Archaeological Approaches to Multiple Burials and Mass Graves in Early Medieval Europe

Abstract

CURRENT ARCHAEOLOGICAL LITERATURE dealing with mass graves and multiple burials in early medieval Europe does not apply a consistent definition of either term. Such features are described according to several factors, including the articulation of the skeletons, the arrangement of the bodies, and the temporality of deposition, but primarily according to the number of people in the feature. Drawing upon examples from Late Antiquity and the early medieval period in Europe and the Mediterranean, this paper argues in favour of a different classificatory approach. It is argued that multiple burials should be understood as continuations of normative burial practices in order to create or maintain social connections with the living community. In contrast, in the case of mass graves this function is overshadowed by the will or necessity to dispose of corpses. Mass graves and multiple burials should be defined not according to the number of bodies in the feature, but rather according to the purpose behind the deposition of the bodies.

Résumé

Approches archéologiques relatives aux sépultures multiples et aux fosses communes au Haut Moyen-Âge en Europe par Janet E Kay et István Koncz

La littérature archéologique actuelle concernant les fosses communes et les sépultures multiples au Haut Moyen-Âge en Europe n’emploie pas une définition homogène de ces deux termes, mais plusieurs facteurs pour les décrire, notamment : l’articulation des squelettes, la disposition des corps et la temporalité du dépôt, mais principalement le nombre de personnes en présence. En s’inspirant d’exemples de la fin de l’antiquité et du Haut Moyen-Âge en Europe et en Méditerranée, ce papier plaide en faveur d’une approche différente pour la classification. Il avance que les sépultures multiples doivent être appréhendées dans la continuité des pratiques de sépulture normatives, pour créer ou préserver les liens sociaux avec la communauté vivante. Dans le cas des fosses communes, en revanche, cette fonction est éclipsée par la volonté ou la nécessité de se débarrasser des cadavres. Les fosses communes et les sépultures multiples doivent être définies non pas en fonction du nombre de corps qu'elles renferment, mais plutôt selon la finalité sous-tendant le dépôt des corps.

Zussamenfassung

Archäologische Ansätze zum Thema der Mehrfachbestattungen und Massengräber im frühmittelalterlichen Europa von Janet E Kay und István Koncz

Die aktuelle archäologische Literatur, die sich mit Massengräbern und Mehrfachbestattungen im frühmittelalterlichen Europa befasst, wendet keinen der beiden Begriffe nach einer einheitlichen Definition an. Zur Beschreibung derartiger Befunde werden verschiedene Faktoren wie die Gelenkverbindungen der Skelette, die Anordnung der Leichen und die zeitbezogenen Aspekte der Beisetzung herangezogen, in erster Linie aber die Anzahl der Menschen im jeweiligen Befund. Anhand von Beispielen aus der Spätantike und dem frühen Mittelalter in Europa und dem Mittelmeerraum plädiert dieser Beitrag für einen anderen Klassifizierungsansatz. Es wird argumentiert, dass Mehrfachbestattungen als Fortführung normativer Bestattungspraktiken zur Schaffung oder Erhaltung sozialer Verbindungen mit der lebenden Gemeinschaft verstanden werden sollten. Bei Massengräbern hingegen ist diese Funktion durch den Willen oder die Notwendigkeit überschattet, Leichen zu beseitigen. Massengräber und Mehrfachbestattungen sollten nicht nach der Anzahl der Leichen im Befund definiert werden, sondern nach dem Zweck, den die Beisetzung der Leichen erfüllte.

Riassunto

Come vengono affrontate in archeologia le sepolture multiple e le fosse comuni dell’Europa altomedievale di Janet E Kay e István Koncz

Le attuali pubblicazioni archeologiche che si occupano delle fosse comuni e delle sepolture multiple dell’Europa altomedievale non adottano una definizione costante per l’uno o per l’altro termine. Le caratteristiche di entrambe vengono descritte in base a vari elementi tra cui l’articolazione degli scheletri, la sistemazione dei corpi e la temporalità della sepoltura, ma principalmente secondo il numero di corpi presenti nel sito. Basandosi su esempi della tarda antichità e dell’Alto Medioevo in Europa e nel Mediterraneo, questo studio porta argomenti a favore di un metodo di classificazione diverso. Si sostiene che le sepolture multiple vadano interpretate come la continuazione di pratiche di sepoltura regolari in modo da creare o mantenere i legami sociali con la comunità dei viventi. Al contrario, nel caso delle fosse comuni tale funzione passa in secondo piano dato il volere o la necessità di sbarazzarsi dei corpi. Fosse comuni e sepolture di massa vanno definite non in base al numero delle salme, ma piuttosto in base allo scopo della sepoltura dei corpi.

The past decade or so has seen the publication of a number of extraordinary multiple burials and mass graves from Late Antiquity and the Early Middle Ages in Europe and the Mediterranean, such as the boat burial of 40 people in two ships at Salme, Estonia, or the analysis of the burial of 14 people in a Roman latrine in Edinburgh, Scotland (for Salme, see Peets et al 2011, 2012; Price et al 2020 and for Cramond see Holmes et al 2018, and most recently, Czére et al 2022). The methods introduced by archaeothanatology, and more recently aDNA analyses in particular, have changed the kinds of questions able to be asked about the individuals buried within ‘mass graves’ or ‘multiple burials’—especially about the reasons why more than one individual was buried in the same physical structure. Interpretative frameworks for this kind of evidence have been particularly visible in scholarly debates about the Justinianic Plague of the mid-6th to mid-9th century ad (McCormick 2015, 2016; Keller et al 2019; Mordechai et al 2019). This article comes at a moment when genomic testing of individuals excavated from late antique and early medieval cemetery contexts allows the testing of the common hypotheses that burials of more than one individual in a single feature happen because of either biological relatedness or contemporaneous or near-contemporaneous death (for the former see Deguilloux et al 2018; O’Sullivan et al 2018; Rott et al 2018, for the latter see Keller et al 2019).

