A Child Burial from Kerch: Mortuary Practices and Approaches to Child Mortality in the North Pontic Region between the 4th Century BCE and the 1st/2nd Century CE

Abstract

This article discusses a poorly studied child elite burial discovered in 1953 at the necropolis of Panticapaeum, situated near the modern city of Kerch, Crimea. A reassessment of previous research is urgently needed since it did not offer an analysis of Bosporan society from the perspective of childhood studies in general and local approaches to child mortality in particular. This fresh approach sheds new light on social structures and transformations within the northern Black Sea region. A broad chronological and geographical perspective is provided in order to detect changing mortuary rituals regarding deceased children in relation to shifting socio-political situations among North Pontic Greek and non-Greek societies. A survey of current social interpretations concerning the (in)visibility of children in the mortuary customs, particularly between the 4th century BCE and the 1st/2nd century CE, is followed by a detailed description of the history of research in the Panticapaeum necropolis. A comprehensive analysis of the grave goods that accompanied the deceased child is also provided. The discussed material suggests that a new form of elite self-representation, expressed through mortuary rites, appeared around the turn of the first millennium. This included a different approach to deceased children, whose ascribed status and expected, yet unfulfilled, social roles were frequently displayed by the family through the funerary ceremony.

1. Introduction

In October 1953, an ancient child burial was accidentally discovered on the territory of the Kerch powerboat-fishing station by an excavator operator named A.I. Sobolev and his assistant A.I. Kosyakin. A short report concerning the discovery, which was accompanied by black and white photographs of the finds, was published in 1959 by the archaeologist, Larisa Chuistova (Chuistova 1959, pp. 239–45). Based on the chronology of the funeral assemblage, the burial was dated to the 1st/2nd century CE. Hitherto, the burial has never undergone closer examination either in relation to other recorded child graves from this period or with regard to the uniqueness of the finds that display a notable cross-cultural character. The mixed Greek-“Barbarian” character of the assemblage suggests that the meaning behind the use of specific objects in funerary rites was clearly not ethnically oriented, which seems to be a frequent feature in funerary contexts of the northern Black Sea region.

This paper aims to build on the aforementioned report by providing a more detailed analysis of the archaeological material discovered in the grave and reassessing the chronology put forward by Chuistova. A catalog of the finds complemented by high-quality color photographs will also be presented. The burial will be discussed in a broader chronological and geographical context, which will highlight the changes in mortuary practices and approaches to child mortality that were adopted in the region in question in both Greek and non-Greek cultural milieus. The late 4th century BCE represents the starting point for the discussion on the visibility of child burials in relation to socio-political situations within societies outside the world of the Greek polis. During this period a visible change can be observed regarding the proportion of inhumed children and adults in North Pontic burial mounds, which will be discussed in more detail further on in this paper. This alone demonstrates that ancient attitudes towards children, as well as their social significance and role in society, varied over time. Changes may have been caused by economic, social, or political factors, which in turn may have influenced not only the display of power and status of the deceased and their family but also the ideas regarding the societal importance of children and their unfulfilled social roles.

Most of the objects from the funerary assemblage in question are stored in the Treasury of the National Museum of the History of Ukraine (NMHU Treasury), which allows for the reproduction of photographs of the gold objects found in the grave. It is important to note that XRF analysis of the enamel and metal from several items has only partially been completed due to the evacuation of the museum’s collection during the Russian invasion of Ukraine in 2022. Nevertheless, the results obtained so far have been included in this study with the caveat that further analysis will be necessary to complete the data once the gold objects from the grave are returned to the NMHU Treasury after the end of the war. Unfortunately, the current location of the glass jug (Chuistova 1959, p. 242, Fig. 9; item no. 11 in the catalog below) has not been ascertained. It is possible, however, that the object is still stored somewhere in the Kerch Museum, which housed the entire burial assemblage up until 1964 when the gold objects were transferred to the newly opened Museum of Historical Treasures of Ukraine in Kyiv.

Additionally, it needs to be explained that the word “Barbarian” used in this study does not refer to a clear-cut division between the Greeks and Others (Scythians, Sarmatians, and other neighboring non-Greek populations), which was rooted in the culture-history approach that was popular for many decades during the 20th century.1 Seeking clear ethnic markers in the archaeological material of the North Pontic region has been proven on many occasions to be futile.2 Moreover, the terminology that is used to describe ancient non-Greek populations in Crimea has repeatedly been proven to be based on modern concepts.3 The terms Greek, non-Greek/Barbarian, Sarmatian, and Scythian will be used in this paper to express differences in traditions, cultural milieus, and place of habitation (the Greek polis and its chora vs. the steppe or territories remaining outside the administrative system of the Greek polis) rather than ethnic identity.

2. The (In)visibility of Children

The growing awareness of archaeologists of the importance and the role of children in past societies is evident in a number of publications that have appeared during the last several decades regarding their visibility in archaeological material.4 The archaeology of childhood, and its ascribed status as seen in archaeological material, has been extensively explored by authors such as Gillian Shepherd and Jane Baxter (Shepherd 2007, 2021; Baxter 2005). Such studies devoted to the Black Sea “colonial” world would be particularly welcome and this paper aims to contribute to the discussion of the importance and visibility of children in ancient societies beyond the Mediterranean. It is noteworthy that the numerous recent studies devoted to childhood in ancient history and classical archaeology tend to focus on text and images rather than bioarchaeology, which seems to be a prevailing trend in current research more broadly. As an example, in (Beaumont et al. 2021) only three out of forty-one studies are devoted to bioarchaeology whereas in (Cohen and Rutter 2007) only one bioarchaeological study is provided (see also Aasgaard et al. 2018; Beaumont 2012).

The reason why children remained for many years beyond the scope of mainstream archaeology was undoubtedly due to the fact that child burials are often underrepresented in ancient cemeteries, which is a general tendency in the Greek world. This trend is also noticeable in Black Sea necropoleis, where the recorded number of child graves does not always correspond to estimated rates of child mortality in past societies, which is believed to be between 34 and 40% for children under 10 years of age (Ērkšķe 2020, p. 162 with further literature). For instance, there are only 12 non-adult burials recorded in flat graves at the Olbian necropolis dating to the late 4th and 3rd century BCE (Papanova 2006, p. 82). It should be pointed out, however, that ascertaining the exact number of adult burials from this period is difficult since Papanova only mentions the discovery of “a great number” of graves dating to the 4th and 3rd centuries BCE during the excavations conducted in 1905–1906 and 1956, and that the Hellenistic necropolis covered an area of approximately 500 ha (Papanova 2006, p. 71). Additionally, she mentions that children were also buried in niche graves; however, she does not specify the number of such depositions, which appear to have been especially frequent during the early Hellenistic period (Papanova 2006, p. 92).

