Growing Up in the Ice Age

Fossil and Archaeological Evidence of the Lived Lives of Plio-Pleistocene Children

April Nowell

• Publisher: Oxbow Books
• Year: 2021
• eBook ISBN: 9781789252958
• Topic: History
• Subtopic: European History
• Index: History

In prehistoric societies children comprised 40–65% of the population, yet by default, our ancestral landscapes are peopled by adults who hunt, gather, fish, knap tools, and make art. But these adults were also parents, grandparents, aunts, and uncles who had to make space physically, emotionally, intellectually, and cognitively for the infants, children, and adolescents around them. Growing Up in the Ice Age is a timely and evidence-based look at the lived lives of Paleolithic children and the communities of which they were a part. By rendering these 'invisible' children visible, readers will gain a new understanding of the Paleolithic period as a whole, and in doing so will learn how children have contributed to the biological and cultural entities we are today.

Chapter 1

Toward an archaeology of Paleolithic children

Without children there is no inter-generational continuity of life and culture. As humans live on earth today, it is an empirical (and banal) fact that children must have been born, raised, grown into adults and had their own children throughout the deep time of human evolution and prehistory.

(Anders Högberg 2018)

Where are all the children?

It is estimated that in prehistoric societies children comprised at least 40 to 65% of the population (Baxter 2005a; 2008), yet by default, in our collective imaginations, our ancestral landscapes are peopled by adults who hunt, gather, fish, knap tools and make art. But these adults were also parents, grandparents, aunts and uncles (however they would have codified these kin relationships) who had to make space physically, emotionally, intellectually and cognitively for the infants, children and adolescents around them. Nonetheless, the archaeological literature has largely been silent about the lives these children lived and the contributions they made because the economic, social and political roles of Paleolithic1 children are often assumed to be negligible at best and unknowable at worst.

This is not a problem that is unique to those who study the Paleolithic period but is an issue that pervades archaeology more generally. Thirty years ago, the Norwegian archaeologist, Dr Grete Lillehammer (1989), published her seminal paper, ‘A Child is Born’, which many archaeologists saw as a call to arms to meaningfully integrate children into archaeological inquiries (Nowell et al. 2020). In the intervening decades, there has been a slow but steady uptake in child-focused studies in archaeology. But why were children understudied in the first place? It should be obvious that children existed in prehistory after all. In answer to this question, I think there are four main reasons that children have traditionally been understudied.

Living populations, archaeological populations and taphonomy

First, there are often fewer of them to study as the number of children uncovered in the archaeological record is not proportional to the number of children in living populations. Let’s consider this in detail. Moving from a living population to an archaeological population is a winnowing process where there is a loss of information at each step of the way (Fig 1.1) (Séguy and Buchet 2013). Let’s imagine a census of all those present in a particular region during a specific range of time, i.e. what paleodemographers call the ‘living population’. It is possible, even likely, that not everyone in that population uses a particular cemetery or burial place. This is particularly true for people who practice a highly mobile hunter-gatherer lifestyle, which, until about 10,000 years ago, was the norm for virtually everyone on the planet (Bocquet-Appel et al. 2005; French 2015; French and Collins 2015; Kretschmer 2015; Maier and Zimmerman 2017; Schmidt and Zimmerman 2019). In this case, the remains of those who die in a given year from, for example, old age, illness, accident or childbirth are distributed over a larger geographic area relative to population density. Even in more sedentary populations, access to a burial ground can be defined based on socio-economic factors (e.g. a pauper’s cemetery vs. Egypt’s Valley of the Kings), religious factors (e.g. a Jewish or Catholic cemetery), ethnic factors (a Chinese cemetery or New York’s historic African Burial Ground) or other determinants (e.g. a Leper’s cemetery or a military cemetery such as Arlington National Cemetery). Those people who have access to a particular cemetery form the ‘burying population’. A subset of the burying population is referred to as the ‘deceased population’. These are the people who end up dying. The specific make-up of this population will vary based on age, sex, socio-economic status and individual pathological condition (Séguy and Buchet 2013, 16). We then move to the ‘exhumed population’. The exhumed population is the proportion of the deceased population that is recovered through archaeological excavation. This number will vary based on whether the cemetery or burial ground is fully excavated or only selectively sampled. It also varies based on differential preservation. Finally, of the exhumed population, only a portion of these individuals will be amenable to analysis such as sex and age determination. This last group forms the ‘analyzable population’. As Séguy and Buchet (2013, 17) observe, ‘[e]ven in optimal archaeological conditions it is never certain that the proportion exhumed is significantly representative of all the components of the buried population.’ And I would argue even less so of the living population.

Figure 1.1: Diagram demonstrating the degree to which an archaeological population is representative of a living population (French, in press; reproduced with permission of J.C. French and redrawn by J. Beller).

