Journal of Anthropological Archaeology 28 (2009) 412–423 Contents lists available at ScienceDirect Journal of Anthropological Archaeology journal homepage: www.elsevier.com/locate/jaa The Eulau eulogy: Bioarchaeological interpretation of lethal violence in Corded Ware multiple burials from Saxony-Anhalt, Germany Christian Meyer a,*, Guido Brandt a, Wolfgang Haak a, Robert A. Ganslmeier b, Harald Meller b, Kurt W. Alt a a b Institute of Anthropology, University of Mainz, Colonel-Kleinmann-Weg 2 (SB II), 55128 Mainz, Germany State Office for Heritage Management and Archaeology Saxony-Anhalt/State Museum of Prehistory, Richard-Wagner-Str. 9, 06114 Halle/Saale, Germany a r t i c l e i n f o Article history: Received 1 April 2009 Revision received 18 July 2009 Available online 15 August 2009 Keywords: Neolithic Europe Fractures Trauma Palaeopathology Mortuary customs Kinship a b s t r a c t The Corded Ware is one of the major archaeological traditions of Late Neolithic Europe. Its burial customs are characterized by single graves but multiple burials also occur. We present a detailed study of antemortem and perimortem trauma in a group of Corded Ware skeletons from four multiple graves and give the most probable interpretation of the site, based upon all available bioarchaeological evidence. The pattern of observed injuries in male, female, and subadult skeletons, including cranial trauma, arrow wounds, and fractures of the forearm and hands points towards a violent event that resulted in the death of all individuals, most probably a raid. In contrast to comparable Neolithic raid sites, there was no complete extermination of the local population and no use of mass graves. The burials have been arranged with care and detailed knowledge about biological kinship ties [Haak, W., Brandt, G., de Jong, H.N., Meyer, C., Ganslmeier, R., Heyd, V., Hawkesworth, C., Pike, A.W.G., Meller, H., Alt, K.W., 2008. Ancient DNA, strontium isotopes, and osteological analyses shed light on social and kinship organization of the Later Stone Age. Proceedings of the National Academy of Sciences of the United States of America 105, 18226–18231]. The combination of clear causes of death and the proven biological relationships among some of the individuals, including a nuclear family, provides new and important insights into Corded Ware mortuary customs and the reasons why and how multiple graves have been utilised. Ó 2009 Elsevier Inc. All rights reserved. Introduction The documentation, analysis, and interpretation of interpersonal violence in the human skeletal record is one of the main topics of palaeopathology and bioarchaeology (e.g. Jurmain, 2001; Walker, 2001; Smith et al., 2007; Judd, 2008) and has profound implications on how archaeological cultures are perceived (Guilaine and Zammit, 2005). Violent behavior has always been one of the ways to resolve intra- and intergroup tensions in human societies and has been described as ‘‘a powerful expression of human social interaction” (Schulting and Wysocki, 2005, p. 107). Skeletal remains which bear the scars of injuries have been found in many locations and time periods, from the Palaeolithic onwards to the present day (e.g. Meyer, 2003; Roksandic et al., 2006; Paine et al., 2007). The European Neolithic is no exception in the history of human violence, although sites which have yielded large numbers of affected skeletons are rather rare. In this paper, we present further evidence for directed interpersonal violence during this * Corresponding author. E-mail address: meyerc@uni-mainz.de (C. Meyer). 0278-4165/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jaa.2009.07.002 period and follow up on the biocultural implications of the injuries recorded for the people of the Corded Ware Culture. One of the most significant points that we want to raise is that multiple graves, like the ones encountered at Eulau, may not be mere convenient structures to dispose of several people who all happened to die within a short time period. Instead, here we have definite proof of simultaneous death, cause of death, and the underlying biological and social relations of the buried individuals. Combined with the genetic and isotopic analyses (Haak et al., 2008), the results gleaned from the analysis of the multiple graves from Eulau reveal new and important insights into the Late Neolithic of Central Europe. Towards the end of the Neolithic, the cultural complex of the Corded Ware forms one of the major archaeological entities of the European continent. First occurring around 2800 cal BC in our study area of the Mittelelbe-Saale, remains of the Corded Ware Culture have been found from Central to Northeastern Europe (Haak et al., 2008). The foremost source of information from this period are the graves and their content as well-preserved settlement structures are rarely encountered. It seems that small villages with houses no larger than about 10 m predominate, and that the Corded Ware people carried out agriculture and animal husbandry C. Meyer et al. / Journal of Anthropological Archaeology 