Mass graves are also, unfortunately, not entirely remote from popular consciousness in 2023. The public understanding of this concept has implications for how the past is studied, and modern perceptions of that term should not be confused with the early medieval concept. Today, is the ‘age of mass graves’ as the result of various crises—including worldwide pandemics, and more regional causes like civil wars—and discussion of these features is concurrent in the growing fields of the archaeology of conflict and violence (Loe et al 2014; Meller and Schefzik 2015; Fernández-Götz and Roymans 2018), and in forensic archaeology of the recent past and the late 20th and 21st centuries in particular (Skinner 1987; Haglund et al 2001; Mitrovic 2008). In the recent global crisis caused by the COVID-19 pandemic, for example, the need to respectfully bury unusually high numbers of individuals in a short amount of time led to the opening of ‘mass graves’ such as those on Hart Island, in New York City (Fig 1; Hennigan 2020; Slotnik 2021). These images of ‘mass graves’ affect current popular consciousness, and therefore scholarship.

Fig 1 COVID-19 mass graves at Hart Island (NYC) on 9 April 2020. Image taken from Slotnik (2021). © AP Images/John Minchillo. Used under license.

Here it is aimed to offer a new approach to the classification and definition of multiple burials versus mass graves in the archaeological interpretation of sites from Late Antiquity and the early Middle Ages in Europe and the Mediterranean, when a typical and expected burial was a single interment. As the primary discussion on ‘multiple burials’ or ‘mass graves’ tends to happen with inhumations, this article focuses on them, though cremations can also have multiple burials (Lüdemann 1994; Lippok 2020; on social connections maintained or created through cremation practices see eg. Williams 2004; 2014; Nugent 2017). The delineation between the terms “multiple burial” and that of “mass grave” typically relies upon the number of bodies found in an archaeological feature, and current usage of the two terms is not consistent, especially across academic traditions. Firstly, this paper argues that it is more helpful for our understanding of past societies to delineate between the two terms based upon the social processes that result in the placement of more than one individual in an archaeological feature. It is also specifically argued, in contrast to dominant trends in current literature, that multiple burials and mass graves should not be defined according to how many individuals are identified in a given archaeological context. Instead, it is argued that multiple burials and mass graves should be defined according to the primary social purpose and symbolic aspects surrounding the deposition of many bodies in that feature. It is considered that multiple burials follow practices that are otherwise typical for single burials (for the time and place), but have more than one person deposited in the feature (though not necessarily at the same time). In contrast, mass graves are the result of events when traditional practices are overwritten or not followed, for any number of reasons including violence and epidemic disease. This is an important intellectual shift from descriptive to interpretative in how such burials are approached and desired accomplishments through studying them. While these examples are primarily from the early medieval period, this approach could also be applied to later medieval graves.

WHAT ARE ‘GRAVES’ AND ‘BURIALS’?

Before discussing the terminology of multiple burials and mass graves, the definition of three broader terms: funeral, burial and grave should first be assessed briefly. The distinctions between these three terms are important to keep in mind in the discussion of mass graves and multiple burials, because they have implications for the level of social engagement and/or connections between the deceased themselves and the deceased and the living community. Funerary actions start with the death of an individual or individuals—or even before that—and can last far beyond the action event through acts of remembrance (Williams 2006; Weiss-Krejci 2013; van Gennep 2019). A funeral is generally understood as an act or a collection of acts (sometimes formalised, sometimes simple) connected to the passing of an individual from life to death (Robb 2013). While burial and grave are frequently used as synonyms describing the archaeological feature, a distinction between these two terms is necessary. A grave is the physical structure in which—even if only symbolically—the body or bodies have been intentionally deposited (Knüsel and Robb 2016, 3); but it does not have to be artificially created just for the purpose of the burial. A burial, in contrast, can be understood as the physical remains of the intentional deposition of a body (interpreted in its immediate archaeological context) (Knüsel and Robb 2016, 3), that is, an incomplete physical record of the funerary practices of which it is a part (Härke 1997, 22). Burial is a structured deposition in the sense that it is created as a result of human agency and action (Garrow 2012). The use of this term to describe an archaeological feature, however, is also an interpretation in itself based on the purpose of the deposition. Not all graves contain burial/s (ie cenotaphs), and not all burials are connected to defined physical structures (ie graves).

There are many different purposes and goals of a burial, which vary through space and time, but for the sake of this argument, here two simplified aspects are emphasised, that the burial of an individual is both a functional necessity and a social act. Burials are generally—but not exclusively—created through highly ritualised acts (funerary rituals). As death is a liminal state, after which the deceased is no longer considered a member of the community, or at least not the same member they were before, communities create such rituals to ease this transition and to help the dead find their place in the afterlife (Metcalf and Huntington 1991: van Gennep 2019). Thus, funerary rituals, as part of the grieving process, help individuals and communities to overcome feelings of loss and distress, as well as to help fend off the ill intent of the dead.