This winnowing process impacts all sex and age classes but the degree to which subadults are affected is still a matter of debate (Lewis 2007). The question is exactly how representative of the living population is the archaeological (i.e. analyzable) population of children. In an influential study, Schofield and Wrigley (1979) wrote that while mortality rates of individuals under 10 years of age are roughly 2.4% in modern industrial societies they are closer to 34% in modern non-industrial societies. According to Lewis (2007, 22), ‘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’. But in many cemetery samples and other forms of burial grounds, subadults represent less than 30% of the analyzable population (Lewis 2007).

This under-representation is usually assumed to be for taphonomic reasons. Taphonomy, from the Greek word taphos (τάφος) meaning ‘burial’ and nomos (νόμος) meaning ‘law’, is the study of burial processes (Efremov 1940). All things being equal, larger, denser bones preserve longer and in better condition in the archaeological record than do smaller, more porous bones – think elephant bones versus hollow bird bones or even a human femur vs. a middle ear ossicle. Not only are children’s bones smaller but their epiphyses (i.e. the ends of their long bones) have not yet completely ossified or fused to the rest of the bone. Their bones are also highly porous, less mineralized and lack tensile and compressive strength. All of these characteristics render them particularly vulnerable to sedimentary pressure, bioerosion through contact with acidic soil and decomposing organic matter, and excavator bias (Nowell 2020; Nowell and Kurki 2020).

Excavator bias refers to the difference in recovery rates of materials based on a variety of factors including excavator experience and choice of methods employed. For instance, unless you screen excavated sediments using a fine mesh, you are much more likely to miss tiny remains such as deciduous teeth or fragments of vertebrae (Pokines and De la Paz 2016). At Drimolen, for example, a 2–1.5 million year old site in South Africa, excavators have uncovered the remains of 80 hominins2 attributable to Homo sp. and Paranthropus robustus. Of the specimens that could be aged, infants (defined in this study as those under 5 years of age) accounted for just under 35%. If those individuals aged between 6 and 10 years of age are included, then immature specimens account for almost half of the hominin sample (Riga et al. 2019). There are important behavioral reasons that likely explain this age distribution but another factor is the choices made by the site’s project directors. Researchers there sieve and water screen all excavated materials including all medium and fine meshed finds. They then spread this material out on a table and meticulously sort through all of it (S. Baker personal communication 2020). This allows them to recover even the tiniest deciduous tooth that would likely have been missed if they did not follow such a rigorous protocol.

Another factor in the under-representation of children in the archaeological record is the fact that sometimes children may be subject to different funerary processes than older individuals due to cultural differences, for example, in attitudes towards ‘personhood’. Differences in personhood mean variations in when (and on what basis) an individual is accorded the status of being a person and, in this case, a member of society requiring burial. Funerary differences associated with children may include interment in remote locations, shallower graves, the absence of grave markers and/ or coffins, no interment at all or interment in burial containers such as jars or in ceramic vessels in caches (e.g. Lally and Ardren 2009; Carroll 2012; O’Reilly et al. 2019).

But questions of personhood are not always straightforward. Eileen Murphy (2011) studies 17th-century cilliní – or children’s burial grounds in Ireland. These burial grounds were primarily for unbaptized infants but were also often the final resting place of ‘other members of Irish society who were considered unsuitable by the Roman Catholic Church for burial in consecrated ground’ (Murphy 2011, 409). Cilliní were often located in deserted church yards or other liminal spaces such as bogs – these spaces are the physical manifestation of being in ‘limbo’ or on the border between Heaven and Hell in the Roman Catholic tradition. More than 1300 cilliní have been identified throughout Ireland by archaeologists. Their location suggests that these infants were ‘spiritually ambiguous’. Nonetheless, Murphy (2011) found expressions of parental grief such as simple grave markers (e.g. stones or stones bearing crosses) and grave goods such as white quartz, sea pebbles, and sea shells as well as an example of a figurine that resembles a baby in swaddling clothes and several ‘jacks’ – perhaps these mementos and toys were for the children to play with or to give them comfort while they waited in limbo. In other cases, neonates and infants were buried surreptitiously under church eaves so that they could be ‘baptized’ by dripping rainwater (Craig-Atkinson 2014). These examples underscore that differential burial doesn’t necessarily mean lack of care and emotion but it can contribute to an under-representation of children in the archaeological record.

