Knopf.

Burnston, S. A. 1982. Babies in the well: an underground insight into deviant behavior in eighteenth-century Philadelphia. The Pennsylvania Magazine of History and Biography 106, 151-186.

Gowland, R. L. and Chamberlain, A. T. 2002. A Bayesian Approach to Aging Perinatal Skeletal Material from Archaeological Sites: Implications for the Evidence for Infanticide in Roman-Britain. Journal of Archaeological Science 29, 677-685.

Calza, G. 1940. La Necropoli del Porto di Roma Nell'Isola Sacra. Rome: La Libreria dello Stato, for the Reale Istituto di Archeologia e Storia dell'Arte.

Greene, J. A. 2003. "Infanticide in the Ancient Mediterranean World." Paper presented at the Hood 34

ETRUSCAN INFANTS: CHILDREN’S CEMETERIES AT TARQUINIA, ITALY AS INDICATORS OF AN AGE OF TRANSITION Miles, D. (ed.). 1984. Archaeology at Barton Court Farm, Abingdon, Oxon. Oxford Archaeological Unit Report 3 (CBA Research Report 50). Oxford Archaeological Unit and the Council for British Archaeology.

Museum Conference, Constructions of Childhood in the Ancient World (November, Dartmouth College). Hanson, A. E. 1994. A Division of Labor. Roles for Men in Greek and Roman Births. Thamyris. Myth Making from Past to Present 1 (2), 157-202.

Morris, I. 1992. Death-Ritual and Social Structure in Classical Antiquity. Cambridge: Cambridge University Press.

Herring, D. A., Saunders, S. R. & Boyce, G. 1994. Bones and Burial Registers: Infant Mortality in a 19th-Century Cemetery from Upper Canada. Northeast Historical Archaeology 20 (1991), 54-70.

Nevett, L. C. 1999. House and Society in the Ancient Greek World. Cambridge: Cambridge University Press.

Hooper, B. 1975. The Human Bones. In B. W. Cunliffe, Excavations at Portchester Castle, I: Roman. London: Society of Antiquaries Research Report 32, 375-377. Jackes, M. K. 1993. On Paradox and Osteology. Current Anthropology 34 (4), 434-439.

Orton, C. R. and Hodson, F. R. 1981. Rank and Class: interpreting the evidence from prehistoric cemeteries. In S. C. Humphreys and H. King (eds.) Mortality and Immortality: the anthropology and archaeology of death. Academic Press: London, 103-115.

Jarva, E. 1981a. Tombe infantile nel Lazio e a Ficana nell'età del ferro e in epoca arcaica. In, Ficana: Una Pietra Miliare sulla Strada per Roma [exhibition catalogue]. Rome: Libreria editrice, 141-147.

Plutarch (Lucius Mestrius Plutarchus, c.46-120CE). The Life of Numa, in The Parallel Lives. Cambridge, Mass.: Harvard University Press/The Loeb Classical Library, Volume 1 (1914), 307-393.

Jarva, E. 1981b. Area di Tombe infantile a Ficana. Archeologia Laziale (Quaderni del centro di studi per l'archeologia etrusco-italica) 5: 269-272.

Pomeroy, S. B. 1997. Families in Classical and Hellenistic Greece: Representations and Realities. Oxford: Clarendon Press.

Kelly, N. U. 1975. Some Socio-cultural Correlates of Indian Sex Ratios: Case studies of Punjab and Kerala. The University of Pennsylvania: Philadelphia: The University of Pennsylvania: Doctoral dissertation.

Prayon, F. 1975. Frühetruskische Grab- und Hausarchitektur. Ergäzungsheft 22 (of the Mitteilungen des Deutsches Archaeologisches Institut, Römische Abteilung). Heidelberg: F. H. Kerle Verlag.

Konigsberg, L.W. and Frankenberg, S. R. 1992. Estimation of Age Structure in Anthropological Demography. American Journal of Physical Anthropology 89, 235-256.

Rawson, B. 1991a. Introduction. In B. Rawson (ed.) Marriage, Divorce and Children in Ancient Rome. Oxford: Clarendon Press for the Humanities Research Centre, Canberra, 1-5.

Lagia, A. 2003. "Notions of Childhood in the classical polis: Evidence from the bioarchaeological record." Paper presented at the Hood Museum Conference, Constructions of Childhood in the Ancient World (November, Dartmouth College).

Rawson, B. 1991b. Adult-Child Relations in Roman Society. In, B. Rawson (ed.). Marriage, Divorce and Children in Ancient Rome. Oxford: Clarendon Press for the Humanities Research Centre, Canberra, 7-30. Ridgway, D. 1992. The First Western Cambridge: Cambridge University Press.

Little, L M. 1999. Babies in Well G5:3: Preliminary Results and Future Analysis [Abstract]. American Journal of Archaeology 103, 284.

Greeks.

Rotroff, S. 1998. "The Bone Well (G5:3) in the Northern Kolonos Agoraios of Athens: Artifacts and some conclusions." Paper presented at the Annual Meeting of the Archaeological Institute of America.

Lomas, K. 1996. Greeks, Romans and Others: problems of colonialism and ethnicity in southern Italy. In J. Webster and N. Cooper (eds.) Roman Imperialism: PostColonial Perspectives. Leicester: Leicester Archaeological Monographs, 135-144

Rotroff, S. 1999. The Artifacts from Well G5:3 and Some Conclusions concerning the deposit [Abstract]. American Journal of Archaeology 103, 284-285.

Mallegni, F., Fornaciari, G. & Tarabella, N. 1980. Studio antropologico dei resti scheletrici della necropoli dei Monterozzi (Tarquinia). Atti della Società Toscana di Scienze Naturali, Memorie (for 1979), Pisa. Series B, Volume 86, 185-221.

Rousseau, J.-J. 1762. Émile, ou De l'éducation (in 4 volumes). A La Haye: Chez Jean Néaulme, librarie [Paris: Duchesne]. Russell, J. C. 1958. Late Ancient and Medieval Population. Transactions of the American Philosophical Society 48 (3). Philadelphia.

Mays, S. 1993. Infanticide in Roman Britain. Antiquity 67, 883-888. 35

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD

Soren, D. 1999a. Introduction to Part Three [The Infant Cemetery, pages 461-651]. In, D. Soren and N. Soren (eds.). A Roman Villa and a Late Roman Infant Cemetery. Excavation at Poggio Gramignano, Lugnano in Teverina. Bibliotheca Archaeologica 23. Rome: "L'Erma" di Bretschneider, 463-464. Soren, D. 1999b. The Infant Cemetery at Poggio Gramignano. In D. Soren and N. Soren (eds.). A Roman Villa and a Late Roman Infant Cemetery. Excavation at Poggio Gramignano, Lugnano in Teverina. Bibliotheca Archaeologica 23. Rome: "L'Erma" di Bretschneider, 477-530. Soren, D., Fenton, T. and Birkby, W. 1999. The Infant Cemetery at Poggio Gramignano: Description and Analysis. In, D. Soren and N. Soren (eds.). A Roman Villa and a Late Roman Infant Cemetery. Excavation at Poggio Gramignano, Lugnano in Teverina. Bibliotheca Archaeologica 23. Rome: "L'Erma" di Bretschneider, 477-530. Steel, T. 1988. The Life and Death of St. Kilda. Revised from the 1975 edition (first publ. in 1965). London: Fontana. Still, G. F. 1931. The History of Paediatrics; the Progress of the Study of Diseases in Children up to the End of the XVIIIth Century. London: M. Milford for Oxford University Press. Vargiu, R. and Becker, M. J. 2005. Appendice: Studio Antropologiche dei Resti Scheletrici Umani. [pp409-411] in M. Cataldi “Sulle ‘Tombe a Buca’ di Tarquinia,’ Dinamiche di Sviluppo delle Città nell’Etruria Medidionale: Veio, Caere, Tarquinia, Vulci (Atti del XXIII Convegno di Studi Etruschi ed. Italici, Oct. 2001, Roma, Veio, etc.). Pisa e Roma: Istituti Editoriali e Poligrafici Internazionali, 395-411.

36

Thrown Out with the Bathwater or Properly Buried? Neonate and Infant Skeletons in a Settlement Context on the Dürrnberg bei Hallein, Austria Raimund Karl¹ and Klaus Löcker² ¹ Bangor University ² ZAMG - Central Institute for Meteorology and Geodynamics Ludwig-Boltzmann-Institute for Archaeological Prospection and Virtual Archaeology societies we draw on belong to the so called IndoEuropean family of languages and cultures (Mallory 1991, 66-109), this assumption seems reasonably justified, and allows to employ “Galton’s problem” (Krauße 1999, 349) to our advantage. It also seems justified to assume this since at least the early La Tène period, much of Central and Western Europe was part of an information exchange network, which will have allowed for even stronger similarities to develop (Karl 2005). This provides a particular advantage over ordinary ‘ethnographic analogies’ (Eggert 1988, 265) and over the ‘African analogies’ which ‘may well be as relevant, perhaps more relevant, than the Irish sources of a millennium later’ (Collis 1994, 31). In the case of our sources, a cultural comparability of the sources is based on proven similarity (in e.g. the linguistic record), not just assumed comparability based on (alleged) similar levels of technological and social evolution, as is typical for the use of ‘ordinary’ ethnographic analogies (Bernbeck 1997, 104-8), though the latter alone would, arguably, allow for valid analogies to be drawn from the European Early Medieval period to create models for much of the European Iron Age (Karl 2006). As such, unless one assumes a priori and ex silentio that different European societies necessarily differ more fundamentally from each other than from any other society of roughly equivalent technological or sociological development elsewhere, the use of European sources is at least as relevant, and probably even more relevant, than the usual ethnographic analogy.

Introduction In the Central European Iron Age, infant burials are largely absent from regular cemeteries, while infant skeletons frequently appear in settlement contexts (Stöllner 1999, 11; Burmeister 2000, 74-8). In the past, this has led to arguments that these infants were simply carelessly disposed, literally thrown out with the trash (for a discussion see Wahl 1994). While such a practice of carelessly deposing of dead infants can not be ruled out, as a general explanation, this seems to us like throwing out the baby with the bathwater. Based on the features excavated on the Ramsautal settlement on the Dürrnberg bei Hallein (Stöllner 1999), we would like to propose a different explanation. To arrive at this alternative explanation, we will mainly concentrate on non-archaeological European source material, some of which comes from considerably later periods, while much is drawn from other Iron Age European societies. In using these sources we do, to some extent, commit the ‘sin’ of eurocentrism, and we are fully aware of the risks involved in trying to apply information gathered from historical sources describing different but contemporary, or possibly related, but much later European societies to much earlier societies in a different area of Europe (for a criticism of such approaches, especially the use of ‘Celtic’ sources, see Collis 1994; 2003). Nonetheless, we think that such sources can and should be used as analogies, much like ethnographic parallels would be, to create possible models of Iron Age Central European belief systems and societies. Of course, such models cannot be seen as necessarily accurately describing Iron Age societies in Central Europe. However, they can be tested against the evidence to assess their viability (von Glasersfeld 1992, 18-31), and it can thus be established whether they should be seen as one possible explanation of the evidence (amongst other possibilities, arrived at by using different sources), or have to be discarded.

With this in mind, and with the aim of reconstructing Iron Age belief systems regarding infant burial, we want to look at our specific case, the infant skeletons found in settlement contexts on the Dürrnberg bei Hallein near Salzburg in Austria. The features Due to the planned construction of drainage for the Glannerbach, excavations became necessary in the Ramsautal on the Dürrnberg bei Hallein in 1988, to rescue settlement features located by, but not excavated, during earlier works (Moosleitner and Penninger 1965; Zeller 1984a; 1984b). In two campaigns until 1989 an area of 230 m² was examined. Because of the wetland character of the site excavation methodology was based on experiences taken from southern German and Swiss wetland settlement site research. The wetland conditions have resulted in excellent preservation of organic materials, especially the manmade Iron Age drainage ditches which divided the settlement site into artificial

Our choice of sources to build our alternative model is based on the assumption that cultural analogies or even homologies have already been established between some of our sources and the Central European Iron Age (e.g. in the field of linguistics), increasing the statistical probability that other cultural analogies or homologies will have existed as well (Karl 2004a; 2004b). It is also based on the assumption that most of the cultures we use as providers of source material were either closely related to each other, or at least had close contact with each other through long periods of European prehistory, and thus share many similar cultural practices. As most of the 37

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD the Lt B1/B2 transition, six to Lt B2 and one to Lt C (Stöllner 1999, 8). The total period of infant deposition in this occupation zone therefore lasted for roughly 150 years.

‘islands’ and contained high amounts of well-preserved organic finds. The analysis of the features (Stöllner 1991; 1996, 22835) and the construction of a stratigraphic sequence (Harris-Matrix) established 9 settlement layers with a total number of 7 partially multi-phased occupation sites, with a total of 24 building phases. Occupation lasted from Dürrnberg (D) II A (= Lt A) to D II C (= Lt C), with only one occupation site – house 3 – lasting into the final phase of occupation. No late Hallstatt period features were discovered. Most of the buildings were built with wooden foundation beams and clay floor layers, partially of substantial thickness.

During the analysis of the features it became apparent that those infant skeletons had been found exclusively in the context of building features. Some skeletons or skeletal fragments were found in the entrance area of buildings (Skeleton 2 and 3 in house 5a – fig. 1; Skeleton 9 in house 3b – fig. 2) or along the building axis (Skeleton 1 – fig. 1). Others were located below the floor layers (Skeletons 5, 6, 7 and 8 in house 3b – fig. 2), between house and drainage ditches (Skeleton 4 in house 3b – fig. 2) or within the drainage ditches surrounding a house (Skeleton 11 near house 3b and Skeleton 10 near house 3f – fig. 2). Those skeletons interred within buildings were obviously interred during the construction of the respective building phase, as no features excavated specifically for the internment of the skeletons could be identified. Such infants must have been ‘available’ during the construction of floor layers (Stöllner 1999, 11). Where it was still possible to observe, burials were oriented along the walls of the buildings or along fences in their immediate vicinity.

Within the excavated area, fragmented and complete skeletal remains of 11 infants (foetus or neonate), aged between 0 and three months, were recovered (see table 1). A twelfth infant skeleton was aged between six to 12 years (Wiltschke-Schrotta 1999). Six of the depositions were recognised during the excavation, the remaining 6 were identified during the assessment of the finds. Five of the skeletons were found in the area of house 3b, two in the area of house 5a, and one each from house 5b, the area between house 3b and 5c, from house 3c and house 3f. This allows us to date three skeletons to Lt B1, one to

Table 1: Infant skeletons from the Ramsautal excavations 1988-9 Skeleton 1 Skeleton 2 Skeleton 3 Skeleton 4

Neonatus, 0-3 months Fetus, IX lunar month Neonatus, 0-3 months Neonatus, 0-3 months

Skeleton 5 Skeleton 6

Fetus – Neonatus, IX ½ lunar months – 3 months Fetus, IX ½ lunar months

Skeleton 7 Skeleton 8 Skeleton 9

Neonatus, 0-3 months Neonatus, 0-3 months Neonatus, 0-3 months

Skeleton 10 Fetus – Neonatus, X lunar months – 3 months Skeleton 11 Fetus, IX – X lunar months

Skeleton 12 Infans II, 6-14 years

Ra 89, Q8, Bef. Q8/5/2b (house 5b), Sk/e-d23/18 Ra 89, Q8, Bef. Q7/5/5 (house 5a), Sk/e22/11 Ra 89, Q8, Bef. Q7/5/5 (house 5a), Sk/e22/12 Ra 89, Q10, Bef. Q7/5/1c (between house 3b and 5c), Kn/g25/53 (possible animal bite damage visible) Ra 89, Q8, Bef. Q7/4/6f (infant skull within house 3b). Possibly matching Skeleton 7, found at same coordinates. Ra 89, Q10, Bef. Q7/5/1c (within house – below house 3b and above house 4) Ra 89, Q8, Bef. Q7/4/6f (within house 3b) see Skeleton 5. Ra 89, Q10, Bef. Q9/4/7 (within house 3c) Ra 89, Q9, Q9/4/7b (below foundation beams GS1/H3 – entrance area [?]). Ra 89, Q10, Bef. Q8/4/3 (below wooden barrel associated with house 3f) Ra 89, Q9/10, Q8/5/1, Kn/d29/50 (within drainage ditch/sedimentation layer southeast of house 3b, Skeleton 3) Ra 89, Q7, Bef. Q9/5/7, Kn/a21/12 (below house 5a – sedimentation layer) possibly dislocated in the Iron Age.

38

THROWN OUT WITH THE BATHWATER OR PROPERLY BURIED?

Fig.1: Infant skeletons in contexts of house 5a and 5c.

39

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD

Fig. 2: Infant skeletons in the contexts of house 3b and 3f.

40

THROWN OUT WITH THE BATHWATER OR PROPERLY BURIED? More recently, at least two more infant skeletons were recovered during the excavation of the floor layers of a building constructed in Lt A onto a terrace built into a Hallstatt D2-tumulus on the Hexenwandfeld, another settlement zone on the Dürrnberg bei Hallein (Zeller 1995).

Most of the infant skeletons recovered at Iron Age settlement sites have come from the interior of houses, from boundary ditches, or from ditches in the settlement interior, most of which can be interpreted as boundaries, separating an ‘interior’ from an ‘exterior’ or an unprotected ‘outside’.

What is generally remarkable is the high number of skeletons and skeletal fragments interred in what can be described as boundaries (8 of 11). Drainage ditches and fences can be seen as boundaries separating the immediate area from the ‘outside’, as the limits of the space associated with one particular building. In entrance features, foundations or under walls, the ‘liminal’ character of the space selected is even more obvious. This particular selection of places for the deposition of deceased infants indicates, to us, more than simple deposition and cannot be adequately explained without reference to the broader belief systems on the Dürrnberg in the Iron Age.

Questions The presence of such large numbers of infant skeletons in settlement contexts raises a couple of questions. First and foremost, of course, there is the question whether the infants were just discarded with the trash, or whether they were properly buried in a special rite for infants. Then we also have to consider the selection processes. Why were mainly infants buried in settlement contexts and why were they buried at or in boundaries? From this, we might be able to reconstruct early La Tène belief systems on the Dürrnberg in Hallein and in the wider European Iron Age. Careless disposal or proper burial?

Comparable features Earlier, it has frequently been assumed that these skeletons in settlement contexts are a result of the careless disposal of dead infants, possibly victims of infanticide (on infanticide see Krauße 1998), or even that they represent child sacrifices, placed as foundation burials for the buildings (see Wahl 1994). The latter seems unlikely in cases where high numbers of skeletons are found in the context of completely ordinary houses in completely ordinary settlements, and has, therefore, not been considered for the Ramsautal infants.

To analyse this phenomenon, comparable features in other regions of Iron Age Europe have to be considered. Similar features can be observed in Austria, for example at the settlement site of Walpersdorf Nord 1973, where 3 foetal or neonate skeletons have been found, one within a postbuilt structure (Ramsl 1998, 32). In a La Tène sunken floor feature on the Leopoldsberg an infant skeleton was discovered at floor level, oriented along the walls of the building (Tuzar 1991). The infant skeletons found in Walpersdorf Süd, Franzhausen, Flur Kokoron also are associated with a La Tène sunken floor feature (Neugebauer and Gattringer 1984, 99).

But also the first explanation has recently been questioned. Thomas Stöllner, the excavator of the Ramsautal site, has argued that these skeletons have to be seen in the context of a complex set of burial rites and beliefs (Stöllner 1999, 11). Karin Wiltschke-Schrotta, the physical anthropologist assessing the skeletal material from the Dürrnberg, observes that no evidence has been found on the skeletons to determine cause of death, making it near impossible to distinguish whether we are observing the simple deposition of deceased infants, a ritual deposition, or simply the proper burial rite for children (Wiltschke-Schrotte 1999, 126).

From Switzerland, comparable features are known from the late Hallstatt settlement in Gamsen, where 17 infant skeletons were recovered, from the settlement site at Scuol – Munt Baselgia with two infant skeletons, and from the La Tène period settlement Ollon, where an infant skeleton was discovered in a pit associated with a building (Berger 1993). In Germany, similar features have been observed in the Heuneburg settlement where three complete infant skeletons - some in the ditches of the defence system - and a further 13 partial infant skeletons have been excavated (Wahl 1995). In France, 22 or 23 infant skeletons were found in the La Tène settlement of Gailhan, in the interior of houses, oriented along the walls (Dedet, Duday & Tillier 1991). Even higher numbers of infant skeletons are known from British settlement sites. For example, 33 from the fortification ditches at Maiden Castle (Whimster 1981, 250) and 13 in two separate groups excavated from ditches at the settlement site at Rotherley (Whimster 1981, 25, 47). Other British examples include seven of 38 infant skeletons from ditches at the settlement site at Gussage All Saints (Whimster 1981, 28, 242; Keepax 1979, 161) and ten infants from Yambury Castle (Whimster 1981, 28, 248). Many more examples could be added to this listing.

The conscious selection of particular places, which can be found both in the Ramsautal settlement and elsewhere, is of course, an argument speaking against the careless disposal of infants – after all, at least some care seems to have gone into the selection of an appropriate placement, rather than throwing a dead infant out with the bathwater in which it had (been) drowned. An additional argument for burial rather than careless deposition can also be found in Hamilton’s (1964a; 1964b) principle of kin selection: to ensure the survival of the genome, individuals are more likely to behave in an altruistic fashion towards another individual the closer they are related to each other. Some form of care between parents and their child, even if deceased, is thus more likely than careless disposal. On top of that, breast-feeding stimulates the emission of the binding hormone Oxytoxin, leading to 41

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD ‘common Celtic’ god Lug (Maier 1994, 215) – but also receives his ‘heroic name’ only after an initiatory adventure, later receives weapons by deceiving his uncle, the king of Ulster, and finally gets his chosen wife after a couple more initiatory adventures. Perhaps equally interestingly, Cú Chulainn is even ‘conceived’ thrice, first he is magically given as a foster child to his later mother, but dies from illness, then is magically conceived by his mother but is stillborn due to her becoming ill during pregnancy, and only when conceived a third time, grows into adulthood (Kinsella 1990). Yet another Irish text, Cormac’s Glossary, describes the colomna áis, the ‘columns of life ages’ as noídenacht = infant age (derived from noídiu, which in turn is probably derived from a common Celtic *ne-weyd-u:s, “not-knower”, see Thurneysen 1946, 212; Vendryes 1960, N 20-1; Stüber 1998, 115-6, which is similar to the meaning of Latin infans, “not-speaker“), maccdacht = childhood, gillacht = youth, etc.

an even stronger attachment between breast-feeding mother and child. Also, both Barley (1995) and Luhmann (1977; 2000) argue that in premodern (and to a lesser extent, even in modern) societies, the afterlife usually is constructed in a fashion corresponding to living society. As a result, the ‘two worlds’ are often seen as interdependent, with the otherworld either as a model for this world or vice versa. Resulting from such a basic correspondence between life and afterlife, which we assume may well have existed in the Iron Age societies inhabiting the Dürrnberg (see also Zeidler 2005), is a socially constructed need to bury the deceased in a way reflecting their position in life, their age, and their level of initiation into society, to ensure that the dead are well cared for in the ‘otherworld’. Although every individual case definitely has to be considered in its own context, we think that, at least in the majority of cases of infant skeletons found in settlement contexts, we are observing a special burial rite. This rite seems to have been reserved, at least in much of central Europe, for infants, particularly the youngest ones, of up to no more than a couple of months after birth and including stillborn babies.

Similar age structures can also be identified in Roman society, with explicit age classes of infans up to seven years, then impuber until 14, then puber (Gaius, Inst. I, 196) etc. – which may actually have been the model upon which the age classes of Cormac’s Glossary were built. Where child mortality is concerned, Plutarch (Numa 12) advises his fellow countrymen that no one should grieve for a child of less than 3 years, of course indicating that child mortality must have been pretty high and that people did grieve. Fulgentius (Helm 1898) is even more to the point where he describes child burial in early Rome, indicating that children who had not lived more than 40 days were buried under the canopy, called the suggrundaria.

Why just infants? It is, of course, a well known fact that age is an important structuring factor in many premodern (and, perhaps to a lesser extent, in modern) societies. For Central Europe, Burmeister (2000, 208-11) has recently argued that age was one of the central factors in determining status in late Hallstatt societies. Where alpine Iron Age societies are concerned we also have direct evidence for age as a structuring factor in Livy, who refers to seniores, elders, opposed to a iuventus, youngsters (Dobesch 1980, 53-4 and 188-90; 1996). On a more general level, a special age group below the iuventus may be indicated by Caesar’s remark that the Gauls “…consider it shameful if a son of boyish age is seen in public with his father…”(B.G. VI, 18.3).

This gives us a reasonably consistent pattern of early European age groups, with the first group, the ‘nonknowers’ or ‘non-speakers’ receiving special treatment where burial is concerned. Given the similarity with the Roman practices attested in Fulgentius, we may assume that a similar practice might have existed on the Dürrnberg bei Hallein as well. Such a special burial rite many have been required as there may have been religious restrictions prohibiting infants from being buried in the regular cemetery; perhaps because infants before a certain age were considered as non-persons, or at least not as full members of society. Additionally, there may well have been economic restrictions on infant burial, especially if infant mortality was high, as it may have been in the unhealthy high alpine wetland environment of the Ramsautal. If regular burial was quite expensive, which would seem reasonable to assume, the means of any given family might simply have been too restricted to afford a ‘regular’ burial in the cemetery. On top of that, the strong attachment between parents and infant might have created an emotional need to keep the dead baby close, perhaps even to allow the child to be conceived again and be reborn in another body at a later date.

Later ‘Celtic’ texts, bearing in mind all the problems that may come with their use, also provide some information on initiations into different social age groups. Such initiations are, for instance, well attested in the Welsh Mabinogi (MacCana 1992; Gantz 1976), especially in the ‘fourth branch’ of the story, Llew Llaw Gyffes. In this story, Llew, the hero, is cursed by his mother, Arianrhod, never to receive a name unless she should give him one, never receive weapons unless she gave them to him, and to never be wed to a human wife. Only through the help of his witty magician uncle Gwydion, Arianrhod can be tricked into giving a name and weapons to the boy, who ultimately also gets a wife, conjured up from flowers by the uncle, thereby becoming a full member of society (Gantz 1976). Incidentally, Llew is usually seen as the Welsh version of the ‘common Celtic’ god Lug (e.g. Birkhan 1997, 602). A quite similar pattern is found in the Irish epic Táin Bó Cúailnge, where the main hero of the story, Cú Chulainn, is not only fathered by a certain Lug mac Ethlenn – usually seen as the Irish version of the 42

THROWN OUT WITH THE BATHWATER OR PROPERLY BURIED? Kelly 1988, 141), reaching as far as the boundary of the church lands.

Why at boundaries? Of course, when burial coincides with boundaries, van Gennep’s (1960) concept of liminality is bound to be brought up as a reason. However liminality is a very fuzzy concept, and even though death is doubtlessly some sort of transition across a boundary, liminality can’t explain why adults were mostly buried in regular cemeteries outside of the settlement in much of Iron Age Central Europe. As such, we think it is necessary to assess the liminal character of settlement boundaries in some greater detail.

Enclosed farms are a common phenomenon in much of Iron Age central Europe (e.g. Kristiansen 1998, 260 and 350-3), except in places where not enough space is available to set out such enclosed farmsteads. The latter applies to the crammed conditions in the Ramsautal on the Dürrnberg bei Hallein (Stöllner 1999), where the house and the immediate space associated with it up to the next drainage ditches may well have taken over the same role that the enclosure would have had elsewhere. Of course, we cannot necessarily assume that a similar idea of sanctuary was already associated with Iron Age enclosed farmsteads, but it seems not at all unlikely either and could explain burial within settlements, particularly within the boundaries, as a means of bringing deceased children, which did not yet have the right to be buried on the ‘regular’ cemetery, into the ‘magically’ protected area of the farmstead.

Of course, we do have reasonably good evidence from Roman belief that boundaries were under special protection by many gods and spirits, with all boundaries under the protection of Terminus, the god of boundaries and boundary markers. Even more strongly protected were house and farmstead, with Cardea protecting the hinges of the doors, Forculus protecting gates, Limentius protecting the doorstep, and the two-headed Janus guarding doors and all kinds of gateways in general (Lecouteux 1987, 24). It is clear that infants, buried at boundaries, would also be under the protection of these gods and spirits.

A tentative reconstruction of Iron Age belief on the Dürrnberg bei Hallein Based on the evidence discussed we might, as a summary, attempt a tentative reconstruction of Iron Age belief regarding infants in the Ramsautal settlement on the Dürrnberg, and quite likely in many other Central and Western Iron Age societies as well.

Another concept that might apply can again be found in early medieval ‘Celtic’ societies, where the enclosed farmstead and the house seem to have been under special, permanent, magical protection (Kelly 1988, 110 and 1401). The concept of protection, a very important legal concept in medieval Celtic law, is referred to by the cognate terms Old Irish snádud and Middle Welsh nawdd, indicating a common Celtic origin of both terms and associated meanings (Kelly 1988, 140; Vendryes 1974, S 146-7). Protection usually needs to be granted to all persons which have no independent standing in the community, because anyone who has not yet formally been accepted into the community is not protected by the law (Kelly 1988, 140). However, this ‘legal’ protection is ultimately a spiritual concept: for instance, climbing across the fence of an enclosed farmstead is seen as an injury to the metaphysical persona of the farm owner. Consequently, violating a person’s protection requires payment of his honour-price, the amount representing both the social status of a person as well as their personal integrity, which can be lost by, for instance, breaking magical taboos (Kelly 1988, 126 and 182-3).

Infants up to a certain age, perhaps of an age group called – where Celtic languages were spoken – by a term derived from common Celtic *ne-weyd-u:s, ‘not-knower’, were not considered full members of the community. The ‘liminal’ status, between being physically born and when they were first initiated into the community, possibly by receiving a name, lasted for a relatively short period up to a couple of months after birth, although the precise length may well have varied considerably from society to society or even from village to village. If infants had died before this initiation, they could not be buried in the regular cemetery, as they had not yet become members of the community, but rather were considered in a similar way as ‘foreigners’ would have been. However, parents would still have cared about their child, and, even though not yet fully adopted into society, would bury the child in a special rite restricted to such ‘not-knower’-infants. Reflecting their social positions as ‘foreigners’, such infants would be buried in a place where they were considered to be under the protection of their family, defined by the limits of the dwelling or farmstead, which additionally might well have been under the special protection of gods and spirits protecting these boundaries.

A quite similar concept can be found in early medieval Germanic society, where the enclosed farmstead and dwelling are sanctuaries (Wenskus 1961, 372; Lupoi 2000, 380). The concept of the sanctity of the house, or Hausfrieden in German, is again a spiritual protection, crossing the fence is a spiritual attack, which is why the hagazussa, the woman riding on the hedge – the hag – is dangerous: she breaches the magical protection associated with the farmstead. It can hardly be seen as much of a surprise that the sanctuary provided by both early medieval ‘German’ and ‘Celtic’ farmsteads was later transferred onto the church and the associated asylum in and around a monastery or church (Lupoi 2000, 380;

Thus, rather then having been thrown out with the bathwater, we conclude that the infant skeletons found in settlement contexts, on the Dürrnberg bei Hallein specifically, and in the Central European Iron Age more generally, were – at least in the majority of cases – buried in a special rite reserved for the youngest of infants, 43

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Helm, R. (ed.) 1898. Fabii Planciadis Fulgentii V.C. Opera. Leipzig: Teubner.

infants still in a liminal space between physical birth and social acceptance into society.

Karl, R. 2004a. Erwachen aus dem langen Schlaf der Theorie? Ansätze zu einer keltologischen Wissenschaftstheorie. In E. Poppe (ed.), Keltologie Heute. Themen und Fragestellungen. Akten des 3. Deutschen Keltologensymposiums – Marburg, März 2001. Münster: Studien und Texte zur Keltologie 5, 291-303.

Bibliography Barley, N. 1995. Tanz ums Grab. München: Deutscher Taschenbuch Verlag. Berger, L. 1993. Säuglings- und Kinderbestattungen in römischen Siedlungen der Schweiz – ein Vorbericht. In M. Struck (ed.) Römerzeitliche Gräber als Quellen zur Religion, Bevölkerungsstruktur und Sozialgeschichte. Mainz: Archäologische Schriften des Instituts für Vorund Frühgeschichte der Johannes Gutenberg-Universität Mainz 3, 319-328.

Karl, R. 2004b. Celtoscepticism, a convenient excuse for ignoring non-archaeological evidence? In E. Sauer (ed.) Breaking Down the Boundaries: The Artificial Archaeology-Ancient History Divide. London & New York: Routledge, 185-99. Karl, R. 2005. Master and Apprentice, Knight and Squire. Education in the ‘Celtic’ Iron Age. Oxford Journal of Archaeology 24/3, 255-71.

Bernbeck, R. 1997. Theorien in der Archäologie. Tübingen & Basel: Franke UTB. Birkhan, H. 1997. Kelten. Versuch einer Gesamtdarstellung ihrer Kultur. Wien: Österreichische Akademie der Wissenschaften.

Karl, R. 2006. Segmentäre Gesellschaften oder Feudalstaaten? Das irische Frühmittelalter und die Interpretation des archäologischen Befundes. In St. Burmeister and N. Müller-Scheeßel (ed.), Soziale Gruppen – kulturelle Grenzen. Die Interpretation sozialer Identitäten in der Prähistorischen Archäologie. Tübinger Archäologische Taschenbücher 5, Münster: Waxmann Verlag, 233-56.

Burmeister, S. 2000. Geschlecht, Alter und Herrschaft in der Späthallstattzeit Württembergs. Münster: Tübinger Schriften zur ur- und frühgeschichtlichen Archäologie 4. Collis, J.R. 1994. Reconstructing Iron Age Society. In K. Kristiansen and J. Jensen (eds.) Europe in the First Millennium B.C. Sheffield: Sheffield Archaeological Monographs 6, 31-9.

Keepax, C.A. 1979. The Human Bones. In G. J. Wainwright (ed.) Gussage All Saints: An Iron Age Settlement in Dorset. London: Department of the Environment Archaeological Report 10, 161-71.

Collis, J.R. 2003. The Celts: Origins, Myths & Inventions. Stroud: Tempus.

Kelly, F. 1988. A Guide to Early Irish Law. Dublin: Early Irish Law Series vol. III.

Dedet, B., Duday, H. & Tillier, A.-M. 1991. Inhumations de foetus, nouveau-nés et nourrissons dans les habitats protohistoriques du Languedoc: l'example de Gailhan (Gard). Gallia 48, 59-108.

Kinsella, T. 1990. The Tain. 15th impression, Oxford: Oxford University Press. Krauße, D. 1998. Infantizid. Theoriegeleitete Überlegungen zu den Eltern-Kind-Beziehungen in ur- und frühgeschichtlicher und antiker Zeit. In A. Müller-Karpe, H. Brandt, H. Jöns, D. Krauße & A. Wigg (eds.) Studien zur Archäologie der Kelten, Römer und Germanen in Mittel- und Westeuropa. Festschrift für Alfred Haffner. Rahden/Westfahlen: Internationale Archäologie - Studia honoria 4, 313-52.