The visibility of child burials in the archaeological material varies considerably over time and space, which may have been caused by changing social conditions (Shepherd 2018, p. 521) and most probably, changing attitudes towards childhood itself. It has been observed that economic and political changes could have had an impact on mortuary rites, an example of which is provided by Olbia. As a consequence of the crisis that the city faced during the 3rd century BCE, the number of earthen crypts decreased in favor of rock crypts that could be constructed using debris from the city (Papanova 2006, p. 106).

The high number of perinatal and infant deaths in antiquity, that often stands in stark contrast with the relative scarcity of archaeological evidence, suggests that deceased children who did not receive a formal burial must have been disposed of in a different way, which is now undetectable in the archaeological record. As an example, the evidence from Athens discussed by Morris demonstrates that less than 10% of the burials recorded in Kerameikos between ca. 925–725 BCE belonged to infants and children, whereas between 725 and 700 BCE this figure increases to 53%; the same pattern is also visible throughout Attica and Argos (Morris 1992, pp. 78–80). The shift in the way children were treated after death is readily apparent and cannot be simply explained by factors such as differences in the speed of child bone decay or the rapid growth of the population from 725 BCE onwards.

Additionally, a 2nd-century BCE deposit of the bones of over 450 infants, neonates, and late-term fetuses, discovered in the so-called Agora Bone Well in Athens, demonstrates that at least in certain periods very young individuals were disposed of beyond the cemetery. The bone analysis has demonstrated that the majority of the deceased infants died of natural causes at or shortly after birth, which rules out other scenarios for the deposit such as an epidemic, famine, human sacrifice, or selected infanticide. It has been assumed that the individuals were buried in the well without receiving conventional funerary rites due to their young age and/or low social status (Liston et al. 2018, pp. 1, 138–40).5

2.1. The Word of Greek Apoikiai

The visibility of children in Black Sea Greek necropoleis also varies depending on the given period and the area of settlement. This diversity of mortuary practices appears to be a general phenomenon throughout the Greek world, including the mainland and Greek apoikiai (Greek “colonies”).6 These practices may have reflected differences in cults or religious beliefs that were popular in a specific polis at a certain time, especially in the case of apoikiai where the cultivated practices were an amalgam of traditions brought from the mother city and other regions of the Greek world from where the settlers originated (Sorokina and Sudarev 2000, pp. 193, 199). An example of this is the practice of child enchytrismoi (burials in clay containers). Burials of this type have been recorded on 18 occasions in Archaic and Classical Olbia even though this tradition does not seem to have its roots in Olbias’s mother city, Miletos (Sorokina and Sudarev 2000, p. 195). The popularity of child enchytrismoi is visible at the settlement of Berezan (Borysthenes) where 150 such burials (which constitutes ca. 31% of all child burials) have been recorded.7 Over the years inhabitants of Berezan moved to the larger urban center of Olbia, which resulted in the complete abandonment of their settlement by the end of the 5th century BCE. Thus it is possible that these migrants introduced the practice of child enchytrismoi to Olbia.8

Even though it is difficult to identify a direct link between a specific cult and a change in mortuary practices performed in the Black Sea region, it should not be excluded that such a link may have existed. The popularity of a particular cult during a certain period can be observed in Olbia during the late 6th and 5th centuries BCE, when the Orphic-Dionysiac cult was attested in the city (Guldager Bilde 2008, p. 30; Porucznik 2021a, p. 54). The cult was associated with transcendence and utopian thinking and may have played an important role in the development of group identity among settlers as a response to the stress associated with the process of migration (Guldager Bilde 2008, p. 33; Porucznik 2021a, p. 109).

Another frequent funerary tradition that appears in Black Sea Greek necropoleis is constituted by child burials that follow the same pattern as adult burials, including cremation. Deceased children are usually accompanied by the same types of grave goods as adults. This tradition seems to have been especially popular in Black Sea Milesian apoikiai such as Olbia, Apollonia, Tomis, Istros, and Odessos on the western Black Sea coast (Sorokina and Sudarev 2000, p. 198), as well as in Roman Chersonesus (Sviridov 2023, p. 336). It is possible to assume, as Sorokina and Sudarev argue, that this tradition may have been connected with religious beliefs, according to which the soul of a child was perceived in the same way as that of an adult, requiring, therefore, the same funerary rituals (Sorokina and Sudarev 2000, p. 198).

It has been pointed out that child burials were located in a specific part of the necropolis of Olbia during certain periods.9 During the archaic period, as Petersen argues, lower-status children were interred in a reserved burial site (Sector И) whereas higher-status children were buried in family plots, which demonstrates that criteria such as age may have overlapped with other aspects such as wealth or the social role applied to the deceased by the family (Petersen 2010, pp. 60, 68). Furthermore, the evidence from Panskoye I, a rural settlement of the so-called distant chora of Olbia situated in north-west Crimea and dated to 400–270 BCE (Stolba and Rogov 2012, pp. 40, 56, 73–74), demonstrates that attitudes to children and their role in society may have also depended on a given lifestyle. Kurgan burials of both children and women are well represented in the archaeological material, which may suggest that their social role in rural societies was more important than in the more male-oriented cities (Petersen 2010, p. 197). Such observations regarding gender and age relations in Black Sea communities suggest that, firstly, there was a difference in the way that children from urban centers were treated compared to those from rural backgrounds, and secondly, there might have been a disproportion between these two zones based on gender. Both issues undoubtedly deserve further analysis.