28 (2009) 412–423 on a developed level (Hecht, 2007). In contrast, graves are known in high numbers, either as singular finds or loosely clustering in small cemeteries (Buchvaldek, 1967). Overlapping in time with the Bell Beaker culture, the Corded Ware comes to a close about 2050 cal BC and is followed by the Únětice Culture and the Early Bronze Age. The burial rites of this culture, which have direct consequences for the amount of information available to bioarchaeological interpretation, follow a rather strict pattern in our region of study. The dead are usually placed in a flexed position in single graves. Males rest on their right side, females on their left side, both sexes facing south. As a result, men have their head to the west, with the heads of women pointing to the east. Exceptions to this rule are rare. It is assumed that most of the graves originally possessed a small mound, which is often unrecognizable today due to centuries of agricultural activity. Multiple Graves, the simultaneous inhumation of several bodies, are a rather rare but constant feature, more so than in other Neolithic cultures of the region (Wahl et al., 1990). As one of the most important and widespread Neolithic cultures in Europe the skeletal remains of the Corded Ware people have been a focus of scientific attention from an early stage. Questions as to their place of origin and their morphological affinities to other groups played the major role in the earlier investigations. Craniotypology, followed by craniometry were the primary tools of the anthropologists of that age (e.g. Heberer, 1938; Schwidetzky, 1978) and other aspects of anthropological analysis were rarely written about. Recognized palaeopathological changes, whose significance and bioarchaeological interpretation are the main concern of this paper, were only mentioned in passing or presented in short notices. The majority of these short communications deal with trepanations, which are not uncommon in Corded Ware crania (e.g. Bach, 1962/1963, 1963). Fractures and wounds presumably caused by accidents or interpersonal violence form another but related topic, which has received considerably less attention (e.g. Bach, 1963; Feustel et al., 1966). Modern studies explicitly dealing with palaeopathology are quite rare for the Corded Ware (Schutkowski et al., 1996; Dittmar and Teegen, 2003; Dresely, 2004), and attempts to get results at the population level for the whole of the Corded Ware are even rarer (e.g. Bruchhaus and Neubert, 1999). Although this paper is essentially another case study, albeit with a rather high number of individual cases, the quality and condition of the skeletal remains presented below and the high number of recognized violent injuries allow us to draw conclusions which are in fact highly relevant for Corded Ware mortuary customs and society as a whole. The main objectives of the present paper can be stated as the detailed documentation of the violent injuries encountered and their interpretation in a bioarchaeological context as well as highlighting the importance of interdisciplinary studies when dealing with complex burial situations. Furthermore, regarding the undisputable evidence for Neolithic warfare in its broadest sense in Central Europe, we still only have very scatchy data indeed, as only very few sites with a higher number of violently killed individuals are published in detail (e.g. Wahl and König, 1987). As the Neolithic spans several millennia, it is of the highest importance to add more substantial evidence to the still emerging picture of Neolithic violence and warfare. Multiple burials of the Corded Ware Culture Compared to other major Neolithic cultures without truly collective burial rites, Corded Ware multiple burials appear quite frequently. About 44% of graves in two cemeteries located in the Tauber Valley in southern Germany contain more than one individual, whereas this figure is somewhat lower in another Tauber Valley cemetery with 17%. In other Corded Ware cemeteries, such as 413 Vikletice in Bohemia, multiple burials number about 10% (Dresely, 2004; Menninger, 2008). Well-published cemetery sites from other major Neolithic cultures usually yield no or very few multiple inhumations, rarely exceeding about 3% if they are present at all (e.g. Lambach, 1993; Nieszery, 1995; Spatz, 1999). In most cases these graves include only subadults or a combination of subadult and adult individuals (Bulla, 1998; Meyer et al., 2004). The extremely high number of double and multiple burials in the Tauber Valley can be explained by the fact that most of these graves were actually used as small-scale collective burials, which are known, in larger scale, from other Neolithic cultures (Meyer et al., 2008). The obvious difference between true multiple and collective burials is the number of inhumation events. Most Neolithic collective burial places are constantly accessible during their functional life-time, and placement of new bodies is carried out repeatedly. True multiple burials are characterized by the