Dobesch, G. 1980. Die Kelten in Österreich nach den ältesten Berichten der Antike. Das norische Königreich und seine Beziehungen zu Rom im 2. Jahrhundert v. Chr. Köln – Wien – Graz: Böhlau. Dobesch, G. 1996. Überlegungen zum Heerwesen und zur Sozialstruktur der Kelten. In E. Jerem, A. Krenn-Leeb, J.W. Neugebauer & O.-H. Urban (eds.) Die Kelten in den Alpen und an der Donau. Akten des Internationalen Symposiums St. Pölten, 14.-18. Oktober 1992. BudapestWien: Archaeolingu, 13-71.

Krauße, D. 1999. Der Keltenfürst von Hochdorf: Dorfältester oder Sakralkönig? Anspruch und Wirklichkeit der sog. Kulturanthropologischen Hallstatt-Archäologie. Archäologisches Korrespondenzblatt 29, 339-58.

Gantz, J. 1976. The Mabinogion. London: Penguin. Kristiansen, K. 1998. Europe before History. Cambridge: Cambridge University Press.

Hamilton, W.D. 1964a. The genetical evolution of social behaviour I. Journal of Theoretical Biology 7, 1-16.

Lecouteux, C. 1987. Geschichte der Gespenster und Wiedergänger im Mittelalter. Köln – Wien – Graz: Böhlau.

Hamilton, W.D. 1964b. The genetical evolution of social behaviour II. Journal of Theoretical Biology 7, 17-32.

44

THROWN OUT WITH THE BATHWATER OR PROPERLY BURIED? Luhmann, N. 1977. Funktion Frankfurt/Main: Suhrkamp.

der

Tuzar, J. 1991. Neue archäologische Ausgrabungen auf dem Leopoldsberg bei Wien. Wien: Archäologie Österreichs 2/1, 47. van Gennep, A. 1990. The Rites of Passage. London : Routledge.

Religion.

Luhmann, N. 2000. Die Religion der Gesellschaft. Frankfurt/Main: Suhrkamp.

Vendryes, J. 1960. Lexique étymologique de l'irlandais ancien. Lettres M N O P. Dublin – Paris: Dublin Institute of Advanced Studies.

Lupoi, M. 2000. The Origins of the European Legal Order. Cambridge: Cambridge University Press. Mac Cana, P. 1992. The Mabinogi. Cardiff: University of Wales Press.

Vendryes, J. 1974. Lexique étymologique de l'irlandais ancien. Lettres R S. Dublin – Paris: Dublin Institute for Advanced Studies.

Mallory, J.P. 1991. In Search of the Indo-Europeans. London: Thames & Hudson.

von Glasersfeld, E. 1992. Konstruktivität der Wirklichkeit und des Begriffs der Objektivität. In H. Gumin and H. Meier (eds.) Einführung in den Konstruktivismus. München: Veröffentlichungen der Carl Friedrich von Siemens Stiftung 5, 9-40.

Moosleitner, F. & Penninger, E. 1965. Ein keltischer Blockwandbau vom Dürrnberg bei Hallein. Mitteilungen der Gesellschaft für Salzburger Landeskunde 105, 47-87. Neugebauer, J.-W. & Gattringer, A. 1984. Rettungsgrabungen im Unteren Traisental im Jahre 1984. Vorbericht über die Aktivitäten der Abt. f. Bodendenkmale des Bundesdenkmalamtes nach der Verkehrsfreigabe der Kremser Schnellstraße S/33. Wien: Fundberichte aus Österreich 23, 97-128.

Wahl, J. 1994. Zur Ansprache und Definition von Sonderbestattungen. In M. Kokabi and J. Wahl (eds.) Beiträge zur Archäozoologie und prähistorischen Anthropologie. Stuttgart: Forschungen und Berichte zur Vor- uund Frühgeschichte von Baden-Württemberg 53, 85-106.

Ramsl, P.C. 1998. Inzersdorf-Walpersdorf. Studien zur späthallstatt-/latènezeitlichen Besiedlung im Traisental, Niederösterreich. Wien: Fundberichte aus Österreich Materialhefte A 6.

Wahl, J. 1995. Die Menschenknochen von der Heuneburg bei Hundersingen. In E. Gersbach (ed.) Baubefunde der Perioden IVc–IVa der Heuneburg. Heuneburgstudien 9. Mainz: Römisch-Germanische Forschungen 53, 365-383.

Stöllner, T. 1991. Neue Grabungen in der latènezeitlichen Gewerbesiedlung am Dürrnberg bei Hallein. Ein Vorbericht. Archäologisches Korrespondenzblatt 21, 25569.

Wenskus, R. 1961. Stammesbildung und Verfassung. Das Werden der frühmittelalterlichen Gentes. Köln – Wien – Graz: Böhlau.

Stöllner, T. 1996. Bergbau und Gewerbe am Dürrnberg bei Hallein. Ein Beitrag zur Siedlungs- und Bergbauarchäologie auf dem eisenzeitlichen Dürrnberg. In E. Jerem, A. Krenn-Leeb, J.-W. Neugebauer & O. H. Urban (eds.) Die Kelten in den Alpen und an der Donau. Akten des Internationalen Symposiums St. Pölten, 14.-18. Oktober 1992. Budapest-Wien: Archaeolingua, 225-43.

Whimster, R. 1981. Burial Practices in Iron Age Britain. Oxford: British Archaeological Report British Series 90. Wiltschke-Schrotta, K. 1999. Die Kinderskelette aus der latènezeitlichen Siedlung Ramsautal/Dürrnberg. In Erich Pucher (ed.), Archäozoologische Untersuchungen am Tierknochenmaterial der keltischen Gewerbesiedlung im Ramsautal auf dem Dürrnberg (Salzburg). Rahden/ Westfahlen: Dürrnberg-Forschungen 2, 113-127.

Stöllner, T. 1999. Bemerkungen zu den archäologischen Ergebnissen der Untersuchungen im Ramsautal 1988 bis 1989. In E. Pucher (ed.) Archäozoologische Untersuchungen am Tierknochenmaterial der keltischen Gewerbesiedlung im Ramsautal auf dem Dürrnberg (Salzburg). Rahden/Westfahlen: Dürrnberg-Forschungen 2, 1-15.

Zeidler, J. 2005. Cults of the Celts: A new approach to the interpretation of the religion of Iron Age cultures. In R. Karl & J. Leskovar (eds.) Interpretierte Eisenzeiten – Interpreted Iron Ages. Akten der ersten Linzer Gespräche zur interpretativen Eisenzeitarchäologie, 29.10.1.11.2004, Linz. Studien zur Kulturgeschichte von Oberösterreich 18, Linz: Oberösterreichisches Landesmuseum, 171-9.

Stüber, K. 1998. The Historical Morphology of n-Stems in Celtic. Maynooth: Maynooth Studies in Celtic Linguistics 3.

Zeller, K.W. 1984a. Keltische Gewerbebauten auf dem Dürrnberg bei Hallein. Unpubl. MA thesis, Marburg/Lahn.

Thurneysen, R. 1946. A Grammar of Old Irish. Dublin: Dublin Institute for Advanced Studies.

45

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Zeller, K.W. 1984b. Latènezeitliche Gewerbebetriebe auf dem Dürrnberg bei Hallein. Veröffentlichungen aus dem Vorgeschichtlichen Seminar Marburg 3, 199-214. Zeller, K. W. 1995. Das Hexenwandfeld auf dem Durrnberg bei Hallein. Salzburg Archiv 20, 19-30.

46

The Children in the Bog Grete Lillehammer, Museum of Archaeology, University of Stavanger It is better not to beg [ask for something] Than to sacrifice too much [to the gods]: A present given always expects one in return. It is better not to bring any offering Than spend too much on it. (Hávamál (145), in Mauss 2001, 3) The disposal of children in water Over a long period of time the disposal of humans in bogs was part of a common tradition among the peoples around the North Sea. However, the archaeological bogfinds of human bodies and skeletal remains have many origins. Human remains have been discovered all over Europe; from Crete in the south to Norway in the north and Russia in the east (Ebbesen 1986, 20); especially in Denmark, the Netherlands, and North Germany; to a lesser extent in Britain and Ireland (Dieck 1972; Stead et al. 1986; Turner and Briggs 1986; Turner & Scaife 1995; van den Sanden 1996; Coles et al. 1999; Bradley 2005). In the 1980s, European museum collections had over 1,500 human remains (Ebbesen 1986), representing both sexes and a wide range of ages, including babies (Turner and Briggs 1986, 156-161). Some of the finds have been dated by pollen analysis, C14 and associated artefacts (Turner and Briggs 1986), covering a wide span of time from the Neolithic to recent times (Dieck 1972; Glob 1969; Turner and Briggs 1986; Pieper 2002). The majority of the finds date from the late pre-Roman to Roman Iron Age (Bradley 2005; see also Monikander 2010, table 5).

The exposure of the human body to a variety of elements such as air, fire, earth and water is well attested within the mortuary record (cf. Montandon 1934, fig. 29). Amongst these, the immersion of the human body in water is of special interest to the archaeological study of children. With regard to the estimated mortality rate of children in prehistoric societies (Chamberlain 2000) and the question of the unusualness of those children buried in the same way as adults (Parker Pearson 2003), the phenomenon refers outsiders to other realms of social expression (Crawford 2000, 2010; Lillehammer 2002, 2005, 2008a, 2008b, 2010a). From a Scandinavian perspective, among these are the places less visible in the environment (Björkhagar 1995), such as the wetland places (Formisto 1993; Berggren 2010; Monikander 2010; Bergerbrant 2011) or their relationships to settlements (Ullén 1994) and rock carvings (Thedéen 2002; Wahlgren 2002). In a nature and culture relationship, water is the breeding area of many species in nature. In developing young humans, from the fertilized egg to the ultimate birth of offspring, water is an essential part of the liquid contained in the nurturing space of the embryo, within the uterus of the maternal body. Within many past cultures water is often associated with perceptions of life and death, to rituals and rites of passage of the body focusing on fertility in humans as well as in nature (Metcalf and Huntington 19911979), and to supernatural worlds (Bradley 2005). In the landscape the wetland was a passage to another world (Kaliff 2001). There is an element of interface between the real world and the metaphysical one; the water in lakes, rivers and bogs was bringing human beings in closer communication with the gods (Brink 2001). The places of fertility sacrifices and offerings were often some distance away from the settlement, or relocated to central places, often to the farm of the chieftain (Fabech 1994), or to wells, the place of a miniature wetland (Kaliff 2001). In South Scandinavia these activities can be evidenced as far back as the Late Mesolithic (Berggren 2010) and the Early Neolithic (Stjernquist 1998), but are a typical expression of practices during the Bronze Age and the Early Iron Age (see Kaliff 2001). On the long-term scale, and as deposition of wealth in the cultural landscape changed from the 5th and 6th century CE (Fabech 1991, 291-294), depositions in liminal areas such as the wetland continued until the medieval period (Hedeager 1999a; 2010, 172173).

Due to the natural conditions of the Norwegian bogs the human finds are unusual, in the form of skeletal remains and hair. A revised distribution list recorded fifteen individuals from nine localities in 2011 (Sellevold in press, fig. 7; 2005). The skeletal remains date from the Neolithic to the Medieval period and include both children and adults. Five of the finds date to the preRoman period and are located to the same region. The Norwegian material consists of either the bones from the whole skeleton or the cranial parts of the body. In one case the cranial part is missing. Causes of death have not been established, although no traces of violence have been found on the bones. Therefore there is no circumstantial evidence indicating what may have caused death. The individuals may have come to the bogs for a variety of reasons. According to the conclusions of the human osteologist and the archaeologist who first revised the Norwegian finds in 1991, little could be gained by searching for one single or uniform theory on bog bodies (Sellevold and Næss 1991, 441). Then how to approach the archaeological find of bog bodies of children in general? On one side are the contextual factors such as circumstantial placing, retrieval of the finds, deposition of the bodies and associated objects. On the other are the biological and social factors 47

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD such as age, sex, gender, genetic anomaly and pathological evidence based on the analyses of their bones in relation to the contextual evidence. These are aspects of vital importance in order to explain the character of the human individual in the archaeological find. In this relationship the individual is classified according to the condition of the corpse versus the skeletal remains and their contextual content on the basis of the archaeological evidence. The definition indicates an interdisciplinary field of research covering a wide spectre of disciplines.

slowly changing cultural features of long durée such as ideologies and world views (Bintliff 1991, 7).

In the following paper I am going to demonstrate why this approach is important to the archaeological interpretation of the bog-children. I am searching for cultural theories on ancient child treatment in order to explain the contextual evidence, looking for alternative interpretations of the remains in a landscape perspective and, in particular, the wetland environment. A specific case from Norway serves as a starting point to approach the children and the theories based on Scandinavian bogfinds. A variety of literary sources, such as Ancient World narratives, Norse mythology, legends and sagas, Scandinavian medieval laws, Nordic folk belief, Scandinavian place names and Norwegian folksongs are included in the analysis. The interval between my paper at the Kent conference in 2006 and this publication has seen an advancement in the literature dealing with childhood and children in the past, which is of particular interest to approach wetland children in a wider culturalhistorical perspective (for overviews, see Brockliss and Montgomery 2010; Mustakallio and Laes 2011). On the various levels of mortuary behaviour and child treatment, the ideology, attitude and practice may have continued or changed, respectively, covering a wider area in time, space and structure. The written sources are to be applied and treated cautiously as the fragmentary remains of the past and a critical approach towards the types of literature is recommended (for further discussion, see Vuolanto 2011).

Fig. 1. Bone remains of approximately four infant cranials from early Iron Age (calibrated CE 90/120410/430) from a bog at Bø, Hå municipality, Rogaland county, south-west Norway (Sellevold and Næss 1991). (Photo: Terje Tveit, Museum of Archaeology, University of Stavanger). The little children – wetland children and bog body theory The Norwegian bog bodies include a single find of infants representing the only children in the collection. In 1984 the remains were found by a farmer in a bog at Bø, Hå municipality, Rogaland County, south-west Norway. The experts, a zoologist and a human osteologist, who first examined the content, concluded that the bones were from various parts of the cranial of at least four children (Fig. 1). The size and development of the bones indicated newborn infants, but their biological sex could not be established on the basis of the macro-analysis. The bones had been disposed before the flesh disintegrated. In infants the cranial bones would have separated easily had the bones not been disposed in a manner that made them look like coconuts. There was no evidence of the skeletal remains from the rest of the corpses; the condition in the bog could have resulted in the disappearance of weak bones and cartilages. No traces of illness, disease or injury afflicted upon the bodies were found on the bones. Therefore the cause of death of the infants is stated as unknown (Sellevold 1987; Sellevold and Næss 1991).

In this approach towards child treatment in the past, a deliberate attempt is made with regard to considering material and immaterial culture interwoven with practice on a long-term scale. On the critical level of analysis, the literary sources are used as comparative models and analogies in order to bring together and highlight the phenomenon of wetland children in the archaeological record. In this respect my concern is continuity and discontinuity in the cultural transference of practice and the transformation into writing of traditions passed down in myths, legends, sagas and folklore from one generation to the next. To a certain extent textual sources may contain fragmentary reminiscences of ancient narratives with a degree of relevance to finding models about historical events and practices of the past. Similarly, laws and regulations may hold contradictory information about the past, in order to prevent unwanted or unlawful practices from continuing in the future. By doing so, I claim the Norwegian remains of bog-children as part of the ancient cultural-historical traditions of practice, which are consistent with the long-term forces of dominant and

The bog find had been discovered at a marginal place in the outland at a higher altitude. From the spot in the present there is an outstanding view downhill of the landscape and further beyond to the North Sea (Fig. 2). Between 1984 and the delivery of the find to the museum in 1987 the area had been drained and cultivated, but the bog had been small at the time of the discovery. The local name of the place was ‘Tviodlo’, referring to two natural springs formerly swelling in the bog. The bones were found together with as little as 30 cm between them at approximately 1 metres depth, at the bottom of the ditch in an area half a meter in diameter, and at the level between the bog and the layer of clay beneath. No 48

THE CHILDREN IN THE BOG artefacts accompanying the skeletal remains were found in the digging. The C-14 dating of five of the bones gave the calibrated age between CE 90/120 and CE 410/430 (see Haavaldsen 1989; Lillehammer 1989; Sellevold and Næss 1991).

period of infants and 14/15 years of age. The infants have been found together with adults or they have been located to the same bog. These infants represent the majority (41%) in the material evidence of bog children in the survey. Among these finds are the skeletal remains of fetus and neonates which have been discovered mainly with other objects at localities classified as sacrificial bogs (Bergerbrant 2010, table 1 and 2; see also Monikander 2010, 79-81, table 5, 86-91, table 7–8, appendix 1). Compared with the adult remains, the lack of violence on most of the children’s bones is significant (Monikander 2010, 90). An osteological study of a Finnish Iron Age bog locality dated between 5th century BCE and 7th century CE (Monikander 2010, 83-84) indicates that the disposal practice included a high amount of children. Among the skeletal remains are 37% children and four children are infants under the age of one year (Formisto 1993, 85, fig. 16). The finds of bog corpses are part of a wider pattern connected with Iron Age use of watery locations (Bradley 2005). Many of the best preserved bog bodies have had violent deaths, capturing the imagination of archaeologists. Corpses held in position by pieces of wood have involved an awareness of the supernatural pointing at criminals and outcasts (Bradley 2005, fig. 3.1; van den Sanden 1996). In his book ‘The Bog People’ P. V. Glob selected a collection of individuals to support his theory on bog bodies from the Iron Age. The bodies shared a similar fate; the archaeological remains were the evidence of ritual sacrifice to a goddess of fertility. Glob pointed at the long tradition of disposing of food vessels and objects of great value in Danish bogs (Glob 1969; Turner & Briggs 1986). His sacrifice theory has been linked with bog ore and precious iron extracted from the depths of the earth. A connection in time and place has been made between contemporary bog ore digging and the performance of sacrificial rites at places hiding the earth’s riches ruled by Mother Earth (Fischer 1980). This is an argument of economical importance scarcely to be overestimated (Ström 1986).

Fig. 2. Location of the Bø-find is facing south and looking down onto the lowland and the North Sea. Modern cultivation has transformed the bog environment to a pasture in the outland. (Photo: Aage Pedersen, Museum of Archaeology, University of Stavanger). The bog-find refers in general to early Iron Age (BCE 500-CE 550); in particular to the Roman period (CE 0400) and the earliest part of the Migration period (CE 400-550) in Norway. It relates to the outland of one of the central farmsteads in the region, and to the outskirts of prehistoric settlements which, during the Roman and Migration periods, reached an elitist level of richness and power (cf. Myhre 1978, fig. 19). The circumstantial evidence points to a disposal of the infants prior to, at death or not long after their deaths. Whether their partial remains of cranial bones represented decapitated corpses or not, the children had been placed in an environment where there were natural springs of water. At the time of the disposal the springs were located on marginal land at a distance from the settlements in the landscape. With respect to the circumstantial evidence of the find a variety of interpretations are called for with regard to discussing disposal of children in water and bog body theory in general, which include the achievement of socio-cultural and political goals in Iron Age society (BCE 500-CE 1030).

The views on Danish bog bodies as the remains of executed criminals, sacrificial victims or as both, are divided into two camps (Munksgaard 1984, 121; Turner and Briggs 1986; Hultgård 2001, 537). The research has led the interpretation of bog bodies to develop into a sacrifice-theory (Thorvildsen 1952; Glob 1969; Fischer 1980) and a punishment-theory (Dieck 1965). In 1986, five hypotheses were presented concerning the deposition of human bodies in bogs, on the basis of the text of Tacitus in Germania part I (Tacitus 1923, 22). As an historian on religion, Folk Ström (1986, 234-136) concluded that the analysis of peat bog bodies, based on Tactitus, lead to a dead end. Alternatives had to be found, and he came up with the following list:

In 1983, Danish bog-finds included the skeletal remains of seven children and sub-adults (N=19, 4%) from two disposals C-14 dated respectively to pre-Roman period (BCE 475) and late Germanic period (CE 675) The youngest children (neonates and infans I) were found together with adults, two of which were male (Sellevold et al. 1984, 64-65, 241-242, table 9–10–1), and the preRoman male was also with the remains of a newborn and a container ((Thorvildsen 1952, 39-40; Sellevold et al. 1984, 242, table 9-10-1, Bergerbrant 2011, table 1). With the exception of the Finnish material, this pattern of disposal is partly confirmed by some recent surveys of wetland finds from the Iron Age in Northern Europe. These finds include 17 individuals between the biological

    49

Human sacrifice Death penalty Sacrifice and penalty Burial

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD 

bogs were situated close to the pathways of people in their everyday life (Christensen 2003).

Apotropaic measures (threats of ghosts)

As seen in the list above there is a third possibility representing the cases of both sacrifice and penalty (cf. the third). These are references to Germanic death penalty ‘as of a sacral nature’ representing sacrificial religious rites as the Gods have been severely angered. There is no solid basis for this theory, but in a modified form it remains incompatible with disposals after death (Ström 1986).

In her study on sacred topography Lotte Hedeager (1999a) has stressed the perceptions of the landscape defined by people themselves, and shared in a common cosmology of mental division and constituted by physical boundaries reflecting the organization of the landscape. The division of the world in the Old Norse cosmological concept of Midgard/Asgard and Utgard reflects the division between infield and outfield in the organisation of the landscape (Gurevich 1985, 47; Hedeager 1999, 246). In drawing the attention to the facts and perceptions connected to the wetland, Monikander (2010, 24-38, 9899) has concluded that it was a borderland between life and death. In order to bind their possible influence on the living, people may have been separated from other dead people and put in wetlands.

In her discussion of bog-finds Elizabeth Munksgaard (1984, 122) has claimed that the human sacrifices from the early Iron Age are essentially different in character from that of bog bodies. The remains of human sacrifice are always accompanied by domestic animals, pottery or sometimes parts of wagons, whilst the places in the landscape are sunken bogs, which most likely were open lakes at the time of the religious activity. Among the skeletal remains from the sacrificial bogs are found the bones of children. The fact that children really never occur in the finds as bog bodies argues against the sacrifice-theory being applied on bog bodies. The few examples from Germany are possibly the circumstantial evidence of accidents, as the bodies are found in raised bogs that defiled the valuable land.

If differences are seen in time and place between the deposition and the associated content of human bodies and skeletal remains in lake and peat bog, compared to their location far or adjacent to the habitat or the pathways of people, these differences are vital to explain the type of activity taken place in the landscape. Therefore the following circumstances have to be added to the Ström list:

A review focussing on the relationships between social organisations, religious ceremonies and regional variations in Southern Scandinavia has led to a division of the sacrificial finds from wetlands into three categories: fertility sacrifice, human sacrifice and booty sacrifice. Fertility sacrifice is carried out by the individual, family or society and includes deposits of pottery, food, vessels, animals, tools and ornamental garments. Human sacrifice is the collective act of deposing individuals in peat bogs. Depending on the pH value of the bog, the bodies or skeletons appear also in booty sacrifices and most show signs of violent death. Booty sacrifice is performed after victory over an intruding enemy. The archaeological finds contain a representative choice of the equipment of an Iron Age army (Fabech 1991, 301).

   

Murder Suicide Accident Natural death

The discussions are relevant to the classification of the Norwegian find. The location of the find is a sunken bog, formerly the location of two springs. The disposal of the infants in the spring was only in the form of bodies or body parts and the causes of death of the newborn are unknown. By taking into account the circumstantial evidence in relation to the theories of sacrifice or punishment, it is clear that the Norwegian find did not hold the remains of animals or other objects accompanying the infant bones. According to ancient tradition in the Hebrew Scripture, animals could be replaced in the substitute of children as sanctuary gifts (Stavrakopoulou 2010, 23). The substitute aspect is difficult to substantiate as the lack of objects in the Norwegian find may speak in favour of historic circumstances, which relate these remains to punishment rather than supporting a sacrifice theory.

The no-children argument by Munksgaard (1984) reveals the difficulties in analysing similarity and difference between the circumstantial evidence of humans and human skeletal remains from wetland areas compared to bodies, to raised and sunken bogs, and to fertility, human and booty sacrifices. It has been suggested that the Scandinavian bog-finds from the early Iron Age represent women, men and children who have been killed deliberately as the result of punishment for crimes committed, penalty, or sacrifice to the gods such as battle tributes of victory at war (Hultgård 2001, 537, 545). According to Bergerbrant’s survey (2011), children were also disposed in the wetlands due to burial or accidents. The intentions and ideologies behind the practices are ambiguously difficult to penetrate due to the fragmentary and secondary state of the literary sources (Hultgård 2001). However, it has been noted that the sacrificial

By asking the question of whom, for and by whom the punishment was enacted, the evidence of the newborn infants speaks for itself: Newborn children are not adults who may determine their own lives and their ability to stand up for themselves is very limited. As minors, infants have no authority over their social conditions (Barth 1976; Lillehammer 1989). However, they do represent the potential of something new (Lillehammer 2002, 2005; 2010a, b), and biological social and cultural expectations may have related them to connections either 50

THE CHILDREN IN THE BOG promising or threatening within the society. Noteworthy is the fact that the newborn infants in the Norwegian find were not found with adults, but that their skulls were found together. These factors suggest the activities of disposal may have taken place either at the same time, or with only short intervals between them. The contextual evidence suggests both single and/or repetitive events that could indicate isolation and dislocation of individuals in relation to social groups in the early Iron Age society.

overviews, see Bøe 1966; Steinsland 2005; Hedeager 2011, 100-104). Various forms of ritual communication with the other world through sacrifice or in burial are well attested during the Iron Age (Hedeager 2011, 103), and in the literature the sacrifice of children generally in relation to religion. These descriptions refer to myths and legends originating during or after the transition from the Norse religion to Christianity in the 11th century CE. In this relationship there are difficulties with the historical sources in determining between sacrifice as a gift to the gods and infanticide as a regulation of population growth (Näsström 2001, 26–28, 45). The part sacrifice played in the cults to the gods is not easy to define (Davidson 1993, 97), and though it was considered common and played an important role, its role in the Norse religion remains enigmatic (Hedeager 2011:100, 103).

The first interpretation suggests an event that required as many as possibly five newborns at one time. In a single stroke this was either a vast tribute contributed by society, or a heavy penalty on an individual. However, in regard to infants, other circumstances have to be considered. Could the bog-find represent the bones of stillborn, twins, and even quins? These questions point to a variety of logics represented in traditions from many parts of the world (DeMause 1974; Boswell 1991; Lohrke 1999; Scott 1999; Milner 2000; Bremmer 2007; Ceruti 2004, 2010). In a cross-cultural perspective, abortion, birth of twins and infants with body deformities or disabilities (Devlieger 2000) have been put in the discussion on the social difference and anomalies in the inclusion and exclusion of children. Amongst the ancient peoples of Scandinavia the various cultural customs of infant treatment (Mejsholm 2009) and abandonment of children (Pentikainen 1968; Vuolanto 2011), circumstantial practices necessitated by anomalous conditions have been noted, especially in relation to childbirth and birth control in the cases of rape and in times of famine (Näsström 2001; Steinsland 2005; Mejsholm 2009).

The sacrifice of infants, defined also as sacrificial infanticide, is connected to public manifestations of beliefs and vows and practiced according to religious obligations to Gods/deities or spirits (see Scott 1999, 81, 89). The ideology behind child sacrifice appears to be related to fertility and to creating or maintaining a balance between cosmological powers and humans in times of crises and therefore not thought to be an ordinary event. The sacrifice of the firstborn child is known from the Old Testament in relation to fertility rites (Stavrakopoulou 2010; Francis 2010). In Scandinavia the prehistoric bog-finds of humans, animals and objects have been linked to religious beliefs in and worship of several female deities (Globe 1969; Näsström 2001, 2004, 55–56, 2010; Steinsland 2005:147–149; Monikander 2010, 34). In this respect we are dealing with the complex relationships of origin and transformation attributed to both male and female powers in the Iron Age. At the centre of the discussion is the relationship between the fertility goddess ‘Nerthus,’ referred to in Germania part II by Tacitus (Tacitus 1923, 48) in the 1st century CE, and the Vanir deities. In the Norse mythology (CE 700/750–1300), the three goddesses Freya, Frigg and Saga, and Urd are all connected with childbirth, children and wetland places of the earth.1

The disposal of infant bodies in places such as latrines, wells, house foundations, post-holes, hollows, pits and ditches have been recognized in the archaeological evidence (see Sellevold et al. 1984; Wahl 1991; Lee 1994; Cueni 1995; Scott 1999; Lillehammer 2000, 2008a, 2008b, 2010a; Lohrke 2004; Bielke-Voigt 2008; Mejsholm 2009; Crawford 2010). As pointed out by Crawford (2010, 60) there are a number of problems in interpreting the archaeological evidence. In this respect, one important question concerns how to define the similarity and difference between sacrifice and birth control in the disposal of dead children outside the sphere of grave burials both inside and outside the vicinity of prehistoric settlements. With respect to the variety visible in the archaeological remains of children, it seems relevant to ask how newborn children were valued and acted upon as individuals in the power-related hierarchies between parents, relatives and other groups in society

1 The fertility goddess Nerthus lived on a sacred island in the middle of a sacred sea. The natural habitats of three female Norse deities Frigg, Saga and Urd are connected with the wetland. Frigg lived in Fensaler (the houses by the fen (Näsström 2001, 82, 2004), or the sea (Branston 1955, 88), Saga in Søkkvabekk (the sunken beck or stream (Branston 1955, 88, 156, Näsström 2001, 82, 2004), and Urd, Verande and Skuld, the three norns in the well beneath the ash-tree Yggdrasil, the world tree (Branston 1955, 88–81; Näsström 2001, 82, 2004). In the discussions on the origins of the Norse mythology, Nerthus is linked with Frigg as well as Freya (Branston 1955, 158–159, 162, 301). Frigg is the wife of Odin (Branston 1955, 158–159, 204, 259) (see also note 6), the mother of the gods (Branston 1955,117, 158–159) and of Balder (Branston 1955, 124, 127, 159). She plays an important role in relation to love, sexuality and fertility (Branston 1955, 158). Frigg was thanked for giving help in birth (Bæksted 2002, 61, 75) and couples as an alter ego with the fertility goddess Frøya (cf. Branston 1955, 133, 135, 156). Saga is one of Frigg’s shadows and her eke name. Her home in the beck is undoubtedly the same place as Frigg’s dwelling Fensalir (Branston 1955, 156–157). Among the norns, Urd is the greatest of the goddesses of fate (Branston 1955, 208, Bæksted 2002, 287). She decides on the length of life becoming to a child (Steinsland & Sørensen 1994, 32). Her name is linked with both death (Henriksen 1993, 61) and one of the moon’s phases, the period of the crescent (Branston 1955, 208–209).

The practice of sacrifice – the powerful children The cultural practice of human sacrifice is mentioned in regard to the Iron Age people of Northern Europe including the Vikings in Scandinavia (Turville-Petre 1964; Hultgård 2001), and based among others on written narratives about official pagan cults among the Swedes in Rus 10th century CE by the Arab Ibn Rustah and ca. CE 1070 by the German bishop Adam of Bremen (for 51

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD During the Iron Age a successive change of world view took place between CE 400-500, from the collective fertility Vanir cult of female deities to an individuallyinclined practice linked to the Asir cult of Odin, the oneeyed god and all-father and the principal group of benevolent Norse deities in the pantheon. The ruler of war and death, and above all the great shaman of inspiration and magic, he performed the legendary act of hanging himself from a three in exchange of almighty wisdom (1999b, 79-81, 2011; see also Davidson 1993, 78). In the mythical cycle of Odin, the Hunn king Attila amalgamated with the already existing pan-German god Wotan/Odin and Odin’s Asir with his entourage of warriors (Hedeager 2011, 227-228). Concurrent with the development of an aristocratic society based on a warrior ideology (Gräslund 2005, 377-378; Fabech 1991), therefore a new symbol-system emerged which was linked to a fundamental change in social practices and social memory in the sense of what was considered reasonable and what was unreasonable (Hedeager 2011, 16-17).

religions. According to the Christian doctrine mankind is a subject to God’s omnipotence, and the believer is God’s obedient child or slave. The Norse language uses the concept siðr in the meaning of those manners and customs that the individual had to conduct as an inevitable part of the cult community. The relationship with the gods was a pact of friendship between man and deity, the god as being the best and faithful friend (Steinsland 1990b, 127, 129). Also worth noticing is the central role of sacrifice in Christian ideology; particular the crucifixion of Christ (Maskel 2010, 43). The transition from a pagan to a Christian ideology in the attitude and practice of child treatment may overshadow the scholarly debate. The references in the medieval literary sources concentrate on turning against infanticide, motivated foremost by the saving of children’s souls (Mejsholm 2009, 82, 98-99). The practice of infanticide – the unwanted children Infanticide, abandonment and the abortion of infants are known from the Graeco-Roman and Early Medieval world (Crawford 2010) and beyond (Milner 2000). Infanticide, the concept of killing of newborn children, has been defined somewhat unclearly, as a primitive method of birth control or removing excess children either through actively putting them to death or allowing them to die (Encyclopædia 1974; on active and passive infanticide, see Crawford 2010, 66, note 42). In this relationship, to explain the ideology, attitude and practice behind the social exclusion of children, we have to differentiate between child abandonment and deliberate infanticide (see Vuolanto 2011). In the discourse on ancient child abandonment in Scandinavia, the focus is on how to apply and interpret the textual sources written down after the introduction of Christianity, and on the Christian influence upon the pagan right to carry the child out. In what manner these activities was based on preChristian traditions have, to some extent, been disputed and are difficult to penetrate on the basis of literary sources (see Vuolanto 2011). According to Mejsholm’s (2009, 90) comprehensive study of infant treatment in the Scandinavian transition period to the early-Christian era, the pagan practice of carrying out children is not to be doubted.

With regard to sacrifice, there were strict rules attached to the performance of the act. Two types of human sacrifices have a specific position in the Norse myths: children and kings. The sacrificial ritual was a purification process of the world kept in accordance with the norms of perfection, and the sacrificed was usually a young man, woman or child. Parents offered their sons and daughters (Peel 2010, 5) or kings sacrificed sons as gifts to the god Odin (Näsström 2001, 26-29). By regarding sacrificed children in this perspective, the child treatment is linked to socio-religious identity and value in death on the same level of included members of a social group in the society (Lillehammer 2008b). It points to a society where contradictive structures were kept together by rituals to prevent the social system falling apart; i.e. where dangers were threatening to human life, the family or the community, and children were used as weapons in the ideological strive for dominance (cf. Douglas 1966). The sacrifice of children as part of fertility rites in the Iron Age has been contested on grounds of the interpretation of sources (Hultård 2003). The historical myths are legends usually told in order to describe what happened in the past, but often found to have a base in references to the critical circumstances of maintaining power, such as during war or in the ruler’s desire for a long life, victory and peace (Näsström 2001, 26-28, 4546). As the narratives about myths and legends were later transformed into writing in the medieval period after Christianisation, the texts may represent Christian polemics to contrast pagan chaos with the Christian order of the world (Näsström 2001, 44-45). A significant similarity and difference existed between the two

Firstly, the Norwegian and Icelandic medieval laws give examples (Mundal 1988, 11, note 4), which throw light upon the social aspects of inclusion and exclusion of newborn children in early Christian society. These can be traced in the dialectical variations of concepts, which describe the process of dislocating the children. The concepts differ between the child being carried out (bera út), slain out (slá út) or thrown out (kasta út). The concepts are linked with acts, which describe the process of letting the child being lost (spilla), in the meaning of getting the child killed by someone (Mundal 1989, 130). The concepts form the background for explaining the Norse type of infanticide. It is an act of actively killing the newborn, and therefore it differs from the practice of

The most important clues to these wetland relationships are the norns and their connections with the ash-three Yggdrasil, with childbirth and the creation of life and with the tree as the symbol of life. The well of the norns is a symbol of the spring, the resource supplying the roots with saps of life in order to preserve the tree to ensure continuance and maintain life in the universe (Branston 1955, 81–82).