A funeral is also a social act, a performance that has an important role in creating or reinforcing a sense of sameness and communal identity or existing social structures between participants. Every death changes both personal and societal dynamics and tears the fabric of the community. The funeral is a place for social performance that gives the community an opportunity to express ideas about the deceased and themselves through the funerary representation (Morley 2017). These ideas are mainly enacted by the group responsible for the burial but are also influenced by the sanction and expectations of the wider audience; as such, a burial reflects a selected version of the relationship between the deceased and the burying group.

A dead body, however, is also a potential public health hazard that can spread disease and contaminate water supplies as well as raise a series of psychological issues, such as fear of the dead or complications in the grieving process—though responses to these issues may vary culturally (Metcalf and Huntington 1991; Shear 2012). An unburied cadaver can be an unpleasant sight; it can smell or attract unwanted animal attention; and it can create a generally disagreeable environment. Most societies therefore establish protocols for dealing with dead bodies, which can include strict regulations governing the physical construction of graves and the establishment of cemeteries, or even legal responsibility for the act of burial, notably in cases where this responsibility is not clear. As an example, for regulations in the Roman World and Late Antiquity see John Bodel (2000) and Sarah E. Bond (2013). This is especially true in the case of a catastrophic event resulting in mass mortality that can easily overwhelm and/or interrupt established protocols.

Even the functional aspect of burials can be shaped by cultural and religious beliefs, as well as societal perceptions of death and the dead. The funerary ritual could include periods of exposure of the body (ie, on a bier), and these could also play an integral part of the literal deposition of the body in certain ways as well, an example being the Tibetan ‘sky burials’ (Mullin 1998). In other cases, the goal might not have been the deposition of the bodies, but rather the conscious exposure or other use of body parts for punishment, deterrence, or as warning. Most mortuary treatments take place somewhere in between entirely functional or entirely performative goals, based on the emphasis given to these aspects by the group responsible for creating the burial. Both of these aspects are present in every burial and operate as two ends of the same spectrum.

CURRENT TERMINOLOGY OF ‘MASS GRAVES’ VS ‘MULTIPLE BURIALS’

In current archaeological research both multiple burials and mass graves are generally viewed as nonnormative burials of more than one individual (Lüdemann 1994). A clear distinction of the two terms is hard to find, especially since they are sometimes used either as synonyms or as overlapping or interchangeable terms (Sutherland 2000). To make the distinction even more challenging, multiple burials have been sometimes interpreted as a subcategory of mass graves and vice versa. Various other terms with partly similar meanings have also been suggested instead of or as a synonym of certain types of multiple burial. These features are also sometimes named according to the cause of interment, such as ‘plague pits’ or ‘massacre sites’ and ‘family tombs’.

There are various definitions for multiple burials in the different traditions of archaeological research (Boulestin and Courtaud 2022). They are generally understood on a synthesis of archaeothanatological traits of the archaeological feature and bioarchaeological traits of the human remains and focus on two main criteria: the number of individuals buried and the chronological framework of those depositions. Most broadly, multiple burials are features or structured depositions that contain the remains of more than one individual (Duday et al 1990; Stoodley 2002; Sprague 2005, 75; Castex et al 2014, 299; Knüsel and Robb 2016, 3; Boulestin and Courtaud 2022). The most common has two individuals, often referred to as a double burial, which is sometimes used as a separate, individual term (Miniaci 2019, 10). While the lower limit is two, the upper limit is much harder to define and usually is not addressed separately, but in relation to the definition of the lower end of mass graves.

A recurring theme in the definitions of multiple burial in the archaeological literature is the temporality of these features, and the use of certain terms differs according to the length of time during which more than one individual was interred in a single feature. According to some of the definitions, bodies in a multiple burial were deposited simultaneously as a result of one single event (Duday et al 1990; Blaizot 1999, 192; Castex et al 2014; Knüsel and Robb 2016, 3). The term ‘collective burial’, in contrast, is suggested for deposits created over a longer period of time or multiple events with the re-opening of a feature (Castex et al 2014; Knüsel and Robb 2016), while ‘plural burial’ or ‘group burial’ is used in cases where such distinction is not possible (Leclerc and Tarrête 1998; Duday 2008, 50; Czére et al 2022). These terms, however, are not widely accepted and/or used among early medieval archaeologists and are sometimes used with different or even opposite meanings (Lüdemann 1994; Rott et al 2018; Boulestin and Courtaud 2022). Some call attention to the different possible temporalities of ‘multiple’ burials but make a distinction without the need to introduce new terms for successive or contemporary temporal uses of the same feature (Stoodley 2002, 106; McCormick 2015, 350–1). The most intricate categorisation of multiple burials based on these two parameters (the body count within the feature and the temporality of their deposition) comes from Gianluca Miniaci, who distinguishes between ‘primary simultaneous’ (burials containing more than one individual interred at the same time), ‘secondary simultaneous’ (simultaneous relocation of non-contemporaneous deceased to the same burial context), ‘delayed simultaneous’ (bodies collected over a period of time and buried together) and ‘sequential’ (re-opened contexts for later interments; Nachbestattungen in German literature). This last term is essentially a cognate for ‘collective’ burials and refers to graves that were re-opened multiple times in order to allow the interment of additional bodies (Miniaci 2019, 6–9). The use of the terms ‘multiple burial’ (Mehrfachbestattung) or ‘multiple grave’ (Mehrfachgrab) has also been used to differentiate between interments where the bodies are within the same context (the former), or, although deposited in the same context, were deliberately separated by thin soil walls or within separate coffins (the latter) (Lüdemann 1994).