simultaneous inhumation of two or more individuals in a grave, which has been prepared for this single event and is closed up permanently afterwards, precluding further use. This contrasts with known Neolithic mass graves, where the bodies form a careless and jumbled heap (Wahl and König, 1987; Meyer et al., 2004), but are nevertheless interred simultaneously. Bodies in multiple burials are usually arranged according to the relevant burial customs of the period. Two main questions often arising when dealing with multiple graves, are: (1) Why did the individuals buried together die at roughly the same time? (2) Are they biologically related? The answer to the first question often has to remain speculative but can sometimes be answered by osteological analysis. In the case of the Eulau graves, it has been possible to establish a likely cause of death (see below). The question of kinship often invites speculations, especially when adults and children are encountered in a single grave. The most frequent interpretation of such a scenario is that of parents interred with their offspring, and an adult female associated with a subadult is almost certainly regarded as a mother with her child (e.g. Buchvaldek, 1967; Wahl et al., 1990). In most cases final proof for all these assessments is unavailable. In our case, it actually has been possible to prove genetic relationships for some of the individuals buried in the four multiple graves via the analysis of mitochondrial and nuclear DNA. The most relevant result has been the identification of a nuclear family and a likely pair of siblings (Haak et al., 2008). This amount of detail has up to now not been available for the reconstruction of mortuary practises of the European Neolithic. The multiple burials from Eulau, Saxony-Anhalt, Germany At Eulau, near the town of Naumburg, excavations in the area of a large open gravel mine revealed four closely grouped multiple burials (Figs. 1 and 2). Associated artefacts and overlapping radiocarbon dates between 2675 and 2495 cal BC put these graves into the context of the Corded Ware Culture (Haak et al., 2008). Two of the graves contained four individuals, one grave held three bodies and the last one contained the remains of two individuals. There was no archaeological evidence of later disturbance or re-use of the graves, and the size of the grave-pits closely corresponds to the amount of space needed for the placement of the bodies. These facts indicate true multiple burials. The skeletons were lifted as blocks from the field and are presently on permanent display at the State Museum of Prehistory in Halle (Saale), Germany. As a consequence, the skeletons had to remain largely in situ during most of the analyses and only small bones or truly essential elements could be temporarily removed. It was therefore necessary to reach a compromise between a detailed osteological study and the best possible preservation of the original context, the latter being given priority. Following standard osteological procedures 414 C. Meyer et al. / Journal of Anthropological Archaeology 28 (2009) 412–423 Fig. 1. In situ photographs of the four Corded Ware multiple burials from Eulau, Germany. (A) Grave 6-0099 with Inds. 1–4. (B) Grave 6-0090 with Inds. 5 and 6. (C) Grave 60098 with Inds. 7–10. (D) Grave 6-0093 with Inds. 11–13. (White and Folkens, 2005), the age and sex diagnosis resulted in the identification of two adult males, three adult females, and eight subadults (Table 1). The morphological sex diagnosis was validated and complemented by DNA analyses including the sexing of the subadults (Haak et al., 2008). Trauma analysis also followed established guidelines (Lovell, 1997; Rogers, 2004). Radiographic documentation was carried out when possible and deemed necessary. Osteological results Case descriptions are only given for those individuals who show evidence of healed or unhealed trauma. A complete osteological catalog is beyond the scope of this article and will be published at a later date along with more detailed archaeological results for the site. The different traumatic lesions are described in the neces- sary detail though, to enable an objective evaluation of our interpretation (Lovell, 2008). Although some injuries presented below are unambiguous signs of interpersonal violence, others may well be the result of accidents instead. This has to be kept in mind when attempting a bioarchaeological reconstruction. Grave 6-0099 Ind. 2. This 4–5 year old boy has a small and healed depressed fracture of the frontal bone, measuring about 0.8  0.5 cm. As fractures of this type are common results of childhood accidents, it is not believed to be a result of interpersonal violence, which can, of course, not be ruled out entirely (Schulting and Wysocki, 2005). Ind. 3. This 40–60 year old man suffered from several injuries, some received years before death, some perimortem. All lesions 415 C. Meyer et al. / Journal of Anthropological Archaeology 28 (2009) 412–423 Fig. 2. Map of Central Europe with Germany (dark grey), the state of