UNDERSTANDING YOUNG PEOPLE BETTER

Through the prioritisation of both osteoarchaeological and quantitative methodologies, knowledge about Roman-period childhood has moved away from a reliance on the surviving corpus of Classical writing and other kinds of evidence from the imperial centre. A significant trait of the current scholarship dealing with juvenile, and especially infant, bodies in the Roman-period northern provinces has been the tendency to emphasise their mutability as a category of being. This is probably due to the greater number of infants recovered during excavation. As Alison Moore notes, ‘[a]s a transitional being … the infant was inherently ambiguous’.Footnote ¹¹ This is an understanding based in part on the finding of many younger bodies at boundary locations such as ditches and the like, but the approach is modified here through discussion of wider evidence. It is suggested instead that juvenile bodies – both infant and child – had distinct and defined roles in funerary ritual, and that these roles changed during the period.

Such a modification is partly encouraged by Marianne Hem Eriksen's work on infant burials in Northern Europe during the first millennium b.c.e. Hem Eriksen contends that the ‘ontological reality was that they [infants] related more closely to vibrant things than humans’, an argument that helpfully points to how different past attitudes may have been to our own.Footnote ¹² Her point is that infant bodies were handled in a manner akin to objects and were utilised as such in depositional practices. This crossover between human and object deposition becomes the more appreciable when evidence for other kinds of past juvenile interment is evaluated. Although younger bodies may not always have been treated like objects, they can be utilised to make a range of statements.

For instance, when infants were placed in pots within the first-millennium c.e. middens of KwaZulu-Natal, South Africa, they acted as a material ‘testimony to the fertility of the lineage’, a lineage of which they were an equal part.Footnote ¹³ At another point, lineage may have been less important than the recognition of dead juveniles as a distinct ‘other’. In Bronze Age Ireland, small communities fixed on ‘no obvious divisions between types of children’ in death – they were recognised as a homogenous bloc, and there was a need to dissipate the ‘troubling’ fact of their death.Footnote ¹⁴ This homogenisation was in contrast to the Bronze Age in Hungary where juveniles were differentiated according to age, small children not being cremated in the Encrusted Ware culture. Elsewhere in Hungary, children were distinguished from adults and placed in the shallowest graves.Footnote ¹⁵ These brief examples underline that juvenile bodies were distinguished from adult ones, and their bodies manipulated in ways sensitive to other forms of need and understanding.

Another illustration of how juveniles were categorised differently through space and time is provided by their utilisation within the visual expression of Roman imperial ideology. Beryl Rawson highlighted how images of children came specifically to symbolise ‘imperial generosity and care for the population’ in the reign of Trajan. As part of this move, children were increasingly depicted on coins: arms outstretched towards the standing figure of the emperor.Footnote ¹⁶ Whether these ideas were ultimately pervasive or not, they circulated physically in Britain. An aureus found at Ribchester is similar to that Rawson discusses, as are other related types.Footnote ¹⁷ It is also significant that the proportion of gravestones commemorating children rose from the first to second centuries c.e. across north-western Europe.Footnote ¹⁸ Remembering children epigraphically gave families the ‘chance of recognition’. Crucially, it provided a means through which family status was expressed.Footnote ¹⁹

Outside the realm of grave monuments, wide-scale, indigenous LIA or Roman-period artistic tradition of juvenile representation is missing from Britain and indeed human representation is rare. But the absence of juvenile depiction is particularly marked in this province. Emma Durham has catalogued 1,000 metal figurines from Britannia. Only 4.5 per cent were of Cupid, that is, of classicised and mythologised depictions of the infant.Footnote ²⁰ Feeding infants were depicted, though obviously only as part of a larger figure.Footnote ²¹ As a specific form, therefore, infants were present in the corpus of handheld art, though their bodies were idealised. They were not specific to place, and many of these images were dependent on ideas stemming from outside Britain. The absence of children from most forms of visual art is conspicuous, therefore, and paralleled by their relative rarity within the burial record.

These examples highlight that many different considerations of infants and children were formed and fixed in different societies across the LIA and into the Roman period. Although these ideas were themselves subject to change across time, it does not mean that they did not have stability at certain points. This is important with regard to considering the empirical evidence from LIA to early Roman funerary contexts in what follows. The core of this paper, then, is to measure some of the extant understandings of previous scholarship against the largest empirical survey of this aspect of burial culture to date. Following the next section's description of the dataset that underpins this research is, first, an evaluation of the trends in juvenile burial, and second, an overview of the way in which juveniles were differentiated in death. These two exploratory sections provide the background to a synthesis that reviews the change in juvenile body valuation through time.