The basic definition of multiple burials overlaps with that of mass graves, and the latter term also generally refers to a deposition of more than one individual in the same archaeological feature. In forensic archaeology, a field that regularly deals with both multiple burials and mass graves in the context of human rights abuses, a mass grave contains, ‘at least half a dozen individuals’ (Skinner 1987, 268). Other archaeological definitions are less specific, and may mention, ‘a number of individuals,’ (Sutherland 2000, 36), a ‘large number of subjects’ (Duday 2008, 53), or ‘large numbers of articulated individuals’ (Knüsel and Robb 2016, 3). Similar to Mark Skinner’s definition, Michael McCormick draws the line at five individuals between multiple burials and mass graves in his catalogue of late antique mass burials (McCormick 2015, 33, 351). Probably the most interesting approach is proposed by Miniaci, who concludes that a mass grave, ‘is not contained in the sheer size (more or less than × individuals) but in the ultimate accessibility of countable individuals’ (Miniaci 2019, 10), an idea that originates from linguistics and suggests that the focus is not on the exact number, but on the ‘opposition of one’, ie many (Acquaviva 2017).

Beside the number of interred bodies, the various definitions of mass graves frequently mention a series of other characteristics as well, such as articulation and placement of the bodies (in an ordered or disordered state), and the underlying cause of the need to bury multiple individuals in one feature; an event of mass mortality such as battle, epidemic or natural disaster for example (Knüsel and Robb 2016, 3). In particular, articulation and placement of the bodies is an aspect that is regularly considered as a deciding factor between multiple burials and mass graves. It has been suggested that multiple burials are similar to single burials in this regard, that the remains are more likely articulated and placed in a more orderly fashion than in case of mass graves, for which a non-systematic placement of bodies is characteristic (Sprague 2005, 102). According to Skinner, a mass grave (unlike a multiple burial) contains bodies that usually have been placed indiscriminately and tightly together with no visible intent to provide spacing between individual bodies (Skinner 1987, 264–8). Roderick Sprague similarly states that multiple burials contain articulated bodies, while, ‘a mass grave consists of disarticulated bodies or only articulated fragments’ (Sprague 2005, 74). Disarticulated bones and normative behaviour, however, are not mutually exclusive, and this disarticulation is not emphasised (Knüsel and Robb 2016) or entirely missing in other definitions (Sutherland 2000, 36; McCormick 2016). Overall, the distinction between the two terms is always based on the number of bodies interred within the same feature; the additional traits are only used to emphasise perceived or real differences.

In the discussion that follows, suggested new ways of thinking about the terms mass grave and multiple burial are introduced and explained. It is argued that neither the temporality and the articulation of the burial events, nor the number of individuals found in an archaeological context should be the defining characteristic between either of these two terms; rather, it is more helpful for archaeological interpretation of such contexts to delineate according to whether or not the burial events follow normative social funerary practices for their particular cultural and chronological contexts.

REDEFINING APPROACHES—SOCIAL REASONING INSTEAD OF DESCRIPTIVE TERMINOLOGY

The important distinction between multiple burials and mass graves, argued here, is that the living community placed individuals within multiple burials with a similar effort to construct social identities through normative burial practices as they would for typical single burials in that time and place. A multiple burial therefore, represents a connection between the interred individuals and the burying community. It is a term that emphasises the highly symbolic placement of more than one individual within a feature, rather than the primarily functional need to dispose of many bodies. This could be in a formal grave within a standard cemetery but could also be in reused natural or anthropogenic features or other contexts. The living community still followed typical social practices surrounding the burial of the deceased, and the circumstances of those burials deliberately emphasise the relationship of the deceased to that living community.

The placement of numerous bodies in the same feature for a multiple burial could indicate contemporaneous death and deposition, but could also simply be the expression of social connections such as kinship between the individuals (Lüdemann 1994; Stoodley 2002; Rott et al 2018). These connections were emphasised by the living community in funerary actions—even during instances of social stress. At the 6th–7th-century site at Hérange, France, for example, four individuals were simultaneously buried in a multiple burial. The results of aDNA analysis proved that three of them (a woman and two young children) were maternally related, so the contemporaneous multiple burial of at least these three related individuals, if not all four, could be understood as the expression of this strong social tie (Fig 2; Deguilloux et al 2018). Another example (although not a multiple burial per se), where biological relatedness and possibly social kinship is expressed spatially was observed in the 6th-century cemetery at Szólád, Hungary. Here burials of an adult male and a young child were found inside a large circular ditch. Paleogenomic analysis identified them as uncle and nephew (Amorim et al 2018; Fig 4). The 6th–7th–century cemetery at Aschheim-Bajuwarenring, in southern Germany, comprised a large number of multiple burials; out of the 456 graves there were 29 double burials, four burials containing three bodies, one burial containing four and one with five. Their dating (Gutsmiedl-Schümann 2010, 112–13), demographic profile (Staskiewicz 2007), and the positive identification of Yersinia pestis in at least some of the multiple burials (Keller et al 2019), suggests a connection between the Justinianic Plague and the appearance of a high number of multiple burials at the site in the second half of the 6th century. Many of these burials, however, are richly furnished and arranged according to typical social practices at the time (Fig 3), and in general do not show negligence or atypical handling of the plague victims (Gutsmield-Schümann et al 2017, 239–41). These multiple burials at Aschheim could be considered a reaction to the simultaneous death of individuals (probably with strong social ties) caused by the pandemic, but they also demonstrate a clear continuation of the normative burials at a time marked by a mortality crisis. At a certain point they might even have become standardised during the pandemic, as 8% of the deceased (or higher, if only the second half of the 6th century is counted) were buried in multiple burials (Gutsmiedl-Schümann 2010, 112). Victims of both the First (Keller et al. 2019) and the Second Plague Pandemic have been found in standard single inhumation burials (Gilchrist and Sloane 2005, 208–209; Cessford et al. 2021) as well.