Children as ‘distorters’ of the archaeological record

The second reason that children have been understudied in the archaeological record is that children’s play and their ‘unconventional’ use of material culture were, until recently, believed to introduce a randomizing and distorting element into the archaeological record (Baxter 2005a; see for example, Hammond and Hammond 1981). As Jane Baxter (2008) argues, for many archaeologists children were not only unknown, they were unknowable. Rather than trying to use material culture to reconstruct the lives of children, children were often used as a cautionary tale (David and Kramer 2001). Researchers would look to the ethnographic record to show how children’s behavior might skew archaeological interpretations. For example, children might move ‘items from their “proper” places or places of adult use and discard’ (Baxter 2005b, 78 my emphasis). Children’s play was also a way of explaining the unexplainable. For example, miniatures found in the archaeological record might be toys or a poorly constructed pot might have been made by a child. These were often ad hoc explanations rather than the basis for hypothesis testing and scientific inquiry.

But let’s unpack this assumption a bit. What do we mean when we say children’s play and unconventional use of material culture ‘distorted’ the archaeological record? The word ‘distortion’ has an interesting history in the study of archaeological theory. Approximately 40 years ago, there was a famous debate between Lewis Binford and Michael Schiffer, both professors of archaeology and both influential writers and thinkers in the realm of archaeological method and theory. The debate concerned the so-called ‘Pompeii premise’. The Pompeii premise is a term that was first coined by Robert Ascher in 1961 in reference to the famous archaeological site of Pompeii, where everyday life was essentially frozen in time when the volcano, Mount Vesuvius, erupted in AD 79, burying the Roman city under layers of volcanic ash and pumice. This catastrophic event provided archaeologists with an unparalleled opportunity to reconstruct the lived lives of those ancient Romans and their relationships to objects with which they engaged and the spaces they inhabited. Everything was found ‘in situ’ as we say, or ‘in place’. Unfortunately, the vast majority of archaeological sites are not, as Binford (1981) observed, ‘mini Pompeiis’. Only Cerén, a 1400-year-old Mayan village site in San Salvador that was also destroyed by a volcanic eruption, and Ozette, a 200-year-old water-logged Makah village site in Washington State, come close. All other sites (and even these ‘Pompeiis’) are subject to a variety of taphonomic processes.

Both Binford and Schiffer were well aware of the impact of taphonomic processes on an archaeologist’s ability to reconstruct the everyday lives of people in the past. Schiffer (1975) referred to them as N-transforms (natural transformation processes such as wind, earthquakes and rodents burrowing) and C-transforms (cultural transformation processes such as looting, the human re-use of building materials or humans digging a basement that breaks through a previous occupation layer). Schiffer was interested in how artifacts moved from their ‘systemic context’ (use in people’s everyday lives) to their ‘archaeological context’ (in which they are discovered and excavated by archaeologists). Further, artifacts could be in either primary context or secondary context. Artifacts in use-related primary context were recovered from the place they were acquired, made or used, while artifacts in transposed primary context were deposited by human activity outside where they were acquired, made or used, for example, middens or historic latrines that doubled as garbage dumps (Fig 1.2). Most of the objects uncovered by archaeologists at Pompeii, Cerén and Ozette were in primary use-related context.3 According to Schiffer, artifacts are in secondary context when the provenience, associations between objects and features and the sedimentary context (matrix) have been altered by natural or cultural transformational processes.

Figure 1.2: A profile wall of a latrine at Fort Wellington, Prescott, Ontario, a 19th-century military fort. This latrine was divided into three sections – one for officers, one for enlisted men and one for women and children. Not only was the latrine used as a toilet but it doubled as a garbage dump. Its water-logged/anaerobic environment facilitated the preservation not only of ceramics and children’s slate boards but of organics too, such as a delicate tablecloth and a woman’s boot. The artifacts in this location would be in transposed primary context according to Schiffer (Photo: April Nowell).

This particular debate centered around whether C-transforms ‘distorted’ the archaeological record or more precisely, the archaeologist’s ability to make inferences from static artifacts existing in the present (archaeological context) about dynamic cultural systems in the past (systemic context). For example, does sweeping up flaking debris after an episode of stone tool making and moving the debris to a midden constitute a ‘distortion’ of the archaeological record? Schiffer (2010, 40) would say it depends on your research question. You can look at sweepi...

Table of contents

• Cover
• Halftitle
• Title
• Copyright
• Contents
• Acknowledgements
• Foreword by Jane Baxter
• 1. Toward an archaeology of Paleolithic children
• 2. Birth and the Paleolithic ‘family’
• 3. Toys, burials and secret spaces
• 4. Stone tools, skill acquisition and learning a craft
• 5. Children, oral storytelling and the Paleolithic ‘arts’
• 6. Adolescence in the Ice Age
• 7. Paleolithic children as drivers of human evolution
• Appendix 1. Chronology of the Paleolithic and timeline of fossil hominins
• Appendix 2. Table of subadult fossils in the Plio-Pleistocene (perinatal–ca. 10 years)
• Appendix 3. Table of subadult fossils from the Plio-Pleistocene (ca. 10 years–20 years)
• Bibiliography