THE DATASET: ITS SIZE

Nearly 5,000 graves were evaluated for this study. It is probable that most were deposited in the period 100 b.c.e.–c.e. 200 in a study area comprising ten counties in southern England. The size of this area allows both comprehensive data collection and an opportunity for a discussion of regionalities in the archaeological record (fig. 1), and represents the most exhaustive collection of data from this region since the work of Whimster and Philpott.Footnote ²² Reports, both published and unpublished, from the vast majority of excavated sites active in the timeframe were consulted and included a mix of newer and older excavations. All graves that were feasibly created in this three-century period are included. Others might have been incorporated that were dated (slightly) differently in the primary reporting, but for the input of the information to the dataset, dates were mostly as originally established. A timeline of burials emphasises again the extent to which younger bodies were only rarely dedicated (fig. 2).

FIG. 1.

Distribution of juvenile graves across the study area.

FIG. 2.

Graph showing numbers of graves buried per decade.

Of the total of 4,817 bodies interred, 3,029 were assigned an age. The age identifications range in quality from the observations of past excavators and cataloguers to the scientific examinations of present-day osteologists.Footnote ²³ There are therefore innate disparities in the evidence quality, and the exact age-profile of the burial population is not recoverable. That said, the total number of funerary assemblages that included plausible infant or child remains stands at 564 (12 per cent of the overall number of bodies, and 19 per cent of those that have been aged). This total might itself be an under-representation of the past figures. Though neonates are more often identified as such in archaeological reports – even when other cremated or inhumed bodies are not further examined – they can be missed and are not always recovered. The demographic structure of the burial population is inherently skewed, therefore.Footnote ²⁴

The corpus is evaluated quantitatively in the next section. The fluctuation in the quantity of these kinds of burial over three centuries will be considered alongside the evidence for what kinds of settlement they were interred next to. As an aid to this analysis, sites have been divided into three types: rural, nucleated, and primary, the latter comprising both LIA formations such as Verulamium and Roman-period urbanised developments like London.Footnote ²⁵

THE DEPOSITION OF JUVENILES OVER THREE CENTURIES

Juvenile graves were a minority occurrence within the already minority rite of grave-making which was not the main option for the disposal of the dead for most of the population during the LIA and Roman period. Nor does it appear that there were any significant quantitative changes in the creation of infant- and child-containing deposits in the first two centuries c.e., despite the conquest of 43 c.e. Although the dating of Iron Age burials is undergoing redefinition, it is probable that this recent research will extend rather than limit this sense of continuation.Footnote ²⁶

During the Roman period, the deposition of known assemblages that included children in the study area never rose beyond around 30 a decade. This is something like a tenth of each decadal total (figs 2–3). Moreover, the numbers of cremated juveniles only grew in proportion to the total number of cremation graves being created and remained at a steady c. 5 per cent of that total over three centuries. Evident from this is that the imperial occupation had a negligible effect on the overall limited interest in creating physical memorials around juvenile remains.

FIG. 3.

Histogram showing numbers of juvenile grave being made per half century.

These apparent continuities, however, disguise striking elements of change. When the numbers of infants and children are compared against the totals for identifiable bodies, there is a noticeable rise in infant deposition from the middle of the first century c.e., a rise which continues into the second century (fig. 3). Part of this may be due to the large numbers of neonates recovered from recent excavations at Springhead, Kent, and Camp Ground, Cambridgeshire, but there were advancing totals in other areas as well.Footnote ²⁷ Another change was in the number of cremated juveniles placed in spaces associated with nucleated settlements, such as Baldock, Hertfordshire (fig. 4), which increased during the second century c.e.Footnote ²⁸

FIG. 4.

Histograms showing numbers of types of juvenile burial at different kinds of sites per half century.

Drilling into the data reveals other trends, showing for example that the inhumation of children and infants on physically smaller settlements – whether nucleated or rural – was more common than around Roman-period primary centres (figs. 3–4).Footnote ²⁹ Here preservation conditions are significant. In their shallow graves, infants are much more likely to be disturbed, especially on sites lying beneath modern conurbations.Footnote ³⁰ This image is reinforced by the evidence of the well-excavated cemeteries at larger centres where there are also smaller numbers of children and infants. As with the already mentioned trend that saw cremated juveniles becoming more prevalent in non-urban-associated spaces, there appear to be different practices at the primary centres compared with other forms of settlement, although the number of inhumed children found in urban cemeteries seems to grow during the second century c.e. This growth is associated with the expansion in inhumation burial more generally.

There is another aspect to this evidence: infant and child bodies were often combined with adult-aged bodies in the same grave. The origins of this (albeit again minority) practice lay in the LIA. The neonate scapula placed with a flexed female-sexed adult in a ditch at Biddenham Loop 8, Bedfordshire, and the double cremation burials at Hinxton, Cambridgeshire, are good examples.Footnote ³¹ It is a feature of the latter practice that the cremated bodies were found in the same urn, as if these bodies could not be disassociated in death. Other combinations were possible, however. For example, the differently aged bodies had been separated out into two urns in Roman-period Grave 4088, Harlington Road, Bedfordshire.Footnote ³² This suggests that bodies may not have been laid together on the pyre, or, indeed, they may have derived from different pyres.

