Making skull cups: Butchering traces on cannibalised human skulls from five European archaeological sites

Highlights

• Cut marks on skulls are often associated with a process of scalping, defleshing, and dismemberment.

• Cut marks on skull cups are distributed in clusters localised to specific areas of the skulls.

• Frequency and clustering of cut marks are related with the intentional preparation of skull cups.

• A high concentration of anthropogenic modifications on skulls may be related to symbolic treatment.

Abstract

The presence of skull cups (bowls made from human calvaria) is considered evidence of the ritualistic treatment of human bodies. These artefacts are characterised by careful manufacturing which can be taphonomically observed in bone surface modifications (BSM) as cut marks and percussion marks. These BSM show morphological similarities across Upper Palaeolithic, Neolithic, and Bronze Age assemblages. This study is focused on the analysis of the frequency and spatial distribution of cut marks on skull cups from Gough's Cave (UK), Herxheim (Germany), and El Mirador Cave (Spain), as compared to the frequency and spatial distribution of modifications on human skulls (non-skull cups) from TD6.2 of Gran Dolina (Spain) and Fontbrégoua (France), with the aim of identifying a common pattern related to a symbolic background. Nearest neighbour analysis and Kernel analyses were used to identify the distribution pattern of anthropogenically induced modifications. The results indicate that the frequency and distribution of cut marks on human skulls modified into skull cups are unique and are most likely to be the result of meticulous cleaning of skulls. A similar frequency and distribution pattern of modifications was also observed on skulls from Fontbrégoua, possibly related to the collection of skulls as war trophies. No parallels with the treatment of skulls of Homo antecessor at TD6.2 of Gran Dolina were observed. We suggest that the treatment of human skulls for ritualistic purposes therefore results in a consistent pattern of modification.

Introduction

The ritual treatment of skulls has been recorded in numerous archaeological sites of different chronologies and geographical areas. Evidence of skull manipulation can include peri- and post-mortem decapitation; skull mask production, skull caching and secondary depositions, decorated carved skulls, and skull cups are among the most common forms of manipulation, and extend around the world from the Upper Palaeolithic until the Contemporaneous age (eg. Campillo, 1976; Villa et al., 1986a, 1986b; Le Mort and Gambier, 1991; Owsley, 1994; Massey and Steele, 1997; Ostendorf-Smith, 1997, Ostendorf-Smith, 1995; Botella et al., 2000; Verhoeven, 2002; Schaafsma, 2007; Bello et al., 2011, 2015; Boulestin, 2012; Boulestin and Henry-Gambier, 2012; Boulestin and Henry-Gambier, 2019, Carod-Artal, 2012; Green, 2012; Jeunesse, 2012; Jammo, 2014; Santana et al., 2019). In past societies, human skulls were honoured because it was believed they possessed vital powers or life force, or they were collected as proof of superiority and authority (Verhoeven, 2013; Jammo, 2014). Enemy skulls collected during warfare also demonstrate specific treatment. An example is the human bone assemblage from the Iberian oppidum of Ullastret (Spain), where skulls were perforated with iron nails, suggesting that they were included as part of a costume displaying the skulls of dead enemies (Verhoeven, 2013).

Different taphonomic proxies help us to recognise possible ceremonial practices. The presence of skulls without associated postcranial remains suggests the possible intentional removal of these elements (Hurlbut, 2000 and inter alia). For example, in the Neolithic site of Fontbrégoua (France), Villa et al. (1986b) and Courtin (2000) interpreted the absence of skulls in one of the three deposits as a possible ritual associated with war trophies. In another geographical and chronological context, Massey and Steele (1997) described the presence of three isolated human skulls in a Mayan pit in Belize, with only some associated cervical vertebrae. These skulls were cut-marked, suggesting that the heads were scalped and the flesh was removed after decapitation.

The most common anthropogenic modifications in association with ritual treatment in the archaeological record are scalping and the intentional breakage in the completion of specific morphologies as skull cups. Scalping is well documented in pre-historic and historic North American assemblages (Seeman, 1988; Miller, 1994; Owsley, 1994; Ostendorf-Smith, 1995, 1997; 2003; Murphy et al., 2002; Toyne, 2011), and is identifiable by a specific distribution of cut marks. According to the description of Ostendorf-Smith (1997: 246) the pattern usually consists of a series of cut marks made in a somewhat circular pattern on the crown of the head. They are most commonly found along the hairline region of the frontal bone, on the mid-section of the parietal bones, or on the suprameatal crest, around the temporal bone and the nuchal crest of the occipital bone. Trophy skulls from the Great Plains region of North America are also characterised by holes drilled on the top or sides of the skulls to secure cords for suspension (Owsley, 1994). Archaeological excavations at Göbekli Tepe, a transitional Neolithic site in southeast Turkey, have revealed several fragmented human bones recovered from fill deposits of buildings and adjacent areas. Three partially preserved human skulls carry artificial modifications of a type previously unidentified in assemblages of the same period. Taphonomic research documented four types of intentional modifications: one drilled perforation, three cases of carvings, the application of colour, and smaller cut marks (partly or not related to carvings). Gresky et al. (2017) have suggested that these remains and modifications could indicate a cult of the skull in the Early Neolithic of Anatolia and the Levant. Likewise, Haddow and Knüsel (2017) suggest other anthropic activity identifiable on burials from Çatalhöyük (Turkey), where removal and recombination of the skeletal elements with an eventual reintegration or substitution, was identified. Other sites such as Azraq 18 (Jordan), also reveal anthropic manipulation of the heads on several burials with painted or plastered skulls (Bocquentin and Garrard, 2016). Otherwise, similarly to Göbkely Tepe, Bocquentin and Aoudia-Chouakri (2009) associate the absence of the lower part of a skull and three perforations with its use as a mask. According to the authors, this skull from Fäid Sounar II (Capsien, Algérie), can linked ritually as a war trophy or as part of a funerary rite.