2.2. Preservation of Child Bones

The preservation of infant and child bones is a problematic issue that impedes the visibility of children in the archaeological material. Child remains, both inhumed and cremated, are frequently difficult to interpret in terms of gender, age, and the original position of the body in the grave. Factors such as acid soil are known to prevent the preservation of bones (Carroll 2011, p. 109). The natural conditions of the North Pontic region in antiquity do not seem to be favorable for the preservation of child bones. Moreover, as Polin argues in his study of the Gaimanova Mohyla, in the chambers of Scythian catacombs that had access to air, child skeletons decompose completely, even if they were isolated from the soil. Therefore, a careful and thorough exploration needs to be conducted at a given site in order to properly identify skeletal remains, a fact that was not taken into consideration during Soviet times (Bidzilya and Polin 2012, p. 147 note 60). This may have also been the case with the child grave from Kerch as Chuistova mentions that “the skeleton was almost completely decayed”; however, she provides no further details regarding the possible presence of any organic material at the site (that could potentially be analyzed today) (Chuistova 1959, p. 239).

This demonstrates how little interest archaeologists had at the time regarding more complex analyses of the age and gender of the deceased child, focusing more on grave goods and their value.10 The lack of remains also makes it impossible today to identify the biological sex of the deceased; however, it is generally difficult to ascertain the sex of pre-pubescent individuals at the best of times. The funerary assemblage may only suggest the gender of the deceased, which may or may not correspond with their biological sex. Additionally, the expression of gender is not only culturally determined but it also, in certain instances, may not be clearly defined by the funerary assemblage. Indeed, the grave goods discovered in the child burial from Kerch do not appear to be gender specific and thus cannot be used to determine whether the buried child was male or female. While DNA analysis also shows potential in determining the biological sex of deceased sub-adults, the lack of bone remains in the grave does not allow such a method to be applied.

2.3. The Age and Ascribed Status

Identifying the age of individuals whose skeletal remains have not survived is extremely difficult and usually based on the size of the grave and/or the objects or elements of clothing that accompanied the deceased. According to Chuistova, the Kerch burial was arranged inside a stone cist while the dimensions of the grave were 0.48 m × 0.89 m with a depth of 0.6 m, which allowed the archaeologists to interpret it as a child burial (Chuistova 1959, p. 239). One of the objects discovered in the grave appears to be a baby rattle, which may suggest a very young age of the deceased, presumably 2–3 years of age (?) (item no. 4 in the catalog below). However, it cannot be excluded that the child was older and placed in the grave in a contracted position,11 which would make it even more difficult to identify its age.12

One also needs to bear in mind that the division of childhood into several periods was by no means uniform throughout antiquity, either in Greek or in other cultural milieus.13 Moreover, the perception of age is relative and may differ depending on whether one applies purely biological criteria or a more social perspective. A given system of division applied by society undoubtedly had an impact on attitudes to child death (especially of perinatal, newborn, and infant children) by determining the moment in which a child becomes part of society and consequently, becomes “visible” in the funerary context. However, there are recorded cases in which inherited social status and wealth determined the treatment of a child after death rather than age.14 Such an applied social status seems to be visible in upper-class child burials in the Mausoleum of Scythian Neapolis. From the late 2nd century BCE until the end of the 1st century BCE, the Mausoleum functioned as a necropolis for the city elites (who were interred there during this period rather than in the cemetery where the rest of the population was buried). In several cases, the deceased children were equipped with gold face covers. Such objects were made specifically for funerary purposes and they frequently appear in rich burials in both Greek and non-Greek cultural milieus between the 2nd century BCE and 3rd century CE.15 As a rule, face covers were present in adult male burials, which suggests that in this cultural milieu, the covers may have been used to indicate the male identity of members of the local elite. Consequently, the face covers that accompanied the deceased children may have served to impose such an identity on deceased underage individuals (Porucznik 2021b, pp. 866–69).

The definition of elites in this paper has been adopted from Mordvintseva’s comparative analysis of Crimean burial assemblages dating between the 3rd century BCE and the 3rd century CE, in which the so-called prestige objects are identified as markers of power and status (Mordvintseva 2020, pp. 30–32).

2.4. Outside the Greek Polis

A similar example of the manipulation of “adult” objects in funerary assemblages can be found in a 4th century BCE Scythian burial of a noble child (aged 2–3 years) from Tovsta Mohyla (Tolstaya Mogila) (Mozolevs’kiy 1979, pp. 110–11). Interestingly, the 4th and 3rd centuries BCE represents the period in which infants and older children are visible in Scythian burials, while they are absent in older kurgans (Kokorina 2023, p. 28). The deceased child from Tovsta Mohyla, who most probably belonged to the royal family, was accompanied by a small “child” rhyton, a small goblet, and a bow that were made specifically for the funeral. Also, the child’s clothing was adorned with smaller versions of gold decorations that were normally used on adult clothing (Kokorina 2023, pp. 30–31; Klochko 2019, pp. 667–70; Mozolevs’kiy 1979, p. 142). Kokorina’s thorough analysis of the semiotics represented through the funerary assemblage has allowed her to reconstruct the social position of the child within Scythian society, a position that was undoubtedly inherited (Kokorina 2023, p. 29 with Tab. 1; pp. 32–39). The gender of the child is not clearly defined by the funerary inventory, which, as Kokorina argues, may have been caused by the fact that the deceased had yet to undergo initiation into the adult world, which would take between the ages of 12 and 15 (Kokorina 2023, p. 36). Additionally, Klochko’s analysis of the burial inventories (including decorative elements of child costumes, earrings, necklaces, bracelets, and rings) from Scythian child graves of that period demonstrates that the combination of clothing and ornaments that accompanied deceased children had a strong symbolic meaning. The objects had apotropaic functions, but they were also used to display the inherited social status and a specific age group to which the deceased may have belonged (Klochko 2019, pp. 665–75).

Child burials have also been recorded in non-elite kurgans, mainly in those located in the “Vodovod” and ‘Sad’ groups near Hlinaia (Glinoye) on the left bank of the Lower Dniester dating from the late 4th to the 2nd century BCE. Over half of the recorded kurgans from this region (29 out of 55) included child burials, 18 of which contained exclusively child skeletons whilst three appear to have been child cenotaphs.16 The reason behind such a change in the funerary rites concerning deceased underage individuals among the steppe population at the turn of the 4th century and the 3rd century BCE could be explained by the change in Scythian lifestyle from nomadic and semi-nomadic to predominantly sedentary, which in turn had an impact on mortuary rites (Sinika et al. 2018a, p. 70). It should not be excluded that this change in lifestyle impacted Scythian attitudes towards children, who were active members of society and whose social roles (both acquired and assigned) started to be displayed after death. Notably, the importance of children as active and productive members of the community has been proven on several occasions in ethnographic studies of mobile pastoral societies, including southern Mongolia (Michelet 2016; Torimiro et al. 2003).