These combination-burials point to the prevalence of curation and the storage of juvenile remains. Curation, though, is more usually encountered in prior scholarship when dealing with disarticulated human remains.Footnote ³³ Yet, with these juvenile graves this was a process which took different forms, and might include the preservation and non-cremation of deceased juveniles until another (perhaps related) individual had passed away.Footnote ³⁴ The deceased and cremated juvenile might also have been preserved until it was placed with another body in a burial. Apart from the late first century b.c.e. – for which much of the material derives from Hinxton – the level of multiple deposition hovers between a fifth and a third of the total cremated juvenile population. This is probably an under-representation. Older excavations were less likely to note the prevalence of many-bodied cremation burials. However, the relatively low number of multiple depositions as a proportion of the total cremation juvenile population suggests those multiple depositions that are identified represent deliberate practice rather than expedient use of the same pyre as an adult. Again, this evidence shows that the incorporation of juveniles into physical memorials was carefully considered and controlled.

It is probable that there were many distinct socio-cultural regions within south-eastern England in both the LIA and Roman period.Footnote ³⁵ The burial record shows significant regional variation in the treatment of juveniles. The inhumation of children sits comfortably within LIA burial traditions centred on some localities in Kent and Hertfordshire, such as Mill Hill and Baldock, respectively (see again fig. 1).Footnote ³⁶ Outside these two counties, where cremated juveniles are encountered in the LIA they were within multiple-bodied burial deposits; for example, at Hinxton, or Noak Hill, Essex.Footnote ³⁷ Many parts of the study region do not produce juvenile burials, or indeed any burials at all. When single inhumations were occasionally placed around settlement spaces in East Sussex in the last century b.c.e., or the first half of the first century c.e., only adults were interred.Footnote ³⁸ The same is largely true of Bedfordshire and Cambridgeshire.Footnote ³⁹ Even in the first 50 years post-conquest, the only inhumations of juveniles in Cambridgeshire were those placed in graves with adult-aged bodies at Duxford.Footnote ⁴⁰ The distribution maps additionally point to the extent to which larger LIA cemeteries, notably King Harry Lane in Hertfordshire, display greater variety in their buried populations.Footnote ⁴¹ The cremated children buried at this last graveyard outnumber those at any contemporary site in the study region.

Looking beyond the overarching stability in the rate of juvenile deposition across the first two centuries c.e., it appears that regional rates of deposition were transformed by Roman-period occupation. The cremation of juveniles developed particularly in Essex, focused largely, although not wholly, around the Roman-period establishment of Colonia Claudia Victricensis (Colchester). In areas that had seen no previous LIA juvenile burials, some now began to appear. This included Greater London, where, by the end of the second century c.e., the relative quantity of child inhumations interred began to set it apart from other places. But Surrey and Norfolk also saw the interment of juveniles, often associated with nucleated settlements like Staines and Scole.Footnote ⁴² Kent, meanwhile – a county that was remarkably different in terms of its burial traditions during the LIA – might appear from one perspective to have had a very steady rate of deposition (fig. 1). But these stable numbers disguise the uneven chronological and spatial distribution of juvenile cremation in this area. Rural communities appear to have laid out formalised cemetery spaces sparingly. Once they had done so, there was no prerequisite that juveniles should be included. In the instances where they were, as at Thorne, it was likely to be in a double burial with an adult.Footnote ⁴³

In Hertfordshire, juvenile cremation became more common during the second century c.e., with nucleated settlements becoming the focus for such practices. This increase is not found in Essex or Kent where levels of juvenile cremation deposition remained largely unchanged from the first century c.e. Similarly, where it was also already established in community practice, the inhumation of children remained pretty much static in most regions after the end of the first century b.c.e. This was in contrast to the rate of deposition of infant inhumations which fluctuated in number between half centuries. Even though this variable rate of deposition may be due to modern preservation conditions, as with Kent's rural cemeteries, surely it is also a product of these kinds of burial being added to practices that only occurred sporadically in the past.

This quantitative assessment has emphasised that social investment in dead juveniles was constrained, suggesting a particular set of circumstances that dictated when their graves were created. As evidence of this, it is clear that the interment of children became more significant for nucleated settlements during the period. Medium-sized sites may have given some groups enough social space to express themselves in a different mode to elsewhere. Neonate deposition increased also, but was very erratic, and may have been utilised in particular ways. Importantly, as already suggested, a frequent feature is the close association of a young body with an older one, as will again be shown in the next section.

THE ARRANGEMENT OF JUVENILE BURIALS

The above demonstrates that the data corroborate seemingly long-lasting socio-cultural stabilities from the LIA into the early Roman period in Britain.Footnote ⁴⁴ However, closer examination of the dataset suggests that continuity is not quite as overarching and consistent and that there were different scales of interest in this youngest segment of the deceased. The form and composition of juvenile graves were also highly variable, suggesting that they may have served a host of different purposes.