Skull cups have been recognised in European prehistoric assemblages dating from the Upper Palaeolithic to the Bronze Age (Le Mort and Gambier, 1991; Bello et al., 2011; Boulestin, 2012; Solari et al., 2012; Boulestin and Coupey, 2015; Saladié and Rodríguez-Hidalgo, 2017; Santana et al., 2019) (Fig. 1). The meticulous breakage of the skulls from Gough's Cave (Upper Palaeolithic, United Kingdom), suggests that the modifications were not necessarily driven by the need to extract the brain for nutritional purposes, rather they were precisely and intentionally produced to shape the skulls into containers or drinking bowls (Bello et al., 2011). The intentional breaking of skulls for the manufacturing of skull cups is currently a diagnostic element for inferring ritual connotations in the treatment of human corpses from European prehistoric assemblages (Bello et al., 2015). Similar modification patterns to those recognised in Gough's Cave have been observed at Le Placard (Upper Palaeolithic, France; Le Mort and Gambier, 1991; Boulestin and Henry-Gambier, 2019), Herxheim (Neolithic, Germany; Boulestin and Coupey, 2015), Cueva de la Carigüela (Neolithic, Spain; García-Sanchez and Carrasco-Rus, 1981), Cueva de El Toro (Neolithic, Spain; Santana et al., 2019), El Mirador (Bronze Age, Spain; Cáceres et al., 2007), and Cueva de Txispiri-Gaztelu (Bronze Age, Spain; Ruiz de Gaona, 1945). In most of these sites, the special treatment of skulls was associated with cannibalistic events.

The number of recognised occurrences of prehistoric human cannibalism in the old world has increased in the past few years (Villa et al., 1986a, 1986b; White and Toth, 1991; Fernández-Jalvo et al., 1996; Patou-Mathis, 1997; Botella and Alemán, 1998; Botella et al., 2000; Andrews and Fernández-Jalvo, 2003; Maureille et al., 2004; Barroso and de Lumley, 2006; Rosas et al., 2006; White and Toth, 2007; Cáceres et al., 2007; Boulestin et al., 2009; Carbonell et al., 2010; Bello et al., 2011; Saladié et al., 2012; Solari et al., 2012; Bello et al., 2015; de Lumley, 2015; Boulestin and Coupey, 2015; Rougier et al., 2016; Santana et al., 2019). The increase in findings of human assemblages promotes a greater understanding of this behaviour and enables us to identify different aspects of it. Prehistoric cannibalism has been linked to intergroup violence, possible periods of starvation, and funerary contexts. Ritualistic behaviour can be associated with the first and last of these situations. In archaeological assemblages, however, it is not always possible to identify the ritualistic treatment. Many ceremonies with symbolic connotations do not necessarily become archaeologically visible. In addition, different treatments may be equifinal with events in which the consumption of human flesh is not related to any other deeper feelings (Saladié and Rodríguez-Hidalgo, 2017). Some researchers have tried to define features of archaeological assemblages that could be considered ritualistic. For Villa et al. (1986b: 144), a secondary burial of human remains is an indication of ritualization. The deposition of human bodies in a different context than other faunal remains has also been considered evidence of a ritualistic treatment of cannibalised bodies (Villa et al., 1986b; Villa, 1992; White and Timothy, 1992; Bello et al., 2016). Other findings, such as the presence of shaped bones, engravings, or the aforementioned skull cups, have been accepted as evidence of a ceremonial component with a marked ritualistic background (Villa, 1992; Wallduck and Bello, 2016; Bello et al., 2017; Saladié and Rodríguez-Hidalgo, 2017).

The skulls that appear in cannibalised contexts are usually characterised by the presence of abundant cut marks. The disposition and frequency of bone modification can be assessed through different proxies. In recent years, different researchers have developed evaluation systems in order to analyse the location of the modifications through geostatistical tools (Nilssen, 2000; Marean et al., 2001; Abe et al., 2002; Parkinson, 2013, 2018). This approach essentially treats each anatomical element as a ‘map’ onto which the surface modifications can be recorded. In this paper, we aim to assess whether it is possible to identify a pattern specific to the manufacture of skull cups by comparing evidence from different prehistoric cannibalistic assemblages in Europe. To this end, we have compared the frequency and distribution of cut marks on skull fragments from TD6.2 (Gran Dolina) (Saladié et al., 2012), Gough's Cave (Bello et al., 2011), Fontbrégoua (Villa et al., 1986b), Herxheim (Boulestin and Coupey, 2015), and El Mirador Cave (Cáceres et al., 2007; Saladié, 2009) (Fig. 2). Cut marks were spatially plotted as polylines over bone templates in ArcGIS, which allowed us to evaluate their presence and distribution in different views of the human skull. It has been proposed that all samples, except those from TD6.2, were involved in rituals or even cannibalistic events. In three of the sites (Gough's Cave, Herxheim, and El Mirador) the elaboration of skull cups was recorded.

Our main aim is to treat the cut mark distribution over the skulls through the spatial statistics methods. In that case, cut marks are treated as objects with spatial characteristics under a delimitated area and the skull surface is used to evaluate the distribution of cut marks identifying spatial patterns. This method can help us to recognise a special manipulation of human skulls in different archaeological contexts, helping to solve the problematic of interpreting wrongly the disposition of cuts and its functionality.

This methodology allows to identify specific anthropic modifications related to human behaviour and describing statistically the significance, in that case, of the accumulation of cut marks in specific areas and probably related to ritual events in contexts where the human cannibalism had also been identified.