52

THE CHILDREN IN THE BOG taking care of foundlings (Mundal 1989, 130), and the sagas mention several cases.2 What, then, were the attitudes towards the practice of infanticide in Norse society? In Germania part I written in ca. 98 CE, a time corresponding with the Norwegian bog-find, Tacitus (Tacitus 1923, 22) informs us that the death penalty for adults, men as well as women, was the drowning in a morass. The punishment was looked upon as a shameful act in the society (Ström 1986, 231). In the Norse world practicing infanticide was regarded as shameless among the rich Icelanders (see Mundal 1989, 132-133: also Mejsholm 2009, 82–83; Vuolanto 2011, 12, 18 note 51 and 53), and therefore not on the level of severe crimes requiring the penalty of death.

family structure counted children born by wives, frills (legitimate lovers of the husband) and the household slaves as legitimate, and the father, parent and family could have had a say depending on circumstance and tradition. Circumstances included social class and gender (Mundal 1989, 123-126; 1988) (Table 1), and both men and women took part in the act, freeborn as well as slaves. The literature (for overview, see Hovstad 1956) refer to the biological, social and economical motives behind infanticide, such as disability, poverty, hunger, population growth, family honour, jealousy and revenge.

Secondly, it is important to admit that the motives behind these practices are based on oral traditions passed and written down hundreds of years after the introduction of Christianity. Several of the Norwegian medieval laws contain prohibition acts against the cultural practice of infanticide. The texts instruct people on how to ensure the survival of newborns (Mundal 1988, 10, note 3). The need to recommend methods to ensure survival may indicate the cultural continuity of infanticide (Mundal 1988, 11-12). The literature does not deal with the normal cases of infanticide, as the content describes the methods, which deal with saving the children (Mundal 1989, 122) (see note 2). As with the biblical legend on the daughter of pharao’s saving of baby Moses floating on the river Nile, the idealistic model of this type of Christian practice is reflected in the Norwegian medieval folk song Agnus dei (the Lamb of God) where the Virgin Mary finds a child who has floated ashore and takes care of it. As an intersession the child asks the Virgin to defend its parents for fear the deed would lead to their expulsion from heaven (Liestøl et al. 1946, 13-14, 113).

CIRCUMSTANCE

CLASS AND GENDER

Disability

All classes

Poverty and hunger

The poor, servants, slaves, the fatherless

Family honour

Freeborn women of the elite

Population surplus

Females and males

Table 1. The circumstances related to the practice of infanticide according to the Norse sagas and the medieval Scandinavian laws (based on Mundal 1988, 1989). The newborns in the spring – rite of sacrifice versus infanticide In order to come to terms with the somewhat confusing and contradictive aspects appearing in the analysis, and to explain the context and circumstance of the Norwegian bog-find, we need to discuss the type of placement in the location and the problematic use of the term sacrifice in relation to ritual in the literary sources. We will consider how these aspects are implicit or integrated in the analysis of mortuary treatment of the bog-children, and how this may influence the explanation of the bog-find.

In the Norse tradition of childbirth, the duty of the parents was to decide on keeping the newborn or not. The poems and sagas mention water as part of an initiation rite in the social process of welcoming the new child into the family. The initiation ritual was the act of ‘pouring the water’, and the ceremony was followed by the father (or sometimes the mother) giving a name. In the sagas the kings were naming children of the aristocracy (Näsström 2001, 115-116, 2002, 71). How severe, then, was the practice of exclusion before the child had become socially alive (Beausang 2005, 119)? The extended Norse

Narratives about human sacrifice of children occur in the legends and myths of the Norse, and seem to indicate continuity in the ideology of Iron Age society. In these narratives children function as idealistic models, given away as valuable gifts to deities. The importance of water has been established to define and explain the practice of children in bog-finds, and the practice of performing human deposals in lakes seems to culminate between BCE 300-CE (Näsström 2001, 54). In part this period of the early Iron Age overlap the dating of the Norwegian bog-find, but children in bogs have also been attested to the late Iron Age (CE 500-1000) in the archaeological evidence (Monikander 2010, 80, table 5).

2 In one of the tales a newborn girl is not given a name since her parents have decided to get rid of her. A man finds the infant, takes care of her and gives the child a name. The parents later recognize the child, but they are not allowed to get rid of the girl and put an end to her. According to the rules this would mean the same as committing murder upon the child after it had been sanctioned with water (Näsström 2001, 115–116). The child found by someone who felt sorry for the baby, or by someone rich and wealthy. The child is placed between two stones and a slab put on top, and a piece of pork is put in the mouth before the people go away. Or the piece of pork is put in the mouth, then a shelter is made beneath a root of a tree, the child is carried out and left. This type of caretaking has been compared with the handling of slaves (Mundal 1989, 123–124). Behind these contradictive motives there are the thoughts of letting the child die without the participation of the parents, or in spite of their participating in the event (Hovstad 1956, 347).

The Norwegian bog-children were discovered in a wetland place, formerly a spring in the landscape, which were located away from the settlements. Seemingly, the newborns had not been accompanied by adults or material objects at the disposal. However, they were exposed to the element of water. A watery placement is reflected in the early Medieval literary sources dealing with infanticide. According to the Medieval Norwegian 53

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Christian laws, the carrying out of children was forbidden (Mundal 1988, 10, note 4), but people were instructed to place disabled children under heaps of stones by the sea at the high mark of water (Mundal 1989, 132). The practice of watery placement for children differ from the treatment of the dead of all kinds stated in the Medieval Norwegian prohibition acts; i.e. covering corpses by stones or digging them down into the earth (Mundal 1989, 132). Similarly, the sacrifice of humans in springs or wells has been considered a religious activity by Tacitus (see also Sundquist 2007, 121). The discussion regarding the relationship between pagan fertility beliefs, fertility goddesses (cf. note 1) and Christian polemics towards continuing folk custom are centred around the use of springs within the landscape for various offerings (Bøe 1967, 58; Näsström 2001, 82-84, 2003, 26-72; Sundquist 2007, 116, 120-122; see also Semple 2011).

have to ask whether the bodies of the newborns had been ritualized; i.e. decapitated or not? Could the rest of their corpses have been buried elsewhere? According to the study of Iron Age Man in Denmark, only children of high social rank were given burial treatment in contrast with children of lower ranks. These children were not disposed in graves, and only a few have been discovered in pits, wells and bogs (Sellevold et al. 1984, 282). Therefore we will approach the ideology and practice of mortuary treatment of the newborns on the general level of inclusion and exclusion of human beings as biological and social bodies in early Iron Age society. In a life and death perspective, given the time gap between the date of the bog-find and the literary sources applied in the analysis, it is important to consider both the development of settlement patterns and ideology, attitude and practice of mortuary treatment of the Iron Age society in the Jæren region of south-west Norway. At the time of disposal of the newborns in the spring, the coastal region was transformed by cultural change due to extended agricultural and settlement expansion and population growth CE 200 and to CE 530/550. Due to the position of ‘northway’ – the seaway along the coast of Norway –, the development of an expanding warrior society with a chiefdom organisation representing a redistributive coastal economy, reflected a manifold trading system of production surplus from local and regional areas in exchange with luxurious goods from the Roman areas on the continent (Myhre 1978; 2003, 7072). A set back in climate and settlement occurred around the time between CE 531 and CE 536 and again after the Justinian plague (CE 541-542) (Myhre in prep.). Following the breakdown of the chiefdom organisation and restructuring of the farm settlements in the 7th century CE, a state society with expanding centralised settlement patterns and petty kingdoms was established in the Viking Period (CE 700/750-1030) (see Myhre 1978, 2003).

In her review on the use of the term sacrifice in anthropology and other academic fields Åsa Berggren has discovered some problematic issues related to implicit Judeo-Christian values, unclear categorization, and the inclusion of a greater variety of rituals. Sacrifice is often regarded as a universal and taken for granted. It is sometimes used as a covering term which may homogenize our understanding of varying practices. The offering concept and the term ‘ritual deposition’ have been brought forward to compensate for the awareness of problems among scholars, but also shown to be problematic, such as in the use of the term ritual; i.e. actions considered ritual by the participants may fall outside of the definition (Berggren 2010, 378-379, see also Bell 1992). In general an initiation rite is a critical event for the individual, the family and the society; a crossing into the world of the family and a symbolic act of accepting or refusing its humanity, gender and affiliation. The practices surrounding the circumstance of childbirth have much in common with formalised religious and ritual behaviour (Beausang 2005, 126). The first time a newborn was put to the breast to be fed may have represented an important social rite to mark the choice between whether the infant should live or die (Mundal 1989, 131) as the Medieval Norwegian laws indicated clearly the proper time to get rid of the newborn. The correct time to remove the baby was the hour after delivery before the first feed.

In the world view of the Nordic religion, the dead were seen as possessing a higher degree of numinous power stronger and different than the living (Schjødt 2008, 2010, 175). The Iron Age society was based on family estates held together with a belief system in which the grave cult of family ancestors and the descendants played an important part. With regard to mortuary treatment, the Iron Age population practiced a selective burial custom of family representation mainly in the form of cremation graves in burial mounds. A comprehensive study of burial treatment of children in the early Iron Age society in Norway indicates a socio-cultural selection of ranking based on the stratification of social status, position and worth of children in the society (Vik 2007). In the region from CE 200 onwards, variations in the form of inhumation graves, and also to some extent collectively in the representation of cemeteries, added to the burial custom which continued until the introduction of Christianity during the Viking period (Lillehammer 1996a, 1996b; Bukkemoen 2007). In the ancestor burial cult, re-use of burial mounds occurred (Lillehammer 1996a; Bukkemoen 2007), and during the late Iron Age

The question of manipulation in order to create and transform individuals from one state of being to the other through rites of passage elucidates why the context of the bog-find is difficult to explain on the religious level of sacrifice. The critique of the application of terminology in relation to concepts of sacrifice, offering and ritual makes it clear that the bog-find may represent the outcome of ritual or non-ritual behaviour following childbirth, but which material remains have been difficult to observe at the time of retrieval or found without circumstantial evidence in the archaeological material at the time of analysis. Based on the circumstantial evidence, and granted the lack of objects in the Norwegian bog-find, we 54

THE CHILDREN IN THE BOG burials also occurred in the ruins of earlier deserted farm houses (Lillehammer 1996b; Thäte 2007).

present the precocious child, a type of not-realistic narratives about children (Mundal 1988, 10), but which reflect the ideals of children and childhood (Lillehammer 2008b, 98). In representing the family offspring and the potential adults of the future, children were perceived as something becoming in faculty and function (Lillehammer 2010b, 13; for case studies, see LewisSimpson 2008). In the Norse lifecycle, the value set upon childhood was low (Mundal 1988, 10). The sacrifice of newborns may have served as the powerful act of connecting the valuable qualities of the firstborn’s right in the family with the cosmological powers of origin and creation (cf. Helms 1998, 78–79; Lillehammer 2008b). But, if childhood was valued low in the life cycle of the early Iron Age population, the newborn had the lowest potential capacity compared to the physical and mental faculties and functions of older children. In a sociofunctional perspective, the sacrifice of a newborn may have been the easiest way out; i.e. cheating the gods.

Notwithstanding the cultural variations and change in settlement pattern and mortuary treatment, continuous relationships existed in the Iron Age between a family settlement system of ancestor warrior estates and the practicing of a selective burial custom of family members. In periods or at events when normal conditions in the society or the family were threatened by crises, it is relevant to consider the question of similarity and difference in the mortuary treatment of newborns and other age groups in the early Iron Age society. Population growth may have lead to structural transformations such as diminishing land availability, violent competitions between families or interhoushold inequalities; comparable with what happened later in Iceland CE 1000-1050 (see Beck et al. 2004, 837). In a strict sense, both the practices of sacrifice and infanticide form part of a selection process of human offspring that are motivated by the overall goal of the household, family or society in order to survive. The early Iron Age in the Jæren region was a time of cultural change, and children’s bodies, or their fragmented corpses, could have been disposed of otherwise than through burial.

Is it possible to explain some of the newborns in the Scandinavian bog-finds to representing unwanted children? The occurrences of children in the wetlands of Northern Europe, and in particular those with the evidence of newborns (cf. Sellevold et al. 1984; Monikander 2010), indicate various positions of ‘being’ in the world. Infanticide is the ultimate exclusion of a form of otherness by dislocating newborns from the strings of biological and social bondage and material care. In the Norse society, a variety of beliefs and motives existed behind the act of exclusion from the initiation rites of new family members. As people brought unwanted children out far away in the fields and forests, it is claimed that the deserted locations were probably the result of restrictions set against the practice by the medieval Christian law (Mundal 1989, 131).

In the Norse poem Rigstula, Hávamál (i.e. saying of the high one) gives the rules on how to conduct oneself and live a proper life. By breaking the rules on the level of practice adults may commit crimes by crossing the limits of acceptable conduct. But in times of crises there is the duty of gift exchange connected with keeping the rules in order to prevent the worst from happening, or hoping for a change of circumstance. The meaningful code of behaviour on exchange and contracts of gifts (see stanza 145 in the introduction) points out an obligatory relationship between the voluntary given and the reciprocated (Mauss 200, 3). In the ultimate matter of child sacrifice, the ritual means ensuring the contact with the gods through gifts in order to stave off or prevent accidents and pests, or to soften up the gods in other ways (Näsström 2001, 123). In looking upon infant treatment in general, there is a contradictive aspect in the recognition of ‘sameness’ or ‘otherness’ by considering the biological and mental capacity and the social and cultural identity of the infant depending on perceptions of circumstance of and attitude towards children and childhood in the society. A strive for balance of normality is seen in the practice of behaviour. According to Scott’s study of infancy and death (1999, 89, 127), societies manipulate codes of death and give them cultural meanings and acceptance which are associated with fertility and life. Sacrificed infants were ‘given up’ and tended to be older compared to the neonatal victims of domestic infanticide. Her study demonstrates an in-between position of the infant as something biological and social between agency, change and continuity.

Fig. 3. The steep mountain Utburden located at the mouth of a deep fjord at the coast Ryfylke, SW Norway. (Photo: © Roy Mangernes). The social displacing of children outside the vicinity of settlement is found in the distribution of Scandinavian place names. A type of place name distributed in the eastern and western parts of Sweden refers to ‘Barnakälla’, ‘Barnasjön’, ‘Barnatjärn’, i.e. children’s spring, lake, and pool. According to Swedish folklore the

In the Norse world, the literary motives in poems about gods and heroes and prose about myths (Snorre-Edda) 55

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD newborns were thought to originate from lakes, bogs, brooks and the likes in the natural environment, or lakes and pools were the places children had been drowned by an animal, ‘Bäckahesten’ (the brookhorse) (Monikander 2010, 34).

neighbouring regions of the early Iron Age bog-find, two sets of high mountains with the place name ‘Utburen’ are known, one by the side of a deep valley, the other looking down at the steep side of a deep fjord (Fig. 3) and situated on the far outskirts of the largest excavated village location at present in Norway. The pre-historic settlement area of farmsteads is dated between the Bronze Age and Late Iron Age (BCE 1500/1400-CE 600/650) (Løken 1998, 107; 2005, 282-283, fig. 44). In the long-term aspect of 2000 years to suggest a link between settlement and mountain in the practice of carrying out of children in the landscape seems relevant.

Another type of place name is ‘Utburen’, a Norwegian place name referring to the Norse concept ‘utburð’ (carry out) in relation to children being carried out in the landscape. The place name has a variety of connotations to Nordic folklore on the dead child being. The meaning of the concept varies in the circumstance of time and place in the encounters of the dead-child being, and in the origin, form, behaviour and removal from the scene. The concept equally refers to an aborted foetus, or to a child who has been carried out, abandoned, drowned, or unburied (Pentikainen 1968, 192-194), to a child baptised without a Christian baptism, and the ghost of such a child (Fritzner 1973, 815), or a child who dies without being christened (Heggstad 1958, 762). Also, the ‘utburd’ is a ‘troll’ (ogre, monster) in Norwegian folklore linked to child murder or the ghost of a clandestine childbirth. Sometimes the afterbirth was looked upon as the twin of the child. If the afterbirth was not burned, it could transform into an ‘utburd’. Many legends tell the tales of the ‘utburd’ crying and whining, or hanging itself on people’s back and asking for a name to be given in order to get peace (A and G 1991, 193).

By looking closer into the Nordic dead-child tradition in a landscape perspective the following spatial patterns emerge (Pentikainen 1968, 192-195, table 10) (Table 2). A major place linked to the encounter of the dead-child being is the deserted environment of a forest or a wilderness. The word refers also often to a hiding place of the child’s body or a lake or water. Less represented are the places under a stone or a tree, a cattle shed or a deserted house. The least frequented place of encounter is the hill, such as ‘Utburden’. The range and scope of these locations of encounter support the distribution indicating the disposal places of unwanted children in the landscape, and which are mentioned in the Norse and early Christian literary sources. PLACE OF ENCOUNTERS

The dead-child images in the narratives, and whose ghost figures in a diversity of forms and shapes, especially animals, are perceived as representing beings of otherness in the world. In the transformation they are empowered with a fearful capacity to interfere in the human world of the living (Lillehammer 2008b, 101–102, table 1). This embodiment of a transformative existence in the different ways of otherness of the body and self in ‘being in the world’ (Hedeager 2011, 61-98, 99-104) is based on the ideas of a human/animal duality of soul-splitting and shape-changing central to the Nordic pagan tradition (see Mundal 1974; Steinsland 1990a; Hedeager 2011, 81-85), and which in particular has been explored in the development of Migration Period animalistic art (Kristoffersen 1995). The Norse saga Flatøybok deals with the childhood legend about a warrior to the first missionary king Olav Tryggvasson (CE 968-1000), Torstein Oksefot, who as a child was carried out, but found dying and taken care of by a farmer (see note 2). The saga relates the story about the little child Torstein who once stumbled and fell over his alter ego, a polar bear cub (Bæksted 1984, 200).

NUMBER OF REPRESENTATIONS

Forest, wilderness

15

Hiding place of child’s body

12

Lake, water

10

Under stone or tree

7

Cattle shed or deserted cottage

6

Supra normal place name

3

Hill

2

Table 2. Place of encounter and number of presentations of the dead-child being distributed in the Nordic traditions (based on Pentikainen 1968). Conclusion The Norwegian bog-find of the early Iron Age is of an earlier date than the main body of literary sources brought forward here in order to explain the context of the find. The find is classified among the archaeological evidence of prehistoric human skeletal remains of wetland children from bogs in Northern Europe. In a Norwegian perspective, the bog-find is unique and the date gives rise to a variety of questions. On a long-term scale, compared with ideologies, attitudes and practices related to mortuary treatment of newborn children, the analyses have shown the contextual evidence to connect generally to wetland places of both sacrifice and infanticide in the location of the natural environment: a spring in the landscape. The marginal location of the infant remains in the landscape link the archaeological find to an environment represented in the traditions of Norse

The circumstances of childbirth, nature of death, murder, and supernatural beliefs in apotropaic measures are connected with the word ‘utburd’. The meaning is linked to the time, place and circumstance of being a social outsider, and not the opposite, as an insider and the one included into the human world by the performance of ritual acts. In particular it is relevant to speak of a common dead-child tradition of continuation covering the extreme northern and western parts of the Nordic area (Pentikainen 1968, 190-191). From two of the 56

THE CHILDREN IN THE BOG literature, Nordic dead-child belief, Scandinavian place names and Norwegian medieval folksongs. As wetland places of both sacrifice and infanticide of children overlap to some extent, the interpretation of the bog-find is difficult. The infant remains in the bog were discovered at the margins of a settlement area of central position in the late Roman and Migration Iron Age in the region. The disposals occurred at the onset of and during a period of cultural change in settlement pattern, population growth and mortuary treatment. If human sacrifice was an event caused by crises and not a common event in society, the amount of four to five newborns immersed in the outland spring may indicate circumstances of ritual communication in offering newborns to the gods rather than concealing them from every day walks of the living. By reconsidering the circumstantial evidence in relation to alternative cultural theories, the performance of familial sacrifice by giving away valuable offspring as tributes to the gods at times of unbalance and stress in society is possible to model. The main point of the sacrifice argument is the ritual act of immersing the newborns into a spring of water. The location is regarded as sacred in this relationship and considered a liminal place of transition on the travel through the landscape to the other world.

cosmic order, they may return also for transformation and rebirth in the fertile and nurturing hidden depths of nature’s womb guarded in the spring by a powerful deity (see note 1). As these hypotheses do not disconnect society’s social memory from the use of liminal places in the environment for various purposes, the discourse on sacrifice versus infanticide in representing a social divide between normality and abnormality of children and child treatment becomes less futile and open for further discussions. The bog-find of newborns from the early Iron Age is an important missing link in the discourse on continuity and discontinuity of the cultural practices of infant treatment in the Iron Age of Northern Europe during times of change. In a landscape perspective on a broader scale, in order to explain the appearance and background of wetland children from the territorial and seasonal boundaries and divisions of bogs, marshes, fens, wells, springs, pools, lakes, brooks, streams, rivers and seas, the inclusion of comparative and interdisciplinary studies of children remains inside and outside the vicinity of houses, settlements and natural environments in the landscape are recommended. Analyses of the infant bones, in order to confirm and establish their biological sex, and the origin and nutrition of mothers, have seriously to consider the unique material evidence in relation to scientific methods in the present.

The Nordic dead-child belief seems to suggest circumstances where this motif was not fulfilled, which may indicate a location connected with the practicing of infanticide directed upon unwanted children. This argument is supported by the facts that the newborns were dislocated from the vicinity of settlement and not disposed together with adults and/or material objects. The location in the outland spring connects them as outcasts also to a liminal place in the landscape. In considering liminality a state between danger and purity (Douglas 1966), a third position is possible to suggest; the outland spring representing a transformative boundary between landscape and people, connects the newborn children to dimensions which signify them as powerful beings inbetween nature and culture (Lillehammer 2000, 2008b, 2010a, 23, 2010b, 15).

Acknowledgements My sincere thanks go to Brigitte Lohrke (posthumously) and Siv Kristoffersen for discussions and sharing of references, to Berit Sellevold for giving in press an updated version of the Norwegian bog finds, to Bjørn Myhre for my reading in manuscript a chapter on the cultural history of Hå municipality, to Fredrik Fahlander for keeping me in touch with the Swedish discourse, and most of all to Mike Lally for insisting to know more about the Scandinavian ‘stuff’ than addressed in the Kent-paper. References A and G 1999. Utburden. Aschehoug og Gyldendals Store Norske Leksikon. Oslo, Kunnskapsforlaget.

A strong relationship between humans and nature is seen in the belief system of the Iron Age. This is found to be integrated in the questions underlining the narratives of children and child treatment in the relationships to existence and place on a long term scale: Where do children come from and where do they go to? In the narratives of place they are connected with watery locations in the natural environment; in the narratives of the dead-child their ghosts are represented in the shape of animals. From the perspective of rites of passage (van Gennep 1965), newborn children have passed the threshold of birth, but their existence are in the state of betwixt and between. Their nature of otherness separate and consign them outside the normal structures of society. In the handling of death (Bloch and Parry 1987), they may return permanently and naked from whence they came; to invisibility, to darkness, to the wilderness (Turner 1991, 95). Or to create a balance in the world’s

Arwidsson, G. 1967. Källa. Kulturhistoriskt Lexikon för Nordisk Medeltid från Vikingatid till Reformationstid. Malmö, Allhems Förlag, 53–57. Barth, F. 1976. Forskning om barn i sosialantropologien. Forskningsnytt 21 (5), 25–27. Beausang, E. 2005. Childbirth and Mothering in Archaeology. Göteborg, GOTARC Series B. Gothenburg Archaeological Theses 37. Beck, R. A., Bolender, D. J., Brown, J. A., & Earl, T. K. 2007. Eventful Archaeology. The place of space in structural transformation. Current Anthropology 48 (6), 833–860. 57

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Beilke-Voigt, I. 2008. Kindergräber in frühgeschichlichen hausbefunden Norddeutschlands und Dänemarks – Bauopfer oder bestattungen? In C. Kümmel, B. Schweizer and U. Veit (eds.) Körperinszenierung – Objektsammlung – Monumentalisierung. Tübinger Archäologische Taschenbücher 6, Münster, Waxmann, 537-562.

Past Monograph Series 1. Oxford and Oakville, Oxbow Books. Bukkemoen, G. B. 2007. Alt har sin pris. Stedsidentitet og sosial diskurs på Jæren i eldre jernalder. In L. Hedeager (ed.) Søreiser og stedsidentitet. Jæren/Lista i bronsealder og eldre jernalder. University of Oslo, Oslo Archaeological Series 8, 135-310.

Bell, C. 1992. Tirual Theory, Ritual Practice. New York, Oxford University Press.

Bøe, O. 1966. Menneskeoffer. Kulturhistoriskt Lexikon för Nordisk Medeltid från Vikingatid till Reformationstid. Malmö, Allhems Förlag, 527–528.

Bergerbrant, S. 2011. Barn i mossar. In F. Fahlander (ed.) Spåren av De Små. Arkeologiska Perspektiv på Barn och Barndom. Stockholm University, Studies in Archaeology 54, 161-179.

Bø, O. 1967. Källa. Kulturhistoriskt Lexikon för Nordisk Medeltid från Vikingatid till Reformationstid. Malmö, Allhems Förlag, 57-59.

Berggren, Å. 2010. Med Kärret som Källa. Om Begreppen Offer och Ritual innom Arkeologien. Vägar till Midgård 13. Lund, Nordic Academic Press.

Bæsted, A. 1984. Nordiske Guder og Helte. København, Politikens Forlag. Bæksted, A. 2002. Nordiske Guder og Helter. Myter og Sagn fra Førkristen Tid. Oslo, Aschehoug.

Bintliff, J. 1991. Annaliste/Structural Approach to Archaeology. In J. Bintliff (ed.) The Annales School and Archaeology. Leicester University Press, Leicester, London, 1-33.

Ceruti, M.C. 2004. Human bodies as objects of dedication at Inca Mountain shrines (north-western Argentina). World Archaeology 36, 1, 103-122.

Björkhagar, V. 1995. Gravskicket under barnaåren. En studie av östgötska gravar från övergången mellan bronsålder och järnålder. In B. Johnsen & S. Welinder (eds.) Arkeologi om Barn. Societas Archaeologica Upsaliensis, Occasional Papers in Archaeology 10, 43-56.

Ceruti, M. C. 2010. The religious role of the children in the Andes – Past and present. In G. Lillehammer (ed.) Socialisation. Recent Research on Childhood and Children in the Past. AmS-Skrifter 23, 125-133.

Bloch, M. and Parry, J. 1987. Introduction: death and the regeneration of life. In M. Bloch and J. Parry (eds.) Death and the regeneration of life. Cambridge, Cambridge University Press, 1-44.

Chamberlain, A. 2000. Minor concerns: A demographic perspective on children in the past. In J. Sofaer Derevenski (ed.) Children and Material Culture, 206– 112. London, Routledge.

Boswell, J. 1991. The Kindness of Strangers. London, Penguin Books.

Christensen, C. 2003. Jernalderens offermoser. In L. Jørgensen, B. Storgaard and L. Gebauer Thomsen (eds.) Sejrens Triumf. Norden i Skyggen af Det Romerske Imperium. København, Nationalmuseet, 346-354.

Bowie, F. 2002. The Anthropology of Religion. An Introduction. Reprint. Oxford, Balechwell Publishers.

Coles, B., Coles, J., and Schou Jørgensen, M. 1999. Bog Bodies, Sacred Sites and Wetland Archaeology. Exeter, Wetland Archaeological Research Project.

Bradley, R. 2005. Ritual and Domestic Life in Prehistoric Europe. London, Routledge. Branston, B. 1955. Gods of the North. London, Thames and Hudson.

Crawford, S. 2000. Children, grave goods and social status in Early Anglo-Saxon England. In J. SofaerDerevenski (ed.) Children and Material Culture, 169– 179. London, Routledge.

Bremmer, J. N. 2007. The Strange World of Human Sacrifice. Peeters, Leuwen.

Crawford, S. 2010. Infanticide, abandonment and abortion in the Graeco-Roman and Early Medieval World: Archaeological Perspectives. In Brockless, L. and Montgomery, H. (eds.) Childhood and Violence in the Western Tradition. Oxford and Oakville, Oxbow Books, 59-67.

Brink, S. 2001. Mythologizing landscape. Place and space of cult and myth. In M. Stauberg, O. Sundqvist and A. van Nahl (eds.) Kontinuität und Brüche in Religionsgeschichte. Festchrift für Anders Hultgård. Ergänzungsbände zum Reallexikon der Germanische Altertumskunde. Band 31. Berlin, Walter de Gruyter, 76112.

Cueni, A. 1995. Ein Säuglingsskelett aus seiner mittelalterlichen Latrinengrube in Schaufhausen. Bulletin der Schweizerischen Gesellschaft für Anthropologie 1, 33-37.

Brockliss, L. and Montgomery, H. (eds.) 2010. Childhood and Violence in the Western Tradition. Childhood in the 58

THE CHILDREN IN THE BOG Davidson, H. Ellis 1993. The Lost Beliefs of Northern Europe. London and New York, Routledge.

Glob, P. V. 1969. The Bog People. London, Faber and Faber.

DeMause, L. 1974. The evolution of childhood. In L. DeMause (ed.) The History of Childhood. New York, The Psychohistory Press, 1-74.

Gurevich, A. 1985. Categories of Medieval Culture. London, Routledge & Kegan. Gräslund, A.-S. 2005. Symbolik för lycka och skydd – vikingatida amilutthängen och deras rituella kon-text. In K.A. Bergsvik and A. Engevik (eds.) Fra funn til samfunn. Jernalderstudier tilegnet Bergljot Solberg på 70-årsdagen. Universitetet i Bergen Arkeologiske Skrifter 1, 377-392.

Devlieger, P. J. 2000. The logic of killing disabled children: Infanticide, Songye cosmology, and the colonizer. In J. Hubert (ed.) Madness, Disability and Social Exclusion: The Archaeology and Anthropology of Social ‘Difference.’ One World Archaeology 40. Routledge, London and New York, 159-167.

Hedeager, L. 1999a. Sacred topography. Depositions of wealth in the cultural landscape. In A. Gustafsson and H. Karlsson (eds.) Glyfer och arkeologiske rum – en vänbok til Jarl Nordbladh. Göterborg University, Gotarc Series A volume 3, 292-252.

Dieck, A. 1965. Die Europäischen Moorleichenfunde (Hominidenmoorfunde). Neumünster, Wacholz. Dieck, A. 1972. Stand und Moorleichenforschung. Korrespondenzenblatt 2, 365-368.

Aufgaben der Archäologisches

Hedeager, L. 1999b. Skygger av En Annen Virkelighet. Oldnordiske myter. Oslo, Pax Forlag.

Douglas, M. 1966. Purity and Danger. An Analysis of the Concepts of Polution and Taboo. Routledge, London.

Hedeager, L. 2011. Iron Age Myth and Materiality. An Archaeology of Scandinavia AD 400–1000. London and New York, Routledge.

Ebbesen, K. 1986. Døden i Mosen. Copenhagen, Carlsen Forlag.

Heggstad, L. 1958. Utburðr. Gamalnorsk Ordbok med Nynorsk Tyding. Oslo, Det Norske Samlaget.

Encyclopædia 1974. Infanticide. The New Encyclopædia Britannica. 15 Edition. Chicago, Encyclopædia Brittanica.

Helms M. W. 1998. Access to Origins: Affines, ancestors and aristocrats. Austin, University of Texas Press.

Fabech, C. 1991. Samfundsorganisation, religiøse ceremonier og regional variation. In C. Fabech and J. Ringtvedt (eds.) Samfundsorganisation og Regional Variation. Norden i Norden i Romersk Jernalder og Folkevandringstid. Beretning for 1. Nordiske Jernaldersymposium på Sandbjerg Slot 11–15 april 1989. Aarhus Universitetsforlag, Jysk Arkæelogisk Selskabs Skrifter 27, 283-304.

Henriksen, V. 1993. Verdenstreet. Mennesker og Makter i Odins Tid. Oslo, Aschehoug. Hovstad, J. 1956: Barneutbering. Kulturhistoriskt Lexikon för Nordisk Medeltid. Malmö, Allhems Förlag, 347-348. Hultgård, A. 2001. Menschenopfer. In R. Müller (ed.) Reallexikon der Germanischen Altertumskunde 19, 533– 546.

Fabech C. 1994. Reading society from the cultural landscape. In P.O. Nielsen, K. Randsborg and H. Thrane (eds.) South Scandinavia between Sacral and Political Power. The Archaeology of Gudme and Lundeborg. København, Arkæologiske Studier X, 169-183.

Haavaldsen, P. 1989. “Pojken må förast ut i kärret”. Et uhyggelig funn fra Hå som forteller om skyggesidene ved livet i fortida. Frá Haug ok Heiðni 1989 (1), 174–176.

Fischer, C. 1980. The bog corpses from Bjældskovdal. Kuml 1979, 40–44. Francis, J. 2010. Childhood, sacrifice and redemption. In L. Brockliss, L. and H. Montgomery (eds.) Childhood and Violence in the Western Tradition. Childhood in the Past Monograph Series 1. Oxford and Oakville, Oxbow Books, 28-32.

Kaliff, A. 2001. Ritual and everyday life – the archaeologist’s interpretation. In M. Stauberg, O. Sundqvist and A. van Nahl (eds.) Kontinuität und Brüche in der Religionsgeschichte. Festchrift für Anders Hultgård. Ergänzungsbände zum Reallexikon der Germanische Altertumskunde. Band 31. Berlin, Walter de Gruyter, 442-463.

Fritzner, J. 1973. Utburðr. Ordbog over Det Gamle Norske Sprog. Tredje bind. Oslo-Bergen-Tromsø, Universitetsforlaget.

Kristoffersen, S. 1995. Transformation in Migration Period animal art. Norwegian Archaeological Review 28 (1), 1–17.

Formisto, T. 1993. An Osteological Analyses of Human and Animal Bones from Levänthula. Stockholm.

Lee, A. K. 1996. Attitudes and prejudices towards infanticide: Carthage, Rome and today. In J. Sofaer Derevenski, (ed.) Perspectives on Children and 59

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Childhood. Department of Archaeology, Archaeological Review from Cambridge 13 (2), 1994, 65-79.

Beiträge zur Archäeologie und Geschichte des 1. Jahrtausends. Bd. 9. Rahden/Westf, Leidorf.

Liestøl K., Semb, K. and Svinndal, Å. 1946. Norske Folkedanser. I. Danseviser. Oslo, Noregs Boklag.

Løken, T. 1998: Hustyper og sosialstruktur gjennom bronsealder på Forsandmoen, Rogaland, Sørvest-Norge. AmS-Varia 33, 107–121.

Lillehammer, G. 1989. A child is born. The child’s world in an archaeological perspective. Norwegian Archaeological Review 22, 89–105.

Løken, T. 2005. Siedlungs-, Gehöft- und Hausformen. Reallexikon der Germanischen Altertumskunde 28, 282– 293.

Lillehammer, G. 1996a. Death, family and gender – life’s starting point? Kvinner i Arkeologi i Norge 21, 61–82.

Maskell, D. 2010. Child sacrifice in Early Modern Europe: Text, image and fantasy. In L. Brockliss and H. Montgomery (eds.) 2010. Childhood and Violence in the Western Tradition. Childhood in the Past Monograph Series 1. Oxford and Oakville, Oxbow Books, 38-44.