For example, some infant bodies were included within assemblages containing unusual combinations of object and bone.Footnote ⁴⁵ At a Cambridgeshire rural site, 31 Tunbridge Lane, Bottisham, two neonates were placed in a ditch during the early to middle part of the second century c.e.Footnote ⁴⁶ These bodies were accompanied by an exotic range of animal bone including a corvid with one and an eel and corncrake with the other. It is perhaps not coincidental that the only other eel specimen recovered from a grave in this study area was also found in a burial including a juvenile.Footnote ⁴⁷ The 31 Tunbridge Lane assemblages fit neatly within the frameworks suggested by Alison Moore and Robin Fleming, especially given that they incorporated non-domesticated animal species.Footnote ⁴⁸ These ditch-bound assemblages were deliberately created in liminal spaces, with the care evident in the unusual selection of animal bone also suggesting a particular purpose.

The use of liminal spaces for burial in this region was not reserved solely for juveniles. Bedfordshire and Cambridgeshire witnessed the burial of adult-aged corpses in ditches throughout the LIA and the early Roman period.Footnote ⁴⁹ Moreover, several recent excavations in Bedfordshire noticed that midden material was placed on top of these bodies.Footnote ⁵⁰ The choice evident in the bones included at 31 Tunbridge Lane does not support an interpretation as a midden. This evidence shows, though, how across these two counties there arguably existed an older practice surviving into the second century c.e. that was reliant on ditches, human bodies and non-normative (in modern terms) disposals of material. Even when some deceased infants were utilised by a group for a socially impactful display, the rituals employed were similar to those used in the deposition of adult remains.

This example highlights the local rootedness of much practice even when communities made the unusual decision to deposit younger bodies. We should not think, however, that there was only one way of utilising infant remains in the study area. Geographically close together, and perhaps connected with a particular routeway, there were sites where infants were placed in close association with cremation burials, the interred neonate appearing to accompany the cremated adult.Footnote ⁵¹ As already noted, there may be a longer legacy to the idea that juvenile identities could be conjoined with those of adults, given that double burials were much more common amongst the first set of LIA cremation graves which included younger bodies. This particular evolution of what appears to be the same idea is remarkable in that the actual physical twinning of infant inhumation burials with cremation graves was only a development of the second century c.e. At certain sites, the placement of an infant alongside the cremated adult was a regular occurrence. For example, single infants were buried in close proximity to nearly all of the cremations at a site in Godmanchester, Cambridgeshire, although osteological evidence to identify the sex of the cremations is lacking.Footnote ⁵² In other places, it appears to have been reserved only for some deposits. In the case of Walden Road, Essex, the neonate was buried close to the only cremation burial that included a glass cup, a rare type of accessory vessel.Footnote ⁵³

Similar kinds of tradition existed among diverse communities, therefore. It may be significant that the adult body in the neonate-accompanied Grave 73 from Walden Road was probably female. Likewise, the double LIA cremation burial with a mirror from Noak Hill, Essex, contained an adult female. The clear representation of the infant at Walden Road as both physically separate (not placed in the same grave) and physically different from an adult female (in not being cremated) highlights that there had been a shift since the LIA. No longer was it necessary to include adults and infants together; instead, it was increasingly important to make a distinction between them. The understanding and representation of small bodies in death was clearly subject to alteration through time. The evidence for the differentiation of juvenile graves below repeatedly underlines the same feature.

Another example of change within seeming consistency is the increasing presence of bangles and beads in juvenile graves from the start of the second century c.e. Some of these graves are of a possible later date, but the incorporation of these kinds of objects began earlier.Footnote ⁵⁴ Most of the time, the use of these dress accessories appears to have been limited to graveyards associated with primary centres and nucleated settlements. Although not an exclusive patterning, this suggests that the corpse was differently laid out and arranged at such places. Significantly, none of the metal bangles was melted or scorched, an indication of them having been placed with their respective cremations during the final deposition, rather than worn by the body on the pyre. There was a potential delay and process through which these bodies were given dress items in death, as if this reference to clothing was symbolic in itself and happened further along in the funerary process.

The presence of large numbers of beads in these juvenile graves was uncommon. Two cremations from Colchester only contained one glass melon bead each.Footnote ⁵⁵ One infant inhumation from the banks of Walbrook in London had been similarly interred with a single example.Footnote ⁵⁶ A stark comparison is the anklet of 330 glass beads that adorned a (possible) male-sexed adult inhumation from Plot III, Eastern London.Footnote ⁵⁷ There also appears to have been a distinction between juveniles of different ages. Infants had fewer beads, children more. No inhumed infants have been recovered with bangles; in fact, most inhumations with bangles from the study area have been aged somewhere between 1 and 12 years. There was clearly a resonance to children's bodies which drew association to such objects. Indeed, Maureen Carroll has argued that colourful collections of beads, amulets and pendants had apotropaic importance in juvenile graves across the western parts of the Roman Empire.Footnote ⁵⁸ As will be seen again, the distinction that society seems to have made between younger bodies may have begun around the age of 1. The child that wore an (apparent) ivory bangle on its lower left arm was 2 or 3 when it was buried at Cottington Road, Kent.Footnote ⁵⁹ This is also not the only life-stage that is seen. Based on present evidence, the western regions of the study area saw cremated older children of about 7 years old buried with two bangles, at least from the second century.Footnote ⁶⁰ These bangles could hint at a further stepping-stone in the life course and would have marked these deceased children out from their younger contemporaries.