Saxony-Anhalt (light grey) and the site of Eulau (star) depicted. Table 1 Sex, age, genetic kinship, and evidence for ante- and perimortem skeletal trauma for all individuals. Proven kinship within a grave indicated by , genetic data from Haak et al. (2008). Ind. Kin Sex Age (y) Antemortem injury Perimortem injury Grave 6-0099 1 2 3 4     F M M M 35–50 4–5 40–60 8–9 – – Left humerus, left radius, left hand – – – Right hand; left hand Cranial trauma (Os occipitale) Grave 6-0090 5 6 F – 25–35 4–5 – – Arrow wound (lumbar vertebra 4) – Grave 6-0098 7 8 9 10 F – M F 30–38 0.5–1 7–9 4–5 – – – – Cranial trauma (Os parietale) – – – M – – 25–40 4.5–5.5 4–5 Left radius, left ulna, left hand – – Right ulna; right radius; right hand – – Grave 6-0093 11 12 13   are found along the left arm, including the humerus, radius, metacarpals and phalanges. Approximately at mid-shaft, the humerus displays an exostosis about four centimeter in length (Fig. 3). The location corresponds to the insertion of the m. deltoideus. An unspecified trauma to this muscle followed by ossification appears to be the most likely explanation (Mann and Hunt, 2005). Another old injury can be found at the distal radius. The bone had suffered an oblique to spiral fracture about three quarters down the shaft 416 C. Meyer et al. / Journal of Anthropological Archaeology 28 (2009) 412–423 Fig. 3. Large exostosis of the left humerus of Ind. 3 most probably due to myositis ossificans. Scale bar length is 3 cm. Fig. 6. Perimortem spiral fracture of a manual phalanx of Ind. 3. Scale bar length is 1.5 cm. Fig. 4. Left radius of Ind. 3 with evidence of a healed fracture (left). CT image with a slightly different orientation (right). Visible fracture line in the top half of both images is postmortem damage. Scale bar length is 3 cm. and has healed with a slight medial displacement and negligible misalignment (Fig. 4). The more proximal end of the fracture space was located on the medial side of the bone. The third and final healed injury involves the proximal interphalangeal joint of the fourth digit. Bony ankylosis has fused the two bones through the joint, resulting in complete immobilisation. Both bones now share a common medullary cavity (Fig. 5). Most likely this finger was subjected to compression forces which resulted in an impacted fracture of these phalanges. As all of these injuries show a similar degree of healing and complete remodelling of cortical surfaces, it may well be that they were received at the same occasion. Further traumatic lesions, which can be deemed perimortem, involve both hands. The right hand suffered fractures to the metacarpals of the second (distal), third (proximal), and fourth digit (proximal). Also affected is one phalanx, probably of the fifth digit (Fig. 6). Fractured bones of the left hand include the metacarpals of the third (distal) and fourth (distal) digit, as well as the basal phalanges of the second (proximal), third (medial), and fourth (proximal) digits. All these fractures of the small bones of the hands are either oblique or spiral fractures. Destruction of the left frontal skull might also be due to perimortem violence, but this could not be securely assessed as the bones had to remain in situ. The same can be said for the destruction of the skull of associated Ind. 1, which showed no further signs of injury. Soil pressure, which is the evident reason for the destruction of most subadult skulls in this sample, might be an alternative explanation. Ind. 4. This 8–9 year old boy suffered a comminuted penetrating wound to the back of the skull, creating a roughly triangular hole measuring about 1.5  2.5 cm. Concentric and bending fractures can be seen in the direct vicinity of the impact site (Fig. 7). Small bone fragments projecting into the interior of the skull were still visible and adhering to the wound margins on examination. Fig. 5. Finger bones of Ind. 3 with complete bony ankylosis, most probably due to an old fracture. Scale bar length is 3 cm. Grave 6-0090 Ind. 5. A transverse flint arrowhead is firmly lodged in the fourth lumbar vertebra of this 25–35 year old woman. The projectile point entered the vertebral body from the front and left side, indicating a ranged attack from this direction. Neither CT-scans nor macroscopic examination of the dry bone surfaces revealed any bony reaction in the area surrounding the stone tip, indicating that the wound was received around the time of death (Ryan and Milner, 2006). A second transverse arrowhead was found lying between the ribs of the left side and might be evidence of another injury. C. Meyer et al. / Journal of Anthropological Archaeology 28 (2009) 412–423 Fig. 7. Perimortem cranial trauma on the occipital bone of Ind. 4. Arrows indicate penetrating fracture margins. Scale bar length is 3 cm. No bone damage was noted, but the poor preservation of the ribs would certainly have masked any evidence of trauma (Fig. 8). Grave 6-0098 Ind. 7. This 30–38 year old woman received two blows to the right upper portion of the cranial vault (Fig. 9). The first wound is located on the right parietal, halfway along the sagittal suture and 1.5 cm lateral from it. The outer dimensions of this penetrating and comminuted cranial fracture measure about 6.0  1.8 cm. Radiating fractures follow the contours of the vault. The interior of the wound area is made up of four larger bone fragments which have been fractured in a cross-like pattern. The intersection point of the fracture lines has been depressed into the cranium for approx. 