Lillehammer, G. 1996b. Død og Grav. Gravskikk på Kvassheimgravfeltet, Hå i Rogaland, SV Norge. Death and Grave. Burial Rituals of the Kvassheim Cemetery, Hå in Rogaland, SW Norway. Museum of Archaeology, AmS-Skrifter 13.

Mauss, M. 2001. The Gift. Form and Reason for Exchange in Archaic Societies. Routledge Classics. London, Francis & Taylor.

Lillehammer, G. 2000. The world of children. In J. Sofaer Deverenski (ed.) Children and Material Culture, 94–105. London, Routledge.

Mejsholm, L. 2009. Gränsland. Konstruktion av tidig barndom och begravningsritual vid tiden för kristnandet i Skandinavien (Borderland. Construction of Early Christian Childhood and Burial Rituals during the Christianisation in Scandinavia). Uppsala University, Occasional Papers in Archaeology 44.

Lillehammer, G. 2002. Archäologie und kindheit. In K. W. Alt and A. Kemkes-Grottenthaler (eds.) Kinderwelten. Anthroplogie – Geschichte – Kulturvergleich. Köln, Böhlau Verlag, 74-94. Lillehammer, G. 2005. The archaeology of children. Kvinner i Arkeologi i Norge 24, 18–35.

Metcalf, P. and Huntington, R. 1991 [1979]. Celebrations of Death: The Anthropology of Mortuary Ritual. Cambridge, Cambridge University Press.

Lillehammer, G. 2008a. Something about children. In L. H. Dommasnes and M. Wrigglesworth (eds.) Children, Identity and the Past, 96–112. Cambridge, Cambridge Scholar Publishing.

Milner, L. 2000. Hardness of Heart: The Stain of Human Infanticide. Lanham, New York, Oxford, University Press of America.

Lillehammer, G. 2008b. Transforming Images: Exploring Powerful Children. Childhood in the Past. An International Journal, 94–105.

Monikander, A. 2010. Våld och Vatten. Våtmarkskult vid Skedemosse under Järnålder. Stockholm University, Stockholm Studies in Archaeology 52.

Lillehammer, G. 2010a. Archaeology of children. Arcqueología de la infancia. Complutum 21(2), 15–45.

Montandon, G. 1934. L’ogonénése culturelle. Traité d’ethnologie cyclo-culturelle et d’ergologie systématique. Paris, Payot.

Lillehammer, G. 2010b. Introduction to socialisation. In G. Lillehammer (ed.) Socialisation. Recent Research on Childhood and Children in the Past. University of Stavanger, AmS-Skrifter 23, 9-19.

Mundal, E. 1974. Fylgjemotiva i norrøn litteratur. Skrifter fra Instituttene for nordisk språk og litteratur ved Universitetene i Bergen og Oslo 5. Oslo, Universitetetsforlaget.

Lewis-Simpson, S. (ed.) 2008. Youth and Age in the medieval North. The Northern World 42. Leyden Boston, Brill.

Mundal, E. 1988. Forholdet mellom born og foreldre i det norrønne kjeldematerialet. Collegium Medievale 1, 7–26.

Lohrke, B. 1999. Mädchentötung in der Merovingerzeit? Überlegungen anhand von Gräberfeldern des alamannsichen Raums In S. Brather, C. Bücker and M. Hoeper (eds.) Archäologie als Sozialgeschichte. Studien zu Siedlung, Wirtschaft und Gesellschaft im frühgeschichlichen Mitteleuropa. Festchrift für Heiko Steuer zum 60. Geburtstag. Rahden/Westf., Leidorf, 113118.

Mundal, E. 1989. Barneutbering. In H. Gunneng (ed.) Kvinnors Rosengård. Medeltidskvinnors Liv och Hälsa, Lust och Barnafödande. Stockholm, Centrum för kvinnoforskning vid Stockholms universitet 1, 122-134. Munksgaard, E. 1984. Bog bodies – a brief survey of interpretations. Journal of Danish Archaeology 3, 120– 123.

Lohrke, B. 2004. Kinder in der Merovingerzeit. Gräber von Mädchen und Jungen in der Alemannia. Freiburger 60

THE CHILDREN IN THE BOG Mustakallio, K. & Laes, C. (eds.) 2011. The Dark Side of Childhood in Late Antiquity and the Middle Ages. Unwanted, Disabled and Lost. Childhood in the Past Monograph Series 2. Oxford and Oakville, Oxbow Books.

Schjødt, J. P. 2010. Ideology of the ruler in Pre-Christian Scandinavia: Mythical and ritual relations. Viking and Medieval Scandinavia 6, 161–194. Schjødt, J. P. 2008. Initiations between Two Worlds: Structure and Symbolism in Pre-Christian Scandinavian Religion. The Viking Collection 17. Odense, University Press of Southern Denmark.

Myhre, B. 1978. Agrarian development, settlement history, and social organization in Southwest Norway in the Iron Age. New Directions in Scandinavian Archaeology. Studies in Scandinavian Prehistory and Early History 1, 224-271. . Myhre, B. 2003. The Iron Age. In H. Helle (ed.) Prehistory to 1520. The Cambridge history of Scandinavia. Volume I. Cambridge, Cambridge University Press, 60-93.

Sellevold, B. J. 1987. Funn fra Bø, Hå, Rogaland. Innberetning til Arkeologisk museum i Stavanger, juni 1987. Topographic Archives, Museum of Archaeology, University of Stavanger. Sellevold, B. J. 2005. Håland gnr 64 bnr 15, Hå k., Rogaland. Undersøkelse av menneskebein funnet i myr. Rapport Antropologisk undersøkelse 4-2005. Topographic Archives, Museum of Archaeology, University of Stavanger. Sellevold, B. J. in press. Myrskjeletter – sjeldne førromerske funn fra Hedmark. Published in Oslo, Viking 2011.

Myhre, B. in prep. Førhistoria. Hå kulturhistorie. To be published in 2014. Nordland, O. 1967. Offer. Kulturhistoriskt Lexikon för Nordisk Medeltid från Vikingatid till Reformationstid. Malmö, Allhems Förlag, 514–524.

Sellevold, B. J., Lund Hansen, U. and Balslev Jørgensen, J. 1984. Iron Age Man in Denmark. Prehistoric Man in Denmark III. København, Det Kongelige Nordiske Oldskriftselskab.

Näsström, B.-M. 2001. Blot. Tro og Offer i Det Førkristne Norden. Oslo, Pax Forlag. Näsström, B.-M. 2002. I livets skeden. Om passageriter i fornskandinavisk religion. In K. Jennbert, A. Andrén, and C. Raudvere (eds.) Plats och Praxis – Studier av Nordisk Förkristen Ritual. V¨gar till Midgård 2. Lund, Nordic Academic Press, 69-85.

Sellevold, B. J. and Næss, J.-R. 1991. Mennesker i myr. Gunneria 64, 429-442. Semple, S. 2011. Sacred spaces and places in preChristian and conversion period Anglo-Saxon England. In H. H. Hameron, D. A. Hinton and S. Crawford (eds.) The Oxford Handbook of Anglo-Saxon Archaeology. Oxford: Oxford University Press, 74-763.

Näsström, B.-M. 2003. Quellheiligtümer und Quellkult. In R. Müller (ed.) Reallexikon der Germanischen Altertumskunde 24, 26–29.

Stavrakopoulou, F. 2010. Child sacrifice in the Ancient World: Blessing for the beloved. In L. Brockliss, and H. Montgomery (eds.) Childhood and Violence in the Western Tradition. Oxford and Oakville, Oxbow Books, 22-32.

Näsström, B.-M. 2004. Vid gudarnas bord – heliga platser och offerplatser i förskandinavisk religion. Oslo Archaeological Series 2, 52–67. Näsström, B.-M. 2010. Frøya. Den store gudinnen i Norden. Oslo, Pax Forlag. Parker Pearson, M. 2003. The Archaeology of Death and Burial. Phoenix Mill, Sutton Publishing.

Stead, I. M., Bourke, J. B. and Brothwell, D. 1986. Lindow Man: The Body in the Bog. London, British Museum Publication.

Pentikainen, J. 1968. The Nordic dead-child tradition. Nordic Dead-Child Beings: A Study in Comparative Religion. FF Communications No. 20. Helsinki, Academia Scientiarum Fennica.

Steinsland, G. 1990a. Antropologiske og eskatologiske ideer i førkristen nordisk religion. Problemet om kognitiv kontinuitet eller diskontinuitet mellom de to idekompleksene. Collegium Medievale 1 (3), 59-71.

Peel, C. 2010. Gutas Saga. The History of the Gotlanders. Viking Society for Northern Research. Texts Series 12. Exeter, Short Run Press.

Steinsland, G. 1990b. The change of religion in the Nordic countries – A confrontation between two living religions. Collegium Medievale 2 (3), 123-135.

Pieper, P. 2002. Moorleichen. Reallexikon der Germanischen Altertumskunde. Zwanzigster Band. Berlin, Walter de Gruyter, 222–229.

Steinsland, G. 2005. Norrøn Religion: Myter, Riter, Samfunn. Oslo, Pax Forlag. Steinsland, G. and Sørensen, P. Meulengracht 1994. Menneske og Makter i Vikingenes Verden. Oslo, Universitetsforlaget.

Scott, E. 1999. The Archaeology of Infancy and Death. Oxford, BAR Publishing. 61

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Stjernquist, B. 1998. The basic perception of the religious activities at cult-sites such as springs, lakes and rivers. In L. Larsson and B. Stjernquist (eds.) The World View of Prehistoric Man Kungl. Vitterhets Historie och Antikivitets Akademien. Stockholm, Konferenser 40, 158-178.

Vik, B. 2007. Barn og Identitet. Ei Analyse av Barnegraver frå Eldre Jernalder i Norge. Unpublished Master Thesis in Archaeology. Institute of Archaeology, University of Bergen. Vuolanto, V. 2011. Infant abandonment and the Christianization of Medieval Europe. In K. Mustakallio and C. Laes (eds.) The Dark Side of Childhood in Late Antiquity and the Middel Ages. Oxford and Oakville, Oxbow Books, 3-19.

Ström, F. 1986. Bog Corpses and Germania. In G. Steinsland (ed.) Words and Objects. Towards a Dialogue between Archaeology and History of Religion. Oslo, Norwegian University Press, 223-229.

Wahl, J. 1991. Menschliche Knochenreste aus mehreren römischen Brunnen aus Pforzheim. Fundberict BadenWürttemberg 16, 509–525.

Sundquist, O. 2007. Kultledare i Fornskandinavisk Religion – A Collection of Papers. Occasional Papers in Archaeology 41. Uppsala, Uppsala University.

Wahlgren, K. Hauptman 2002. Bilder av Betydelse. Hällristningar och Bronsålderslandskap i Nordöstra Östergötland. Stockholm Studies in Archaeology 23. Stockholm, Bricoleur Press.

Tacitus, C. 1923. Folk og Seder i Germania. Frå latin ved Halvdan Koht. Klassiske bokverk i norsk umsetjing utgjevne av Studentmållaget 5. Oslo, Det Norske Samlaget. Tedéen, S. 2002. Livscykel och landskap: Om bronsålderrösen och skärvstenshögar i Södermanland. Tidsperspektiv. Tidskrift för Arkeologisk Samällsperspektiv 1–2 [2001], 62-82. Thorvildsen, E. 1952. Menneskeofringer i Oldtiden. Kuml 1952, 32–48. Turner, V. 1991. The Ritual Process. Structure and Antistructure. Ithaca, Cornel University Press. Turner, R. C. and Briggs, C. S. 1986. The bog burials of Britain and Ireland. In I. M. Stead, J. B. Bourke, and D. Brothwell (eds.) Lindow Man: The Body in the Bog. London, British Museums Publications, 144-176. Turner, R. and Scaife, R. 1995. Bog bodies: New Discoveries and New Perspectives. London, British Museum Press. Turville-Petre, E.O.G. 1964. Myth and Religion of the North. The Religion of Ancient Scandinavia. London, Weidenfeld and Nicolson. Thäte, E. S. 2007. A question of priority: Houses and barrows as places of burials in Late Iron Age Scandinavia. Anglo-Saxon Studies in Archaeology and History 14. Oxford, Oxbow, 183-193. Ullén, I. 1994. The power of case studies. Interpretation of a Late Bronze Age settlement in central Sweden. Journal of European Archaeology 2 (2), 49–61. van den Sanden, W. 1996. Through Nature to Eternity: The Bog Bodies of Northwest Europe. Amsterdam, Batavian Lion International. van Gennep, A. 1965. The Rites of Passage. London: Routledge and Keegan Paul.

62

Parenting, child loss and the cilliní of Post-Medieval Ireland Eileen Murphy

was entirely different and I believe the reason behind this change was the birth of my own daughter, Abigail, two years previously. I was now a mother myself and when I looked at the skeletons this time it was like an electric shock had hit me and I became aware of the powerful emotions and feelings that may have surrounded the burial of each of the dead babies that I was studying. I am of the opinion that archaeologists are often guilty of ignoring the powerful physiological responses that are associated with pregnancy, birth and certainly motherhood. As archaeologists we are trained to be objective in our work and to ensure that we do not let our own Western ethnocentric 21st century experiences impinge upon our interpretations. However, we know that cilliní were being used until the mid-20th century; the final examples of the monument type went out of use less than 50 years ago. I felt sure that my own thoughts and emotions towards these burial grounds could not be all that different to those of my recent ancestors.

Introduction Cilliní, or children’s burial grounds, were the designated resting places for stillborn and unbaptised children who were considered unsuitable for burial in consecrated ground. Although traditionally associated with the burial of unbaptised infants, oral lore has also identified the mentally disabled, strangers, the shipwrecked, criminals, famine victims, and people who had committed suicide as individuals who would also be buried within cilliní (Hamlin and Foley 1983, 43). Locations for this class of burial ground were diverse and included deserted churches and graveyards; archaeological sites including megalithic tombs, secular earthworks and castles; natural landmarks and boundary ditches; sea or lake shores and cross-roads (Ó Súilleabháin 1939). The sites appear to have proliferated during the Post-Medieval period as a result of the stringent reforms enacted by the Roman Catholic Church during the Counter-Reformation (Donnelly and Murphy 2008).

The exclusionary nature of the sites has been interpreted as a reflection of the ‘ambiguous category of the dead infant’ with the ‘liminal state of the infant soul’ spending all eternity in Limbo being reflected in the ‘spiritually marginal location of their burial’ (Finlay 2000, 408). I believe that this conclusion is too simplistic and the objective of this paper is to re-evaluate the archaeological evidence associated with the burial of dead infants within cilliní and to place the findings within the context of oral accounts of women’s experiences of childbirth and the death of young infants in early modern Ireland. Although the collective Irish society may have paid little, if any, attention to cilliní I would like to propose that others, in particular the parents of the dead infants would have had a different perspective which placed great importance on these sites. Rather than regarding them as marginal and liminal places they would have been very much in the thoughts and minds of these women and men.

In recent years this class of monument has received academic attention in a range of studies throughout Ireland and a number of cilliní have been excavated, including Castle Carra, County Antrim, and Tonybaun, County Mayo. This previous research has largely tended to explore the regional distributions of the sites (e.g. Crombie 1990; Dennehy 1997; Donnelly et al. 1999), their associated folklore (e.g. Ó Súilleabháin 1939; O’Connor 1991), and the osteology of the individuals buried within the sites (e.g. Murphy and McNeill 1993; Hurl and Murphy 1995; Lynch 1998). Recent work has also attempted to relate the topographical locations of cilliní to van Gennep’s (1909, 10) Rite of Transition or the liminal phase. This work has proposed that the apparent marginal locations of these monuments in conjunction with the nature of the mortuary ritual accorded to the individuals buried within cilliní is a reflection of the liminal qualities of Limbo, where the souls of unbaptised children are considered to reside in the afterlife (Dennehy 1997; Finlay 2000).

Baptism Within Christian theology the necessity for infant baptism is based on the teachings of Jesus, who stated: ‘truly, truly, I say to you, unless one is born of water and the Spirit, he cannot enter the kingdom of God’ [John 3.5] (Lawler et al. 1976, 468). Writing in the 4th century, St. Augustine of Hippo stated that the souls of unbaptised children were condemned to hell because of their original sin (Orme 2001, 124). The harsh doctrine of St. Augustine was later modified by the Medieval Church which introduced the concept of Limbo – limbus inferni – which can be translated as a place on ‘the edge of hell’ (ibid.). Most theologians followed the beliefs of St. Thomas Aquinas who taught that unbaptised infants would definitely not have any personal suffering after

I first encountered cilliní in 1995 when I undertook the osteological analysis of the infant remains recovered during the first season of excavation at Castle Carra, County Antrim, which had been reused as a cillín. I was fascinated by the concept of this site type and excited at the opportunity to examine a corpus of extremely well preserved infant skeletons. But, I had no real appreciation of the emotive nature of this type of burial ground – I was approaching my research in a scientific, objective manner, which is of course commendable, but only up to a point. In 2002 I was given a second assemblage of infant skeletons from Castle Carra to analyse, but this time, however, my attitude towards these dead infants 63

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD see if it is possible to gain an insight into the forces and emotions that may have lain behind the creation of children’s burial grounds.

death. It is believed that although these children would be deprived of the ‘the blessed vision of God because they have died without grace; God will bless them with natural happiness’ (Lawler et al. 1976, 529). Within this theological context, it is not surprising to find that urgency was given to the baptism of Christian newborn babies in order to prevent them going to Limbo should they die.

Geographical location of cilliní In most discussions of cilliní the idea that these individuals were excluded and marginalized from society is a central theme. The fact that these individuals were buried in unconsecrated burial grounds separate from the wider society would tend to suggest this is obviously the case. Many of their burial locations – town land and field boundaries, field corners, adjacent to water, within bog land or woodland, at cross roads and road sides, on hilltops and within ancient monuments – have been interpreted as an indication of the liminal and marginal nature of these burials. It is considered that the liminal nature of the geographical location of such burial grounds is a reflection of the liminal qualities of Limbo (Crombie 1990, 57-62; Dennehy 1997, 59-64; Finlay 2000; Dennehy and Lynch 2001, 22). While I accept that these individuals were treated in an exclusionary manner by the broader society it could be argued that families would have had natural emotional ties to their dead relatives, particularly their dead children. As such, they may have selected prominent places in the landscape for the burial of these individuals – places that would not be forgotten or disturbed.

Fionnuala Nic Suibhne (1992) has provided information from a number of oral accounts which clearly indicate the importance of baptism in early modern Ireland. In County Tyrone it would appear to be the case that private baptisms by laypeople took place the moment a child was born and this would then be followed by an official christening in the church (Nic Suibhne 1992, 60). One informant recounted that baptism would occur ‘as soon as possible after a child is born. One time they wouldn’t let clear on them. They’d have it christened before dark and none would hardly let it into the next day’ (Nic Suibhne 1992, 18). A woman in Donegal related that she used to get someone with a car to call to her house on the day she gave birth so that the godparents could bring the newborn baby to the church to be baptised (Nic Suibne 1992, 58). Folklore accounts pertaining to unbaptised children have indicated that the souls of unbaptised infants are frequently depicted as weak lights (O’Connor 1991, 66). An informant from County Antrim, for example, recounted a tale concerning a mother of an unbaptised child who had died who had seen ‘a wee dim light’ during the night. When she told the priest about her vision he had told her to pray very earnestly for the soul of the child. The reason for this was that the child needed her prayers since it was not as high up in heaven as other people (Nic Suibhne 1992, 66).

The documentary and archaeological evidence has indicated that specific locations – archaeological monuments, close proximity to a church or natural landmarks – were deliberately selected as suitable locations for cilliní. Emer Dennehy’s thorough study of cilliní in County Kerry has indicated that 48.6% (125/257) of the sites were associated with pre-existing monuments (Dennehy 1997, 37). A notable proportion of the burial grounds were associated with sites that were obviously ecclesiastical in nature, including Early Medieval ecclesiastical enclosures (42 instances) or Medieval churches or graveyards (16 instances) (Dennehy 1997, 38-39). A study of cilliní in the Iveragh Peninsula of County Kerry found a similar trend and it was proposed that people had deliberately selected sites with an earlier religious function because of their former sanctity (O’Sullivan and Sheehan 1996, 323). Dennehy (2001, 22) has even noted that within an ecclesiastical site there appears to have been a deliberate selection of the most sacred aspect of the site for burial at the expense of the apparently more profane areas. She observed, for example, that the majority of burials were clustered in the vicinity of the leacht and gable shrine at the cillín at Killoluaig, County Kerry.

From the preceding accounts it is clear that baptism was of paramount importance to Roman Catholics in early modern Ireland to prevent the souls of dead infants spending all eternity in Limbo. It is within this context of theology and folklore that a necessity arose for the development of children’s burial grounds in Ireland. To date, the majority of dating evidence attained from burials within cilliní has indicated that they proliferated during the Post-Medieval period and it has been proposed that they arose as a result of the stringent reforms that took place within Roman Catholic Church during the CounterReformation (Donnelly and Murphy 2008). As stated earlier there has been an implicit assumption that the exclusionary nature of these sites is an indication that a negative attitude was shown by society towards the individuals buried within the cillín. Although this may be true of collective society in its broadest terms as a result of Church teachings it would not appear to be the case at a more individual level. In the following section of the paper the archaeological evidence will be examined to

Although it can be proposed within a theoretical context that the sites of cilliní were deliberately selected because of their liminal qualities, their sitting can also be explained in more practical terms. It is known that within

64

PARENTING, CHILD LOSS AND THE CILLINÍ OF POST-MEDIEVAL IRELAND

Fig. 1: General view of Castle Carra, County Antrim (Photo: Declan Hurl, Crown Copyright). for burial within a consecrated graveyard. The cillín was located just across the road from the family home, however, and the parents chose to bury their dead baby there so that they could be close to him. It has been proposed that since the cillín was situated adjacent to an early church site this may have further added to its appropriateness for the burial of the baby.

18th century Ireland typical sites of deposition for victims of infanticide included water, dung heaps, soil, privies, public places, waste ground and indoors. Indeed, most of the infants remained unburied (Kelly 1992, 14). The use of an ecclesiastical enclosure, a ring fort or any archaeological monument on the landscape would afford the cillín with physical protection since it would be clearly marked on the landscape. This is in clear contrast to the unprotected burials of victims of infanticide. Prehistorians have written at length about monumentality and Richard Bradley (1993, 2), for example, has observed that ‘monuments are about memory: they join the past to the present’. It is accepted that monuments can change their meaning from period to period without a change to their form. Unless a monument is physically destroyed it is practically impossible to eradicate it from human experience (Bradley 1993, 5). If the unbaptised children found buried within cilliní had been predominantly buried in random locations throughout the landscape then it would be valid to believe that their parents and families were intent on forgetting about them, as would appear to have been the case for victims of infanticide. While society at large may have tried to marginalise these unconsecrated burial grounds, it is possible that they were very much in the hearts and minds of individuals within these same communities.

Finlay (2000, 412) has stated that ‘the host sites were selected because of their marginal location in the contemporary landscape’. Excavations at Castle Carra have revealed evidence of Late Mesolithic activity in addition to the Late Medieval castle (Hurl and Murphy 1996, 21). The site is located on the north Antrim coast overlooking Cushendun Bay and it commands a spectacular view (Figure 1). This was evidently a place which attracted substantial human attention in the past, and the location of a cillín in this spot does not seem to correlate with the idea of a marginal site. Even the creation of a cillín within ‘marginal bog land or woodland’ can be explained in this manner. It has been proposed that this type of land is consistent with liminality since it is often shunned by humans and animals because of its barren quality (Dennehy 1997, 61). An example of a cillín in such a location within County Kerry is the site at Killoluaig (mentioned above), which is described as being ‘located on the edge of marginal land on a raised island in the bog (O’Sullivan and Sheehan 1996, 332 as quoted in Dennehy 1997, 62). The use of poor land for the location of a cillín would be a sure way to ensure that the remains were not disturbed by later agricultural activities, and it may have represented a deliberate choice by the families who, rather than wanting

Crombie (1990, 56) recounts a story which suggests that individuals did hold cilliní in high regard and considered them to be an appropriate place for burial. The last burial in the cillín at Carrownaseer North, County Galway, was reported to have occurred during the 1940s when the present landowner’s baby brother was buried within the cillín. The baby had been baptised and hence was eligible 65

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD practical purposes it would seem necessary for the graveyard to be clearly demarcated on the landscape and the individual graves to be marked. This situation would certainly appear to be true of a substantial proportion of cilliní.

to forget about their dead relatives, wanted to ensure that their bodies were able to rest in peace without danger of disturbance. It has been suggested that leprosaria were deliberately located in liminal positions within the urban landscape because leprosy sufferers were marginal or excluded members of society (Gilchrist 1992 as quoted in Hadley 2001, 50). It has, however, been observed that these hospitals were not hidden away, and that they were in reality highly visible within the landscape of urban environments (Hadley 2001, 50). I believe that a similar case can be argued for cilliní – while the broader society may have selected these sites to have liminal locations they were still very much focal points on the landscape.

Cilliní in County Kerry are frequently found to contain grave markers, often in the form of small stone uprights which generally lack any form of inscription. In a survey of 88 cilliní in the county, Dennehy reported that 72.7% (64/88) of sites had one or more grave markers, with 21 sites having between 80 and 250 markers and three sites having over 300 markers (Dennehy 1997, 31-32). This situation is mirrored in County Galway, where grave markers frequently occur within cilliní which are enclosed (Crombie 1990, 25), and Aldridge’s (1969) brief article on cilliní in County Mayo makes reference to at least eight sites with grave markers, one of which was cross-incised. Cilliní in both Counties Kerry and Galway were found to occasionally contain graves with head- and foot- stones (Dennehy 1997, 34-35; Crombie 1990, 26). In 2003 Joanna Nolan of Mayo County Council excavated the cillín at Tonybaun, County Mayo, where she identified that a number of burials were associated with very distinct grave markers, which took the form of rectangular stone settings (Figure 2). These were associated with the burials of young infants and adults alike (Nolan 2006, 93).

Mortuary ritual at cilliní Finlay (2000, 413) has stated that ‘the character of these sites and the type of burial deposits does not encourage visitation as an active act of remembrance’. I would disagree with this statement since both the topographical features and the excavated remains would tend to suggest that in many cases real effort had been made to demarcate each individual grave. Anthropological studies of modern graveyards have shown that one of the best ways for people to grieve about the loss of their loved ones is to frequently visit their grave (Francis et al. 2005). For

Fig. 2: An example of an infant’s grave marker at the Tonybaun cillín, County Mayo (C103) (drawn by Paddy Ryder). 66

PARENTING, CHILD LOSS AND THE CILLINÍ OF POST-MEDIEVAL IRELAND 2004, 67; McKerr et al. 2009). As such, the only real difference between the grave memorials of those within a cillín compared to those who lie in a graveyard of the period is the consecrated nature of the latter (Figure 3).

Deirdre Crombie (1990, 28-29) has reported inscribed grave markers from at least 21 cilliní in County Galway. In some cases, the stones bear crosses, while in others the grave marker takes the form of an iron cross. Most of these appear to date to the late 19th century and they commemorate adults and young children alike. It is difficult to know if the infants commemorated with memorials were unbaptised or if they were buried within a cillín for some other reason. One example from the cillín at Lavally in Galway may indicate that unbaptised infants were at least occasionally provided with inscriptions. An inscription on a limestone slab and surmounted with a small Latin cross stated: ‘In memory of two infant daughters of Mary and Thomas McKinney, of Culdaff, County Donegal, who died 31 January 1883’ (Crombie 1990, 28). The absence of names for these twin girls may be an indication that they had died prior to their baptism. It has been stated that ‘the simplicity and frequent absence of grave-markers serves to emphasize the anonymity of the infants and individuals interred within, turning the sites into passive memorials rather than places of active remembrance’ (Finlay 2000, 419). The presence of even the most rudimentary grave markers, however, would tend to suggest that the location of the dead individual within the cillín was being marked. Furthermore, the occurrence of inscribed grave markers within cilliní in County Galway is a clear indication that many of the children and adults buried within these burial grounds were being actively remembered, at least by their families. One of the few archaeologists to have considered this aspect of cilliní is Michael Gibbons who has visited most of Connemara’s cilliní. He has recounted his experience of one of his first visits to a cillín and recalled: ‘It was hard to make out the graves because there was a lot of lichen, primroses and beautiful heather. I remember one grave in particular which stood out above all the rest because it had a rush cross with a red bow on it. I remember standing watching it and thinking there’s some woman out there, an old woman now, climbing up here, probably on the anniversary of the death of her child, going to the trouble of scrambling up the hillside to mark this child’s grave. This was not some anonymous death; she was still grieving. She was still acknowledging that this child existed and had a presence, no matter how short. She had put her marker to her child on this little grave in the landscape and this marker was such a powerful symbol’ (Gibbons as quoted in Gannon 1999, 136).

Fig. 3: Nineteenth century grave markers in the Roman Catholic cemetery associated with St. Patrick’s Church, Derrygonnelly, County Fermanagh. (a) General view of the grave markers in the landscape, (b) close up of painted and unpainted iron crosses (Photos: Patrick Murphy). A further characteristic of cilliní is the presence of white quartz on the graves. Crombie (1990, 27) and Dennehy (1997, 36) have observed this practice in Counties Galway and Kerry respectively. Excavations at cilliní have also indicated that white quartz was deliberately associated with these burials. Quartz was associated with the burials at Tonybaun, County Mayo (Nolan 2006, 98), and at Caherlehillan, County Kerry (Sheehan 1996, 40; 2000, 89). Liam de Paor recorded that each of the infant burials at St. Michael’s monastic enclosure, Inishcaltra, County Clare, contained a ‘handful’ of quartz pebbles in addition to a long stone pebble (de Paor 1974, 5). Tom

The majority of individuals living in Post-Medieval Ireland and buried within consecrated ground would not have been able to afford an elaborate and inscribed headstone. One of the predominant funerary markers of this time for the poorer members of society was a simple metal or wooden cross, which would have afforded the individuals interred beneath with a similar degree of anonymity to those people buried within a cillín (Mytum 67

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD a similar manner to those within a consecrated burial ground.

Fanning (1981, 74) reported that concentrations of quartz and sea pebbles were recovered from the interior of the cist-like structures which formed the graves of the cillín at Reask, County Kerry. A number of explanations have been suggested to account for the association of white quartz with cilliní. Crombie (1990, 27) noted that quartz and sea shells have been used for decorative purposes in a variety of cultures since prehistoric times, while Dennehy (1997, 36) proposed that it may have been used as a symbol of purity. Gannon (1999, 136) suggests that the quartz was symbolic of the innocence of the children or hope that ‘these children who had nothing in life might have something in the afterlife’. The tradition of using white quartz stones on modern graves in Irish graveyards has developed both because of the attractive appearance of the stones and their symbolism. Indeed, it is interesting to speculate that the tradition of using white quartz in modern graveyards parallels its use in cilliní. If this is the case then it is also probable that the Catholics who were burying their dead within cilliní were using quartz as a further way of conferring a religious dimension to these burials. One of the clearest archaeological examples of infanticide within the archaeological record has involved the discovery of the commingled remains of almost 100 neonatal infants in a sewer dating to the Late Roman and Early Byzantine periods at the site of Ashkelon in Israel (Smith and Kahila 1992). The casual nature of the burial of these infants has been interpreted as an indication that a lack of regard was shown towards the dead infants (Smith and Kahila 1992, 669). In complete contrast, burials within cilliní tend to respect each other and there is evidence that a certain amount of care was shown during the burial process. Excavations at the cillín at Castle Carra, County Antrim, revealed that the young children interred in the building’s interior were buried in shallow scoops in the sand. A number of the babies appear to have been deliberately surrounded by stones which demarcated the margins of their graves. The orientation of the burials was varied, but the children appeared to have been carefully laid out, either lying on their backs or on their sides. An older child of 1.5 to 2.5 years was buried in an extended supine position with its head to the west, as would be expected of a Christian burial (Hurl and Murphy 1996, 21). John Sheehan (1994, 41; 1995, 43; 1996, 41) reported the presence of both stone-lined and coffin burials at the cillín at Caherlehillan, while at Reask, also in County Kerry, the individuals buried within the cillín were interred in cistlike structures (Fanning 1981, 74). Excavations at Tonybaun, County Mayo, revealed that a number of infants in the cillín were buried in wooden coffins (Nolan 2006, 95), and the remains of wooden coffins have also been identified during excavations at the cilliní at Illaunlaughan, County Kerry (White Marshall and Walsh 1998, 106), and Johnstown, County Meath (Clarke 2002, 15). The use of coffins and stone-lined graves would tend to suggest that an effort was being made to respect the individuality of the infants and others buried within a cillín. Again, this mortuary process can be interpreted as an indication that the burials within cilliní were treated in

Fig. 4: Stone figurine recovered from an infant grave during excavations at the cillín of Reask, County Kerry (after Fanning 1981, 124; re-drawn by Libby Mulqueeny). It is generally uncommon for Christians to be buried with grave goods, although individuals have been excavated who have been found to have been accompanied by a variety of objects. A number of the Post-Medieval individuals from Christ Church, Spitalfields in London, for example, were found to have been buried in association with objects, including wedding rings, glass phials and wooden trinkets. Some of the objects were considered to have been deliberately placed within the coffins, while others may have been accidental inclusions or items of jewellery worn by the deceased (Reeve and Adams 1993, 89). During the excavations at Reask, County Kerry, one infant grave was discovered which contained a small stone figurine, which the excavator considered to resemble an infant in swaddling clothes (Figure 4). This burial was also found to contain four púirthíní or jackstones (Fanning 1981, 127-128). A further two sets of púirthíní were found at the site, at least one of which was considered to have been associated with a cillín burial (Fanning 1981, 138). The recovery of the figurine and game stones at Reask could quite feasibly be interpreted as a sign that affection had been shown towards the dead children by whoever had buried them. Indeed, I recall an experience from my own childhood where my four and a half year old brother, Stephen, was buried with some of his favourite toys – ferocious looking plastic dinosaurs. His brother and sister 68

PARENTING, CHILD LOSS AND THE CILLINÍ OF POST-MEDIEVAL IRELAND about the little box, a few prayers were said, and then we all turned listlessly away, leaving the lonely, unfledged soul to its eternity …’ (ibid., 84-85).

wanted the toys to be included in his coffin so that he would have something to play with in heaven. These objects were not amulets, symbols of tribal affiliation, or an indication that the child was part of a pagan belief system; they were simply toys and were included within the coffin as a sign of the love and affection shown to the child by his siblings.

Finlay (2000, 413) makes reference to Sugrue’s (1993) study of funerary traditions associated with cilliní in the Iveragh Peninsula. She states that ‘the burial would take place at night with little or no ceremony’, although – citing the same work – O’Sullivan and Sheehan (1996, 323) record that the burial ‘sometimes took place at night’. Also citing Sugrue’s (1993, 40) work, Cecil (1996a, 181) states that burial usually took place during the daylight, although if it was an illegitimate child or an early miscarriage it would be buried at night ‘so that no one would know’. It is unfortunate that Sugrue’s important study remains unpublished and is currently unavailable for consulation. Cecil (1996a, 182) also states that ‘the burial of infants usually took place between the hours of sunset and sunrise’. Robin Flower’s (1944) account, however, is a clear indication that at least some babies were buried in the cillín during the day – in a similar manner to those who would have been interred within the consecrated ground of the graveyard.