Distinctions between children and infants are also apparent in relation to the presence of shoes in the grave. It is not that infants were definitely buried without such artefacts; it is just that inclusion of shoes with these kinds of body was extremely rare apart from double burials where the shoes could well have been associated with the adult. It is likely, after all, that infant feet did not require hobnailed shoes. However, there is a number of child burials that, interestingly, did include them. In a similar manner to adult burials, these shoes were not always on or near the feet in the grave. In one of London's northern burial plots, for instance, at least one shoe was placed across the upper legs of a juvenile inhumation.Footnote ⁶¹ Nearby, two shoes were placed in different positions within the grave slot.Footnote ⁶²

Yet differences between juvenile and adult burials are much easier to see than those between children and infants. Communities used a host of mechanisms through which juveniles were symbolised. On the whole, it appears that fish were as usual in juvenile burials as they were in adult ones. The evidence here is derived exclusively from more recent excavations. Yet this evidence for fish contrasts with a near-absence of cattle bones, except in instances where the juvenile was placed with an adult in a cremation grave.Footnote ⁶³ Size was perhaps a distinguishing factor. The unusualness and smallness of fish or eels were perhaps preferred for unusual sorts of funerary assemblages which centred on physically smaller bodies.

In terms of both material culture and animal remains, it appears that effort was put into creating assemblages that revolved around creating the ‘juvenile’ as a distinct concept. Clear examples include the foal placed with an infant at East Stagsden, Bedfordshire, as well as the dismembered lamb with an infant at Camp Ground.Footnote ⁶⁴ There is also the use – especially in the second century – of smaller pottery vessels in many juvenile graves. The double burial from Harlington Road mentioned previously physically articulated the different sizes of the bodies interred by placing the adult in a jar and the infant in a beaker. This makes sense in that cremating adult bodies yields larger amounts of ash and bone, but similar practices have been found elsewhere. At Wallington Road, outside Baldock, a child was placed in a miniature jar.Footnote ⁶⁵ Similarly, a child from Coggeshall, Essex, was inhumed with two petite flasks.Footnote ⁶⁶ Miniature objects had wider ritualistic value in Roman-period Britain, and were also placed in adult burials.Footnote ⁶⁷ For example, outside Canterbury, one child's cremation had a miniature jar, and so did an adult's cremation.Footnote ⁶⁸ But they are more common in juvenile burials. On the whole, the evidence shows that deliberate attempts were made to reference a juvenile's youth through placing smaller objects with them.Footnote ⁶⁹

This impression is compounded by intermittent indications that emotion may well have governed which objects were put into juvenile graves, whereas in the case of adult graves this behaviour can be less obvious. At Pepperhill, a tettine – a spouted possible milk bottle – had been placed outside the coffin.Footnote ⁷⁰ It was handmade and soot-stained. It is entirely possible that this object had been fashioned specifically for, and used by, the deceased. In this light, emotion was actively conveyed at the graveside as the bottle was positioned outside the coffin. Relatedly, a rock – a favourite flint ‘face’ – was included in the burial of a child, or perhaps teenager, from Wallington Road.Footnote ⁷¹ Another seemingly significant stone was found within a juvenile's lead coffin dating to the third century c.e. from Butt Road, Colchester.Footnote ⁷² There is evidence for sensitivity elsewhere in the Empire too – a small ceramic toy ball came out of a cremation burial from Nijmegen-Hatert in the Netherlands.Footnote ⁷³

Emotion is additionally symbolised by the use of the so-called dea nutrix – the Nurse – and other mother-type figurines in child burials. While pipeclay figurines were also given to adults in death, there is something powerfully protective in the careful placement of one of these nursing figures on the chest of a child at Baldock.Footnote ⁷⁴ A ‘mother goddess’ type was also included alongside others at Arrington, Cambridgeshire, and the inclusion of three Venuses in Grave 392, Eastern London, may be a reference to the triple goddess – maid, mother, matron – that was worshipped across Europe in the Roman period.Footnote ⁷⁵ One of these Venus figurines is smaller and of a different type to the other two, matching the relative youthfulness of the maid compared to the mother and matron.