The child burial from Kerch is most probably connected with and/or influenced by the “Barbarian” cultural milieu (including the Late Scythians), which begs the question of whether there were also any social or political changes in the region that could have impacted mortuary rites. Importantly, as will be discussed below, excavations of the necropolis at Glinishche have revealed that the burial was located in an area in which other child burials have been discovered, which points to increased visibility of children during that period. Mordvintseva’s analysis of markers of prestige in funerary assemblages reveals that there was indeed a change in the socio-political structure of the Crimea at the turn of the 1st century BCE/1st century CE (Mordvintseva 2017, p. 193). Consequently, during the 1st and 2nd centuries, CE Crimean cemeteries such as those of Scythian Neapolis, Ust’-Al’ma, Zavetnoe and Vilino started to resemble elite necropoleis, which points to the existence of more than one non-Greek center of power in the region, unlike in the previous period when Scythian Neapolis seems to have been the only center of political power in the Crimea (Mordvintseva 2017, p. 194). This suggests that local elites transformed their means of self-representation and the display of power and wealth through mortuary rites. The evidence from Ust’-Al’ma (1st–3rd centuries CE) demonstrates a considerable number of child graves17, among which a separate category constitutes the so-called “double” burials that include the burial of a child and an adult, usually a woman. Analogies have been recorded at the cemetery in Scythian Neapolis and Zavetnoe (Puzdrovskiy and Medvedev 2015, p. 254). A separate child sector is also visible at the non-elite cemetery at Opushki near Simferopol dating to the 1st century BCE–3rd century CE (Stoyanova 2012, p. 6, Fig. 2).

3. The Archaeological Context of the Child Grave from Kerch

As mentioned above, the child burial in question was discovered on the site of the Kerch powerboat-fishing station in October 1953. Chuistova mentions that the grave was found accidentally by two workmen who were digging a trench for a water pipe. Unfortunately, when the archaeological staff arrived at the grave, they discovered that it was already damaged. Since the finds from the burial had to be “removed from a pile of mixed soil”, it was impossible to ascertain the skeleton’s position and finds in situ (Chuistova 1959, p. 239).

The burial was located in a stone cist made from four limestone slabs set on the edge, covered with another slab. The child’s skeleton was almost wholly decayed, and it was most likely laid with its head to the east. Fragments of thin wooden planks suggest that the grave may have had wooden decking (Chuistova 1959, p. 239). The burial assemblage included a torque, elements of a necklace, a bracelet, a ring, and a glass jug.

Based on the location of the grave, it is assumed that it was situated within the boundaries of a 1st/2nd century CE cemetery that served the city of Panticapaeum in the Glinishche section of the modern city of Kerch (Chuistova 1959, p. 242)18 and was explored during the 19th and early 20th centuries. The boundaries of the Panticapaeum necropolis were ascertained at the beginning of the 20th century during excavations led by Vladislav Shkorpil, the published results of which were used by Michail Rostovtzeff in his description of the necropolis (Rostovtsev 1925, p. 215). Later on, the topography and planimetry of the necropolis were refined and analyzed in detail by Galina Tsvetaeva (Tsvetaeva 1951). In fact, these boundaries are only nominal, since the main part of the necropolis was explored in the late 19th and early 20th centuries, a time when the approach to excavations was barely scientific and resembled treasure hunting. Extensive and systematic excavations were either not carried out at all or (as in the case of the Shkorpil’s excavations) they were not recorded on a map. A topographic plan of the excavations or drawings of the excavated graves were also not prepared. In addition, during the second half of the 19th century (after the discovery of the first burials) the necropolis was regularly plundered by the local population, which reached catastrophic proportions.

Another factor that affects the current state of research involves the previous practice of transferring finds from archaeological sites to central imperial museums or to private collections, which was still widespread at the beginning of the 20th century. Consequently, most of the archaeological material from the Panticapaeum necropolis was relocated; only small quantities of “doublet” finds were allowed to be kept in the local museum (Bykovskaya 2004, pp. 503–4). As a result, the artifacts discovered at this rich Black Sea necropolis have been scattered across several museums and are only partially preserved today. These museums include the Hermitage, the Kerch Museum, and the State Historical Museum in Moscow, whilst the Staatliche Museen of Berlin and the British Museum hold items labeled as being “from Kerch”, which also possibly come from the Panticapaeum necropolis. Unfortunately, the finds from the burial ground have never been published as a comprehensive list of materials from a single archaeological site.19 Analyzing these materials today is also difficult since a verbal description is often the only basis for dating the excavated burials. Moreover, the descriptions sometimes appear vague, and certain details remain incomprehensible to the modern reader. One can only assume how much information has been lost due to such undetailed and inaccurate descriptions.

The total number of excavated burials dating to the 1st–3rd centuries CE is enormous: more than a thousand have been opened since the beginning of the 20th century, the majority of which date to the 1st and 2nd centuries CE (Maslennikov 1990, p. 20). Most of the recorded tombstone stelai that include reliefs and inscriptions from this period have been found in the territory of Glinishche. These burials can be associated with the local aristocracy as well as military and commercial elites, together with members of their families (Tsvetaeva 1951, p. 81). Marble tombstones with representations of Greeks from Asia Minor, who were apparently engaged in trade and military affairs in the Bosporus, have also been found. Interestingly, these individuals were depicted wearing Greek clothing, unlike the local inhabitants who were often portrayed on stelai in local, non-Greek attire, although they bore Greek names (Tsvetaeva 1951, pp. 83–84). These representations undoubtedly draw attention to the mixed cross-cultural character of the necropolis.

Local treasure hunters20 claimed that there was once a burial mound at the site where the grave in question was discovered. Therefore, Chuistova suggested that the child may have been interred inside this tumulus as there are analogous burials recorded at the necropolis of Panticapaeum, including in the territory of the so-called Sobachyi kurgan (Shkorpil 1911, pp. 90–91).