1.5 cm and all fragments are tilted towards this central spot. The edges of the bones display internal bevelling, indicating a penetrating force from the outside to the inside of the skull. 417 Fig. 9. Skull of female Ind. 7 with two cases of perimortem penetrating blunt force trauma to the right parietal bone. Arrows indicate penetrating fracture margins. The second wound is located anterior to the first, the coronal suture forming the anterior border of this large defect which measures approximately 6.0  2.0 cm. Again, a number of small bone fragments are tilted towards a central spot in the interior of the wound, the deepest point of fragment intrusion into the cranial vault being 2.0 cm. Internal bevelling is evident along the visible rims of the fracture. One fracture line of this wound terminates in the defect described above, indicating that the anterior fracture occurred later than the posterior one (Puppe, 1914; Lovell, 1997). No signs of healing or remodelling are evident, indicating that both blows were delivered perimortem. Grave 6-0093 Ind. 11. This 25–40 year old man is the most robust individual of this series, but due to adverse conditions during excavation and preparation of the block also the least well preserved of the adults. Neverthe- Fig. 8. (A) Transverse flint arrowhead in the left chest area of Ind. 5. (B) Another transverse flint arrowhead lodged in the fourth lumbar vertebra of Ind. 5. (C) CT scan of the same vertebra showing the morphology of the cutting edge of the tip. Scale bar length is 2 cm. 418 C. Meyer et al. / Journal of Anthropological Archaeology 28 (2009) 412–423 Fig. 10. Remodeled joint surface of the left distal radius of Ind. 11 with evidence of trauma-induced osteoarthrosis. Scale bar length is 3 cm. less, several injuries could be detected. Like the other male (Ind. 3) this man has a healed fracture of the left radius. The styloid process has been proximally displaced for up to 0.5 cm and has healed in this position. Osteoarthrosis developed as a direct result of this trauma and the accompanying alteration of the original joint morphology. It affected most of the radial articular surface, which appears extensively remodelled (Fig. 10). The displaced fragment of the original articular surface remains largely unchanged. The distal articular surface of the left ulna appears remodelled as well. Further traumatically induced changes affect the scaphoid, which suffered a transverse fracture of the proximal half. The two (or possibly more) resulting fragments did not fuse during the healing period (Fig. 11) and the smaller fragment(s) could not be located during the osteological analysis. The fracture surface of the remaining scaphoid body has been completely remodelled. Injuries received perimortem are found on the right forearm and hand. Both, radius and ulna suffered oblique and comminuted fractures. The separated distal part of the limb, including the hand, has been displaced proximally and medially. Also fractured are the metacarpals of the second (medial), third (medial), and fourth (distal) digit (Fig. 12). All fractures are oblique. There is no sign that the observed displacement of elements resulted from postmortem processes or soil movements. Fig. 12. Right hand and distal forearm of Ind. 11 with displacement of the distal fragments of ulna and radius as well as fractures to some metacarpals. Probable perimortem fractures of the smaller bones are indicated by . Note also clearly postmortem damage to bones on the top left. Scale bar length is 3 cm. Discussion Analysis of cranial and postcranial skeletal trauma remains a major focus in palaeopathological research and the most direct evidence for interpersonal violence in the archaeological record. Fractures often result in gross anatomical changes and are therefore rather easily identified (Jurmain, 2001; Lovell, 2008). Trauma is also very well suited to and used as one aspect of bioarchaeological reconstructions of past populations, as it can give insights into the risks of everyday life and episodes of violent behavior in human history. This is especially true when a larger number of people apparently died at the same time and of similar causes (e.g. Wahl and König, 1987; Willey, 1990; Meyer, 2003; Bennike, 2008). This appears to be the case at Eulau. The burials form true multiple graves, without evidence of postburial disturbance or later re-use of a grave. It could be demonstrated that five out of the 13 skeletons analysed show clear signs of perimortem violent injury, among them most of the adults and one male child. There is