The use of pipeclay figurines has been linked to the movement of legio XX in the first century c.e.Footnote ⁷⁶ During the second century these figurines were diffused more widely, whatever the actual origins of those who utilised them. They had far-reaching connections, since the objects came from Gaul and Germania. Famously, the ‘Child's Grave’ from Colchester contained comic figurines indicative of some form of cultural subtlety.Footnote ⁷⁷ Even if this was not the grave of a child, the signification of the triple goddess at London is indicative of something like this, too. Another example underlines the point: three of the figures at Arrington have been viewed as representing that young person's potential life course.Footnote ⁷⁸ The risus infant figure is believed to symbolise the babe, the laughing boy becomes toddler, and the spinario, the youth. Together they may have idealised the wished-for life progression of the deceased. Using a Gallic comparison, this suggestion becomes stronger if the bearded adult male with Phrygian cap also in the Arrington grave takes the symbolised life stage beyond the young, to the mature, adult.Footnote ⁷⁹ In all of these instances, cultural resonances were woven about and around the bodies of children. It was as if the circumstance of their burial gave stimulus to these kinds of display.Footnote ⁸⁰

Here it is important to note again that the dea nutrix had been encased in a wooden box and set down on the chest of the child at Baldock. At Scole, Norfolk, a wreath of box leaves and nightshade was similarly placed. Conjecturally, what might have been important in these examples was the idea of care and rejuvenation, since box is an evergreen plant. There is also a nurturing aspect to the dea nutrix, a statuette of a female suckling at least one infant.Footnote ⁸¹ The physical connection between object and chest could also have been important, pointing to the effort of communities to (re-)animate the child's heart and lungs symbolically, or at least to protect them and emphasise their importance. These ideas seem to have been significant enough to have been expressed into the mid-term of the study period. The burials from Baldock and Scole had wooden sheds built around them, for example. Possibly the graves in these places were left open to be shown to the wider community. Clearly, the burial of a child was exceptional enough in these local environments for the assemblage to deserve housing.

Baldock and Scole were sprawling roadside settlements defined by major roadways. As with the 31 Tunbridge Lane infants, the rituals employed in both cases were associated with other traditions of burial. The child at Scole was interred next to a roundhouse that was derelict or out of use – in other words, and as with several of our other instances, it was in a liminal area. At Baldock, the inhumation of children was a long-standing element of the local funerary scene. Both places were probably small enough for the making of a child grave to have been a major event with long-lasting repercussions for the community – hence the construction of a cabin around them. The hut containing Scole's child's burial became the foundation for a cremation cemetery. It is true of course that the use of infants as or within foundation deposits, and especially as foundations for buildings, is a widely known feature of the period, as at Bishop's Avenue, Kent, for instance, during the late first to mid-second century c.e.Footnote ⁸² The constructors of Scole and Baldock must have been aware of this practice when they put together the wooden buildings around the graves. Their successors effectively followed suit in using spaces close by for further funerary activity.

Communities were united in their appreciation of the atypical circumstances that governed juvenile deposition, and found creative means of emphasising the special nature of these graves. This impression is compounded by the use of gorse as fuel specifically for the pyre of younger bodies at Colchester. These burials were clustered within the cemetery, and their spatial proximity was also accentuated by a half of the same coin being put into two graves. Both were further presented with a lamp that was apparently lighted and left burning as the grave was filled.Footnote ⁸³ A similar clustering of juvenile graves occurs to the north of London in the second century c.e.Footnote ⁸⁴ Similar evidence exists elsewhere too; younger cremated bodies were grouped together in Fishergate, York, for example.Footnote ⁸⁵ In sum, there was a sharp disassociation between the adult and the juvenile. The evidence highlights that communities were responsive to the distinct pathways that juvenile deposition required.

Overall, the wealth of information now available emphasises the diversity and richness of past understandings of juvenile bodies. Though their relative paucity in the archaeological record is intriguing in itself, the form their burial took where we do have evidence is equally noteworthy (fig. 5). The deposits are distinctive enough to suggest that their buriers were concerned to set out clear claims and statements in their making, while also referencing traditions current in their area. A last set of examples confirms this point. Children were sometimes placed in graves that were quite ornate in the context of Roman-period western Europe. Geldiston, Norfolk, had a cremated child that had been buried in a glass urn and within an oak cist.Footnote ⁸⁶ Likewise, on Mersea Island, Essex, another child had been cremated and its fragments slipped into a glass jar.Footnote ⁸⁷ This burial would have lain next to an impressive burial mound that dominated the skyline of the island when viewed from the mainland opposite.Footnote ⁸⁸ If their burial was only occasional in the period, then the role juveniles had within funerary ritual, as this review of the evidence has suggested, becomes the more significant.

FIG. 5.

Table showing the likelihood of certain objects and animals being dedicated alongside juveniles.

SYNTHESIS: GENERAL PATTERNS

Accumulating evidence, including that from the dataset which underpins this paper, has captured the distinctiveness of regional manipulations of juvenile bodies. In this last section, the material will be drawn together so as to understand better the trajectory of deceased juvenile treatment during the course of the study period. Through a discussion of the changing form of grave assemblages over time, the shifting importance of juvenile remains and the roles they played will be made explicit. The view mounted here will be roughly chronological. Moving from the first century b.c.e. to the second century c.e., figs 6–7 show how infant and child bodies utilised in death were idealised.

FIG. 6.

A chronological visualisation of the ways in which infants were memorialised by their communities. The widths of the lines give an indication of the numbers involved.

FIG. 7.

A chronological visualisation of the ways in which children were memorialised by their communities. The widths of the lines give an indication of the numbers involved.