Three sets of possible twins were discovered during the excavation of the cillín at Tonybaun, County Mayo. In one case the babies appear to have been deliberately buried so that one was held within the arms of the other, a finding which moved the excavators greatly (Joanna Nolan pers. comm.). This is another case where affection would appear to be showing through in the archaeological record. Whoever buried these two children appears to have deliberately positioned them in this manner, presumably so they could remain as close to each other in their grave as they would have been in the womb. A contemporary account of burial in a cillín Oral history and contemporary accounts of the burial of individuals within cilliní are also a useful source of information when studying these burial grounds, although in some cases this information would appear to be contradictory and it may be a reflection that different communities practiced different traditions. In previous studies certain aspects of the funerary ritual have been considered to support the assertion that the people buried within cilliní were treated in an exclusionary and marginal manner (e.g. Finlay 2000, 413). A detailed account of the funeral of an unbaptised baby on Great Blasket Island in County Kerry has been provided by Robin Flower (1944) in his account of life on the island during the early 20th century.

Another factor that is considered significant in the mortuary ritual associated with the burial of infants within cilliní is the fact that the father or a male relative undertook the burial (e.g. Sugrue 1993 as quoted in O’Sullivan and Sheehan 1996, 323). In some regions of Ireland, however, it would appear to be the case that it was unusual for women to attend any funeral. Cooper Foster (1951, 21), for example, recounted that women did not attend funerals in Ulster because it was considered to be ‘not their place’. The account of Robin Flower (1944), however, would tend to suggest that men, women and children all attended the burial of an unbaptised baby at a cillín.

‘A little procession was coming from the top of the village, and from every house, as it passed, the men, women and children came out to join it … A few words told me all. A new-born baby had died, and the father had come to us the night before for wood to make the coffin. He walked now at the head of the procession through the rain, with the little box that he had knocked together from that raw, unhallowed wood under his arm. We too went out and joined the company. It wound through the scattered houses of the village, always increasing; the men wearing their hats of felt, the women with their shawls drawn close about their heads, and all in a speechless trance of sorrow or respect …

It is also possible that the absence of women from the burial of dead infants is simply related to other rituals and practices associated with childbirth at this time. Irish women would have ‘lay in’ and been confined after a birth. The length of time a woman spent in confinement appears to have ranged from a set number of days, up until a number of weeks, depending on the state of the mother’s health. In Donegal, for example, women reported that confinement generally occurred for approximately nine days but that it could last as a long as two weeks (Nic Suibhne 1992, 21). As such, even with the birth of a living child, the mother would not have attended baptism and the child would generally have been brought to the church by its godparents (Ballard 1985, 68). During the period of confinement and prior to undergoing ‘churching’ – a purification ritual which was undertaken in a church by a priest – the woman would have been prevented from carrying out her normal activities – even washing and brushing her hair (Ballard 1985, 69). It is interesting to note that some oral accounts indicate that the unchurched woman was considered to be unlucky and have the ability to bring misfortunate onto others. This negative status within the community has

We turned into a little promontory of the cliff beyond the houses, and stopped in an unkempt space of dank, clinging grass, with stones scattered over it here and there. A man with a spade had dug a shallow grave, and there, amid the sobs of the women and the muttered prayers of the whole assembly, the father with a weary gesture laid away his child. The earth was shovelled back, closing with hardly a sound 69

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD birth of twins at eight months it was clear to Anne D’Ewes and her husband that the children had little chance of survival. It was recorded that the babies’ parents ‘enjoyed them but a few hours’ to their ‘great grief and sorrow’. One of the twins survived for three days and Simonds D’Ewes stated that he was ‘well hoping of the continuance of the life of my little Adrian’ (Houlbrooke 2000, 234-235). Although Houlbrooke could find little documentation pertaining to stillborn or newborn infants it should not be assumed that there was no sorrow at their loss. The death of babies and children appears to have been such a devastating loss for parents in 19th century England that the Church of England felt it necessary to produce literature which would help console the bereft parents and explain why a just God would allow children to die so young (Jalland 1999, 238).

been paralleled to that of the unbaptised infant. Since childbirth was such a private and mysterious event, the woman who had given birth was regarded as being outside society and outside the Church. She would only be accepted back as a member of the community and the Church if she underwent the churching process (Nic Suibhne 1992, 21). It should be born in mind that after a difficult and emotional birth, the mother of a dead baby may have been neither mentally nor physically fit enough to walk to a local cillín to bury her dead baby. Oral accounts of pregnancy loss in early modern Ireland In Post-Medieval Ireland the birth of a child would have been the cause of great celebration. In the early 20th century in North Antrim a custom known as ‘blessed fasting’ was observed when a child was born. It was necessary to entertain and provide food and drink to anyone who called to visit after the birth. Although this would have often entailed real hardship amongst the poor, such provision was a tradition and a matter of pride. To these earlier generations children were not only blessings ensuring the continuity of a family lineage but they were regarded as material assets since they represented potential labourers and providers for the parents in their old age (Cooper Foster 1951, 9).

Having experienced pregnancy and birth three times myself I am well aware of the physiological processes involved. The mother’s body is awash with hormones which generally all serve to help ensure that she bonds with her baby and is in a position to care for it. Humans are biological creatures and modern scientific research has revealed how hormone levels change in females during pregnancy and after birth to encourage maternal behaviour. Indeed, research has now also shown that males are not immune to this type of physiological response, and many expectant fathers have been found to undergo similar hormonal changes which appear to encourage paternal behaviour (Storey et al. 2000; Berg and Wynne-Edwards 2001). Most of these aspects of humanness are instinctive and beyond the individual’s control, as post-natal depression attests. As such, I think it is very difficult to accept that, certainly, the mothers of dead infants would have been able to act as if there had never been a baby. From a perusal through modern literature on miscarriage it is clear that the overriding emotions experienced by parents who have suffered a miscarriage are those of intense grief and loss (e.g. Jones 1990). Studies have indicated that the grief following a miscarriage could be as intense as that following a stillbirth or the death of a baby shortly after birth (Peppers and Knapp as quoted in Cecil 1996a, 190). Christine Moulder (1998, 4) has summarised a range of common mourning symptoms and behaviours that have been observed in women who have suffered a stillbirth. These include being woken in the middle of the night by the baby crying; still feeling the baby kicking after it has been born; being concerned that the baby is warm and comfortable in its grave; having a fear of going mad and contemplating suicide.

The importance of children in Post-Medieval Ireland is further emphasised by the way that a women attained a higher level of status when she married relative to her situation if she remained unmarried. This is emphasised by the manner in which an unmarried female was never referred to as bean but remained a cailín, or was known as a bunbhean or a bean dhíomhaoin, terms which have been interpreted to mean that she was baseless or redundant in function. Marriage was considered to be of such importance since it was only then that a woman had the official permission of her society to have children. As such, it appears to be the case that to marry and have children was the aim of all women at this time (Nic Suibhne 1992, 13). An underlying assumption within some of the previous research on cilliní appears to have been the existence of a lack of emotion towards unbaptised infants by their parents and families. Based on evidence derived from Sugrue (1993), O’Sullivan and Sheehan (1996, 323) stated that dead infants were not mourned in any of the traditional ways. It has been proposed that Medieval parents restricted emotional investment in young children because of the high probability that they would die during infancy, thereby making it easier for them to accept their loss (Stone 1977 as quoted in Jalland 1999, 237). This assertion has been rejected by some scholars, however, and research on Late Medieval and Post-Medieval documents has suggested that high child mortality rates did not equate to reduced parental affection (Jalland 1999, 237). Ralph Houlbrooke (2000, 234-236) has provided a number of examples from 17th century England where parents had grieved for babies who had died after even only a few days of life. In 1633 after the

Individuals would probably have been able to put on a ‘brave face’ because that is what their society dictated. It is highly unlikely that the people living in Post-Medieval Ireland were that different to the island’s current population and in many cases, underneath a brave exterior, they would probably have been grieving. Prior to Vatican II in 1962-65, which changed church regulations concerning the unbaptised and Limbo, the sorrow of losing a child in this life was probably exacerbated by the parents’ fears about what would 70

PARENTING, CHILD LOSS AND THE CILLINÍ OF POST-MEDIEVAL IRELAND They’re buried here and there around the island and the person that goes there now wont know that any body was ever buried there because there was never any sign but on the graves except a couple of stones got on the strand left lying in the form of a cross on top of the grave. For the length of the summer the ground is covered in white daisies’ (Nic Suibhne 1992, 71).

happen to their child in the afterlife. In the most up-todate version of the Catholic Catechism, Limbo is not even mentioned and it is stated: ‘With respect to children who have died without Baptism, the liturgy of the Church invites us to trust in God’s mercy and to pray for their salvation’ (Catholic Church 1995, 289). The emotional turmoil and feelings of guilt would presumably have been even greater for parents whose children had died prior to baptism – not only would they have lost their newborn baby, but they could not even have the comfort of knowing that it had gone to heaven.

In an anthropological study of pregnancy loss in mid 20th-century Ireland, Roseanne Cecil (1996a, 179) noted that there was a dearth of ethnographic writing on pregnancy loss, but she was not surprised at this finding since many aspects of women’s lives have escaped thorough documentation in anthropological research. This situation was considered to partly reflect the interests of ethnographers in the past, in addition to their more restricted access to female informants. She was also of the opinion, however, that women’s attitudes to their pregnancy losses and their willingness and ability to talk about them were the main contributing factor to this lack of information. If a woman’s primary role in society was regarded as the production of children, then the loss of a pregnancy or the birth of a stillborn baby would be regarded as a failure. Furthermore the association of pregnancy loss with a considerable amount of physical pain and blood loss may make it a difficult subject to discuss. Indeed, Cecil has poignantly stated that: ‘The feelings concerning simultaneous birth and death, the death of one who never was, may be virtually impossible to convey’ (Cecil 1996b, 2).

Anne O’Connor (1991) has provided an overview of the folklore pertaining to unbaptised children in Ireland. In general, the children are considered to reside in a separate place or otherworld where they live as small children as opposed to newly born infants. Under particular circumstances the souls of unbaptised children return as ghosts, often in the form of weak lights (O’Connor 1991, 66). She also recounted a number of specific folk stories which would appear to confer negative qualities upon the unbaptised children. These include the tradition that a person who walks on the grave of an unbaptised child will be overcome with starvation and die unless they eat something immediately, or that the person who walks on such a burial will get lost (O’Connor 1991, 70-71). In addition, there was a belief that walking on the grave of an unbaptised child would result in the development of a skin rash (ibid, 72). In contrast, however, in her anthropological study of the female life cycle in early modern Ireland, Fionnuala Nic Suibhne (1992, 69) has stated that:

Cecil’s subjects were twelve elderly women from the north of Ireland whose babies would have died in the period between the 1940s and 1960s. The women ranged in age from 65 to 89 years, and the sample included both Catholics and Protestants (Cecil 1996b, 183). She observed that many of the women undervalued their own worth and the value of their personal experiences (Cecil 1996b, 179-180), presumably because of the nature of the society in which they had lived most of their lives. At the time of their loss, some of the women were given support from their husbands and were able to talk a little about their feelings, but in most cases the husbands had not offered any support. Although some of the women displayed a matter of fact approach towards their miscarriage, others were considerably distressed (Cecil 1996b, 189). One woman, for example, stated: ‘It was just we tried to forget about it, but you don’t forget … I cried to myself but nobody knew I did it’ (Cecil 1996b, 185). Discussion within the family about the loss of the baby was limited, but outside the family it was even rarer. One woman is reported to have said that ‘definitely nobody ever said “we are sorry you lost the baby”. No, it just wasn’t mentioned,’ despite the fact that a wake might be held for a dead child (Cecil 1996b, 185, 187-188). Cecil (1986, 186) commented that the loss of a stillborn or newborn baby at that time did not form part of a shared social memory or event. It would appear to be the case that a pregnancy loss was not considered by others to have been an event that could easily be discussed, and the

‘even though there seems to have been an acceptance of the church’s teaching … insofar as people believed the souls of unbaptised children went to Limbo, and insofar as they buried such children in unconsecrated ground, there often appears to have been a strongly felt respect and affection for unbaptised children in the accounts of many women’. She relates how one Donegal informant said that people should not walk on the laftán (grassy bank) where unbaptised children were buried. In a similar way such children were believed to provide a cure for people with epilepsy (Nic Suibhne 1992, 69, 70). She also recounted how it was traditional for people in the Rosses area of County Donegal to bury people on an island – Oileán na Marbh, the island of the dead. An 80 year old woman from the area stated that: ‘It’s likely that the old people thought a lot of this island, and certainly it’s lovely… The old people made out that it would be a lovely place for a young child who’d die unbaptised, because the priest’s weren’t happy to bury them along with the other people …. It is said, or I always heard it, … that it is the second graveyard in the world which will be called to on the Day of Judgement … May God preserve every poor small creature that is buried there tonight. 71

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD mothers received community.

little

support

from

the

indicates that the cillín itself is still evidently a place of active remembrance. Furthermore, this unbaptised baby boy had not been forgotten and he is still very much in the hearts and minds of his siblings.

wider

A more recent example clearly illustrates the clear sorrow of a mother at the death of her unbaptised babies. In 1994 Mary Salmon from Letterfrack, County Galway, had her lifetime’s wish come true on her 80th birthday when the Catholic Church finally blessed two of her dead children who had been refused a Christian burial 60 years previously. The local priest officially blessed a plot of land by the sea which had seen the burial of over 100 stillborn and unbaptised infants (Clancy 1994 as quoted in Gannon 1999, 147). Mary recalled how, as a twenty year old mother, she watched her husband take her stillborn son away to be buried, noting that ‘I didn’t even get to see the baby … My husband had the child in a little box and took it to the seashore two miles away. Then I lost another baby and it was buried there as well. Hundreds of babies are buried here. We were told they were in Limbo and could not be let into consecrated ground’ (Clancy 1994 as quoted in Gannon 1999, 147.).

Conclusions In her study of pregnancy loss in early modern Ireland Cecil (1996a, 184) observed that for each woman she had interviewed ‘her memories were, on the whole, a very personal and private account of a personal and private event’. As archaeologists we need to be cautious not to assume that a collective societal and religious response is the only one worthy of study or apparent in the archaeological record. In the case of cilliní the archaeological evidence, contemporary accounts and oral lore would tend to suggest that the broader society was responding in a different way to the death of young infants when compared to their close relatives. Although the societal response which excluded these individuals may be more dominant and obvious in the archaeological record it is also possible to extract evidence of a more private and personal nature which can provide an insight into parental attitudes towards the dead infants buried within a cillín.

The previous accounts have focused on the experiences of the mothers of dead infants buried within cilliní and female attitudes towards these burial grounds. It is important to remember, however, that in at least some cases fathers may often have grieved just as much as the dead baby’s mother. In his book investigating various aspects of Gaelic games in Ireland, Keith Duggan (2004, 143-144) recounts a poignant story from the late 1960s about Len Gaynor, a hurling legend from County Tipperary, and his wife Eileen. Their first child, a baby boy called Patrick, died at birth and both parents were distraught. Two weeks later Len was selected to play in a county match, which he did. However, the emotion of the game was too much for him and back in the dressing room, all he could do was to bury his face in a towel and cry. Although well aware of his personal tragedy the rest of the team either did not make the association with his loss or could not broach it at that moment. ‘What’s wrong with you?’ he heard. ‘Where did you get the belt?’ The attitude of even that time was that you moved on. The Gaynor’s did not forget their son, however, and after some 15 years of enquiries they were able to establish where he had been buried in Glasnevin cemetery, Dublin, and they erected a headstone in his memory (Duggan 2004, 143-44).

Acknowledgements I would like to thank Dr. Colm Donnelly, Centre for Archaeological Fieldwork, Queen’s University Belfast, for his support and encouragement during the preparation of this article and for his comments on earlier drafts of the text. I am also very grateful to Joanna Nolan, Mayo County Council, for many stimulating discussions about the cillín at Tonybaun, County Mayo and for her permission to use the illustration displayed in Figure 2, which was drawn by Paddy Ryder. I would also like to thank Michael Gibbons, Walking Ireland, for providing me with valuable information concerning cilliní in Connemara. Thanks are also due to Declan Hurl and Tony Corey, Environment and Heritage Service DOE: NI, for providing me with access and permitting me to use Figure 1; to Patrick Murphy for taking the photographs displayed in Figure 3 and to Libby Mulqueeny, School of Geography, Archaeology and Palaeoecology, Queen’s University Belfast, for redrawing Figure 4.

It should also not be forgotten that the death of an unbaptised infant would have impacted on the child’s siblings. In his biography, Irish sports commentator Micheál Ó Muircheartaigh, who grew up in County Kerry, remembered his brother who died at birth. He recalled the body having been placed in a cradle or cot before it was buried in an unconsecrated graveyard at the site of a number of ogham stones. Although his father and a neighbour were the only ones to attend the burial, the rest of the family witnessed it from a neighbouring hill. The graveyard has since been consecrated and the Ó Muircheartaigh family chose the cillín as the spot to welcome in the new millennium on the morning of 1st January 2000 (Ó Muircheartaigh 2004, 2). This story

References Aldridge, R. B. 1969. Notes on Children’s Burial Grounds in Mayo. Journal of the Royal Society of Antiquaries of Ireland 99, 83-7. Ballard, L. M. 1985. “Just whatever they had handy”. Aspects of childbirth and early child-care in Northern Ireland, prior to 1948. Ulster Folklife 31, 59-72. Berg, S. J. and Wynne-Edwards, K. 2001. Changes in Testosterone, Cortisol, and Estradiol Levels in Men 72

PARENTING, CHILD LOSS AND THE CILLINÍ OF POST-MEDIEVAL IRELAND Finlay, N. 2000. Outside of life: traditions of infant burial in Ireland from cillín to cist. World Archaeology 31, 407-22.

Becoming Fathers. Mayo Clinical Proceedings 76, 58292. Bradley, R. 1993. Altering the Earth. Stroud: Alan Sutton Publishing Limited. Society of Antiquaries of Scotland Monograph No. 8.

Flower R. 1944. The Western Island. Oxford: Oxford University Press (1978, paperback edition).

Catholic Church. 1995. Catechism of the Catholic Church: Pocket Edition. Dublin: Veritas Publications.

Francis, D., Kellaher, L. and Neophytou, G. 2005. The Secret Cemetery. Oxford: Berg.

Cecil, R. 1996a. Memories of Pregnancy Loss: Recollections of Elderly Women in Northern Ireland. In R. Cecil (ed.) The Anthropology of Pregnancy Loss. Oxford: Berg, 179-96.

Gannon, M. 1999. Consummatus. In Brevi. In P. Gannon (ed.) The Way It Was. Renvyle: Paul Gannon, 135-41. Gannon, P. 1999. Sméaróid. In P. Gannon (ed.) The Way It Was. Renvyle: Paul Gannon, 142-4.

Cecil, R. 1996b. Introduction: An Insignificant Event? Literary and Anthropological Perspectives on Pregnancy Loss. In R. Cecil (ed.), The Anthropology of Pregnancy Loss. Oxford: Berg, 1-14.

Gannon P. 1999. The Way It Was. Renvyle: Paul Gannon. Hadley, D. M. 2001. Death in Medieval England. Stroud: Tempus.

Clarke, L. 2002. An early medieval enclosure and burials: Johnstown, Co. Meath. Archaeology Ireland 16:4, 13-15. Cooper Foster, J. 1951. Ulster Folklore. Belfast: H.R. Carter Publications Ltd.

Hamlin, A. and Foley, C. 1983. A women’s graveyard at Carrickmore, County Tyrone, and the separate burial of women. Ulster Journal of Archaeology 46, 41-6.

Crombie, D. 1990. Children’s Burial Grounds in County Galway. Unpublished MA thesis, Department of Archaeology, National University of Ireland, Galway.

Houlbrooke, R. 2000. Death, Religion and the Family in England 1480-1750. Oxford: Oxford University Press (first published 1998).

Dennehy, E. A. 1997. The Cellúnaigh of County Kerry: An Archaeological Perspective. Unpublished MA thesis, University College Cork.

Hurl, D. P. and Murphy, E. M. 1996. Life and death in a County Antrim tower house. Archaeology Ireland 10:2, 20-3.

Dennehy, E. 2001. Children’s burial ground. Archaeology Ireland 15:1, 20-3.

Jalland, P. 1999. Victorian death and its decline: 18501918. In P. C. Jupp and C. Gitting (eds.) Death in England: An Illustrated History. Manchester: Manchester University Press, 230-55.

Dennehy, E. and Lynch, L. 2001. Unearthed secrets: a clandestine burial-ground. Archaeology Ireland 15:4, 203.

Jones, W. 1990. Miscarriage: Overcoming the Physical and Emotional Trauma. Wellingborough: Thorsons Publishers Ltd.

de Paor, L. 1974. Inishcaltra (Holy Island): monastic site. In T. G. Delaney (ed.), Excavations 1973: Summary accounts of Archaeological work in Ireland, 5-6. Belfast: AYIA, IHS and the UAS.

Kelly, J. 1992. Infanticide in Eighteenth Century Ireland. Irish Economic and Social History 19, 5-26.

Donnelly, C. J. and Murphy, E. M. 2008. The origins of cilliní in Ireland. In E. M. Murphy (ed.) Deviant Burial in the Archaeological Record. Oxford: Oxbow Books, 191223.

Lawler, R., Wuerl, D. W. and Comerford Lawler, T. 1976. The Teaching of Christ: A Catholic Catechism for Adults. Dublin: Veritas Publications. Lynch, L. G. 1998. Placeless Souls: Bioarchaeology and Separate Burials in Ireland. Unpublished MA thesis, University College Cork.

Donnelly, S., Donnelly, C. and Murphy, E. 1999. The Forgotten Dead: The cilliní and disused burial grounds of Ballintoy, County Antrim. Ulster Journal of Archaeology 58, 109-13.

McKerr, L., Murphy, E. and Donnelly, C. 2009. ‘I Am Not Dead but Do Sleep Here’: the representation of children in Early Modern burial grounds in the north of Ireland. Childhood in the Past 2, 110-32.

Duggan, K. 2004. The Lifelong Season: At the Heart of Gaelic Games. Dublin: Town House. Fanning, T. 1981. Excavation of an Early Christian Cemetery and Settlement at Reask, County Kerry. Proceedings of the Royal Irish Academy 81C, 67-172.

Moulder, C. 1998. Understanding Pregnancy Loss. London: Macmillan Press Ltd. 73

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Murphy, E. M. and McNeill, T. E. 1993 (published 1996). Human remains excavated at Doonbought Fort, Co, Antrim, 1969. Ulster Journal of Archaeology 56, 120-38.

Sheehan, J. 1994. Caherlehillan: Early ecclesiastical enclosure and ceallunach. In I. Bennett (ed.) Excavations 1993: Summary Accounts of Archaeological Excavations in Ireland. Bray: Wordwell Ltd, 41-2.

Mytum, H. 2004. Mortuary Monuments and Burial Grounds of the Historic Period. London: Kluwer Academic/Plenum Publishers.

Sheehan, J. 1995. Caherlehillan: Early ecclesiastical enclosure and ceallúnach. In I. Bennett (ed.) Excavations 1994: Summary Accounts of Archaeological Excavations in Ireland, 43-4. Bray: Wordwell Ltd, 43-4.

Nic Suibhne, F. 1992. “On the Straw” and other aspects of pregnancy and childbirth from the oral tradition of women in Ulster. Ulster Folklife 38, 12-24.

Sheehan, J. 1996. Caherlehillan: Early ecclesiastical enclosure. In I. Bennett (ed.) Excavations 1995: Summary Accounts of Archaeological Excavations in Ireland. Bray: Wordwell Ltd, 40-1.

Nolan, J. 2006. Excavation of a children’s burial ground at Tonybaun, Ballina, County Mayo. In J. O’Sullivan and M. Stanley (eds.) Settlement, Industry and Ritual. Dublin: National Roads Authority: Archaeology and the National Roads Authority Monograph Series No. 3, 89-101.

Sheehan, J. 2000. Caherlehillan: Early ecclesiastical enclosure. In I. Bennett (ed.) Excavations 1998: Summary Accounts of Archaeological Excavations in Ireland. Bray: Wordwell Ltd, 89.

Orme, N. 2001. Medieval Children. London: Yale University Press.

Smith, P. and Kahila, G., 1992. Identification of Infanticide in Archaeological Sites: A Case Study from the Late Roman-Early Byzantine Periods at Ashkelon, Israel. Journal of Archaeological Science 19, 667-75.

O’Connor, A. 1991. Child Murderess and Dead Child Traditions. Helsinki: Academia Scientiarum Fennica: FF Communications No. 249.

Storey, A. E., Walsh, C. J., Quinton, R. L. and WynneEdwards, K. E. 2000. Hormonal correlates of paternal responsiveness in new and expectant fathers. Evolution and Human Behaviour 21, 79-95.

Ó Muircheartaigh, M. 2004. Micheál Ó Muircheartaigh: The Autobiography – From Dún Síon to Croke Park. Dublin: Penguin Ireland Ltd.

Sugrue, D. 1993. An Examination of Aspects of Tradition Relating to Some Ceallúnaigh in Uaibh Rathaih. Unpublished MA thesis, University College Cork.

Ó Súilleabháin, S. 1939. Adhlacadh Leanbhí. Journal of the Royal Society of Antiquaries of Ireland 69, 143-51. O’Sullivan, A. and Sheehan, J. 1996. The Iveragh Peninsula: An Archaeological Survey of South Kerry. Cork: Cork University Press.

van Gennep, A. 1909. The Rites of Passage. London: Routledge and Keegan Paul Ltd.

Reeve, J. and Adams, M. 1993. The Spitalfields Project. Volume 1 – The Archaeology. Across the Styx. York: Council for British Archaeology Research Report 85.

White Marshall, J. and Walsh, C. 1998. Illaunloughan, Co. Kerry: An Island Hermitage. In M.A. Monk and J. Sheehan (eds.) Early Medieval Munster: Archaeology, History and Society. Cork: Cork University Press, 102111.

74

The Disposal of Dead Infants in Anglo-Saxon England from c.500-1066: an overview Sally Crawford

associated grave yards proliferated throughout AngloSaxon England.

Introduction The archaeology of Anglo-Saxon archaeology has received detailed study over the last 20 years, in particular the evidence from excavated Anglo-Saxon furnished cemeteries from the 5th to the 7th centuries, before the Christianity had been established as the dominant religion in England (Crawford 1991, 1993, 1999; Lucy 1994; Stoodley 2000; Lee 2008). The place of children within the mortuary ritual of later AngloSaxon, unfurnished cemeteries has been less extensively analyzed, in part because the later Christian cemeteries appear to offer a more inclusive, uniform ritual where all members of society, young and old, were given a less socially differentiated ritual. However, important recent research has provided new emphasis on social differentiation in Later Anglo-Saxon England and the place of children within that ritual (Buckberry 2000; Hadley 2010).

Even this brief overview of Anglo-Saxon burial practice from the 6th to the 11th century highlights that there were significant changes in mortuary ritual, both in terms of where the dead were buried, and what they were buried with. These developments reflect major social and spiritual changes taking place within Anglo-Saxon society, the most significant of which was the introduction of Christianity, a new religion which was responsible for systemic changes to burial practices amongst adults. Christianity, introduced to Anglo-Saxon society in the late 6th century, also brought with it the single clearly documented expression of an infant initiation ceremony or rite of passage in the form of baptism. There is no surviving evidence for any ceremony or ritual, before the introduction of baptism, to mark the infant’s arrival into the wider adult community. As well as providing an overview of the current state of knowledge of infant burial in Anglo-Saxon England, this paper will reflect on the extent to which these changes in attitudes affected infant mortuary treatment at this time.

Broadly speaking, the Anglo-Saxon mortuary ritual went through three readily distinguishable phases. In the 5th and 6th centuries, Anglo-Saxon England was a preChristian society, without written records. The ritual, symbolic and social messages of burial practice at this time remain a subject for debate. The predominant method of disposal of bodies in early Anglo-Saxon England was inhumation or cremation, frequently with accompanying grave goods. Within individual cemeteries, burials displayed great variety in terms of grave structure, body position, and type and number of grave goods. Where grave goods occurred with burials, their clearest message was about gender and tribal identity (Pader 1982; Arnold 1980; Brush 1998; Stoodley 2000).

In this paper, the term ‘infant’ will be used to describe young babies and toddlers up to about two to three years of age. The upper limit of this category is deliberately vague. Given the variable state of skeletal preservation and archaeological reporting from Anglo-Saxon cemetery sites, it would be impractical to try to impose a more rigid biological and chronological definition of ‘infant’ on evidence which will not bear this level of accuracy. In addition, for the Anglo-Saxons themselves, the society under study, the idea of ‘infant’ may have been as much a developmental or social construct as a chronological one. One Old English expression for ‘infant’ was ‘þa unsprecendlic cild’ (the child that does not speak), a definition which could include children up to the age of two or more, and which, for all we know, might have been a literal criterion for an age group or cohort in preChristian Anglo-Saxon society. Unless we know that the early Anglo-Saxons had a definition of ‘infant’ based on chronological age, which we do not, then the rigid imposition of chronological age definitions - based on skeletal development - on the surviving evidence from the burial ritual might be unhelpful, because the archaeological evidence may relate to Anglo-Saxon social perceptions of age and identity, which may not map biological age. An age category which includes all children from birth to 36 months is itself somewhat arbitrary, and may disguise more subtle nuances or other age divisions within that category. A new born baby is not the same as a toddler in terms of social interaction, dependence and self-direction, and more accurate ageing

By the 7th century, large cremation cemeteries had been abandoned. Many folk inhumation cemeteries were relocated, and there were a number of significant changes in the inhumation ritual, almost certainly connected to the Christianisation of Anglo-Saxon England (Boddington 1990; Geake 1992; Stoodley 2007). Deposition of grave goods declined, and grave goods, where they were included in the burial, now signalled association with the elite, rather than tribal affiliation, in an increasingly hierarchical society (Geake 1997). These ‘Final Phase’ cemeteries have been interpreted as the earliest Christian burial grounds, rather than the last manifestation of pagan burial practice, though this view is by no means unchallenged (Geake 1997, Kilbride 2000, Crawford 2004). By the 8th century, the majority of burials were without grave goods: the old furnished burial sites had largely been replaced by new field cemeteries and by cemeteries which were increasingly associated with churches. By the 10th century, parish churches with their 75

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD of individual burials under discussion in this paper will be given where the evidence allows.

between the infant and the buriers (Crawford 2000; Stoodley 2000; Thompson 2004, 9-11).

Pre-Christian Infant Disposal Patterns: The Furnished Cemeteries

While the general trend of fewer grave goods, and weakly gendered or gender-neutral grave goods for infants, is identifiable across early Anglo-Saxon inhumation sites, other patterns are visible at a local level. At Blacknall Field, Pewsey (Wiltshire), burials of children under the age of three at the time of death are characterized by necklaces or neck beads in their graves (Stoodley 2010, 93). Of the 20 infant graves excavated at the site, nine were buried with an iron neck ring, or with glass or amber beads positioned at or near the neck area, indicating a bead-string at the neck. By contrast, older children and adults at the site who were buried with beads mostly had their beads positioned in the area of the lower chest, waist and pelvis (Brugmann 2010, 48). All four neck rings found at Blacknall Field were with infants. Although neck rings are not common in Anglo-Saxon contexts, where they have been found, they have been associated with the burials of younger children or infants as at Lechlade (Gloucestershire), where a 15-18 month old and a one-to-two year old were buried with neck rings. Other neck rings are associated with child burials at Wakerley (Northamptonshire), West Heslerton (Yorkshire), Empingham (Rutland), Flixton (Suffolk), and Ruskington (Lincolnshire) (Eagles 2010, 21).

In the 5th and 6th centuries, infant burials occur within the Anglo-Saxon furnished burial ritual, but it is a fact of mortuary studies for this period that infants and children are generally underrepresented in the cemetery population (Crawford 1991). Children represent only 1015% of the total Anglo-Saxon inhumation mortuary population, when comparable juvenile mortality for nonindustrial populations is closer to 50% (Crawford 1993; Richards 1987; Buckberry 2000). At Sewerby (Yorkshire), for example, of the 59 inhumations excavated from the site, only two are less than two years of age, and this is sufficiently typical not to invite comment (Hirst 1985). There are cases which contradict the general pattern: by contrast, the site of Great Chesterford (Essex) is conspicuous in Anglo-Saxon mortuary studies for its relatively high number of infant burials, as well as animal inhumations. Some of the infant burials at this site are unusual; the infant deposited in grave 99, for example, was furnished with a spear, a small knife and a small iron buckle, and is one of only a few securely attested Anglo-Saxon cemetery contexts in which a child under the age of five was buried with a weapon (Evison 1994). The burial of several infants in a single pit at Great Chesterford is also unusual, and more characteristic of late Romano-British burial practice than early Anglo-Saxon. However, early Anglo-Saxon inhumation cemeteries generally indicate that infant burials represent a selected proportion of dead infants from this period (Crawford 1993). The same would appear to be true of cremation cemeteries. At Spong Hill (Norfolk), for example, the cremation cemetery, which was excavated in its entirety, has been calculated to have been short of infants by about 11% on expected mortality (McKinley 1994, 69). However, foetus/neonates/young infants of less than one year were found in single cremations at Spong Hill, Norfolk, and their inclusion suggests that some infants selected for inclusion in the normative rite were not treated any differently from adults (McKinley 1994, 101).

Earlier Anglo-Saxon inhumation and cremation cemeteries usually do not show any pronounced sign of ‘zoning’ of infant burials, but the general mixture of gender, age and status groups throughout Anglo-Saxon inhumation cemeteries has been used to suggest that burial took place on a ‘household’ basis (Hirst 1985, 20; Down and Welch 1990; Sayer 2009). At Berinsfield (Oxfordshire), the identification of epigenetic traits in the population has made the case for family grouping particularly strong, which in turn suggests that infants buried in these plots were given similar treatment to other family members who were older at the time of death. Inhumation cemeteries tend to be relatively small – at Berinsfield, for example, the excavators suggested that the site was used by three households, on the basis of the epigenetic evidence (Boyle et al 1995, 133). Even at the very large cremation sites such as Spong Hill, there is equally compelling evidence for family or household groups, in the form of single pits containing clusters of cremation urns containing individuals of different ages and sexes (McKinley 1994, 103).