Fig 2 Position of the bodies in the early medieval multiple burial (Grave 41) at Hérange, France. Of the four individuals buried in this feature, A, B and D are maternally related, while C—a juvenile with opposite orientation—is not. After Deguilloux et al 2018, fig 1. © Elsevier and authors. Reproduced with kind permission of the authors and Elsevier.

Fig 3 Individuals 167 (L) and 166 (R) buried with various types of brooches and beads typical of the period in one of the multiple burials connected to the Justinianic Plague from Aschheim-Bajuwarenring, Germany. After Gutsmiedl-Schümann et al 2018, fig 2. © De Gruyter and authors. Reproduced with kind permission of the authors.

Mass graves, in contrast, are first and foremost examples in which the living community was more concerned with the disposal of multiple bodies rather than with treating those bodies in a way that prioritised social connections and relationships. The social connections of each individual are no longer expressed, as they were not buried as individuals, but as part of a group. Here, it is argued that ‘mass grave’, therefore, is a term that emphasises the functional aspect, overshadowing (intentionally or out of necessity) the social aspect. The method or event that caused the large quantities of corpses does not matter—though violence, epidemics, famine, or earthquakes are common causes—and it does not preclude there being some symbolism attached to the likely public disposal of bodies, such as the undignified burial of enemies in war. In case of mass graves, the interred bodies are no longer treated as individuals, but as cadavers, where the living needed to dispose of large numbers of corpses without the time, ability, or intent to provide for normative funerary rites. One such example is the 10th-century ad mass grave at St John’s College, Oxford, in the UK, where between 35 and 38 adult males and juveniles, all of whom died violently, were unceremoniously dumped in a late-Neolithic henge ditch (Wallis 2014; Fig 4). They were found in, ‘positions suggestive of being pushed or falling into the grave,’ while one individual had his arms raised, ‘above his head in an orientation that would be expected from being carried between two individuals’ (Falys 2014, 42).

Fig 4 The reuse of an earlier henge ditch for the convenient disposal of many bodies in the 10th-century mass grave at St John’s College, Oxford. Each individual skeleton in the mass grave is drawn in red. Image taken from Wallis 2014, fig. 3.1. © Thames Valley Archaeological Services. Reproduced with kind permission of the authors.

Such an objectification of the body, severing the connection between the once living individual or the soul and the physical body, is an integral part of most funerary rituals. In mass graves, this separation takes place in an instant without any consideration of the deceased or the grieving community and could happen for at least three main reasons. Firstly, the normal protocols of dealing with the dead could be overwhelmed by the sheer number of corpses. Therefore, because they serve to dispose of many corpses very quickly, mass graves can often lack clear stratigraphic contexts, such as in the 2nd-century ad mass grave (possibly attributed to deaths from the Antonine Plague) at 120–122 London Road in Gloucester, UK (Simmonds et al 2008). Here, the bodies of at least 91 individuals were seemingly thrown into a deep pit in a haphazard manner; their interpretation and identification made more difficult by their largely disarticulated nature and highly variable skeletal preservation (Simmonds et al 2008, 34-39).

Secondly, the deceased could be deliberately denied standard or typical burial, either as a punishment or a sign of disrespect, or because the goal of the deposition was never intended as a burial. For example, alongside the Via Emilia near Modena, Italy, the bodies of three young males were excavated in close proximity to a 1st century BC–1st century ad cremation cemetery. The bodies, showing signs of physical trauma suggesting a violent death, were dumped in an irrigation ditch and were weighted down by bricks to prevent them floating to the surface (Wileman 2015, 62). The context at Modena suggests that the main goal of the deposition was to get rid of the bodies and expected burial practices were ignored; as such even though there were only three individuals in this feature, by our interpretation it could be considered a mass grave because those three people were not buried as individuals, but en masse.

Another such example is the 10th- or 11th-century ad massacre site at Ridgeway Hill, Dorset, where the postcranial skeletons and decapitated skulls of roughly 50 young men were piled without grave goods (or even clothing) in an earlier Roman quarry pit (Loe et al 2014; Fig 5). Isotopic analysis suggests that the men in the mass grave were not local and had most likely come from a wide range of locations (Chenery et al 2014). In this case, therefore, the group who decapitated these men and placed their bodies and heads in separate piles within the mass grave purposely denied marking important social ties or identities, probably as part of the punishment, as a denial of respect (Loe et al 2014, 232). The exposition of the skulls and bodies also could have played an important role in the creation of the post-depositional tableau, emphasised by its location near a major road, in the manner of many Anglo-Saxon execution cemeteries (Loe et al 2014, 233).

Fig 5 The c 50 men decapitated and piled in a 10th- or 11th-century mass grave with their skulls at Ridgeway Hill. Each skeleton is coloured differently for visual distinction. Loe et al 2014, fig 2.4 © Oxford Archaeology and Dorset County Council. Reproduced with kind permission of Oxford Archaeology.