As we have seen, children formed an established part of some burial traditions in the last century of the first millennium b.c.e such as those related to the graveyards at Mill Hill, Kent, and Baldock.Footnote ⁸⁹ Clearly, in these places deceased children were accorded burial, and the placement of their bodies was given what appears to be equal precedence to that of adults. They were not only buried in double graves with adult bodies, but also granted their own graves. This was true at King Harry Lane, too. A comparison can also be made with Bronze Age Ireland. At that time, small societies attempted to mediate the ‘troubling’ nature of child death by incorporating them within their burial places. These graves formally recognised these bodies as similar to older ones in the community.Footnote ⁹⁰

However, in other regions, including those where cremation graves were beginning to occur, there was no such flexibility. Juveniles were interred with adults, or not at all. These double burials may have commemorated a particular event, perhaps even death in childbirth. The use of juvenile bodies as accompaniments to adult ones certainly is well known in (pre)historic contexts where they reinforced familial and social linkages.Footnote ⁹¹ However, certain groups were stopping short of making child- or infant-only graves in cemeteries in Britain. In spaces like Cambridgeshire, or Essex, it seems unlikely that juveniles were viewed as akin to adults. Finally, inhumed infants were uncommon finds in the first century b.c.e. across the study area. At sites where they do appear, their singularity was demonstrated through their inclusion alongside broken objects. The neonates buried together in a pit at Waterstone Park, Kent, had been put into a feature apparently originally used for grain storage. A half of a sliced beaker came from the same pit.Footnote ⁹² Because of the performative damage wrought on these vessels, these graves might emphasise and symbolise the loss some felt. Alternatively, it is possible that these very deliberate assemblages were physical memorials to acts of closure.

Based on present dating, the traditions so far described also existed in the first half of the first century c.e. where more communities started to include juveniles amongst their other memorialised dead. For example, a grave at a small cemetery that was arguably established in this period, Luton Road, Bedfordshire, contained a cremated child.Footnote ⁹³ At Cottington Hill, Kent, a first-century c.e. juvenile was placed in a grave alongside one which contained an adult male aged over 50.Footnote ⁹⁴ Both bodies were tightly shrouded. This gradual spread of practices leading to child commemoration was not normally mirrored in infant deposition, however. When infants appear in the archaeological record for the first half of the first century c.e., their extreme alterity was generally reinforced. The East Stagsden infant was placed with a foal and a pot that had been chopped into halves; a neonate from Biddenham Loop 8 was placed with an incomplete jar.Footnote ⁹⁵ Yet, at the same time, some infants were integrated into the wider burial population in this century. Significantly, King Harry Lane's Grave 471 held an infant in a bowl-urn.Footnote ⁹⁶ This may suggest that the longer-lasting incorporation of children at certain key places in the LIA allowed communities to relax, or even to challenge, the consensus view that infants could not be granted a similar burial rite to other children.

In effect, the Claudian invasion, though not overly affecting numbers of juveniles placed in the ground, coincides with the increasing diversification found in the evidence up to this date. Not only did the kind of communities making juvenile burials continue to broaden, but infants became much more prominent as part of funerary practices that took place in formal cemeteries. It was in the middle to late first century c.e. that Harlington Road received its first cremated children.Footnote ⁹⁷ In Kings Wood, Greater London, a cremated infant was found within a small cremation cluster.Footnote ⁹⁸ Flexed inhumation graves were arranged down the line of a silted ditch at High House, Essex. Two of these were for younger bodies.Footnote ⁹⁹

The increase in numbers of sites that were utilised for burial from the middle of the first century c.e. suggests that localised and smaller groups were adopting funerary practices which recognised the physicality of the dead and gave the departed a tangible, longer-lasting presence in their local landscapes. It can be imagined that at the stage at which these groups started to bury their dead locally for the first time, they also started to reconfigure their responses to different kinds of body, with consequences for juvenile identity. This may have allowed previously hesitant groups finally to integrate younger bodies alongside their other dead in ways that did not dwell on their otherness.

But the situation was not the same everywhere. The nucleated settlements of Hertfordshire continued to draw significant numbers of bodies of varying age. And the populations of Colchester and London did establish difference between their dead of various ages. For example, a juvenile from one of Colchester's western graveyards had been interred in the kind of beaker often associated with the Rhineland, while also having a tettine – an unusual artefact – as an accessory.Footnote ¹⁰⁰ In addition, infants were increasingly placed under, or within, domestic settings in urbanised spaces. They were suddenly much more symbolically foundational; performative destruction is less abundantly evident in the assemblages that they were a part of.

This shift in the use of bodies points to the development of new relationships to the dead. There was a possible move away from juveniles being understood in death as similar to objects once important to the community, that is, objects which were damaged and rendered useless as part of depositional acts of conscious destruction, but that implied wealth, status, or group-strength.Footnote ¹⁰¹ Instead, it is possible that infants in death were becoming significant in affirming and marking community foundation, continuation and preservation. The formation of physical, lived space around these small bodies continued to mark their importance in everyday lives.