In general, some differentiation of child burials from adult was made in inhumation cemeteries in that some artefacts, especially the strongly gendered artefacts such as weaponry and jewellery, were not normally buried with children. Comparably, in cremation cemeteries, children’s cremations included fewer animal bones compared to adult cremation deposits, and they tended not to be buried with larger animals (Crawford 1999; McKinley 1994). Both cremation and inhumation rituals indicate, however, that chronological age, as an aspect of the ritual associated with those infants buried in AngloSaxon cemeteries, was a relatively weak differentiating factor which might be overridden by other considerations. It is possible that these may have included the status of the family, the place of the infant within that family (first born; last born; heir), or the emotional connection

Some inhumation cemetery plans do suggest that deliberate clustering of infants and older children did take place. For example, at Westgarth Gardens (Suffolk), a site containing 55 assessable inhumations and four cremations, the centre of the excavated area had a concentration of four certain or probable infants, four juveniles, and one adult female with an infant (West 1988, 4). There has been much discussion over the extent to which Final Phase cemeteries represent the last gasp of pagan burial ritual, or the earliest expression of Christian burial 76

THE DISPOSAL OF DEAD INFANTS IN ANGLO-SAXON ENGLAND FROM C.500-1066: AN OVERVIEW before church burial without grave goods became the norm from the 8th century (Blair 2005; Geake 1997; Hadley 2000; Crawford 2004). In terms of infant burials, it would not appear that Final Phase cemeteries brought about any significant change in the numbers of dead infants disposed of in the cemeteries, nor in the ritual accorded to them. At Butler’s Field, Lechlade (Gloucestershire), a large cemetery of at least 219 inhumations dating from the mid 5th century to the late 7th or early 8th century has been excavated. The burials fall into two phases – from the mid 5th to 6th century and from the 7th to the early 8th century, respectively. Neither phase shows any sign of zoning by age or gender. 13 infants under the age of two were included in the 138 identifiable Phase 1 burials, while two were included amongst the 57 Phase 2 burials. nine of the 28 undated burials are also of infants. Only one of the five newborn babies could be phased: this infant belonged to Phase 2 and was inhumed in the same grave as an adult female with a workbox and associated items. The cemetery included the burials of two in utero babies, both of which appear to have been given independent burials, although burial 95 was disturbed and redeposited in the grave of an adult female.

accompanied by infants aged between one and three years, including one case (grave 49) where the child was placed in the crook of the arm of the adult, making a mother and child relationship seem particularly plausible (Timby 1994, 17). Some of the relationships between adults and children in double burials seem open to less cosy interpretation. Also at Empingham II, in grave 113, was the burial of a male aged between 17 and 25 years. This individual was buried with a prone child aged three/four years placed under his legs. In grave 119, a child aged two had been placed in a supine position below burial 119C of a male adult whose grave goods included a spear. Two other adult males with weapons were also included in this grave. The possibility that infant burials might have ‘accompanied’ adult burials as a supplementary part of the adult burial ritual is worth consideration, and the assumption that contemporary multiple burials incorporating infants represent the burials of two close family members who died almost simultaneously is not necessarily a given (Crawford 2007). Cultural attitudes towards infant burials, which may make the death of an infant significant, symbolic and meaningful beyond the simple loss of a child, have been highlighted for many other European societies, not least in the papers presented elsewhere within this volume, and a more careful consideration of the location and context of infant burials with adults is due.

Multiple burial While single burial was the normative practice in early Anglo-Saxon inhumation ritual, all age groups and sexes might be given double or multiple burial. Where infants are found within earlier Anglo-Saxon cemeteries they are disproportionately likely to be buried or cremated with other bodies, although individual burial was by far the normal adult burial pattern (Crawford 2007). At Spong Hill, of the 90 double cremations, 70% were of an adult with an immature individual, of which five were female or unsexed adults with foetus/neonates, and 20 were female or unsexed adults with infants (McKinley 1994, 100). At Westgarth Gardens, of the 10 identified infants, two were buried with adults on a site where no other burials were multiple (West 1988)1. At Lechlade (Gloucestershire), six of the 24 burials under the age of two were with other burials either as contemporary burials or redeposited in an adult burial (Boyle et al 1998). One of the burials at this site was of an infant lying just on top of the skeleton of an adult female in the multiple burial grave 1006. The right arm of the woman was wrapped around the child in a close embrace, and it is difficult not to hypothesise that this was the burial of a mother and infant who died together (Boyle et al 1998). Other examples of such burials of women and children or infants come from: Portway (Hampshire), burials 13 and 14; Nassington (Northamptonshire) burials 31 and 31a; and Beacon Hill (Oxford) grave 33 (Cook and Dacre 1985; Leeds and Atkinson 1944; Chambers 1973). Similarly, at the inhumation cemetery of Empingham II, of the 10 double burials, five were of adult females

Pre-Christian Disposal Patterns: Settlement Burials Infants bones have been identified in early Anglo-Saxon settlement contexts, but the inevitable focus of excavations at settlement sites on building structures and on the retrieval of domestic artefacts has made infants relatively invisible, first to the excavators, and then within the site reports. For example, at West Stow (Suffolk), an Anglo-Saxon settlement site consisting of halls and sunken featured buildings dating from the 5th to the 7th centuries, two graves of adults were found within the settlement area (West 1985). In addition, deposits of infant bones were found at 14 separate locations across the site (West 1985, fig 1 and table 1). All the infant bones belonged to infants aged less than one year old, and most were, according to the brief specialist report, newborn. A minimum of four infants were represented by the sample, as this was the maximum of left femurs from the site. Either the remains of four babies had been dispersed widely across the site, or the infant bones may represent up to 14 separate mortuary events. Five of the deposits came from the general occupation layer covering the site, and may be residual, but seven deposits come from the sunken featured buildings, and two come from the ditches, so some deliberate action of deposition is implied by the presence of bone in these sealed contexts. No infants were found in association with the hall buildings. The sunken featured buildings range in date from the early 5th to the 6th centuries. Unfortunately, the site report gives no information on where, within the sunken-featured- buildings, the infant bones were found, so it is not possible to relate the bones to the large

1 In addition to the eight infants discussed in the site report, a note to Table 1 (bone survival) adds an additional two infants with adult burials: teeth from a seven/eight year old with a further adult burial; and a child with cremation no. six (West 1988, 5).

77

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD from a disturbed burial, but the number of bones allows the tentative suggestion that this is a burial within a ditch, but associated with the cemetery (Boyle et al 1998, 43).

number of other Anglo-Saxon artefacts within the sunken featured building assemblages. The infant bones at West Stow are decontextualized, but excavations of infant deposition at other early AngloSaxon settlement sites allows some consideration of the activities associated with the burial event. At Wharram Percy (Yorkshire), excavation of two sunken featured buildings revealed the burial of an infant and an associated deposition of sheep bones on the construction levels of a sunken featured building, found within 3m of the building’s west wall (Milne and Richards 1992, 84-5). Radio-carbon dating placed the infant’s death in the late 6th or early 7th century. The infant was complete and articulated, and was either stillborn or died immediately after birth. The sheep remains ‘indicate a dump of partly butchered bones, possibly in articulated groups, and probably mostly from the same animal, around a year old’ (Milne and Richards 1992, 84). The infant burial appears to have been marked by a round stone or boulder. Fine sieving around the burial revealed minute specks of amber, and an amphora sherd was also found in the vicinity of the burial, although it may have been residual, rather than being deliberately associated with the infant. The infant was carefully placed on a layer of material, up cast from the sunken featured building, which had been used to fill a nearby Romano-British ditch. There was no trace of a grave cut, but the burial was covered by a layer of clay and charcoal very shortly after deposition (Milne and Richards 1992, 20).

Post-Conversion Burial Sites From the later Anglo-Saxon period, isolated infant burials in settlement contexts are extremely rare. Two infants, dating to the 8th century, were found in pits associated with the proto-urban site of Hamwic, Hampshire. In one, pit 5736, the baby was ‘no more than a few weeks old, perhaps newly born’, and it was not clear whether the baby had been disposed of directly into the pit or redeposited from elsewhere. The bone was scattered on the bottom of the pit with soil which may have been cess (Andrews 1997, 204; Morton 1992). By contrast, excavations of the earliest Anglo-Saxon churchyards show much higher proportions of infant burials than were found in the earlier furnished cemeteries, suggesting that the place of infants in the mortuary ritual had undergone a significant change with the Conversion (Crawford 1993). At the church cemetery of Raunds (Northamptonshire), for example, infant burials comprised 18% of the total mortuary population (Boddington 1996, 55) – still by no means a ‘normal’ population, but closer to expected archaeological retrieval rates of infants. Infant baptism was a norm in the 7th century Roman and Frankish church, and was introduced to Anglo-Saxon England as such (Morris 1991, 15; Foot 1992; Blair 2010, 149). Anglo-Saxon law codes emphasise how important and urgent was the performance of this ritual: according to the late 7th century law code of King Ine of Wessex, infants were to be baptised not less than 30 days after birth (Attenborough 1922, 37). There were serious penalties for churchmen failing to do their baptismal duties, and serious penalties for any child not baptised, specifically exclusion from normal burial (Morris 1991). The relatively scarcity of infant burials in non-graveyard contexts in the later Anglo-Saxon period may be due to archaeological recovery, or it may reflect a real absence, which might imply that almost all infants received some form of acceptable baptism, or that infants, whether baptised or not, were normally included within the community burial ground.

At Sutton Courtenay (Berkshire), an adult male was found in the fill of House X and an adult female with an infant was recovered from a pit (Leeds 1923, 169; Leeds 1947, 86). According to the excavator, special attention had been given to this burial group: the earth over the bodies had been stamped down hard (Leeds 1947, 86). At the 6th century settlement site at Eye Kettleby (Leicestershire), one infant had been placed at the base of a sunken featured building, while another was buried within a sunken featured building and accompanied by a pot (Hamerow 2006, 5). Infant bones were also found within a sunken featured building at Barrow Hills (Oxfordshire) (Hamerow 2006, 5). These cases of infant bones associated with settlements are evidence that, whilst the burial of adults within settlement contexts was rare and unusual, infant burial in these locations was, if not common, then at least a significant aspect of infant disposal (Hamerow 2006, 14). Whether their deposition represented ‘special deposits’ or ‘normative burial practice’ for infants remains an open question (Hamerow 2006; Crawford 2008).

It has been noted for earlier Anglo-Saxon cemeteries that infant burials may have been shallower than those of adults, but where burials are intercutting, as occurs in many churchyard and unfurnished burial sites, earlier infant burials will have been disturbed by the digging of later, deeper adult graves, and the bones redeposited at a deeper level, to be retrieved with other human bone during the excavation process (Buckberry 2000). At Worcester Cathedral Chapter House, for example, the excavator noted that almost all the infant bones, a proportion of which were certainly redeposited, have come from the latest levels of intercutting burials, and he has the expectation that further neonate burials will be found during post-excavation analysis (Chris Guy, pers. comm 2005).

Finds of infants in other contexts are extremely rare. A possible case, which may have more in common with the burial of infants in ditches associated with settlements than with burial in cemeteries, comes from Lechlade, where the fill of the pre-Anglo-Saxon ditches around the cemetery site included the fragmentary remains of some Anglo-Saxon human skeletal material. One ditch section included the skull, mandible and part of the torso of a newborn baby. This may represent redeposited material 78

THE DISPOSAL OF DEAD INFANTS IN ANGLO-SAXON ENGLAND FROM C.500-1066: AN OVERVIEW they are ‘non-normative’. These ‘special’ burials do not fall into any particular pattern: they are cases where an infant or infants have been accorded burial outside the normal pattern for all burials at this period. At Raunds, Northamptonshire, all burials are within the churchyard, except for one - a single infant, aged about one to three months according to recent analysis of the skeleton, whose grave was cut into the foundations of the former east wall of the pre-chancel church (Boddington 1996, 21; Hadley 2010, 109). The grave was apparently marked by a post at the foot. It was not possible to tell whether the grave was contemporary with the construction of the chancel (i.e. a foundation burial), or whether it post-dated the chancel. A single intra-mural burial within the timberbuilt church at Cherry Hinton, Cambridgeshire, was also of an infant (Cessford and Dickens 2005, 56 fig. 6), as was the sole burial within the chapel at Burnham, Lincolnshire (Coppack 1986, 39; Hadley 2010, 109). Also worth considering may be the single infant burial under the tower of the Anglo-Saxon church at Bartonupon-Humber (Rodwell and Rodwell 1982). Here, bodies were exhumed from an Anglo-Saxon cemetery before the building of the new late Saxon church, with the exception of an uncoffined child’s burial, which remained undisturbed (Waldron 2007, 29). At the large church cemetery site of St Nicholas Shambles, London, the dead were buried with a variety of associated features, including wooden coffins and stones in the grave, and one single case of ‘Type V’: a burial with charcoal (White 1988). Charcoal burial is a noted, but not particularly common, feature of 10th to 11th century burials in England and on the Continent, and has been associated with high status, usually adult male, burials (Hadley 2000, 208). At St Nicholas Shambles, however, the only burial given this treatment was an infant (burial 5322) (White 1988, 18).

Burials of infants in the Later Anglo-Saxon period showed some variety, perhaps associated with a need to express status or social positioning (Hadley 2000, 2010; Blair 2005). At Worcester Cathedral, for example, at least five infants were buried in coffins, and a further six infants had stones associated with their burials, as did a number of the adult interments (Chris Guy, pers. comm. 2005). A characteristic of infant burials at post-conversion burial grounds, with or without an associated church building, is that, while infant burials may occur across the site, there tends to be identifiable zoning of burials. At Winwick (Cheshire), a large unfurnished Anglo-Saxon cemetery includes an area with ‘a higher proportion of child burials in the north’, where there is also the only possible evidence for a church on the site, which survives as the trace of a small structure of 4m x 8m (Freke and Thacker 1987-8, 33). The churchyard at Raunds (Northamptonshire) was in use for a short period, starting in the mid10th century, and effectively going out of use by the late 11th century. As at Winwick, where infant burials were found throughout the site, there is also a very clear phase of zoned infant burials. At a time of graveyard expansion, the area adjacent to the church walls was brought into use, first for adults, who were buried close to the wall, and then for infants, who were buried right up against the wall as ‘eavesdrip’ burials (Boddington 1996, 9). Infants were almost entirely absent from zones 4 and 5, to the north east and south east of the graveyard. Other excavated sites showing similar patterns of ‘eavesdrip’ infant burial include Rivenhall (Essex), where there was ‘a tendency at all periods to bury infants and young children close to the church walls’ (Rodwell and Rodwell 1985, 101). Infant burials emphasise the footprint of the timber-built church at Cherry Hinton (Cambridgeshire) (Cessford and Dickens 2005, 56). An increased density of infant burials was observed by the walls of St Peter’s, Barton-upon-Humber, and at the timber church of St Andrew’s, Fishergate, York (Buckberry 2007, 124). At the late Anglo-Saxon cemetery site underlying the Norman Chapter House at Worcester Cathedral, infant burials are clustered at the north east of the excavated area. One of the major research problems at Worcester Cathedral has been identifying the location of the pre-Conquest Anglo-Saxon churches (Barker 2005: Crawford and Guy 1997). Given the tendency for Anglo-Saxon infant burials to cluster close to the walls of churches, the presence of an infant ‘zone’ within the burials under the Chapter House may be the single most useful piece of archaeological evidence for the location of an earlier Anglo-Saxon church at this site. Clusters of infant burials may also mark the locations of the ‘missing’ Anglo-Saxon churches at Swinegate, York, and Hartlepool, Cleveland (Buckberry 2007, 124, Crawford 1993, 88).

Yeavering, Northumbria, described by Bede in the Ecclesiastical History (HE II:14) as a royal centre used by King Edwin of Northumbria in the 7th century, was also a site of ritual significance and burial before the period of Anglo-Saxon conversion (Hope-Taylor 1977, 244). During the Anglo-Saxon period, a number of different inhumation rituals were carried out, including radial, unfurnished burials centring on a wooden building or enclosure containing wooden posts dating to between 300 and 500, with the Western cemetery, dating from the middle of the 6th century, consisting of east-west inhumation focussed on building D2 (a possible temple/church?) (Hope-Taylor 1977, 244; Blair 1995). The Western cemetery then appears to have lapsed, to be replaced by a string-grave area and burial within an enclosure, containing multiple inhumations. These appear to have been ‘strictly Christian’. Two graves from this sequence stand out from the norm. Grave AX appears to have been the dedicatory burial of an adult, laid at the threshold of hall A4 and buried with a staff and a goat’s skull. The other is a non-normative burial amongst a group of 16 graves associated with the south-west corner of building D2. The bone preservation from the site was very poor, but enough remained to show that all the bodies had been laid with head to the west, except for the

‘Special’ Burial Within the post-Conversion burial ritual, there are a number of burials of infants which are notable because 79

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD burial of a child, which was buried in a tightly-crouched position in the eastern corner of one normal-sized grave, with an ox tooth resting directly on the grave-floor in the eastern part of the grave. None of the other burials had grave goods.

Beckford (Hereford and Worcester), where two graves with individuals exhibiting signs of leprosy, with rich grave goods, and in a location relatively isolated from the main burial area, have graves of infants in close proximity (Lee 2008).

Physical differentiated adults and infant burial

The settlement of Flixborough (North Lincolnshire) was in use throughout the Middle and Late Saxon periods. Between the late 7th and mid 8th centuries, the whole of the excavated area was inhabited, with five timberframed buildings and a large boundary ditch. Two buildings (6 and 21) had stone-lined ‘soakaways’. In the early to mid-8th century building 20 was replaced by an exceptional structure, building 1a, internally divided with a hearth at one end, and containing four burials along its walls on an east-west alignment. Two further burials were located outside the walls of the building to the south and south-east. All the burials were of juveniles aged between three and 12 years old, with the exception of an adult female, aged 20-30 years, who had been buried with a perinatal foetus at her feet (Loveluck 2001, 86). All the children showed signs of cribra orbitalia, and the female has been diagnosed as having suffered from tuberculosis. The small inhumation cemetery associated with the site contained no burials under the ages of 17, though this cemetery was not fully excavated. The site of Flixborough was initially interpreted as an undocumented monastic centre on the basis of the similarity between the high status artefacts recovered from Flixborough and those from documented monastic sites (Whitwell 1991, 247; Yorke 1993, 146), although the site has also been interpreted as entirely secular in character (Loveluck 2001, 120-121). The burials within and beside building 1a might indicate that this structure was the mortuary chapel of an elite household, separating itself from the main cemetery (Morris 1989, 133; Loveluck 2001, 10), but the presence of a building, a physically differentiated woman, and the infant and juveniles, may suggest a ‘special’ burial group.

At Edix Hill (Cambridgeshire), the inhumation cemetery was in use from about AD 500 to the early 7th century. Bones from this site were in a relatively good state of preservation, allowing detailed analysis of pathology. The site contained a number of burials of people who were suffering from obvious physical defects, but who were included in the normal burial ritual. Grave 84 contained the burial of a female adult, with a neonate (aged less than one month) under her head, as a pillow. Grave 13 contained the body of a woman aged about 25, over whose body had been laid the burial of an infant (aged from one month to one year). These two burials were contemporary, but were separated by a ‘substantial layer of soil’ (Malim and Hines 1998, 48). Both women were suffering from disfigurements which would have been evident to their communities. Both women were given high-status burials; these are the only two female burials to score 11 points (the highest score) in the scales used to rank these burials for analysis (Malim and Hines 1998, 302). Both women have bags or pouches which contained a range of small items, and were buried with probable ‘amulets’ – the crystal and amber beads in grave 13, and the fossil, glass, and animal bone in grave 18. There is a possible association of such collections with ‘special’ women, who may have been considered to have had medical or supernatural powers (Meaney 1989; Webster 1985). The actual grave goods associated with the primary burials in these two graves are very different, but this difference masks a very real similarity in burial ritual which may be summed up by the four outstanding factors linking the burials: amulet items, including a pouch or box; high ranking wealth scores; clear evidence of disease in the primary burial; and the addition of an infant burial to the grave. In this context, grave 42B, belonging to the earlier phase of the cemetery, also warrants closer inspection. This was a woman aged about 25-35, whose skeleton shows changes to her skull and lower legs commonly associated with leprosy. The disease was not particularly advanced, but this woman would have had, at the least, ‘lepratomous nodules over her face and a profuse discharge from the nose’ (Duhig 1998, 177). This woman was amongst the richest in the cemetery, and, as the bone specialist comments: ‘the fact that she had been given a bed-burial casts an interesting light on the attitude of the Anglo-Saxons to disfigurement and obvious disease’ (Duhig 1998, 177). Furthermore, this woman had one of the richest female burials on the site, with an unusual range of associated artefacts, including a bed. Her grave was also used for the burial of two further bodies, one of a child, and one of an adult.

At late Anglo-Saxon Raunds, the grave of a child with a cleft palate, aged eight to nine at time of death, was located in a relatively empty area to the west of the church, as were three associated burials of infants in graves 5169, 5060 and 5057 (Boddington 1996). It has been suggested that cases of physically impaired adults or older children associated with infant burial illustrate the possibility that the sick were being treated, in the burial ritual, as children (Lee 2008, 31). Given the high status of the grave goods associated with the diseased adults (but not with the children) in the furnished cemetery examples, which is particularly obvious at the cemetery at Edix Hill, and given the positioning of infants within the graves (particularly the infant placed beneath the skull of burial 84 at Edix Hill) and around the graves of adults, I would suggest that some disfigured and diseased adults were accorded status as the ‘special dead’, and became the focus of infant burial.

A similar association between infants and the diseased and disfigured in the earlier mortuary ritual is visible at 80

THE DISPOSAL OF DEAD INFANTS IN ANGLO-SAXON ENGLAND FROM C.500-1066: AN OVERVIEW The extent to which disease and disfigurement conferred ‘special’ status to some adults needs further consideration in the context of broader European practice. At the 6th century Early Byzantine settlement at Punta Secca, Sicily, the ‘most surprising and unusual’ discovery was of an elite tomb, uniquely located within a domestic building, containing the bodies of a woman with a congenital deformity of her skull, a foetus, and a child aged about four (Wilson 2011, 263). The woman’s tomb was interpreted by the excavators as that of a ‘holy woman’, whose behaviour or appearance during life, because of her medical condition, was interpreted as divinelyinspired (Wilson 2011, 282).

burials with features which are not replicated in the adult graves within the cemetery. In a few cases, it may be that infants fall into the category of ‘grave good’; their burial is part of the furnishing of the adult burial with which they are associated, and it may be that their value as a symbolic object is also reflected in their deposition within settlement contexts. In the Christian period, a similar division of the infant mortuary population occurred. Some infants were buried in the graveyard with the adult males and females. Just as in the pre-Christian period, these burials imitate adult burials in being given a variety of treatments – coffins, stones and other markers. However, there is another category of infants who are given zoned burial. Where the churchyard has a known associated church, the infant burial area is closely associated with the church building. This pattern is so reliable that the discovery of a zone of infant burials may be the clearest archaeological evidence for the location of an Anglo-Saxon church, as at Worcester Cathedral. The phenomenon of ‘eavesdrip’ burials may be directly linked to the Roman practice of burying infants close to the walls of buildings. The practice of burial by walls was presumed to have disappeared from England with the arrival of Germanic burial practice, though the presence of infants in settlement contexts begs a revision of that evidence, but extra-mural infant burial certainly continued in Romanised areas of Europe in late Antiquity (Crawford 2008). However, as the clear archaeological phasing evidence from the churchyard at Raunds demonstrates, eaves-drip burial was not common until the very end of the Anglo-Saxon period, at the same time that greater emphasis was being placed on the proper baptism of infants (Boddington 1996, 69).

Not all disfigured and diseased adults became the focus of infant burials: at Raunds, the males with probable leprosy are not associated with infants, and nor is the male with tuberculosis and a disfigured leg in grave 5218. The excavator does note one male burial, in grave 5283, which may be the grave of the ‘special dead’, although the site report does not note any exceptional pathology associated with the skeleton in this grave (Boddington 1996). This burial lay in a small area of the churchyard otherwise empty apart from the graves of two infants. The male had a stone grave cover decorated with interlace, and there was evidence to suggest that there was a standing cross at the head of the male grave, and the site report notes that ‘these [infant] insertions into this otherwise vacant plot reflect the special treatment of infants in the same way as the concentration of infant burials against the church walls’ (Boddington 1996, 51). The apparent pattern of infant burials clustered with elite, physically non-normative adults may have been an expression of the same behaviour that led to infant burial close to church walls, because the church was the spiritual representative and locus of the ultimate Christian ‘special’ dead, as well as sometimes being a physical location for the bones and relics of saints and holy women and men.

In addition to infants buried amongst adults, and ‘eavesdrip’ infant burials, there is a further category of post-Conversion burials – those infants who are buried within the churchyard, but whose burial is non-normative, because the burial is in an unusual (and possibly privileged) location, as at Raunds, Cherry Hinton, and perhaps Barton on Humber.

Conclusions Both the pre-Conversion folk cemeteries and the postConversion churchyard cemeteries raise interesting problems of interpretation, not least in the way that there seems to be a divergence in treatment of infants. For early furnished inhumations, infant burial seems to have fallen into two categories – those who were buried in the cemeteries and the large majority who were not. Of those who were not, some, at least, were buried within buildings or close to the settlements, though the current state of archaeological recovery and reporting does not allow any detailed discussion about how infant burials relate to the buildings they were buried in. We are left with a problem – why were some infants given cemetery burial, and why were some not? Within the cemeteries, there are further problems – some infant burials seem to imitate the ‘normative’ burial ritual of adults – they are buried singly, with a variety of grave goods – but some are given non-normative burial, even by Anglo-Saxon inhumation standards, either as multiple burials, or as

Given the long history in the British Isles of the symbolic disposal of selected infants in cemeteries, in settlements, and in other significant locations, and given the apparent continuity of some traditions into the earlier Anglo-Saxon period – particularly the burial of infants within settlements and the inclusion of animal remains with special infant burials – it might seem that Christianity, and in particular, the rite of baptism, which placed the new infant on an equal footing with adults within the Christian community, should have marked a distinctive end to what might be seen, taking the long view, as an extremely long tradition of using the deliberate, nonnormative burial of infants to make statements about society, economy and belief systems. In practice, although Christianity brought with it churchyard inhumation, largely without grave goods, it is clear from this brief overview of the evidence for the disposal of infants that some infant deaths continued to be treated as 81

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Blair, J. 2010. The prehistory of English fonts. In M. Henig and N. Ramsay (eds.) Intersections: the Archaeology and History of Christianity in England, 400-1200. Papers in honour of Martin Biddle and Birthe Kjølbye-Biddle. BAR Publishing: Oxford, 140-177.

exceptional, or different, and that the same concerns and anxieties about infant death that are exhibited in earlier periods are played out within the post-Conversion burial practices of the Anglo-Saxons. Infants were clustered in specific parts of the church; buried by or with adults who had special status, perhaps due to significant or taboo illnesses or deformity, as multiple or secondary interments; and were given special and unusual status within church buildings. The general picture of infant mortuary disposal in Anglo-Saxon England seems to support the idea that the arrival of Christianity, and with it the practice of infant baptism, had an obvious impact on infant burial patterns, not least in dramatic rise in the proportion of infants in the burial population (Hadley 2010, 109). However, a more detailed examination of infant disposal patterns suggests that there were persistent underlying similarities in disposal patterns before and after the conversion to Christianity, which may indicate that Anglo-Saxon cultural attitudes towards dead infants may have been less radically altered by the new belief system, and that the rite of baptism did not mark an immediate change or innovation in attitudes towards the place of infants in the living or dead population. In this context, it may be that baptism did not represent a new, dislocating and foreign concept within Anglo-Saxon England, but rather was a different articulation of an existing sense that the death of infants was different from that of older children and adults, even if Christianity did offer a new way to mark a transition or change of social status and belonging in the newborn child (Fisher 1965).

Boddington, A. 1990. Models of burial, settlement and worship: the Final Phase reviewed. In E. Southworth (ed.), Anglo-Saxon cemeteries: a reappraisal, Stroud: Sutton, 177-99. Boddington, A. 1996. Raunds Furnells: The Anglo-Saxon and Churchyard. English Heritage Church Archaeological Report 7. Boyle, A., Dodd, A. Miles, D. and Mudd, A. 1995. Two Oxfordshire Anglo-Saxon Cemeteries: Berinsfield and Didcot. Thames Valley Landscapes Monograph 8, Oxford University Committee for Archaeology. Boyle, A., Jennings, D., Miles, D. and Palmer, S. 1998. The Anglo-Saxon Cemetery at Butler’s Field, Lechlade, Gloucestershire. Volume 1: Prehistoric and Roman activity and Anglo-Saxon grave catalogue. Thames Valley Landscapes Monograph 10, Oxford Archaeological Unit. Brugmann, B. 2010. Beads. In F. K. Annable and B. N. Eagles, (eds.) The Anglo-Saxon cemetery at Blacknall Field, Pewsey, Wiltshire. Wiltshire Archaeological and Natural History Society Monograph 4, Wiltshire Archaeological and Natural History Society: Trowbridge, 40-42.

References Abbreviations:

Brush, K. A. 1998. Gender and mortuary analysis in pagan Anglo-Saxon archaeology. Archaeological Review from Cambridge 7, 76-89.

HE = Colgrave, B. and Mynors, R. 1969. Bede: Ecclesiastical History of the English People Oxford: Oxford University Press.

Buckberry, J. 2000. Missing, presumed buried? Bone diagenesis and the under-representation of Anglo-Saxon children Assemblage 5 www.shef.ac.uk/assem/5/ buckberr.html

Andrews, P. (ed.) 1997. Excavations at Hamwic: Volume 2. C.B.A. Research Report 109. Arnold, C. J. 1980. Wealth and social structure: a matter of life and death. In P. Rahtz, T. Dickinson, L. Watts (eds.) Anglo-Saxon Cemeteries 1979. Oxford: British Archaeological Reports 82, 81-142

Buckberry, J. L. 2007. On sacred ground: social identity and churchyard burial in Lincolnshire and Yorkshire, c.700-1100 AD. In H. Williams and S. Semple (eds.) Early Medieval Mortuary Practices. Anglo-Saxon Studies in Archaeology and History 14: 120-132. Oxford: Oxbow

Attenborough, F. L. 1922. The Laws of the Earliest English Kings. Cambridge.

Cessford, C. and Dickens, A. 2005. The manor of Hintona: the origins and development of Church End, Cherry Hinton. Proceedings of the Cambridge Antiquarian Society 94, 51-72.

Barker, P. 2005. Reconstructing Wulfstan’s cathedral. In J. Barrow and N. Brooks (eds.) St Wulfstan and his World. Studies in Early Medieval Britain, 167-188. Blair, J. 1995. Anglo-Saxon pagan shrines and their prototypes. Anglo-Saxon Studies in Archaeology and History 8, 1-28.

Chambers, R. A. 1973. A cemetery site at Beacon Hill, Near Lewknor. Oxoniensia 38, 138-45. Cook, A. M. and Dacre M. W. 1985. Excavations At Portway, Andover 1973-1975. Oxford: Oxford University Committee for Archaeology Monograph 4.

Blair, J. 2005. The Church in Anglo-Saxon Society. Oxford: Oxford University Press.

82

THE DISPOSAL OF DEAD INFANTS IN ANGLO-SAXON ENGLAND FROM C.500-1066: AN OVERVIEW Coppack, G. 1986. St. Lawrence church, Burnham, South Humberside. The excavation of a parochial chapel. Lincolnshire History and Archaeology 21, 39-60.

Foot, S. 1992. ‘By water in the spirit’: the administration of baptism in Early Anglo-Saxon England. In J. Blair and R. Sharpe (eds.) Pastoral Care Before the Parish. Leicester: 171-92.

Crawford, S. 1991. When do Anglo-Saxon children count? Journal of Theoretical Archaeology 2.

Freke, D. J. and Thacker, A. 1987-8. The inhumation cemetery at Southworth Hall Farm, Winwick. The Journal of the Chester Archaeological Society 70, 31-38.

Crawford, S. 1993. Children, death and the afterlife in Anglo-Saxon England. Anglo Saxon Studies in Archaeology and History 6, 83-92.

Geake, H. 1992. Burial practice in seventh- and eighthcentury England. In M. Carver (ed.) The Age of Sutton Hoo, Woodbridge: Boydell Press, 83-94.

Crawford, S. 1999. Childhood in Anglo-Saxon England. Stroud, Sutton Publishing.

Geake, H. 1997. The Use of Grave Goods in ConversionPeriod England, c.600-c.850. British Archaeological Reports, British Series, 261, Oxford.

Crawford, S. 2000. Children, grave goods and social status in Early Anglo-Saxon England. In J. Sofaer Derevenski (ed.) Children and Material Culture. London: Routledge, 169-179.

Hadley, D. M. 2000. Burial practices in the Northern Danelaw, c.650-1100. Northern History 36 (2 September), 199-216.

Crawford, S. 2004. Votive deposition, religion and the Anglo-Saxon furnished burial ritual. World Archaeology 36:1 (March), 87-102.

Hadley, D. M. 2010. Burying the socially and physically distinctive Later Anglo-Saxon England. In J. Buckberry and A. Cherryson (eds.). Burial in Later Anglo-Saxon England c. AD 650-1100 Oxford: Oxbow, 103-115.

Crawford, S. 2007. Companions, co-incidences or chattels? Children and their role in early Anglo-Saxon multiple burials. In S. Crawford and G. Shepherd (eds.) Children, Childhood and Society. IAA Interdisciplinary Series Vol. I: Studies in Archaeology, History, Literature and Art. BAR International Series 1696. Oxford: BAR Publishing, 83-92.

Hamerow, H. 2006. ‘Special deposits’ in Anglo-Saxon settlements. Medieval Archaeology 50, 1-30. Hirst, S. M. 1985. An Anglo-Saxon Inhumation Cemetery at Sewerby, East Yorkshire. York: York University Archaeological Publications 4.

Crawford, S. 2008. Special burials, special buildings? An Anglo-Saxon perspective on the interpretation of infant burials in association with rural settlement structures. In K. Bacvarov (ed.) Babies Reborn; Infant/Child Burials in Pre- and Protohistory. BAR International Series s1832. Oxford: BAR Publishing, 197-204.

Hope-Taylor, B. 1977. Yeavering; An Anglo-British Centre of Early Northumbria. London: HMSO Department of the Environment Archaeological Reports 7.

Crawford, S. and Guy, C. 1997. As Normans tore down Anglo-Saxon Cathedrals. British Archaeology 29, 7.

Kilbride, W. G. 2000. Why I feel cheated by the term Christianisation. In A. Pluskowski (ed.) Early Medieval Religion, 1-17. Archaeological Review from Cambridge 17:2.

Down, A. and Welch, M. 1990. Appledown and the Mardens. Chichester: Chichester Excavations 7.

Lee, C. 2008. Forever young: child burial in Anglo-Saxon England. In S. Lewis-Simpson (ed.) Youth and Age in the Medieval North, Leiden and Boston: Brill, 17-36.

Duhig, C. 1998. The human skeletal material. In T. Malim and J. Hines (eds.) The Anglo-Saxon Cemetery at Edix Hill (Barrington A), Cambridgeshire. London: CBA Research Report 112, 154-199.

Leeds, E. T. 1923. A Saxon village near Sutton Courtenay, Berkshire. Archaeologia 73, 147-92

Eagles, B. 2010. Jewellery: neck rings. In F. K. Annable and B. N. Eagles, (eds.) The Anglo-Saxon cemetery at Blacknall Field, Pewsey, Wiltshire. Wiltshire Archaeological and Natural History Society Monograph 4, Wiltshire Archaeological and Natural History Society: Trowbridge, 17-25.

Leeds, E. T. 1947. A Saxon village at Sutton Courtenay, Berkshire: 3rd report. Archaeologia 92, 79-93. Leeds, E. T. and Atkinson, R. J. C. 1944. The AngloSaxon cemetery at Nassington. Northamptonshire Antiquaries Journal 24, 100-128.

Evison, V. 1994. An Anglo-Saxon Cemetery at Great Chesterford, Essex. London: CBA Research Report 91.

Loveluck, C. 2001. Wealth, waste and conspicuous consumption. Flixborough and its importance for mid and late Saxon settlement studies. In H. Hamerow and A. MacGregor (eds.) Image and Power in the Archaeology

Fisher, J. D. C. 1965. Christian Initiation: Baptism in the West:A study in the disintegration of the primitive rite of initiation. London: Alcuin Club Collections Volume 47. 83

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD of Early Medieval Britain: Essays in Honour of Rosemary Cramp. Oxford: Oxbow Books, 78-130.

Stoodley, N. 2007. New perspectives on cemetery relocation in the seventh century AD: the example of Portway, Andover. Anglo-Saxon Studies in Archaeology and History 14, 154-162.