Execution cemeteries, in particular, present an interpretative challenge under our proposed definition of mass graves and multiple burials. On one side of the terminological divide are sites like Ridgeway Hill, where the treatment of so many bodies in such a manner is one the best recognised visual representations of a mass grave. On the other hand, however, are multiple burials like the triple burial of three individuals (117, 118, and 119) at the execution cemetery of Guildown, Surrey (Lowther 1931, 37–8; Reynolds 2009, 57). According to the definitions of multiple burial and mass grave that are laid out in this article, these execution burials could possibly fall into the mass grave category because the funerary rites are nonnormative and were specifically used to mark exclusion from a community—either in their location, or in the burial of a prone or decapitated body (Reynolds 2009, 64–7). A careful application of the definitions outlined here, however, indicates that they cannot be considered mass graves because, a) while the deceased in an execution cemetery were buried with nonnormative rites, those rites themselves, in creating metaphorical and physical distance between the disgraced dead and the living community, perpetuate a series of social rules and conditions by which the living community abides (Reynolds 2009). And b) in the case of certain burials at execution cemeteries, like the example from Guildown, the functional aspect of disposing of corpses did not outweigh the social practices inherent in the multiple burial rite. Execution burials of more than one person may not be typical, but those like the ones at Guildown follow socially proscribed patterns, and therefore fall into this paper’s definition of multiple burial.

Thirdly, at the time of deposition in a mass grave, the deceased may no longer be recognisable as an individual to the living—for example, after prolonged decomposition—or there may be no connection between the deceased individuals and the group doing the burying, which could have been emphasised through nonnormative burial practices. At the 5th-century mass graves found in the late Roman-period fort at Keszthely-Fenékpuszta, Hungary, for example, the remains of at least 70 individuals were found stuffed inside eight kilns (Fig 6). Complete skeletons were only rarely excavated, but a large number of articulated body parts also remained in anatomical connection. The status of the remains suggests that the bodies lay unburied for an extended period of time (possibly days or weeks) and were buried in an advanced state of decay (Straub 2002).

Fig 6 Kiln XXVIIb/1 from the late Roman-period fort at Keszthely-Fenékpuszta, reused as a 5th-century mass grave and containing remains from at least 27 individuals. Drawing and photo after Straub 2002, abb 8.

Under the new definitions proposed here there is no strict numerical component to the dividing line between multiple burial and mass grave (Tab 1). The current terminology limits the number of people in a multiple burial to only a few individuals but considers mass graves to include at minimum five or more individuals—though, of course, the use of the word ‘mass’ implies a much higher number, in the tens or hundreds. Preservation of the osteological material is therefore key in definitions that rely upon numerical classification, especially as many burials discussed in the current literature contained between four and six individuals. (McCormick 2015, 352). Any count of the minimum number of individuals—whether done in the field or in the lab—depends first upon preservation conditions and taphonomic concerns, and second upon the ability to correctly count how many people (as a minimum number of individuals) were found in a given feature during archaeological excavations.

Table 1 Comparison of suggested new definitions for mass graves and multiple burials.

Multiple Burial

Mass Grave

- More than one individual in a grave/feature, in a time/place where normative burial practice is single interment. - Aside from the number of individuals, burial practices follow normative patterns of body position, grave goods, etc. that are otherwise found in single interments - Burial represents a social connection between the deceased individuals in the grave and the deceased individuals and the surviving community.

- More than one individual in a grave/feature, in a time/place where normative burial practice is single interment. - Burial practice does not follow normative patterns of body position, grave goods, etc., that are otherwise found in single interments - Burial does not necessarily represent a social connection (or represents a deliberate break) between the deceased individuals in the grave and the deceased individuals and the surviving community. - Functional aspect of disposing of multiple corpses outweighs socio-cultural aspect

It is argued here that the number of individuals within an archaeological feature should not be decisive in defining them as either mass graves or multiple burials, as long as the manner of their placement within the feature follows the parameters of human agency and the emphasis on functional or social practices outlined above. For example, two 8th-century ship burials excavated at Salme, Estonia, have been described as ‘mass graves’ within the literature based upon the number of individuals buried within them. The two ships—Salme I and Salme II—contained the bodies of 6 and 34 adult males respectively (Price et al 2020; Fig 7). However, considerable care was taken during the burials: the use of ships as grave features, the careful placement of the bodies, the presence of gravegoods, and remains of the funeral feasts suggest that these individuals were laid to rest in a manner that goes way beyond the simple deposition of bodies, and elements of normative practices were retained even after an event of mass mortality (Allmäe 2012; Peets et al 2011, 2012; Price et al 2020).

Fig 7 The 8th-century multiple burials at Salme I and Salme II in Estonia. (a) A possible reconstruction of the position of the bodies in Salme I. Their placements at the oars and helm are understood as reflection of their roles in the ship; (b–c) Position of the bodies in Salme II. In the northeastern part of the ship the remains of at least 34 individuals were arranged into four regular layers (coloured differently on the 3D model) that were separated by thin layers of soil and sand and were partly covered with shields. Illustrations taken from: (a) Allmäe 2012, fig 2; Reproduced with kind permission of the authors; (b–c) After Peets et al 2012, figs 10 and 12.