Regarding the funerary rite, inhumations were predominant in the 1st and 2nd centuries CE. These could be ordinary flat graves without covering, flat graves covered with stone slabs, flat graves with tiles21, flat graves covered with wooden boards or beams, niche graves, stone cists, or earthen/stone crypts. The deceased could be placed on decking, grass or laurel leaves as well as in a coffin or a sarcophagus.22

The diverse characteristics of the necropolis are also reflected in child burials, which were located in all types of graves. Children have been recorded in individual graves (which constitute the vast majority) and in burials alongside adults. The fact that some burials were devoid of grave goods whilst others, including the child burial in question, contained an exceptionally rich assortment of such objects suggests that the deceased most probably belonged to various social classes. The accompanying burial inventory of child graves is also diverse, similar to adult burials recorded at this necropolis. Grave goods that are frequently recorded include different types of glass and clay vessels (which are present in almost every grave), terracotta figurines, coins, astragals, beads (that are possibly the second most common find after pottery in child graves from both parts of the cemetery at Glinishche and Mount Mithridates), and metal jewelry.

It is difficult to precisely ascertain the percentage of child graves at Glinishche. During the period of the most intensive and systematic excavations at the site between 1902 and 1903, when, respectively, 336 (OAK 1904, p. 47) and 217 burials (OAK 1906, p. 49) were excavated, child burials accounted for 57 in 1902 (Shkorpil 1904) and 56 in 1903 (Shkorpil 1905). A large concentration of child graves was located “on the shore of the bay”, which is apparently in the same area where the child burial in question was discovered in 1953. Vladislav Shkorpil alone examined about 50 child burials in this area.23

4. The Grave Goods

The cultural amalgamation that was present in the North Pontic region seems to be reflected in the funeral assemblage itself. Importantly, as we argued above, objects cannot indicate the ethnicity of the deceased. On the contrary, the appearance of specific objects in graves often reflects their circulation between different cultural milieus, as well as their development and conceptual change depending on a given context.24 The burial from Kerch contained grave goods that are frequently recorded in child burials. The burial assemblage includes the following items:

• A torque made of twisted wires that form an open-worked “Herakles knot” in the center, decorated with a carnelian intaglio and featuring the head of a young man (Apollo?) in left-profile with a kithara. Gold, carnelian.25 80.32 g; 164 × 170 mm; cast: 25 × 21 mm (stored in the NMHU Treasury Inv. No. A3C—1719; Figure 1, Figure 2 and Figure 3).
• A bracelet with knotted ends. Gold. 13.62 g; 44 × 48 mm (stored in the NMHU Treasury; Inv. No. A3C—1677; Figure 4).
• A finger ring with a dot-punch inscription “χάρα” (“rejoice!”). Gold. 2.7 g; 15 × 12.5 mm (stored in the NMHU Treasury; Inv. No. A3C—1678; Figure 5 and Figure 6).
• A pendant, possibly a rattle, in the form of a scallop shell with a wide rim and a wire loop. Hollow, made of two identical halves; inside there is a small object that makes a noise when the item is shaken. Gold. 2.64 g; 35 × 26 mm (stored in the NMHU Treasury; Inv. No. A3C—1683; Figure 7a and Figure 8h).
• A pair of double-tube decorations with an ornament in the form of ivy leaves, teardrop- and round-shaped, lined with twisted wire and inlaid with green and blue enamel. Gold, enamel. 3.26 g, 3.06 g; 23 × 14 mm, 24 × 15 mm (stored in the NMHU Treasury; Inv. No. A3C—1720, A3C—1723; Figure 8c,d and Figure 9c,d).
• A pair of round and rhomboid-shaped plaques with loops. The surface is decorated with an ornament of S-shaped scrolls and drop-shaped figures made from thin wire, inlaid with enamel and red stone. Gold, garnet, enamel. 3.46 g, 3.97 g; 31 × 23 mm, 42 × 20 mm (stored in the NMHU Treasury; Inv. No. A3C—1721, A3C—1722; Figure 8a,b and Figure 9a,b).
• A pair of elongated barrel-shaped beads. Gold. 1.05 g; 0.95 g; 11 × 6 mm (stored in the NMHU Treasury Inv. No. A3C—1679, 1680; Figure 8i).
• A drop-shaped pendant with a green insert. Gold, malachite. 1.46 g; 21 × 9 mm (stored in the NMHU Treasury; Inv. No. A3C—1724; Figure 7c and Figure 8g).
• A rectangular pendant with five hemispherical bulges with a wire loop. Gold. 1.75 g; 27 × 13 mm (stored in the NMHU Treasury; Inv. No. A3C—1681; Figure 7b and Figure 8f).
• A hollow spherical pendant. Gold. 2.02 g; d—12 mm (stored in the NMHU Treasury; Inv. No. A3C—1682; Figure 8e).
• A glass jug (Chuistova 1959, p. 242; Figure 10) (the current location of this object is unknown; however, it is probably stored in the Historical and Archaeological Museum in Kerch).

4.1. Metal Composition

Non-destructive X-ray fluorescence analysis (XRF) was carried out with a desktop XRF analyzer ElvaX’03 by Yurij Bulakh at the laboratory of the Kyiv National University of Trade and Economics as part of cooperation between the Museum and the University (

Table 1. X-ray fluorescence analysis.

Object	Number of Samples	Au (%)	Ag (%)	Cu (%)	Other (more than 0.1%)
Torque. Kat. 1. Inv. No. A3C—1719.	6	82.83–86.92	10.12–10.84	2.41–6.34	Fe—0.62; Sn—0.14
Bracelet. Kat. 2. Inv. No. A3C—1677.	3	75.00–77.55	20.65–22.63	1.69–2.13	Fe—0.10–0.32
Finger ring. Kat. 3. Inv. No. A3C—1678.	3	92.25–92.66	5.15–5.41	2.11–2.27	Fe—0.3
Double-tube decoration. Kat. 5. Inv. No. A3C—1720.	5	69.34–70.18	26.75–27.94	2.23–2.78	Fe—0.4; Ni—0.1
Double-tube decoration. Kat. 5. Inv. No. A3C—1723.	5	71.41–67.93	26.20–29.06	2.00–2.99	Fe—0.19
Round-shaped plaque. Kat. 6. Inv. No. A3C—1721	7	76.12–78.12	19.63–21.92	1.32–2.32	Fe—0.42; Ni—0.1; Pb—0.22
Rhomboid-shaped plaque. Kat. 6. Inv. No. A3C—1722.	7	74.94–77.12	19.23–21.12	2.93–3.72	Ni—0.1; Fe—0.22
Drop-shaped pendant. Kat. 8. Inv. No. A3C—1724.	3	80.00–82.01	17.01–18.32	1.21–1.23	Fe—0.22; Sn—0.12

).26 The detection system is a Fast SSD detector. The measurement spot size is about 6 mm². The resolution of the detector is 140 eV for the Kα line of Mn; identification of elements in the range of elements from S (Z = 16) to U (Z = 92) in various concentrations.