Lucy, S. 1994. Children in Early Medieval Cemeteries. Archaeological Review from Cambridge 13:2, 21-34.

Stoodley, N. 2010. The social structure of the AngloSaxon cemetery. In F. K. Annable and B. N. Eagles, (eds.) The Anglo-Saxon cemetery at Blacknall Field, Pewsey, Wiltshire. Wiltshire Archaeological and Natural History Society Monograph 4, Wiltshire Archaeological and Natural History Society: Trowbridge, 87-99.

McKinley, J. I. 1994. The Anglo-Saxon Cemetery at Spong Hill, North Elmham, Part VIII: The Cremations. East Anglian Archaeology Report 69. Malim, T. and Hines, J. 1998. The Anglo-Saxon Cemetery at Edix Hill (Barrington A), Cambridgeshire. London: CBA Research Report 112.

Thompson, V. 2004. Dying and Death in Later AngloSaxon England. Woodbridge: The Boydell Press AngloSaxon Studies 4.

Meaney, A. L. 1989. Women, witchcraft and magic in Anglo-Saxon England. In D. Scragg (ed.) Superstition and Popular Magic in Anglo-Saxon England. Manchester, 9-40.

Timby, J. R. 1996. The Anglo-Saxon Cemetery at Empingham II, Rutland: Excavations carried out between 1974 and 1975. Oxford: Oxbow Monograph 70.

Milne, G. and Richards, J.D. 1992. Two Anglo-Saxon buildings and associated finds. Wharram: a study of settlement on the Yorkshire Wolds, VII. York: York University Archaeological Publications 9.

Waldron, T. 2007. St Peter’s Barton-upon-Humber, Lincolnshire: A parish church and its community. Volume 2, the Human Remains. Oxford: Oxbow Books. Webster, L. 1985. The grave goods. In J. D. Hedges and D. G. Buckley (eds.) ‘Anglo-Saxon burials and later features excavated at Orsett, Essex, 1975’ Medieval Archaeology 29, 8-13, 15-16, 19-22.

Morris, R. 1989. Churches in the Landscape. London: J. M. Dent. Morris, R. 1991. Baptismal places: 600-800. In I. Wood and N. Lund (eds.) People and Places in Northern Europe 500-1600: Essays in Honour of Peter Hayes Sawyer. Woodbridge: The Boydell Press, 15-24.

West, S. 1985. West Stow The Anglo-Saxon Village Volume 1: Text and Volume 2: Figures and Plates. East Anglian Archaeology Report No. 24.

Morton, A.D. (ed.) 1992. Excavations at Hamwic Vol 1: excavations 1946-83, excluding Six Dials and Melbourne Street. C.B.A. Research Report 84.

West, S. 1988. Westgarth Gardens Anglo-Saxon Cemetery, Suffolk: Catalogue. East Anglian Archaeology 38.

Pader, E. J. 1982. Symbolism, Social Relations and the Interpretation of Mortuary Remains. Oxford: BAR International Series 130.

White, W. 1988. Skeletal Remains from the Cemetery of St Nicholas Shamble, City of London. London and Middlesex Archaeological Society Special Paper 9.

Richards , J. D. 1987. The Significance of Form and Decoration of Anglo-Saxon Cremation Urns. Oxford: BAR British Series 166.

Whitwell, J. B. 1991. Flixborough. Current Archaeology 126 (vol 11, no 6), 244-7. Wilson, R. J. A., Hayes, J. W., Sulosky, C. L. and Di Stefano, G. 2011. Funerary feasting in Early Byzantine Sicily: new evidence from Kaukana. American Journal of Archaeology 115, 263-302.

Rodwell, W. and Rodwell, K. 1982. St Peter’s church, Barton-upon-Humber: excavation and structural study, 1978-81. Antiquaries Journal 62, 296-9. Rodwell, W. and Rodwell, K. 1985. Rivenhall: Investigation of a Villa, Church and Village 1950-1977. London: CBA Research Report 55.

Yorke B. 1993. Lindsey; the lost kingdom found? In A. Vince (ed.) Pre-Viking Lindsey. Lincoln: City of Lincoln Archaeological Unit Archaeological Studies 1, 141-50.

Sayer, D. 2009. The 7th century Kentish family: considering the evidence from the legal codes and cemetery organisation. In D. Sayer and H. Williams (eds.) Mortuary Practice and Social Identities in the Middle Ages. Exeter: the Exeter University Press, 141166. Stoodley, N. 2000. From the cradle to the grave: age organization and the early Anglo-Saxon burial rite. World Archaeology 31:3 Human Lifecycles, 456-472 84

Where have all the flowers gone? Bronze Age children’s burials in South East England: initial thoughts Dawn McLaren Introduction specific counties have been selected for study because the area has generally been considered peripheral to and overshadowed by the Bronze Age burial evidence from other areas, particularly Wiltshire, Yorkshire, and the Peak District. This is a deliberate attempt to shift the focus away from these regions and to concentrate on an area with an assumed poverty of information with regard to Bronze Age burial activity and where children are currently believed to be invisible.

This paper is the result of recent investigations into Bronze Age children’s burials in South East England as part of my ongoing research looking at aspects of children’s burials of this period in Britain. The aim of this paper is to introduce some of the main themes of my research in this area, focusing on the counties of Sussex, Surrey and Kent (fig. 1), where ninety-three possible children have been identified within previously recorded burials. These

Fig. 1: Area of study discussed in paper thought of by the adult community and can help us to reconstruct the child’s role and status within society.

The apparent invisibility of children in general within the archaeological record has been previously linked to problems of preservation or failure to identify children’s remains during excavations. Their conspicuous absence was assumed to mean that children were not considered important enough to be given a formal burial. However, the current writer believes this is an attempt to oversimplify the problem.

What is a child? Before further analysis it is essential to understand that caution should be exercised on using the broad terms ‘child’ and ‘childhood’ without any conceptual understanding of their meaning (Derevenski 1994, 8). These expressions are modern, socially constructed terms, which are culturally loaded with perceived ideas on what a child is and what childhood should be. The concept of the child is often viewed as a universal state of physiological development reflecting our biological life cycles. The idea of what a child was in prehistoric societies may have been a far more flexible concept than it is today. Joanna Sofaer has suggested that modern Western society has

The rites afforded to children in mortuary contexts is not a direct reflection of the world of the child itself, but of aspects of culture transferred from the adult society onto the child, in accordance with the way the community regarded them. Children do not bury themselves and what we are observing is not an expression of the child’s world itself but merely a view of it through adult eyes. This affords us the opportunity to glimpse the way children were 85

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Secondly, varying standards in the levels of recording has also placed limits on our ability to analyse effectively available data. Today our excavation methods are designed to gain and record as much information as possible. In the past, records were often sketchy and lacking in detail. When skeletal remains were encountered they were often divided simply into ‘child’ or ‘adult’ categories. Despite this lack of detail, they can still be regarded as a useful source.

marginalised the child’s economic, social and political importance, constructing a stereotype that establishes children as peripheral to adult society (Derevenski 1997, 193). Children, as defined by demographic conventions, encompass all individuals under fifteen years of age (Chamberlain 2000, 207). This concurs with the concept of childhood terminating and giving way to ‘adolescence’ at the onset of puberty. The end of childhood means an end of dependence on the care of the adult population and the inclusion of the individual into the community as a competent member of society. But when does this take place in prehistory?

The final problem is that high infant mortality rates have led some to believe that parents would have divorced themselves emotionally from their young. It has been estimated that up to 50% did not survive past infancy (Chamberlain 2000, 208; Goodman and Armelagos 1989, 227). Despite ethnographic evidence that suggests that in some circumstances ‘adult care-giving varies in response to perceived risks of childhood mortality’ (Chamberlain 2000, 208) we should not fall back on this as an excuse to justify a lack of understanding of children at this time.

For the purposes of my analysis it is proposed that the term ‘child’ should concur with established demographic conventions, to include all individuals up to fifteen years of age. It should be noted that prior to any evidence suggesting a real social boundary existed around this age group, this division is, to a certain extent, arbitrary. However, it is necessary to define a practical boundary within which analysis can take place.

Methods and Approach My initial methods start back at the original reports – the local (e.g. Thanet Archaeological Trust publications), regional (e.g. Sussex Archaeological Collections, Surrey Archaeological Collection and Archaeologia Cantiana) and national. A total of seventy children’s burials have been identified from South East England, from the counties of Sussex, Surrey and Kent. Each county will be considered separately below to determine whether any general trends can be identified.

Problems of Identification Eleanor Scott believes that our marginalisation of the prehistoric child is the result of ‘our own cultural association of infants with the female and the domestic which has effectively rendered them unworthy of serious academic study’ (1992, 82). Because greater attempts to identify and visualise women in prehistory have themselves been a relatively recent development, there has been a general belief that it is beyond the scope of the archaeological record to produce evidence of the child’s world. To state that remnants of the child’s world are too difficult to identify in an already fragmentary archaeological record or that factors of preservation have meant that children have all but disappeared, can be seen as an attempt to oversimplify and dismiss the problem. The reason why so little information on children has been found is due to a lack of systematic study. Despite a recent increase in interest in this subject, children remain peripheral within archaeological interpretations.

Sussex Leslie Grinsell’s 1931 study of the Sussex barrows forms the foundation for analysis of burial traditions. Grinsell noted approximately 1000 round barrows, in similar concentrations to that seen in Wessex (1931; 1934a, 229). Despite the large numbers of barrows less than 150 have been recorded and fewer than 50 have been excavated to modern standards. Garwood’s Sussex Barrows Project has highlighted the systematic destruction of round barrow sites as a result of local land use and he estimates that 70% of the barrows recorded in Grinsell’s study have now been destroyed (Garwood 2002).

There are three main problems that affect our ability to study children in prehistory. The first is that skeletal identification remains a factor; the brittle and delicate nature of children’s bones often mean that preservation is not great, even when soil conditions are optimal, leading to a lack of precision in identification. Dental and skeletal analysis can give quite precise indications of the age of the individual, potentially to within two years, up to the age of fifteen if preservation is good (Lunt 1973; 114). This is due to rapid but predictable osteological and dental changes. After the age of fifteen, identification of age becomes more difficult due to more variable developmental rates (Mays and Anderson 1995, 355; Lunt 1973, 114). It is also difficult to determine the sex of a child and, in many cases, this is not attempted.

But the problem is not just due to the modern plough. Antiquarian activity has also had a severe impact on barrow study in Sussex. Rev. H. Smith observed in 1870 the problem with unstructured, unscientific exploration noting that ‘many barrows have been dug through, and all that can now be learned…is that several urns were found of rude character’. Despite these limitations, spatial patterning of burial traditions can be observed with concentrations of Beaker burials on the fringes of the central chalk lands. Rich Earlier Bronze Age burials cluster in this area and include impressive grave assemblages such as those at Pycombe (Butler 1991) and Hove (Phillips 1857).

86

WHERE HAVE ALL THE FLOWERS GONE? BRONZE AGE CHILDREN’S BURIALS IN SOUTH EAST ENGLAND

Fig. 2: Distribution of Bronze Age children’s burials in Sussex (1.Eastbourne, 2.Baily’s Hill cairn, 3.Alfriston barrow, 4. Bostle ring ditch, 5.Itford Hill, 6.Beddingham barrow, 7. Cliffe Hill barrow, 8. South Malling Barrow, 9. Looe’s Barn Tumulus, 10.London Road cemetery, 11.Ditchling Field, Cold Dean, 12. Ladies Mile Barrow, 13. Hove (isolated burial), 14. Slonk Hill barrow, 15.Church Hill, 16. Chactonbury, 17, Blackpatch, 18. Cock Hill, 19. Claypit ring ditch, 20. West Heath barrow) and Curwen 1935), which will be discussed in more detail later, and the burial of an adult female and child at South Maling barrow (Grinsell 1934a, 261). The majority of child inhumations are secondary burials within mounds (53%). The position of the grave within the mound is unknown in a further five cases.

Forty-nine children below fifteen years of age and a further five tentative children’s burials of Bronze Age date are known from Sussex (fig. 2). These include twenty inhumation burials and 34 burials after cremation from a total of twenty-four sites. All ages of children are represented; from foetus to adolescent but neonates, perinates and foetuses are only observed as unburnt skeletons and are absent amongst the burials after cremation.

The placement of the body has only been noted in seven graves; all were crouched. These include three children from Ladies Mile barrow (Holleyman and Yeats 1960, 13643) and an infant from Hove (Rudling 2000, 11) whose bodies had been placed on their left sides. Only one body was noted as having been placed on their left side: that of a two- to three-year-old child from Ladies Mile barrow (Holleyman and Yeats 1960, 138). In addition to the formal burials within barrows, some more unusual deposits have been noted

If we look first at the inhumation burials in this county it is possible to demonstrate that most are formal burials within barrow and cairns (75%). There are also small quantities of children’s graves within Later Bronze Age settlements (16%) and a single isolated example from Hove (Rudling 2000, 11). Only four burials are confidently assigned to the Earlier Bronze Age through associated material culture (such as the two beaker vessels found with an adult and child at Church Hill Barrow: Curwen and Curwen 1935) or direct dating, such as that from Blackpatch barrow 3 (Drewett 1982, 352). Six children’s graves can be fairly confidently dated to the Later Bronze Age: one, from Cock Hill settlement site by associated vessel sherds and another from Bostle barrow directly dated to 970-790 cal. BC (McKinley 2004, 30). A further nine graves are not closely dated.

Three possible ritual burials have been identified within the assemblage. One example came from Ditchling Field barrow (Holleyman and Yeates 1960, 133-6). Whilst two child cremations conform to other burials within this mound, one further deposit, consisting of the teeth of a young child and an adult skull fragment, was placed underneath an inverted collared urn. This is very unusual and atypical child burial; it has more affinities with a special, ritual deposit than a deliberate formal grave, perhaps representing the deposition of deliberately curated human bone. Similar possible relic interments are observed elsewhere, such as that from Ampleforth barrow 4,

Only two children’s graves were recognised as forming the principal burial on barrow sites. These comprise a beaker associated adult and child burial from Church Hill (Curwen 87

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD at Ladies Mile barrow (Holleyman and Yeats 1960, 138), beaker sherds with a child at Itford Hill (Burstow and Holleyman 1957, 175), an inverted collared urn at Ditchling Field barrow (Holleyman and Yeates 1960, 1336; Longworth 1984, 195) and Deverel-Rimbury urn sherds associated with a child burial at Cock Hill settlement (Ratcliffe-Densham et al. 1961, 89).

Yorkshire (Smith 1994, 101-2) and it is likely to be a more widespread practice than previously recognised.

Fig. 3: Infant burial from Hove, Sussex (after Rudling 2000, 11) The burial of a four- to six-year-old child from a pit at Hove (fig. 3) (Rudling 2000, 11) was situated near to two large flat-bottomed pits, containing special deposits of red deer antler. Although the burial itself is typical in terms of the position of the body and the paucity of grave goods, its proximity to two ritual features makes it unusual. A similar unusual deposit comes from a Late Bronze Age settlement at Itford Hill where a storage pit was found to contain sherds of Beaker, fragments of animal bone, human bone from child, quern fragments and a hammerstone (Burstow and Holleyman 1957, 175). A final possible ritual burial of an infant skeleton comes from the Later Bronze Age ceremonial and occupation site at Eastbourne where several disarticulated human remains have been discovered (Denison 1995, 4).

Fig. 4: Full-sized and small Beaker from Church Hill Barrow, Sussex (Curwen and Curwen, 1935). Inhumations of an adult and child each provided with a Beaker comes from Church Hill barrow (Curwen & Curwen 1935); the adult with a full-sized pot, the child with smaller version (fig.4). This association has affinities with many Scottish burials such as Broomend of Crichie (Davidson 1868) and Nether Criggie (Kirk and McKenzie 1955, 3) in Aberdeenshire (fig.5). The placement of the small sized Beaker beside the body of the child implies the selection of a small pot was a deliberate choice. It is too simple to assume that the small pots found in the graves are children’s toys as this fails to take into consideration why they were interred within the graves of some adults. Their common occurrence within the graves of children does suggest that the adult community selected them specifically because of their small size, perhaps recognising different levels of consumption between adult and child. Although funerary practice was clearly comparable to that of the adult, we can see that some consideration of the needs of a child, as distinct from that of an adult, were acknowledged with the inclusion of this category of small scale item.

On 25% of the children considered here were associated with another individual. In two instances, from South Maling and Church Hill, was the child accompanied by a single adult and at Bostle, two foetuses or neonates were buried together (Grinsell 1934a, 261, Curwen and Curwen 1935; McKinley 2004, 30). At Eastbourne, the recovered unburnt human bones were found to represent at least three adults and a child. The round barrow at Ladies Mile contained five child inhumation burials from two to eight years of age (Holleyman and Yates 1960, 136-143). The oldest child, identified as seven to eight years at death, shares dental characteristics with an adult male inhumation within the barrow and these two individuals may be biologically related. The excavator suggests that this may have been a ‘family’ barrow (1960, 143) but these are secondary burials inserted into the mound on successive occasions making this assertion difficult to substantiate.

In only one cases had the burial been provided with another form of grave good: the inclusion of shells of snail and limpet with an infant burial at Alfriston (Smith 1870, 70). It is unlikely that the limpet shells, a deliberate deposition, represent feasting and may have been included in the grave as amulets or as the child’s collection of precious trinkets.

Grave goods are rare in association with child burials in this county. The most common accompanying artefacts are ceramic vessels which are found in 25% of children’s graves. Such vessels include Earlier Bronze Age Beaker fragments from the grave of a two- to three-year-old child

If we turn now to look more closely at the deposits of burnt children’s bones we can observe that the apparent bias towards burial after cremation can be explained by the acidic quality of the soil in some areas of Sussex. In these 88

WHERE HAVE ALL THE FLOWERS GONE? BRONZE AGE CHILDREN’S BURIALS IN SOUTH EAST ENGLAND soils, unburnt bone is less likely to survive. The bias towards child cremations is consistent with the pattern observed in adult burials of this period.

Like the inhumation burials, deposits burnt children’s bones are mostly associated with barrow, cairns and cemetery-barrows, the majority as secondary burials (39%) and are more frequently associated with those of adults within deposits of cremated bone than in unburnt burials (over 25%). Five such deposits are associated with settlements of Later Bronze Age date and will be discussed further below.

Children from two to sixteen years of age are present within these burials and it is significant to note that no foetus’, neonates or children under one year of age are represented. It is unclear if this is due to a lack of recognition of the small, delicate bones of particularly young children or whether this reflects a genuine absence in the burial record.

In Sussex, it was more common for a burial by cremation to be provided with grave goods and in seven cases these included flint flakes that had passed through the pyre with the deceased. This is entirely consistent with the adult burials, few of which were provided with any accompanying items and where rich grave assemblages are rare. Finlay (1997) and Grimm’s (2000) recent studies of flint knapping techniques suggest that ‘knapping practice begins…during childhood or adolescence’ (Grimm 2000, 54), introducing the possibility that the flints interred within the graves of children were not just made for them but by them.

A greater proportion of the cremation deposits than inhumation burials can be closely dated due to the frequent association of chronological distinctive ceramic vessels. Eight burnt bone deposits are associated with Collared Urns, such as that from Looe’s Barn cairn (Holdgate 1987, 230), indicating an Earlier Bronze Age date. A further two possible Earlier Bronze Age child cremations are present at Bostle Barrow (McKinley 2004, 28-29, table 1). The majority of these cremations are Later Bronze Age in date (18 examples, representing 42% of the dataset). Twelve, such as three from Claypit Lane ring ditch (Chadwick 2006, 13-15) are dated by direct association with Later Bronze Age Deverel-Rimbury-style pots. The remaining eleven interments are not closely dateable.

A significant assemblage of flint tools came from a child’s grave at Blackpatch barrow 1 which included two flint axes (one a roughout), a purposefully broken oval flint knife and a segmented tool (Pull 1932, 65).

Fig. 5: The Beakers from Nether Criggie, Aberdeenshire (after Kirk and McKenzie 1955, 6).

Fig. 6: (left) ‘chalk charms’ from Blackpatch Barrows, Sussex (after Pull 1932, 66); top right: chalk axes, Wiltshire (Copyright trustees of the National Museums of Scotland); bottom right: chalk balls from Stonehenge Bottom (©Trustees of the National Museums of Scotland).

89

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD This barrow at Blackpatch was built directly over an early Neolithic mining shaft (Pull 1932), perhaps indicating a desire to connect with the past. Pull comments that ‘the sole interment for which this mound was raised consisted of the remains of a child, presumably that of some person of very high rank (1932, 65). In addition to the accompanying flint items already listed, the child’s cremated bones were also associated with unburnt bones of a domestic pig and ‘chalk charms’ (fig. 6 left, no. 7 & 8). The recoveries of pig bones, often the remains of joints of meat, are characteristic of Earlier Bronze Age inhumations. Although Russell (2001, 53) later dismissed these charms, it was noted by Pull that similar objects were recovered from other barrows in the area and from the early Neolithic flint mines themselves (Pull 1932, 74). They may have been specifically selected for inclusion within the burial because of this connection with the past. Worked chalk objects are well known, including the chalk axes from Woodhenge, Amesbury (Cunnington 1929, 112, 113, pl 22.1-2), chalk balls from Mount Pleasant, Dorset (Wainwright 1979, 167, fig 75-77) and perhaps, most impressively, are the Folkton Drums associated with the burial of a child in Yorkshire (Greenwell 1890, 14-16). The loss of the objects from the Blackpatch burials makes their significance unclear.

children, were associated with it. In Brück’s review of late Bronze Age burial practices she notes that human remains were found on only forty three settlement sites and 72% were from adults making these child burials very unusual (fig. 7) (Brück 1995, 249).

Six children’s burials are directly associated with settlements in Sussex; at Blackpatch (Rudling 2001, 201), Cock Hill (Ratcliffe-Densham et al 1961) and Itford Hill (Burstow and Holleyman 1958, 175). These include six cremation deposits and three inhumations. All ages of child are represented from foetus to fifteen years. Four are associated with fragments of pottery and at least one was marked by a post associated with a structure (RatcliffeDensham et al 1961, 87-88). The Cock Hill settlement is of particular interest as several human burials, both adults and

Surrey

Many aspects of the mortuary treatment of children in Sussex are unlike that seen in Scotland, where preliminary research has been carried out by the author. There is a more diverse suite of practices including the occurrence of burials within settlements and ritual contexts. These are exceptional and are not representative of the general pattern of burial treatment. Most of the burials seem to conform to practices associated with adult burials and few were associated with grave goods. Although there are some unusual burials in Sussex involving children, it is clear that sites such as Cock Hill settlement and the Ditching Field deposit are particularly unusual and should be considered separately. With the exception of the burial at Hove, the burials from ritual or settlement contexts have all been dated to the late Bronze Age and are likely to indicate that comparing funerary rites from this period to those of the earlier Bronze Age may not be appropriate for identifying general trends.

Only five possible child burials have been recovered from previously recorded excavations in Surrey (fig.8). Two child inhumation and three possible cremation burials are recorded, most are unaccompanied and only one is provided with a pottery vessel. This corresponds with adult burials in the region which are also few and far between, and where the provision of grave goods is also limited.

Fig. 7: Table showing the distribution of human remains within late Bronze Age contexts (after Brück 1995, 248). 90

WHERE HAVE ALL THE FLOWERS GONE? BRONZE AGE CHILDREN’S BURIALS IN SOUTH EAST ENGLAND

Fig. 8: Distribution of Bronze Age children’s burials in Surrey (1. Queen Mary’s Hospital, 2. Hengrove, 3. Merrow, 4. Runneymede, 5. Sunningdale).

Fig. 9: Stoneyfield cremation burial with reused saddle quern fragment (after Oakley et al. 1939, 176).

91

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD simple reuse of a broken tool or a symbolic link between life and death? The juxtaposition of production and fertility, which querns are symbols of, and consumption and death is a common theme in ethnographic examples of complex burial rituals. The broken quern stone may have been used to symbolise death itself (Brück 2001, 152; Heslop 2008, 73-5). Evidence of Bronze Age activity is not as dense as that seen in the surrounding areas and much of the evidence has been destroyed. Although Earlier Bronze Age beakers, from burials, are present in Surrey, the majority are unprovenanced. Common artefact associations with Earlier Bronze Age activity found elsewhere, such as battle axes, axed-hammers, flint knives and bronze implements are known from the county but most are stray finds (Needham 1987). The distribution pattern of these artefacts is overwhelmingly concentrated along the Thames. Collared urn distribution overlaps that of beaker associated objects but are also found across central and western Surrey where Beaker finds are rare. Food vessels and Biconical Urns, pottery types well represented in Wessex, are restricted to West Surrey, completely exclusive to the Beaker zone of influence.

Fig. 10: Biconical urn from Guildford, Surrey, associated with the cremated remains of a child (Needham 1987, 102).

A possible child cremation burial at Merrow in Guildford (Cruse & Harrison 1984, 104) associated with a biconical urn (fig. 10) is therefore significant and illustrates that elements of ‘Wessex’ burial traditions, present in Surrey, were considered appropriate in regards to some adults and children alike. This represents the only Earlier Bronze Age child burial known to the writer to date.

Most of the records of child burials in this county are late Bronze Age in date and derive from both cemetery sites and settlements. Two possible children’s cremations were observed from a barrow at Sunningdale (Gardner 1924, 21). This barrow was associated with a series of late Bronze Age cremations, the majority contained within coarse pottery vessels. Two un-urned deposits were recovered which the excavator interpreted as the cremated bones of children. Yet, the recognition of the bones as those of children appears to be based on the poverty of the burials in relation to others from the site rather than on osteological identification of the skeletal remains. As such, the presence of child cremations at this site cannot be confirmed and should be treated with caution.

This review has highlighted significant problems with the burial record in this county. As Whimster observed in the 1930’s: ‘no period of Surrey archaeology has suffered more from neglect and lack of scientific investigation than the Bronze Age’ (1931, 62). Surrey’s burial mounds have never been subjected to a concerted campaign of investigation with Grinsell’s 1930s survey forming the foundation of our knowledge (1934b). Although they were subject to antiquarian investigation, few records exist to attest to any findings. Understanding Surrey’s burial traditions is hampered by both natural mound formation and the presence of Saxon burial mounds and it is estimated that very few of the original numbers have survived, due to agricultural intensification and urban expansion, especially along the gravel terraces of the Thames.

Two children’s burials have been recovered in association with late Bronze Age occupation sites in Surrey; from a late Bronze Age midden at Runneymede (Needham 1992, 61) and an occupation site at Queen Mary’s Hospital (Adkins and Needham 1985, 17). Here, the child had been placed in a crouched position on a saddle quern and burnt in situ. This was not a straightforward cremation burial and sets this example apart from the rest of the unurned cremations associated with the site. Not only does this not conform to inter-site burial traditions but it is an unusual form of burial for a child in general. Burning bodies in situ is paralleled in Yorkshire (i.e. cremation of an adult male at Ganton: Greenwell 1877, 179) but no further examples involving either adult or child is known in Surrey.

Perhaps the most significant factor is that, even in relatively recent reports of burials from this area, it is very rare for the skeletal or cremated remains to be analysed. It has been hard to find children within the records in Surrey but it is equally difficult to find records of adult burials. Despite the small numbers of children identified from within the published records, the similar lack of identified adults suggests that the paucity of such burials is a result of a modern bias in the record rather than a reflection of past funerary practices.

The unusual association of a burial with a saddle quern is difficult to explain but a similar association is known from Stoneyfield (fig. 9) (Oakley et al. 1939, 175) where a cremation was placed on a broken quern stone. Was this 92

WHERE HAVE ALL THE FLOWERS GONE? BRONZE AGE CHILDREN’S BURIALS IN SOUTH EAST ENGLAND

Fig. 11: Distribution of Bronze Age children’s burials in Kent (1.North Foreland barrow, 2.South Dumpton, 3. Bradstow barrow, 4. Ramsgate Harbour, 5., St.Margaret’s Bay barrow, 6.West Cliff ring ditch,7. Lord of the Manor barrow 1, 8. Lord of the Manor barrow 3, 9. Minster barrow, 10. Ringwould-on-Kingsdown, 11.Eastling Down barrow, 12. Barham Downs barrow, 13. Boughton Aluph, 14. Shrubsoles cemetery, 15. Kingsmead Park, 16. Whiteheath barrow, 17. Great Bargrave barrow). Kent associated with flat graves than barrow sites. The general Moving into the county of Kent we are again faced with lack of grave goods is reflected in the children’s burials, similar problems of preservation and recording. Champion with only five being accompanied. Although not rich in (1982, 31) notes that ‘if the prehistory of Kent has been comparison to Wessex or Yorkshire graves, or even the neglected in favour of later periods of its archaeology, the richest graves in the county, children’s burials from Kent Bronze Age has perhaps suffered worst…There are…no are generally richer than those from Sussex and Surrey. major field monuments, no important ceremonial centres, little pottery and few settlements.’ Unlike Yorkshire and The remains of thirty-four children have been identified Wessex, there is no core assemblage of prehistoric material within the excavation records in Kent involving both from 18th and 19th century barrow exploration with which inhumations and burials after cremation. If we look first at to compare the data from more recent archaeological the rites afforded to children buried by inhumation (twenty investigations. The presence of later Saxon mounds has examples) we can start to draw together a general picture. only confused the matter further. Only 50% of the burials have been examined by an osteologist, allowing estimates of age to be determined. These children vary in age from foetus’ and neonates through to individuals of twelve to fifteen years of age. Most burials are unaccompanied and are only broadly datable to the Bronze Age by their context within diagnostic funerary sites of this date (e.g. barrows).

The area known today as Thanet, in the north east of the county, was an island in the Bronze Age and has a large concentration of over 300 barrows (Jay 1994, 418) (fig. 11). Grinsell’s 1930’s survey noted approximately 170 barrows from the rest of the county (Grinsell 1992) and very few of these have been examined to any scientific extent. It is estimated that only 16 barrows in the Thanet area have been excavated since 1743 (Jay 1994, 418).

Only two children’s graves have been furnished with grave goods, none of which are chronologically distinctive. These comprise an ironstone pebble and marine shells associated with a ‘young person’ at St Margaret’s Bay barrow (Rawlins 1872; Woodruff 1874; Grinsell 1992, 369) and a large fragment of cetacean bone which covered the crouched inhumation of a child at North Foreland barrow

Bronze Age round barrows in Kent are predominately of Bowl Barrow form (Grinsell 1992). Very few of the barrow burials were accompanied by any form of grave good but, where included, the most common association was a pottery vessel. Rich burials are more common in Kent than they are in Sussex and Surrey but are more commonly 93

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Harbour barrow where a child and a neonate had been placed within the same grave (Shand 1999, 20). Similarly a grave associated with a barrow at Eastling Down barrow contained sequential interments of the body of a 5-year-old child and a foetus or newborn baby (Bennett 1997, 319; Anderson, unpublished).

(Diack et al. 2000, 472-3). The inclusion of specifically selected pebbles such as that at St Margaret’s Bay barrow is a recurring feature of children’s burials elsewhere, like that of a three-to-five year old child at West Water Reservoir, in the Scottish Borders (Hunter 2000, 125-126). Rather than being accidental inclusions these pebbles are likely to have specifically selected for inclusion in the grave, perhaps because of perceived amuletic quality.

This barrow, investigated as part of the Whitfield-Eastry bypass excavations, revealed an exceptional series of child graves (Bennett 1997, 319; Anderson, unpublished). Much of the northern half of the barrow had already been lost at the time of excavation but a spread of seven graves containing articulated skeletons and five deposits of cremated human bone were recovered from the southern half of the barrow (Anderson, unpublished; Bennett, forthcoming). Six of the inhumations were identified as children ranging from a foetus/neonate to eleven-to-twelve years of age and two of the deposits of cremated bone were from infants (Anderson, unpublished). The dominance of children’s burials within this barrow is of interest as such a practice cannot be paralleled within the counties under consideration here, but it is similar to others from the south west such as West Overton G.6, Wiltshire (Smith & Simpson 1966). The number of children and the paucity of contemporary adult burials makes it similar to the more recent Irish cillini sites, which were burial grounds often reserved only for children (Finlay 2000).

The presence of a large cetacean bone fragment at North Foreland is more difficult to parallel but the practice of placing large slabs or nodules of flint or chalk over the deceased is known from elsewhere, for example Blackpatch, Sussex (Pull 1932, 71) and Godmersham, Kent (Bradshaw 1968, 253). The closing of the grave can be seen as part of the final act of a funerary rite. The physical act of separation between the living and the dead can be seen as a metaphor for the end of the life of the individual and the separation of the deceased from the living community. All of the child inhumation burials have been recovered from barrows. Where the placement of the body in the grave has been recorded, most have been carefully arranged into a crouched or flexed position, consistent with the placement of adult burials. Only one primary phase burial of a child has been noted in Kent, from St Margaret’s Bay barrow, comprising the crouched body of a ‘young person’ (Rawlins 1872; Woodruff 1874, 26). All other child burials appear to be secondary burials associated with barrow mounds (such as those from Lord of the Manor barrow 1: MacPherson-Grant 1977) or within ditch fills (e.g. the disarticulated remains of three individuals, including two children from Lord of the Manor barrow 1, MacPherson-Grant 1977; Grinsell 1992).

The children from the Eastling Down barrow are of particular interest, not only for the quantity of immature individuals present, but also for the pathological indicators recognised during examination of the skeletal remains. In most case, no pathological indicators are present to indicate the cause of death but one skeleton (skeleton 558), that of an eleven-to-twelve year old had a localised defect of the upper right incisor, interpreted by the osteologists as the result of facial trauma during infancy. An eight-month-old child (skeleton 639) had noticeable pitting within the eye sockets indicative of a condition known as cribra orbitalia which has been linked to iron deficiency but could also be the result of other factors not linked to diet (Anderson, unpublished). Further evidence of ill-health was indicated in the teeth of a five-year-old (skeleton 623) where the lack of wear implied that the child had been consuming a soft diet for a prolonged period of time (Anderson, unpublished).

Most of the graves contain only a single immature individual. Notable exceptions to this come from Ramsgate Harbour barrow where a child and a neonate had been placed within the same grave (Shand 1999, 20). Similarly a grave associated with a barrow at Eastling Down barrow contained sequential interments of the body of a five-yearold child and foetus or newborn baby (Bennett 1997, 319; Anderson, unpublished). All of the child inhumation burials have been recovered from barrows. Where the placement of the body in the grave has been recorded, most have been carefully arranged into a crouched or flexed position, consistent with the placement of adult burials. Only one primary phase burial of a child has been noted in Kent, from St Margaret’s Bay barrow, comprising the crouched body of a ‘young person’ (Rawlins 1872; Woodruff 1874, 26). All other child burials appear to be secondary burials associated with barrow mounds (such as those from Lord of the Manor barrow 1: MacPherson-Grant 1977) or within ditch fills (e.g. the disarticulated remains of three individuals, including two children from Lord of the Manor barrow 1, MacPherson-Grant 1977; Grinsell 1992).

The burial of a possible pregnant female was observed within a multiple burial deposit at South Dumpton Down. Here the bones of a very small child were found in the area of the pelvic girdle of the adult, implying the burial of a pregnant female who may have died from complications relating to pregnancy or childbirth (Perkins 1992, 9). It is unclear whether the child’s body was in utero. An unusual example of possible dwarfism was noted in the skull of a twelve-to-fifteen year old from Lord of the Manor barrow 3 (Perkins 1980). Although believed to be Bronze Age because of its discovery within the ditch of this barrow, it cannot be closely dated.