Another example where the ‘number’ count makes a difference is a multiple burial at the mid-5th- to early 8th-century cemetery at Butler’s Field, Lechlade in Gloucestershire (Boyle et al 1998, 2011). Lechlade comprised 219 individuals inhumed in 199 graves, as well as 29 cremations, and at least 16 graves contained the intentionally deposited remains of more than one person (Boyle 1998, 35–7). The majority of these graves, however, are double burials (Boyle 2011, 156). One notable exception is the apparently contemporaneous five individuals in Grave 81—two adult females, two children (around two to three and six to seven years old), and an infant 15 to 18 months (Boyle et al 1998, 89–91; Boyle 2011, 156). A more thorough examination of the context, however, revealed that the initial burial was that of one of the young adult females, which was later superimposed by the burial of the remaining four individuals (Fig 8). The two women and the elder of the children were all buried with fairly typical funerary objects for Lechlade, including pairs of brooches, bead necklaces, and knives (Boyle et al 1998, 89–91). As part of a larger interpreted pattern of graves based around possibly family relationships at Lechlade, Grave 81 was originally interpreted as a family ‘vault’ (Boyle 1998, 37; Boyle et al 2011, 157) This grave, both in the original site report and following the new criteria as outlined above, is a multiple burial—despite the unusually high number of people interred in the same feature, each individual was buried within the boundaries of standard accompanied practice for the time. If a strictly numerical definition was followed disregarding its temporality and practices, however, Lechlade Grave 81 would be reclassified as a mass grave. Such a redefinition would fundamentally alter the understanding of not only Grave 81, but also the parameters of Lechlade cemetery. Even if such a definition would only be semantic, potentially inappropriate perceptions of the phrase mass grave might be applied and could influence the understanding of this cemetery and its burial archaeology.

Fig 8 Grave 81 at Lechlade which contains the remains of five individuals (scale 1:20). Individual 81.1 was buried below the rest, and the grave was covered with limestone. After Boyle et al 1998, fig 5.15. © Oxford Archaeology. Reproduced with kind permission of Oxford Archaeology

In the suggested definitions, the length of time that it takes for bodies to accumulate in an archaeological feature is also not decisive for delineating between mass graves and multiple burials. Multiple burials can be consecutive and successive (collective burial), intentionally reusing the same grave over and over again, or simultaneous and contemporary (Duday et al 1990; Lüdemann 1994; Leclerc and Tarrête 1998; Castex et al 2014; McCormick 2015; Knüsel and Robb 2016, 3). In some cases, the interpretative line can be complicated by the temporality of the feature, though that does not change the normative or nonnormative practices surrounding the burial of individual people. If the bodies were intentionally deposited to express social connections between the deceased, it is argued this should be interpreted as a sequential multiple burial, even if the period of burial took a long time.

For example, the late antique chamber tombs with small, recessed rooms called acrosolia, which were continuously used over longer periods of time and are generally interpreted as family tombs, or at least burial places of individuals with strong social ties (Seligman et al 1996; Taha 2003: Weiss 2010). At Atara, Israel, 32 individuals—including a young woman with her newborn child (Zias et al 1993)—were found in an undisturbed chamber tomb with three acrosolia (Taha 2003, 90). This site is described as a family tomb and was in use for several generations between the 4th and 5th centuries based upon the accompanying gravegoods, mostly oil lamps, ceramic and glass vessels and dress accessories (Taha 2003, 90, 101). The similarly undisturbed, 5th-century Tomb 200 at Giv’at Sharet, Israel, contained 25 individuals in six acrosolia (Seligman et al 1996). One of these contained eight individuals crowded in a confined space. The fact that there was still adequate space in other tombs nearby might indicate very strong social contacts between the individuals intentionally buried in the multiple burials (or in this case collective burials) in the crowded acrosolia (Seligman et al 1996, 48).

In contrast, if the deposition of more than one individual over a long time in the same feature does not show signs of social symbolism in burial practice (bodies can also occupy the same space by accident, generally referred to as superpositions, without any intentionality), but rather looks to be primarily intended as a way to dispose of bodies, then it should be considered a mass grave. One example of a mass grave with a long period of use is observed in one section of the 1st- to 3rd-century ad catacombs of SS Peter and Marcellinus in Rome that differs greatly from the typical organisation observed in these kinds of underground funerary sites. In contrast to the usual straight galleries comprising loculi, acrosolia, and cubicula, here chambers contained multiple layers of human remains without any kind of visible intentional separation of individual bodies. The results of three-dimensional reconstruction excluded the possibility of simultaneous deposition of fully fleshed and complete corpses, as the total volume of the bodies greatly exceeded the available space in the chambers. This suggests multiple phases of deposition with a clear intention of optimising the use of space, ie interring as many bodies as possible (Castex et al 2014, 310–12). The careful analysis of the context also suggests that a large number of bodies were disposed here in a relatively short time-span (Castex et al 2014, 310-314). The instance of SS Peter and Marcellinus also serves as an example of the variable articulation and arrangement of bodies within these contexts. In this case, the bodies of adults were placed first and then the empty spaces were filled with children—a very systematic practice that resulted in a seemingly disorderly state (Blanchard 2007, 992–3; Castex et al 2014, 313; Fig 9).

Fig 9 Two successive layers of Tomb 18 from the catacombs of SS Peter and Marcellinus: the east-west orientations (a) are superseded by north-south orientations (b). Mission SSPM 2005-2006. Illustrations: After Blanchard and Castex 2007, fig 3. © Antiquity and authors. Reproduced with kind permission of the authors.