A total of 39 analyses were obtained for eight items, the number of samples taken for each product ranged from 3 to 7 (samples were taken of both the internal and external surfaces, as well as from attached elements, decorative elements, and in soldered areas).

Previous studies have established that the copper content in gold from primary and alluvial deposits amounts to 1%, rarely 2% (Zaykov et al. 2015, p. 273; Hribkova and Bulakh 2013, pp. 69–75). In accordance with previous researchers’ conclusions, a threshold of 2% is therefore accepted as the boundary between native (natural) and alloyed gold. The first category includes gold with an admixture of copper of up to 2%, which is due to copper sulfides being included in the gold ores. This metal is sometimes called naturally alloyed. For alloying, additives of copper, nickel, and lead were used, which increased the wear resistance of products (Zaykov et al. 2015, p. 274). A higher concentration of tin was observed in the soldered areas.

As for the use of gold for different elements of the same object, in most cases, metal of the same composition was used. Only in some cases is there a discrepancy in the composition of the metal, particularly for the manufacture of loops and twisted wire.

The items made from precious metal are of a very high quality, which suggests that they were produced in a highly professional workshop. This is evidenced by the close attention to detail, the use of the finest twisted wire from which the ornament is laid, the well-inlayed high-quality enamel, and the neat and invisible soldering of elements.

A more in-depth study of the composition of the metal will likely make it possible to hypothesize the sources of gold based on its microinclusions. To accurately determine the sources of the metal from which the described finds are made, it will be necessary to compare the composition and isotopic characteristics of the metal with samples in primary and placer gold deposits.

The accumulation of results concerning the composition of the metal will provide more opportunities for future research and will add to the discussion on a wide range of issues—from trade relations in antiquity to the specifics of technological processes.

4.2. The Torque

Torques are often perceived as a typically “Barbarian” piece of jewelry that acquired “Greek” features. The emergence of the tradition of wearing these objects among residents of the ancient cities of Crimea is still debated. According to Natalia Pyatysheva, the use of torques by the inhabitants of Chersonesus was associated with the influence of the city’s ethnic Scythian component (Pyatysheva 1956, p. 257). Tatiana Vysotskaya considers the torques from the Late Scythian Ust’-Al’ma necropolis to be an element of Sarmatian influence (Vysotskaya 1994, p. 108; Figure 11). Vitaliy Zubar’ associates them with the Romanized people from the Danube region and attributes the presence of torques in burials from Scythian Neapolis to a Late La Tène influence (Zubar’ 1987, p. 88). Anastasia Stoyanova believes that torques were characteristic of Late Scythian society from the very beginning, whose members essentially continued the Scythian tradition of wearing this type of jewelry (Stoyanova 2011, p. 124). This claim is supported by the fact that early Late Scythian torques, unlike Roman period jewelry, do not have analogs in ancient cities of the Northern Pontic region while being typologically close to the torques from Scythian burials.

The torque (item no. 1 in the catalog) is a vivid illustration of this concept: it is a typically “Barbarian” piece of jewelry that symbolizes high status; however, it is decorated with a Greek “Herakles Knot” and an intaglio. An analogous piece depicting a “Herakles Knot” in the center, made using the same technique, comes from a male burial in a sarcophagus (crypt 2, sarcophagus 2 found in 1975 in Gorgippia (Treister 2003, pp. 45–46; Figure 11). There are no other attested objects similar to this type. Another well-known case of using a “Heracles knot” on torque is a decoration on the loop of a torque made of smooth, round wire from Chersonesus, dated to the second half of the 1st–2nd century CE.27 Numerous torques made from bronze false-twisted wire with smooth ends bent into a loop and hooks are known from the cities of the Crimea and they mostly date the 1st–early 2nd century CE; however, torques of this type survived until the 3rd century CE.28

The carnelian intaglio decorated with an elegant head of Apollo in left profile with a kithara belongs to a series of images of Apollo dating to the late 1st century BCE /1st century CE (Figure 2 and Figure 3). It has been pointed out by scholars that after the Battle of Actium, the image of Apollo became extremely popular in glyptics (Zwierlein-Diehl 2012, pp. 126–27; Gołyźniak 2020, p. 183). All these representations are variations of the head of Apollo from a statue made by Scopas in the 4th century BCE which was transported from Ramnunta (Greece) to Rome in 28 BCE and installed in the Temple of Apollo on the Palatine Hill. Intaglios with Apollo wearing a laurel wreath (Gołyźniak 2020, Kat. 9.799–800, Fig. 594) accompanied by his bow and quiver have also been recorded (Gołyźniak 2020, Kat. 9.781–802, Fig. 595). Iconography similar to the intaglio from the Kerch torque—namely Apollo holding a kithara—appears on intaglios held at the Museo Archeologico Nazionale dell’Umbria (Perugia), the Antikensammlung Berlin (Berlin), the Calouste Gulbenkian Museum (Lisbon), and in a private collection from Germany, that was bought from Sternberg, which dates to the last third of the 1st century BCE (Gołyźniak 2020, Kat. 9.803–806, Fig. 596).