Most of the graves contain only a single immature individual. Notable exceptions to this come from Ramsgate 94

WHERE HAVE ALL THE FLOWERS GONE? BRONZE AGE CHILDREN’S BURIALS IN SOUTH EAST ENGLAND appropriate for children to be afforded prestige-type burials in a county where rich burials were not the norm.

Only thirteen deposits of cremated human bone have been found to contain skeletal elements belonging to immature individuals in Kent. Osteological information and, in particular, assessment of the age of the individuals is limited but the recorded burials include neonates through to children below seventeen years of age. Like the child inhumation burials from this area, few are reliably dated. Only three are Earlier Bronze Age, dated by reference to associated diagnostic pottery types and are discussed in more detail below.

Most of the cremation deposits involving the burnt bones of children are Later Bronze Age in date: three dated by direct association with chronologically distinctive ceramic vessels and a further three by their proximity to other Later Bronze Age graves. These include the burnt bones of an immature individual and a five-year-old child which were both interred within inverted Ardleigh-type urns at Barham Downs barrow 2 (MacPherson-Grant 1981, 171-2, table 1) and a collection of cremated bone of a child no more than ten years of age within a Deverel-Rimbury pot at West Cliff, Ramsgate (Moody et al. 2010, 155, 162-3).

At Ringwould-on-Kingsdown, cremated infant bones were found in association with a small cup (a miniature collared urn?) and a larger cinerary urn, probably a Biconical Urn (Woodruff 1874, 54). Although accessory cups are not uncommon (Longworth 1984, 50), the relationship between the burial of a child and a small pot is interesting and has been a reoccurring them throughout this paper.

A further three graves, two from Eastling Down barrow (Bennett 1997, 319) and one from Kingsmead Park (McKinley 2005, 77) containing cremated bone of infants/juveniles are not closely datable.

A second Earlier Bronze Age example comes from Lord of the Manor barrow 1 where a grave pit was found to contain burnt bones of a child of only a few months of age accompanied by a food vessel and a barbed and tanged arrowhead (MacPherson-Grant 1977). A third possible burial of cremated remains of an immature individual comes from Whiteheath barrow where burnt bones belonging to a very young person or female were protected by an inverted collared urn (Poste 1953, 165, fig. 1:7). A further collection of cremated bone, identified as that of two children, was associated with a single burnt flint tool at Boughton Aluph bowl barrow (Bradshaw 1966, 258). This deposit was recovered from the central, principal grave and may also be Earlier Bronze Age in date.

Like the inhumation burials from Kent, the majority of cremation burials are associated with barrows. Only two, from Whiteheath barrow (Poste 1953, 165) and Boughton Aluph (Bradshaw 1966, 258) are associated with the earliest phase of funerary activity on the site; the rest are all secondary burials. Also consistent with the inhumations is the general lack of associations in terms of both associated grave goods and accompanying individuals. In only one grave, from Kingsmead, were the burnt bones of a child associated with an adult, in this case a possible adult female (McKinley 2005). Only one grave, from Boughton Aluph, contained the burnt bones of two children probably cremated together on the same pyre. Burials associated with late Bronze Age settlements or occupation deposits, a major component of the assemblage in Sussex and Surrey, are also present in Kent. A deposit of cremated bone from an infant/juvenile and an adult female at Kingsmead Park, Allhallows, was discovered in a ditch feature associated with a Late Bronze Age occupation site (McKinley 2005, 77). The burnt bone was unaccompanied and in most respects appeared typical of Bronze Age cremation deposits except for its context within an area of settlement, which makes it exceptional.

The general lack of grave goods associated with child burials in Kent is reflective of broader patterns observed in funerary rites afforded to children during this period. In contrast with graves containing cremated bone, none of the child inhumations were accompanied by ceramic vessels. A total of six of the graves associated with cremated human bone are associated with vessels comprising typical forms for the region including a collared urn, a food vessel, Ardleigh urns and Deverel-Rimbury vessels. A less typical form of accompanying vessel comes from Ringwould-onKingsdown where a Biconical Urn, now lost, was associated with a small cup and an infant cremation (Woodruff 1874, 54). In an adjacent barrow was a rich grave containing a Biconical Urn, faience beads and a slotted incense cup (Woodruff, 1874, 55) none of which would be out of place in a Wessex Series burial. The description of the infant burial urn suggests it was just as elaborate as the latter example and indicates that elements of Wessex burial traditions adopted in Kent were considered applicable to the burial of a child.

Discussion The examination of the burial record from south east England has highlighted three major points. First, it has underlined the importance of associated material culture. The inclusion of a pot was an important part of the adult burial tradition during this period. It is interesting to note how few of the child inhumation burials were provided with a pottery vessel. On the other hand, an association with child cremation burials is common, with eighteen out of thirty four being provided with either a pot or associated with sherds. This is most obvious in Sussex, where it was also more common for children to be provided with flint tools than in the other counties examined. Provisions of other forms of grave goods, such as possible amulets in the form of specially selected pebbles, are more common in Kent.

Burials with Wessex-type elements are restricted to the east of the county which is in marked contrast to that of the other contemporary prestige items, such as metalwork, which was limited to the valleys and the northern plain (Champion 1982, 34). The Ringwould burial conforms to this pattern of distribution and illustrates that it was 95

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD

Attempts to understand the position of children in a community using burial evidence must be looked at critically but it is hoped that once a general understanding of children’s position within Bronze Age society in Britain is better understood, it will give us the opportunity to apply this knowledge in other archaeological contexts.

The burial at Church Hill, Sussex, has wider implications suggesting that, despite being buried with an adult, the child was still recognised and treated as an individual in its own right, through the provision of separate and specifically chose grave goods. This aspect is echoed in the presence of a cemetery-barrow dominated by the burials of children at Eastling Down barrow in Kent (Bennett 1997, 319) and Ladies Mile barrow in Sussex (Holleyman & Yates 1960, 136-143).

Despite patterns shown from this area being patchy, due to the limited data set available, they are valid as similar patterns, such as the inclusion of small pots in children’s graves which can be seen in Scotland, where a larger dataset has enabled general trends to be identified. This systematic search for Bronze Age children’s burials in south east England has revealed that the child’s world is far from invisible.

Secondly, the high infant mortality rate does not appear to have prevented parents from thinking about their children’s immediate or future needs. The care taken over the construction of the burials, the children’s central, primary position beneath some significant barrows, i.e. Blackpatch barrow 1, Sussex (Pull 1932, 65) and Boughton Aluph, Kent (Bradshaw 1966, 258); and the specific selection of grave goods such as small vessels and possible amulets, all suggest that the adult community considered these children as individuals worthy of respect and reverence. The state of childhood during prehistory would have been characterised differently from today, with no toys or child-specific items of material culture being present.

Acknowledgements I would like to thank Mike Lally for the opportunity to present an earlier version of this paper to the Archaeology of Infancy and Childhood conference and for drawing my attention to unpublished material from Kent. My thanks also go to Dr David Clarke and Dr Alison Sheridan for their useful comments on an earlier draft of this paper. I am also grateful to Craig Angus and Marion O’Neil for their assistance with the illustrations.

And finally, to answer the question posed in the title ‘where have all the flowers gone?’ this study has shown that children are under-represented in Bronze Age funerary rites in this area but that not all are missing. Analysis of wider burial traditions in these regions has indicated that this apparent invisibility of children’s burials has more complex reasons than are immediately apparent. Local land-use, lack of scientific, systematic investigation and the paucity of detailed records have restricted information about burial traditions during this period. In Surrey, in particular, where the sphere of Bronze Age activity appears to be limited, the lack of detailed analysis of skeletal remains makes it almost impossible to identify children within the archaeological record. This is a bias formed by the limitations of archaeological investigation rather than reflecting a genuine trend in burial practice.

References Adkins, L. & Needham, S. 1985. Research on a Late Bronze Age Enclosure at Queen Mary’s Hospital, Carshalton. Surrey Archaeological Collections 76, 11-50. Anderson, T. unpublished. Report on the human bones from Whitfield-Eastry by-pass, area 3. Commissioned by Canterbury Archaeological Trust. Bennett, P. forthcoming. Excavations on the line of the Eastry to Whitfield Bypass, 1995. Canterbury: Canterbury Archaeological Trust.

Most burials within this region conform to a basic burial tradition in which adults and children were treated in a very similar manner, indicating that children, in death if not in life, were considered part of the wider community. Analysis of associated material culture has shown that regional and inter-regional variations in burial practices existed and extended to the burial of children. Rich burials in this area are not common but can be seen to extend to funerary traditions involving children. Several of these ‘richer’ burials can be seen to involve elements of ‘Wessex’ burial traditions and indicate that these rites were also considered appropriate for children in some circumstances. A small number of burials, associated material culture, such as the provision of a small pot, indicates that some understanding of childhood as a separate life-stage was acknowledged and understood within Bronze Age society. Despite this, there is significant under-representation of children in the burial record in this area and it is hoped that this analysis will go some way to explaining why.

Bennett, P. (ed.) 1997. Interim report on work carried out by the Canterbury Archaeological Trust. Archaeologia Cantiana 116, 311- 325. Bradshaw, J. 1966. Late Bronze Age barrow, Boughton Aluph. Archaeologia Cantiana 80, 258. Bradshaw, J. 1969. Godmersham. Archaeologia Cantiana 83, 253. Brück, J. 1995. A place for the dead: the role of human remains in Late Bronze Age Britain. Proceedings of the Prehistoric Society 61, 245-277. Brück, J. 2001. Body metaphors and technologies of transformation in the English Middle and Late Bronze Age, in J. Brück (ed.) Bronze Age Landscapes: Tradition and Transformation, 149-60. Oxford. 96

WHERE HAVE ALL THE FLOWERS GONE? BRONZE AGE CHILDREN’S BURIALS IN SOUTH EAST ENGLAND European Archaeology. London: Leicester University Press, 230-212.

Burstow, G. P. and Holleyman, G. A. 1957. Late Bronze Age settlement on Itford Hill, Sussex, Proc Prehist Soc 23 (New Ser.), 167-212.

Finlay, N. 2000. Outside of life: traditions of infant burial in Ireland from cilliní to cist’ World Archaeology 31:3, 407-422.

Butler, C. 1991. The excavation of a Beaker bowl barrow at Pyecombe, West Sussex. Sussex Archaeological Collections 129, 1-28.

Gardner, E. 1924. Bronze Age urns in Surrey. Surrey Archaeological Collections 3, 1-29.

Chadwick, A. M. 2006. Bronze Age burials and settlement and an Anglo-Saxon settlement at Claypit Lane, Westhampnett, West Sussex, Sussex Archaeological Collections 144, 7-50.

Garwood, P. 2002. Round barrows and funerary traditions in Late Neolithic and Bronze Age Sussex. In D. Rudling (ed.) The Archaeology of Sussex to AD 2000. Sussex: University of Sussex.

Chamberlain, A. 2000. Minor concerns: a demographic perspective on children in past societies. In J. S. Derevenski (ed.) Children and Material Culture. London: Routledge, 206-212.

Goodman, A. H. and Armelagos, G. J. 1989. Infant and childhood morbidity and mortality risks in archaeological populations. World Archaeology 21:2, 225-243.

Champion, T. 1982. The Bronze Age in Kent. In P.E. Leach (ed.) Archaeology in Kent to AD 1500. Dorchester: CBA Research Report 48 31-39.

Greenwell, W. 1877. British Barrows: A record of the examination of sepulchral mounds in various parts of England. Oxford: Clarendon Press.

Cruse, R. J. and Harrison, A. C. 1984. Excavation at Hill Road, Wouldham. Archaeologia Cantiana 99, 81-108.

Greenwell, W. 1890. Recent researches in Barrows in Yorkshire, Wiltshire, Berkshire. Archaeologia 52, 1-72.

Cunnington, M. E. 1929. Woodhenge. A description of the site as revealed by excavations carried out by Mr and Mrs B H Cunnington, 1926-7-8. Also of four circles and an earthwork enclosure south of Woodhenge. Devizes: George Simpson and Co. Ltd.

Grimm, L. 2000. Apprentice flintknapping: Relating material culture and social practice in the Upper Palaeolithic’, in J.S. Derevenski (ed.) Children and Material Culture. London: Routledge, 53-71. Grinsell, L. V. 1931. Sussex in the Bronze Age. Sussex Archaeological Collections 72, 30-68.

Curwen, E. & Curwen, E. C. 1935. Two Beakers and an Early Iron Age Urn. Sussex Archaeological Collections 76, 1-5.

Grinsell, L. V. 1934a. Sussex Barrows. Archaeological Collections 75, 217-275.

Davidson, C. B. 1868. Notice of further stone kists found at Broomend, near Inverurie Papermills. Proceedings of the Society of Antiquarians of Scotland 7, 115-118.

Sussex

Grinsell L. V. 1934b. An analysis and list of Surrey Barrows. Surrey Archaeological Collections 42, 26-60.

Denison, S. 1995. ‘News.’ British Archaeology 10, 4-5.

Grinsell, L. V. 1992. The Bronze Age Round Barrows of Kent. Proceedings of the Prehistoric Society 58, 355-384.

Derevenski, J. S. 1994. Where are the Children? Accessing Children in the Past. Archaeological Review from Cambridge 13:2, 7-20.

Heslop, D. H. 2008. Patterns of quern production, acquisition and deposition. A corpus of beehive querns from Northern Yorkshire and Southern Durham. Yorkshire Archaeological Society Occasional Paper No.5.

Derevenski, J. S. 1997. Engendering children, engendering archaeology. In J. Moore & E. Scott (eds.) Invisible People and Processes: Writing Gender and Childhood into European Archaeology. London: Leicester University Press, 192-202.

Holdgate, R. 1987 The Looe’s Barn Tumulus, Saltdean, East Sussex, Sussex Archaeological Collections 125, 230. Holleyman , G. A. & Yeates, C. W. 1960. Excavations in the Ditchling Road area, north of Brighton 1950-57. Sussex Archaeological Collections 98, 133-149.

Diack, M., Mason, S. & Perkins, D. 2000. North Forland. Current Archaeology 168 (Special issue: Kent), 472-3. Drewett, P. 1982. Later Bronze Age downland economy and excavations at Black Patch, East Sussex, Proc Prehist Society 48, 321-400.

Hunter, F. 2000. Excavation of an Early Bronze Age cemetery and other sites at West Water Reservoir, West Linton, Scottish Borders. Proceedings of the Society of Antiquaries of Scotland 130, 115-182.

Finlay, N. 1997. Kid knapping: the missing children in lithic analysis, in J. Moore & E. Scott (eds.) Invisible People and Processes: Writing Gender and Childhood into

Jay, L. A. 1994. Researches and discoveries in Kent, Archaeologia Cantiana 112, 405-422. 97

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD Kirk, W. and McKenzie, J. 1955. Three Bronze Age cist burials in north-east Scotland. Proceedings of the Society of Antiquaries of Scotland 88, 1-14.

Perkins, D. R. J. 1992. Two Archaeological Evaluations; Dumpton Gap and South Dumpton Down, Broadstairs. Broadstairs: Thanet Archaeological Trust.

Longworth, I. H. 1984. Collared Urns of the Bronze Age in Britain and Ireland, Cambridge: Cambridge University Press.

Phillips, H. 1857. Discovery of a tumulus at Hove, near Brighton, containing an amber cup, etc. Sussex Archaeological Collections 9 , 119-24.

Lunt, D. A. 1972. The Dentition in Studies of Skeletal Material from Archaeological Sites. Scottish Archaeological Forum 4, 114-117.

Poste, B. 1953. Memoranda relating to opening of the barrow on Whiteheath, 26th July 1842. Archaeologia Cantiana 65 (1952), 162-66.

Macpherson Grant N. 1977. Excavation of a Neolithic/Bronze Age Site at Lord of the Manor, Haine Road, Ramsgate. Isle of Thanet Archaeological Unit, Publication No. 1.

Pull, J. H. 1932. The Flint Miners of Blackpatch. London: Williams Norgate. Ratcliffe-Densham, H. B. A. & Ratcliffe-Densham, M. M. 1961. An anomalous earthwork of the Late Bronze Age, on Cock Hill, Sussex. Sussex Archaeological Collections 99, 78-101.

MacPherson Grant, N. 1981. Archaeological work along the A2: 1966-1974, Archaeologia Cantiana 94 (1980), 133-83.

Rawlins, F. J. 1872. Tumulus near Walmer, Kent. Proceedings of Society of Antiquaries of London, 2nd Series, 5 (1870-73), 380-2.

Mays, S. A. and Anderson, T. 1995. Archaeological research priorities for human remains from South-East England (Kent, East and West Sussex and Surrey). Archaeologia Cantiana 115, 355-388.

Rudling, D. 2000. In the (way of the) pipeline. Sussex Past and Present 90, 11.

McKinley, J. 2004. Archaeological inverstigations at The Bostle, Bronze Age and Anglo-Saxon cemeteries, Balsdean, East Sussex, 1997. Sussex Archaeological Collections 142, 25-44.

Rudling, D. 2001. Downland Settlement and Land-Use: The Archaeology of the Brighton Bypass. London: University College London Field Archaeology Unit Monograph No. 1.

McKinley, J. 2005. The cremated bone, in C Greatorix Later Prehistoric Settlement on the Hoo Peninsula: Excavations at Kingsmead Park, Allhallows. Archaeologia Cantiana 125, 77-8 (67-81).

Russell, M. 2001. Rough Quarries, Rocks and Hills: John Pull and the Neolithic Flint Mines of Sussex. Bournemouth: Bournemouth University School of Conservation Sciences Occasional Paper 6.

Moody, G., MacPherson-Grant, N., Anderson, T. 2010. Later Bronze Age cremation at West Cliff, Ramsgate, Archaeolgia Cantiana 130, 147-172.

Scott, E. 1992. Images and contexts of Infants and infant burials: some thoughts on some cross-cultural evidence. Archaeological Review from Cambridge 11:1, 77-92.

Needham S. P. 1987. The Bronze Age. In J. Bird & D.G. Bird (eds.) The Archaeology of Surrey to 1540. Surrey: Surrey Archaeological Society, 97-137.

Shand, G. 1999. Ramsgate Harbour Approach Road, in Canterbury’s Archaeology 1998-1999: Canterbury Archaeological Trust 23rd Annual Report, 18-22.

Needham, S. P. 1992. The structure of settlement and ritual in the Late Bronze Age of South-East Britain. In C. Mordant & A. Richard (eds.) L’habitat et L’occupation du Sol à L’âge du Bronze en Europe. Paris: Editions du Comité des Travaux Historiques et Scientifiques, 49-69.

Smith, H. 1870. Notes on Prehistoric Burial in Sussex. Sussex Archaeological Collections 22, 57-76. Smith, M. J. B. 1994. The Excavated Bronze Age Barrows of North-East Yorkshire. Durham: Architectural and Archaeological Society of Durham and Northumberland (= Research Report No.3)

Oakley, K. P., Rankine, W. F. & Lowther, A. W. G. 1939. A Survey of the Prehistory of the Farnham District, Surrey. Guildford: Surrey Archaeological Society.

Smith I.F. & Simpson D.D.A. 1966 Excavation of a Burial Barrow on Overton Hill, North Wiltshire, Proc Prehist Soc 32, 122-55.

Perkins D.R.J. 1980. Site 3. Lord of the Manor (Ozengell) Ramsgate. Interim Excavation Reports 1977-1980, 13-17. The Isle of Thanet Archaeological Unit.

Wainright, G. J. 1979. Mount Pleasant, Dorset: Excavations 1970-71. London: Report of the Rescue Committee of the Society of Antiquarians, London, No. 29.

98

WHERE HAVE ALL THE FLOWERS GONE? BRONZE AGE CHILDREN’S BURIALS IN SOUTH EAST ENGLAND Whimster, D. C. 1931. The Archaeology of Surrey. London: Methuen.

Woodruff, C. H. 1874. An account of discoveries made in Celtic Tumuli near Dover, Kent. Archaeologia 45, 53-56.

99

Ble mae’r babanod? (Where are the babies?): Infant burial in Early Medieval Wales Marion R. Page

in Somerset, a grave that measured 0.76m x 0.35m contained the burial of a six month old (Bell 1990, 76).

Introduction In contrast with the cemeteries of Anglo-Saxon England there has until recently been little archaeological examination of the mortuary practices of the non AngloSaxon peoples who inhabited Britain in the early medieval period. This bias has been for two reasons: the first being that Anglo-Saxon archaeology has a very long history (Lucy 1998, 5-21) and the second being that, in general, the pagan Anglo-Saxon burials of the 5th to 7th centuries CE were richly furnished and consequently highly visible. Archaeologists have therefore been able to examine the wider context of Anglo-Saxon burials through artefact analysis and through social approaches such as that by Lucy (1994), which looked at age-related patterning in mortuary practices within and between sites.

Cemetery types Cemeteries of this period in Wales can be divided into four typological groups (James 1987, 64). The first group contains long-cists and dug graves in, or associated with, prehistoric burial sites. It is assumed, based on this association, that these are the earliest type and that they date to a period before rural cemeteries were well established. It has also been argued that they probably represent a continuation of Iron Age and Romano-British practices (James 1987, 64). The second type exists within small defended enclosures. The best examples of this type are from Pembrokeshire (Caer, Bayvil) and Anglesey (Arfryn). Some of these cemeteries are associated with early medieval stones (EMS). Nancy Edwards (2001, 38) has recently suggested that the primary function of the fifth to 7th century EMS (Nash Williams Group 1 (1950)) was to commemorate ‘the graves of the more important members of early medieval Welsh society, almost entirely men’. Group 1 EMS are inscribed in Ogham or Latin or have a bilingual inscription. The more numerous Group 2 EMS are decorated with simple crosses and are mostly associated with known church and cemetery sites. Some, however, are associated with earthwork enclosures and James (1987, 64) has suggested that if these stones are indicative of cemeteries, rather than single burials, then the number of cemeteries of this type can be expanded to include these sites. Again this type of cemetery contains long-cists and dug graves. Whether the final use of these sites as cemeteries represents the re-use of a settlement site after a period of abandonment, or a change in function of a continuously occupied site, is not yet known (James 1987, 64).

In Wales, social approaches to interpretation are made more problematic because of the nature of the burial evidence from early medieval cemeteries. In general, the Welsh soils are acidic so there is at best, poor, at worst, non-existent skeletal preservation at the majority of these cemeteries. And it is believed that Christianity was introduced to Wales in the Roman period, so an unfurnished west-east supine inhumation was the normative Early Medieval burial rite. Wales also has less excavated cemeteries of any great size that date from the period. Yet despite the problems with the data, this paper will demonstrate that a social analysis of early medieval Welsh mortuary data is still possible using new perspectives and approaches. This paper developed from the results of an MA thesis (Manwaring 2000) that examined the nature of the infant burial rite in the early medieval Welsh cemeteries. Infants are consistently under-represented in communal cemeteries in many countries and from many periods (Scott 1999). One of the objectives of the study was to see if under-representation of infants, classed by this author as fetuses, neonates or babies under the age of one year, could also be identified in early medieval Wales, when death in the first year of life (by analogy with nonWestern rural populations) is likely to have accounted for around 20% of the total mortality rate (Longley 2001, 108). For each cemetery the published excavation report and, where possible, excavation archive was examined for the presence of grave cuts measuring less than one metre. Given the aforementioned absence of skeletal material, this was deemed by the current author as presenting the best method of identification for infant graves. Such a model for identification has been adopted before now. At Capel Eithin cemetery, for instance, one metre was the maximum length of grave cut that the excavators allowed when identifying an infant burial (Smith and White 1999). Outside Wales, at Brean Down

A third, and larger, group of cemeteries from this period are those that have no known earlier or later activity at the site. Again some of these sites are associated with EMS (James 1987, 66), which have often been used to date the cemetery in the absence of other evidence. This is not a reliable dating method as very few stones are found where they were originally sited, and it has been convincingly argued that the presence of Group I EMS does not date a cemetery (Petts 2002, 27-8). The final cemetery group contains burials associated with medieval church or chapel sites and, again, some of these are associated with EMS. Both cist burials and dug graves are found, either through the ploughing of abandoned sites or through excavation in churchyards that continue in use. This represents the most common occurrence of cist burial in Wales (James 1987, 66).

100

BLE MAE’R BABANOD? (WHERE ARE THE BABIES?): INFANT BURIAL IN EARLY MEDIEVAL WALES Dating early medieval cemeteries The dating of early medieval cemeteries in Wales is problematic. That the presence of Group 1 EMS does not securely date a cemetery has been noted above, and the Group 2 stones are now thought to have a longer date range than Nash-Williams original assessment (Edwards 2001, 30). The absence of grave goods in early medieval Welsh burials means that graves cannot be dated through artefact association and it should be noted that cist burial is known to have continued into the later medieval period, for example at Eglwyswrw in Pembrokeshire (Ludlow 1998). Where secure dating evidence is unavailable a cemetery is usually dated to the early medieval period due to its similarity to dated sites; i.e. a cemetery of artefactually sterile extended inhumations that are orientated east-west. Before discussing Welsh cemeteries it must also be noted that the period between the late 5th and the mid-6th centuries produces wide distributions of radiocarbon dates, due to a plateau on the calibration curve between 450 and 530 CE. This means that, even when there is datable evidence, close dating of sites of this period is difficult (Petts 2002, 27).

Fig. 2: Caer Bayvil. Plan of trench showing location of graves. ©Dyfed Archaeological Trust. On the basis of this evidence fifteen out of sixty-one burials at Caer appear to have been those of infants, representing 24.6% of the excavated sample. This is a level of infant mortality that one would expect for a cemetery of this date (Longley 2001, 108). Yet the excavation revealed that the infant graves were the latest in the sequence of burials, and the suggestion was made that they may have been interred after the cemetery went out of use for adults (James 1987, 63). This could mean, given the evidence of the radiocarbon date, that all the infant graves post-date the late 9th century CE and that infants were under-represented in the cemetery population at the time that it was in use as a community burial ground.

The Welsh cemetery evidence

Plas Gogerddan (Ceredigion) Dateable material from Plas Gogerddan shows it to be a multi-period site with evidence for activity beginning in the Neolithic when some carbonized food remains were buried in a small pit (K. Murphy pers. comm). The excavations also revealed three Bronze Age ring ditches, Iron Age burials and the Early Medieval cemetery (Murphy 1992). Wood from a coffin in this cemetery gave a radiocarbon date of 340-620 CE (CAR-1045) (Petts 2000, Appendix 3). The lack of bone survival, along with the shallow depth of some graves and the possible confusion of small graves with other features meant that the exact number of graves was not certain (Murphy 1992, 15). Twenty-two graves were positively identified, of which three were described as being the graves of children. None of these graves was more than 0.94m in length, suggesting that all three may have been infant graves. A further unexcavated grave was only 0.7m long and this is also likely to have been an infant grave. Three adult graves in the cemetery were surrounded by rectangular structures, two of which contained one small, stone-lined pit as well as the grave (Fig. 3). Phosphate analysis revealed that these were not the graves of infants and they were interpreted by the excavators as receptacles for dedicatory or ritual offerings (Murphy 1992, 22). The total of four infant graves represents 18.2% of the

Fig. 1: Cemetery locations. Caer, Bayvil (Pembrokeshire) In 1979 trial excavation by the Dyfed Archaeological Trust at Caer, or Y Gaer, revealed 61 graves (Fig. 2). Very little bone survived, although a single radiocarbon date of 650-890 CE (CAR-291) was obtained for the site (Petts 2000, Appendix 3). Four lintel graves were excavated, two of which were identified as children’s graves. Measuring approximately 0.62m by 0.28m, it is possible that these were both infant graves. Of 13 possible cist graves, not all of which were excavated, four were less than a metre in length and may also have contained infant burials. Forty-four simple dug graves were also excavated, nine of which may have contained infants since all had a grave cut of less than one metre in length.

101

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD were infant burials since they were each only 0.8m in length. They represent 28.5% of the total number of burials, although this figure is clearly skewed by the small size of the cemetery. The association with the barrows makes this one of James’ early type, and its small size suggests either that the focus of burial must have shifted elsewhere very soon after its foundation, or that this was a group of individuals who, for some reason, could not be buried in with the rest of their community.

excavated cemetery population, slightly below what one would expect.

Fig 3: Plas Gogerddan. © Dyfed Archaeological Trust. Tandderwen (Clwyd) Like Plas Gogerddan, Tandderwen is a multi-period site where there was Mesolithic and Bronze Age activity, in addition to the Early Medieval cemetery. It is thought that a Bronze Age barrow within the larger of two ring ditches would still have been visible when this cemetery was founded (Brassil et al., 1991, 87). Thirty-nine inhumations, all orientated west-east, were excavated. Eleven of these burials were surrounded by square ditches arranged in three rows, also on a west-east alignment. Dates centering on the 8th and the 10th centuries CE were obtained from two of these square ditch burials (Petts 2000, Appendix 3). The graves ranged in length from 1.76m to 2.8m and in width from 0.7m to 1.2m. This led to the identification of all the burials as extended adult inhumations. Bone did not survive, but tooth fragments from three graves did not suggest the presence infants or children in adult graves (Brassil, et al. 1991, 64, 89). The excavators caution that it is possible the cemetery has not been completely excavated. If it has, then burial of infants, and indeed children, elsewhere may have been the norm for this population.

Fig. 4: Burials at Trelystan. ©Clwyd-Powys Archaeological Trust. The Eastgate Cemetery, Caerwent (Gwent) Archaeological preservation at this site was poor due to damage from previous excavations (Ashby, 1911; Morgan, 1856) and further subsequent disturbance. Three broad chronological phases were identified in the most recent excavations: pre-cemetery, Roman cemetery/Early-Medieval and post-cemetery/PostMedieval. A minimum of 136 burials were excavated. Almost all graves had contained extended unaccompanied west-east inhumations with the head to the west (Campbell and MacDonald 1993, 78, 80). The excavators grouped the burials into four age groups, one of which was ‘infants’ – classed in the report as individuals under the age of two years – of which 12 were identified. This group is under-represented, making up only 8.8% of the total cemetery population. Some clustering of infant and child graves was observed but was not thought to be significant as the clusters were interspersed with adult burials (Campbell and MacDonald 1993, 87). There were two types of graves: cists/partial cists and simple dug graves. The excavators found little differentiation in burial rite according to age or sex but noted that none of the infant and child graves were stonelined (Campbell and MacDonald 1993, 80, 86). The infant and child mortality rate of 24% at Caerwent was noted as being comparable with other contemporary and

Trelystan (Powys) The small cemetery at Trelystan was located adjacent to two Bronze Age round barrows (Britnell 1982). It comprised the rock-cut graves of five adults and two children, all orientated east-west. There was no dating evidence, as bone survival was poor, but the burials were assumed to date from the early medieval period (Britnell 1982, 162). It is possible that both the smaller graves 102

BLE MAE’R BABANOD? (WHERE ARE THE BABIES?): INFANT BURIAL IN EARLY MEDIEVAL WALES infant burials appear to be focused on prehistoric features at other contemporary Welsh cemeteries.

demographically normal sites. However, this is the infant and child mortality rate; the figure of 8.8% for individuals under the age of two years is much lower, and very young infants are under-represented in the cemetery population at Caerwent.

Atlantic Trading Estate, Barry (South Glamorgan) A total of forty-five east-west burials were excavated at this site (Price 1987, 60). Like Llandough this cemetery may have its origins in the Roman period. All graves were contained within full or partial cists and were cut into wind-blown sand. Samples taken from the graves have given radiocarbon dates that span the late 4th to the early 6th centuries CE, with one 8th century CE outlier (C. Newman pers. comm.). Although cist burial appears to have been the norm, a variety of other burial practices were identified. There was a double burial of an adult and child with pillow stones, hunching of the shoulders of some skeletons suggestive of shroud burial and several graves containing evidence for either wood coffins or for bodies being bound between planks of wood (Price 1987, 60).

Radiocarbon dates for Caerwent span the mid-4th to the early 12th centuries CE (Petts 2000, Appendix 3). The floruit for the cemetery seems to be the 6th and 7th centuries CE; the 12th century date is an outlier (Campbell and MacDonald 1993, 87). Two explanations for this late date, obtained from a burial in the northwestern area of the cemetery, are made in the excavation report. The first is that this area represents the final phase of the cemetery and therefore that all the burials here would yield similar dates. The second is that, since the burial is that of a child aged about 11 years, it represents secondary use of an already abandoned site for someone who failed to qualify for burial in consecrated ground and the occurrence of this practice for the unbaptised is noted Campbell and MacDonald 1993, 88-89). The ‘late’ child burial at Caerwent is not an infant but, although it would have been rare, there is evidence to suggest that in the 12th century a child could remain unbaptised for many years (Crawford 1993, 88). Llandough (South Glamorgan) The cemetery at Llandough is one of the most archaeologically important early medieval cemeteries in Wales because of the degree of skeletal preservation at the site. A total of 858 graves were recorded, by the Cotswold Archaeological Trust, at the site of a mid-7th to 11th century CE monastery founded within the estate of a former Roman villa (Holbrook and Thomas 1996, 74). A total of 773 skeletons from the cemetery could be aged. As with the Roman cemetery at Lankhills it was possible not only to identify and record the graves of infants but also to age their remains with a high degree of accuracy. The graves of six fetuses, 13 perinates and 46 infants were excavated. This total of 65 infants represents 8.4% of the number of individuals that could have the age at death determined, so infants are under-represented (Loe and Robson-Brown 2005).

Fig. 5: Burials at Atlantic Trading Estate. After Price 1987. Only one of the burials at the Atlantic Trading Estate was a very young individual. The site has the appearance of a Christian cemetery and has been likened to Poundbury (C. Newman pers. comm.) However, the number of infant burials reveals a striking difference between the two sites. At Poundbury infants represent 26.2% of the total cemetery population (Farwell and Molleson 1993, 155). At Barry the single infant burial represents a mere 2% of the total number of graves and cannot be an accurate reflection of the infant mortality rate in the living population.

The majority of the infant burials at Llandough were found in areas originally thought to represent later phases of burial at the site, where they cut and overlapped earlier graves (Holbrook and Thomas 1996, 75). In particular there were a noticeably high number of infant graves in Areas II and III to the north of the site (Loe and RobsonBrown 2005, 45). In Area III these infant graves are interpreted as representing the last phase of burial in that part of the cemetery (Holbrook and Thomas 2005, 89). This could be interpreted as evidence of continuing use of the site for infants beyond its use for older members of society. It has also been suggested that there may have been more than one focus of burial at the cemetery and that there may have been a prehistoric feature in the northern part of the site (Holbrook and Thomas 2005, 89). The high percentage of infants in this area is interesting; it will be seen elsewhere in this paper that

Capel Eithin (Gwynedd) The cemetery at Capel Eithin is associated with a Bronze Age cairn and cremation cemetery and a possible Roman signal station (Smith and White 1999, 19, 123). There is no reliable date for the site; the radiocarbon dates obtained range from the 1st to the 11th century CE (Smith and White 1999, 145) due partly to the aforementioned plateau on the calibration curve. A total of 102 graves in the Early Medieval inhumation cemetery 103

(RE)THINKING THE LITTLE ANCESTOR: NEW PERSPECTIVES ON THE ARCHAEOLOGY OF INFANCY AND CHILDHOOD have been excavated, 99 of which were suitable for further study (Table 1). Some burials in the cemetery were in simple dug graves; others were full or partial cists. Bone had not survived in the acidic subsoil but phosphate analysis showed the interpretation of elongated features as dug graves to be correct (Smith and White 1999, 128, 130, 139). Size range >1.6m 1.0-1.6m