While the placement of the bodies is again not a decisive difference between multiple burials and mass graves, it is a good indicator of the normative or nonnormative treatment of the bodies during the deposition. Though the current terminology emphasises the disarticulated or disorganised nature of mass graves—and many examples of mass graves indeed fit this description—that does not mean that mass graves cannot be planned or systematic. A famous late-medieval example of pre-planning for the use of mass graves is the East Smithfield Black Death cemetery in London. City officials, hearing of the great mortality on the Continent and expecting it to cross the Channel to England, dug a mass burial trench 67 m long and 2 m wide in advance of the arrival of the plague in 14th century ad (Grainger et al 2008, 9–12). Whether orderly and planned in advance, or an opportunistic use of a pre-existing feature, mass graves were primarily about the disposal of corpses in a particular time and place, and none of the burials within them followed the typical social practices of the time.

Pushing aside older remains or collecting them in bone depositories to make place for later burials frequently leads to mass grave-like features, especially in cases where primary and secondary deposition of the remains takes place in the same space (Fabian and Goldfus 2004). And, as recent analyses at the ‘mass grave’ at St Wystan’s Church at Repton, Derbyshire, has shown, not all contexts that look like ‘mass graves’ are, in fact, necessarily primary depositions. Here, the disarticulated remains of least 264 people came to light from a 9th-century context within a disused building (Biddle and Kjølbye-Biddle 2001: Jarman 2021, 29–37). Originally thought to be the bones of the Viking ‘Great Army’ or their victims, or the re-interred bones of Anglo-Saxon monks from the churchyard (who possibly died two centuries earlier), recent scientific testing indicates that at the very least, the disarticulated bones were redeposited as a charnel burial in the late 9th century, and are the remains of people who lived in a number of locations across northern, western, and even southern Europe (Jarman et al 2018). The placement of these individuals in the disarticulated mass grave-like feature in the disused building at Repton was in clear contrast to the contemporary multiple burial of four children at the building’s entrance (Jarman 2021, 37–41; Jarman et al 2018). The seeming ‘mass grave’ at Repton, however, is not a primary but a secondary deposition, more in keeping with secondary contexts like ossuaries.

But a seemingly disorderly placement of the bodies could also be used to indicate meaningful differences in multiple burials as well. In the already mentioned Grave 41 at Hérange, the different orientation of the bodies was used deliberately to express difference between maternally related and unrelated individuals (Deguilloux et al 2018). In the Salme I ship, the bodies are again positioned in a seemingly disordered state, but their placement might have mirrored their position during life (Allmäe et al 2011, 118). In Salme II, 34 bodies were placed on top of each other in four layers. Ancient DNA analysis identified four brothers, all buried relatively close, in the same layer (Margaryan et al 2020, supp 89). Their placement suggests that the layers were carefully planned by individuals who knew the deceased and wanted to express social ties (Fig 7). In the chamber tombs with acrosolia at Atara and Giv’at Sharet, both articulated and non-articulated remains are present as result of prolonged use (Seligman et al 1996, 47–9; Taha 2003, 89–90). These acrosolia, however, are (sequential) multiple burials since the same tomb, the same space, is used to emphasise the social connections—usually family or kinship ties—between the recently deceased and the earlier dead.

CONCLUSION

The publication of new sites and techniques has changed the discussions that can be had about multiple burials and mass graves in the late antique and early medieval periods. In defining and understanding multiple burials and mass graves, the human agency in the past that resulted in them is key, rather than the possibilities of archaeological excavations in the present. It is argued that the terminology should primarily depend on the original intent and purpose behind the deposition of more than one individual in the same archaeological feature, rather than what is possible to find in the ground or identify later on. Here, these two terms are not distinguished according to the number of individuals buried in the feature, the temporality of the deposition of corpses, the construction of the grave structure, or the articulation of skeletons. A distinction between simultaneous and sequential deposition of bodies in a feature over time is important and adds a lot to the understanding of a context but should not be used to differentiate between a mass grave or multiple burial.

These terms operate on a spectrum between primarily functional motivations and actions on the one hand and more noticeably symbolic, social practices on the other. Mass graves—whether opportunistically using pre-existing anthropogenic or natural features to dispose of large numbers of bodies, such as at Ridgeway Hill or St John’s College, or carefully planned and arranged in a predetermined, systematic way, such as those in the catacombs of SS Peter and Marcellinus—were primarily used to dispose of the remains of more than one individual. In mass graves, therefore, most normative burial practices that a community would use to ascribe social identities to each individual deceased were overshadowed by functional purposes, and as such they are positioned closer to the functional end of the spectrum. Multiple burials, in contrast, were created using similar social acts to most single burials, and though they contain the remains of more than one individual, the practices of creating social identities within those features was retained—as seen in the cemetery at Aschheim or the Salme ship burials. Therefore, the proposed new definitions rely upon the interpretive nature of the feature as originally deposited (as best as can be recovered) by living communities in the past, rather than upon the counting of individuals in a feature as excavated by archaeologists in the present day.

Acknowledgements

The authors would like to thank Patrick Geary, Ádám Bollók, Merle Eisenberg, Lee Mordechai, Tim Newfield, and Jordan Wilson for their feedback and comments on earlier versions of this article. We also thank the anonymous reviewers. This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement n° 856453 ERC-2019-SyG). This article received Open Access publication assistance funding from Princeton University.