The decoration of the central part of the torque with the Heracles knot and the intaglio is probably a reflection of the same tradition used in the decoration of torques with beads and pendants, which appeared in the 1st century CE. Such cases are known from other locations including the necropolis of Panticapaeum29, Chersonesus30, Late Scythian burial grounds of the Crimea31, Scythian Neapolis32, and the Lower Dnieper region.33 Several analogies can also be found at Sarmatian sites of the North Pontic region, including a golden torque with a carnelian intaglio as a pendant from a female burial near the village of Chugunno-Krepinka, Donetsk region (Mordvintseva and Treister 2007, Cat. 382.6; Simonenko 2008, Taf. 57.44) and a bronze torque with beads and pendants from the girl’s burial in the Zolotyj Mys burial ground near the village of Shyroka Balka, Kherson region (Dzneladze and Sikoza 2022, pp. 369–71, Fig. 33). It is still difficult to explain what caused the emergence of the practice of selecting this type of personal ornament as a grave offering. However, it was most likely due to changes in the way in which the torque as a specific social marker was perceived culturally, or perhaps the catalyst was provided through cultural contact when a typically “Barbarian” form of jewelry acquired traditionally Greek features. Future research is required to properly address this problem.

4.3. The Necklace

A number of the finds from the Kerch burial are presumed to be parts of a necklace, which includes the plaques with loops, the double-tube decorations, the gold beads, and various pendants (items nos. 5–10 in the catalog). Most probably, the grave contained further beads that have not survived. The round and rhomboid-shaped plaques with loops were probably used as a necklace tip, double-tube elements formed its central part and were complemented with beads and pendants.34 This type of necklace usually appeared in rich child burials of the 1st century CE and was carefully analyzed by Mikhail Treister (Treister 2007, pp. 82–84; 2015b, pp. 134–36). Treister drew attention to the frequent combination of double-tube decorations with figured or round tips with a quite stable decor of filigree, enamel, and stone inserts. The originality of this compact group of necklaces and their elements, as well as the fact that finds of such jewelry, are unknown outside the Northern Black Sea region, suggests that they were more than likely produced in Bosporan workshops. Furthermore, similar necklace elements are frequently found at urban necropoleis only (when the context of the find is known), including Panticapaeum35, Chersonesus36, and Phanagoria.37 They are absent both in Late Scythian and Sarmatian tombs. Another significant observation is that most examples were discovered in child graves, which suggests that they could have been produced specifically as child jewelry. Interestingly, like the child burial from Kerch, the necklace in grave 99 on the slope of Mount Mithridates was also accompanied by a ring bearing the inscription “χάρα”.

The shell-shaped pendant (item no. 4 in the catalog) may also have formed part of a necklace, as is indicated by the design of the loop which closely resembles that of the rectangular pendant with five hemispherical bulges. However, this does not exclude the possibility that it could have been used as a baby rattle, as Chuistova assumed, although children’s toys made of gold have yet to be discovered. There are known finds of toys (or items that are interpreted as toys) in the necropolis of Panticapaeum, such as the following: “some kind of wooden toy, covered with thin silver in the form of the club of Herakles” in grave 99 on the slope of Mount Mithridates (Shkorpil 1905, pp. 25–26), a terracotta toy—“torochtushka” in the form of a box with the relief of a figure on top and a clay ball inside.38

Pendants in the shape of scallop shells are known from burials in the Chersonesean necropoleis dated to the 1st–2nd centuries CE: for example, crypt 2158 excavated by Karl Kostsyushko-Valyuzhinich in 1907 (Zhuravlev et al. 2017, Cat. 134) and crypt 113 excavated by Robert Leper in 1912 (Zhuravlev et al. 2017, Cat. 135). The closest parallel in shape, size, and technique is a shell from the Staatliche Museen39, which is decorated using the same technique and style as the necklace elements from Kerch. Unfortunately, the archaeological context of this find remains unknown (Treister 2007, p. 84).

4.4. The Finger Ring

The finger ring with the inscription “χάρα” (“rejoice!”) (item no. 3 in the catalog) also has numerous analogies from child burials in Panticapaeum, Phanagoria, and Chersonesus (Rozanova 1968, pp. 126, 130, note 21; Treister 2015b, pp. 154–55, Tab. 3, 5–6; 49, 3–4; Fig. 34, 1–2; Cat. 11, 158; Zhuravlev et al. 2017, Cat. 327). Around seventeen rings with “χάρα” inscribed on the shields have been found in the North Pontic region. Ranging from 12 to 16 mm in diameter, they are mostly small in size (Arsent’eva and Gorskaya 2019, p. 13), and date to the 1st century CE. Documented examples are either from Panticapaeum or are presumed to have originated there. In 1843, Sergei Kareisha discovered a child’s burial in a sarcophagus to the west of Glinishche with a golden torque around the deceased’s neck, a bracelet with knotted ends, and a ring with the inscription “χάρα”. According to the shape of the sarcophagus, the assemblage dates to the 1st century CE (Chuistova 1959, p. 244). The same type of ring is known from a child’s burial discovered in a sarcophagus on the Northern slope of Mount Mithridates in 1873 (OAK 1876, p. 14), and at least three rings of this type have been unearthed during excavations conducted by Vladislav Shkorpil.40

Michail Treister considers these rings as a “local Bosporan phenomenon” and assumes that they were produced in the Bosporus (Treister 2015b, pp. 154–55). Sergei Tokhtas’ev devoted a special study to the “χάρα” inscriptions that appear on rings. He noted that the wish “χάρα” meant roughly the same as “χαῖρε” if one uses the proper definition—“rejoice!” (i.e., “wearing this ring”) (Tokhtas’ev 2015, p. 203).

The ring was most likely made for a child due to its small size. Additionally, examples of this type of ring are absent in typically “Barbarian”-style tombs. Thus, it is reasonable to consider that such finger rings were a localized Bosporan phenomenon that perhaps reflected a local tradition or cult.

4.5. Other Finds

Bracelets with knotted ends, such as item no. 2 (Figure 4 and Figure 6), represent one of the most common forms of gold bracelet found in the necropoleis of North Pontic Greek cities and Late Scythian and Sarmatian complexes (Treister 2007, p. 144). This type of bracelet is also frequently found in child burials, which seems to reflect its popularity among young individuals. This can be explained by the fact that the size of the bracelet could easily be adjusted to fit a growing child’s wrist. Numerous silver and bronze examples of such bracelets point to their remarkable popularity. This type of bracelet first appears in Thrace and the Western Black Sea region as early as the 3rd century BCE. From the 1st century BCE onwards, it also became a characteristic adornment among members of the Late Scythian culture.