All of these first-century practices continue to be broadly evident in the second century c.e. archaeological record. Around this time, neonates were placed in and around the features that supported the structures of the riverine port at Camp Ground. There are instances of juveniles being central to foundational rituals, therefore. But exceptionality is evident, too. For instance, a cremated older infant filled a glass bottle that had been buried in a box in what was the only grave of its kind in that area of Colchester's cemeteries. Double burials continued to be made at many other locations.Footnote ¹⁰². They could be significantly different to the ordinary run of burials, and continue to serve as grave accompaniments, perhaps for the reinforcement of familial ties. There were undoubtedly shifts in practice also. Inhumed infants were set down alongside cremation graves at some burial places. Although it is not clear which interment happened first, in these cases neonates were useful in extending social linkages. Their displacement from the actual cremation, and movement to a place alongside the grave, shows how their distinct body was valued in its own right, in a way that might have to do with social status. At St Ann's Lane, Godmanchester, there is another example of this partnership being created, one where the significance of the burial itself (placed in a casket, not the ordinary jar), was again underlined through the placement of an infant.Footnote ¹⁰³

During the second century c.e., the growing number of places playing host to juvenile bodies could have led some groups to seek other means through which their children might be distinguished. My qualitative assessment earlier touched on a group of graves that paired children older than neonatal with pipeclay figurines in burials, generating complicated references. The 9-month-old juvenile from Arrington had – because of hydrocephalus – a skull resembling in size that of a 4-year-old.Footnote ¹⁰⁴ Apart from Grave 392 in Eastern London, itself sited in a wealthier plot within the wider cemetery, the placement of this kind of burial often occurred at smaller settlements where such a dedication must have had greater prominence. There is an overlap between graves laid out in this manner and those with dress accessories as they often occurred in similar places. The delineation of such bodies may have played into particular forms of display. Social distinguishability was significant in such instances: at Mersea Island, the aforementioned child was englassed next to a barrow mound.

John Pearce has outlined how a set of graves in north-western Europe came to signal the humanitas of the elites who dedicated them from the second century c.e. In his assessment, this was a substantial move away from commensal practices in the LIA that aimed at establishing the nobility's liberality, towards practices which were more concerned with exclusivity, or which reinforced a Mediterranean-style cultural praxis.Footnote ¹⁰⁵ In the same century, studies show that children were also becoming more significant in imperial iconography, a development matched by a greater epigraphic presence for children. Empire-wide, there was a further shift away from dedications made in city graveyards. Instead, elites became interested in burying their dead in spaces directly associated with their estates.Footnote ¹⁰⁶ The British evidence, especially if some of these figurine-graves are included, could fit this broader patterning. Integrating children into particular funerary practices helped certain groups to forefront the child's – and their own – exceptional status.

The increasing use of nucleated settlements for juvenile burials was apparently linked to these trends. Places like Baldock saw regular juvenile deposition over the course of two centuries. But the fact that second-century children were buried in new numbers across places of this size points to a new familiarity and acceptance of smaller bodies into the cemeteries of these communities. On one level, this was a trend set by urban centres, where children's numbers increased dramatically. But wider social factors should also be taken into account. Communities coming together around these nucleated nodes may have used the placement of children – often traditionally marginalised in death – to mark out the newness of this urban and social space. Furthermore, as with the use of dress accessories and figurines, this placement allowed some groups to emphasise their social prominence. Through creating difference between those they buried and those buried by others, groups may have established themselves in the local social hierarchy by referencing practices found elsewhere in occupied north-western Europe.

An additional significant development in this period was the spreading use of ditches for burial, as with the infants at Tunbridge Lane. This kind of deposition has historical antecedents and was not just found in rural locations in the second century. The cemeteries of Southwark, London, demonstrate similar depositions. The interment of infants on the edge of a formal graveyard again suggests that marginality was key to signifying meaning.Footnote ¹⁰⁷ As in this instance, it might be that communities accentuated through liminality the potentially troublesome or unwelcome nature of early death. One of those placed in such a position at Southwark, BL11, was recovered with a curated, ostensibly heirloom, jar. The curious mix of remains at Tunbridge Lane and this antique vessel at Southwark make clear that destruction was not as important to the ritual surrounding these acts, unlike in earlier centuries. What may have been of more interest to the dedicatees was the envelopment of this young section of the deceased into rituals aiding community growth and longevity.

A last intriguing grave within the same cemetery at Southwark shows that the centrality of lineage may have continued to be important. An apparently flexed female-sexed inhumation was cut by a supine male-sexed young adult which itself had an infant laid on its stomach and a child at its feet.Footnote ¹⁰⁸ The repeated reopening or readjustment of a particular burial speaks to ideas of lineage-creation, while also reflecting processes of remembering. The infant placed on the belly area of the youngish adult was around 9 months old. It has been noted that the Arrington child was also 9 months in age. From this it is possible that, by the second century c.e., a period around the child's first birthday was a stepping-stone in the life course and prompted acts of inclusion. It led communities to commemorate younger bodies in ways less concerned with alterity than with inclusion.