It is difficult to say anything more about the glass vessel based on the low-quality photograph provided in Chuistova’s publication (Chuistova 1959, p. 242; Figure 10). To judge by the silhouette, it may belong to a common type of bottle that appeared in the second half of the 1st–2nd century CE (Lazar 2003, Kat. 6.3.4) and was in use until the 4th century CE (Leljak 2012, p. 127, Fig. 6a). However, in order to provide a more accurate attribution, it is necessary to obtain access to the item or at least a good-quality photograph.

It is noteworthy that in 1st-century CE Panticapaeum a fairly typical set of decorations was used for child graves, consisting of a ring, a bracelet, and a necklace or a torque. Elements of this assemblage could be combined with each other in different variations. The fact that unique “child” forms of jewelry, unknown in adult burials, such as rings with the inscription “χάρα” were produced, suggests that in Bosporan society a high demand existed specifically for child adornments. Undoubtedly, the subject of child jewelry, its use, and its meaning requires further research.

5. Conclusions

There is no doubt that the child burial from Kerch should be attributed to the category of elite burials. This is clearly demonstrated by the jewelry, the torque in particular, that displays the child’s high social status and the sense of belonging to the local elite. The torque is apparently too large for a child and, evidently, it was originally intended for an adult. Nevertheless, it was intentionally placed in the child’s grave, which suggests that it was deposited there due to its symbolic rather than practical meaning, which could include the display of the family’s wealth and social position. Similarly, the intaglio that was used for the torque’s decoration dates to an earlier period than the burial itself, which may suggest the presence of an heirloom and, as such, highlights the elitism of the grave goods. As one may assume, such an expensive stone with high-quality carving was passed down within the family from generation to generation. The funerary assemblage also reveals that the child belonged to a local family. This is indicated by the large percentage of local forms of jewelry used exclusively in the Bosporus, such as elements of necklaces, a ring, and a torque with a Herakles knot. The analysis of the findings demonstrated a cross-cultural character of the objects.41 The “mixed” nature of the jewelry is demonstrated by elements such as the Hellenized “Barbarian” torque, the bracelet typical of Late Scythian sites, and the ring bearing a Greek inscription. The lack of gender-specific objects as well as the presence of a pendant in the shape of a baby rattle may point to a very young age of the deceased individual.

Moreover, the type of grave that included a burial in a stone cist may also indicate that the child belonged to an aristocratic Bosporan family. Although, according to the funerary ritual, such types of burials did not differ greatly from simple flat graves (except for the fact that their construction was more expensive), the grave goods that accompanied individuals buried in stone cists resemble those recorded in tombs with ceilings which were the most common form of burial (Maslennikov 1990, p. 25). Therefore, it can hardly be argued that the deceased buried in stone cists differed significantly in terms of wealth from those buried in tombs with ceilings. The differences in the shape of the grave may have been associated with specific religious cults that were popular within a family. Alternatively, they could have also been used as markers of a particular social group. In Panticapaeum, the highest number of such graves in stone boxes has been recorded in the area of Glinishche, where they constitute approximately 4% of all recorded graves (Tsvetaeva 1951, pp. 78–80).

The analysis of the grave goods confirmed the previously proposed chronology of the child complex from Kerch in the second half of the 1st–the first half of the 2nd centuries CE (Treister 2003, pp. 45–46; Mordvintseva and Treister 2007, Cat. A. 343). It is not without importance that this period witnessed important changes regarding the self-representation of North Pontic elites in both Greek and non-Greek cultural zones. As mentioned above, the self-representation of the Crimean “Barbarian” elites seems to have undergone transformation as a result of the decentralization of power in the region. New contact zones influenced the display of power and status in elite burials, which is visible through the so-called “prestige objects” such as golden funeral wreaths that are well-known from the Greco-Roman world (Mordvintseva 2017, p. 194).

The period between the 1st and 3rd centuries CE is also the time when the so-called consolation decrees were popular, which were issued by Greek cities on the occasion of the premature death of a citizen to console the grieving family. The means of consolation was the decree itself, but it could also include a public funeral, a statue, or a golden wreath. Half of all recorded consolation decrees pertain to deceased children and young individuals who were praised for their exceptional values and their unfulfilled future duties towards the city (Strubbe 1998, pp. 60–62). As a rule, the deceased belonged to local elite families whose members held the most important magistracies and were often local benefactors. This suggests that the occurrence of consolation decrees can be connected with oligarchic tendencies that were characteristic of Greek poleis during the Roman period.42 The phenomenon of consolation decrees seems to have also influenced the North Pontic epigraphic habit. Elements characteristic of consolation decrees can be found in several post-mortem decrees from Olbia dating to the 2nd and 3rd centuries CE.43 One such decree was issued by the city to honor a boy (παῖς) named Dados who was granted a golden wreath and a statue. He “had been snatched away from his parents and from his mother city without mercy”, and he was also “expected to fulfil all liturgies according to the prestige of his family”.44 Another example honors a deceased (and most probably young) man named Karzoazos whose death brought sadness to the whole city.45 The partially preserved IosPE I² 46 may also honor a deceased young man who was similarly expected to fulfil all liturgies towards the city.46

A common motif of such decrees demonstrated that a child’s death was a great loss not only to the parents but to the whole city. In a non-Greek cultural milieu, where epigraphic culture was not practiced, rich child burials may have served a similar purpose by offering elite families the means to express their social status and the importance that family continuity had with regard to the good fortune of the whole community. It should not be ruled out that both Greek and non-Greek elite families may have shared and promoted the similar idea that the death of children should be perceived as a loss for the whole community since these deaths deprived it of future valuable members and possible benefactors who would no longer be able to perform their duties and fulfil their ascribed future social roles. The cross-cultural character of the burial assemblage from Kerch demonstrates that elite self-representation was not based on ethnicity but rather on a shared concept of the agency of elite families and their importance with regard to the well-being of the whole of society. The example of the Opushki cemetery (Figure 11), where a separate section for child graves was identified, may suggest that the idea of treating deceased children in a special way was also shared (or emulated?) by non-elite societies.

The material discussed in this study has demonstrated that the treatment of children after death and the mortuary rites conducted by the living changed over time and were intertwined with a given socio-cultural situation, which may have been affected by political factors and changing power relations in the region. These factors may have prompted specific cultic and religious trends that were reflected in the mortuary rites, but they may also have influenced the modes of displaying one’s power and wealth through the funerary ceremony.