at least one individual with perimortem injuries from each of the multiple graves, in the best preserved burial 6-0099 (Fig. 1a) there are two. The condition of the more than 4000 year old skeletal remains and the fact that they had to remain largely in situ during the analysis mean that more injuries Fig. 11. Unchanged scaphoid bone of Ind. 3 (left) compared to the altered scaphoid of Ind. 11 (right). Part of the bone is missing due to an old fracture. Scale bar length is 3 cm. C. Meyer et al. / Journal of Anthropological Archaeology 28 (2009) 412–423 may be present than could be identified during this study. Although the documentation of skeletal trauma might therefore not be complete, further evidence of trauma would only strengthen the interpretation given here, namely that all of the individuals buried in the multiple graves at Eulau died as a result of a single, violent episode during the Late Neolithic. Further, it is well known that not every violent encounter or even every killing results in observable lesions on the skeleton (Schutkowski et al., 1996; Rogers, 2004; Smith et al., 2007). In many cases there is no skeletal involvement, leading to the conclusion that the wounds which actually can be observed are only the ‘‘tip of the iceberg” (Walker, 2001, p. 584). This is well illustrated by the Neolithic mass grave from Talheim in southwestern Germany, which contained the skeletal remains of 34 individuals. Perimortem injuries of the skull have been found in 59% of this population, as well as some postcranial fractures and arrow wounds. The widely accepted biocultural interpretation of this mass grave is that the almost complete population of a Neolithic hamlet has been killed in a violent attack (Wahl and König, 1987). In this instance, the people without any observable skeletal trauma are believed to have met a violent end as well, as they apparently died at the same time as the persons with clear perimortem injuries. Another illustrative study found that only about a third of historically documented arrow injuries to the human body actually hit bone (Milner, 2005). Therefore the number of lethal injuries in the archaeological record is certainly much higher than can be recognized from osteological analyses alone, especially in the case of arrow injuries. Although massive cranial trauma certainly has a high chance to damage bone and is thus recognized more easily in archaeological human remains, bad preservation of the bones may mask or destroy these signs, which also leads to an underestimation of violence in prehistory. Antemortem injuries When interpreting trauma in the skeletal record, the distinction between wounds received antemortem, perimortem, or postmortem is of the highest importance, as this forms the basis for reconstruction of events (Bennike, 2008). When looking at the Eulau bones, the injuries observed on the left humerus, radius, and phalanges of Ind. 3 (Figs. 3–5) and on the left radius and scaphoid of Ind. 11 (Figs. 10 and 11) clearly occurred antemortem. All are well-healed, indicating that they were received at least some years prior to death. It is important to note that only the two male individuals have healed traumatic injuries and that these injuries are, in fact, very similar. In both cases the left arm was affected with involvement of the distal radius and the hand. In the palaeopathological literature fractures to the distal radius and scaphoid are most often interpreted as a result of falls or other accidents, as is the case in the modern clinical literature (e.g. Grauer and Roberts, 1996; Smith, 1996; Walker, 2001; Lovell, 2008). Although violence cannot be ruled out (Kunter, 1974), the classic mid-shaft parry fractures of the forearm, where the ulna is most commonly affected, are usually seen as better indicators of interpersonal violence (Smith, 1996; Lovell, 1997; Judd, 2008). The parry fracture, as an example of an osteologically visible defense wound, can be explained by raising the arm to deflect an incoming blow during hand-to-hand combat. Fractures of the distal radius are usually explained by a fall on the out-stretched hand. Although such a fall may well have been an accident, it may have also been the result of a violent attack. Chances of stumbling or falling are surely higher while fleeing in panic or trying to defend oneself from an armed attacker. Similar thoughts have recently been voiced by Judd (2008), who aptly called this an interpretational predicament. While trying to deflect blows delivered with Late Neolithic close combat weaponry, injuries such as classic parry fractures of the forearm as well 419 as fractures to other parts of the arms may certainly occur. The crushing force of stone axes and adzes may well lead to fractures and injuries at the actual point of impact, which is not necessarily at mid-shaft of the ulna. Another point rarely taken into account when interpreting Neolithic fighting or combat is the possible use of shields or other defensive weapons. Shields attached to the forearm or just held in hand may certainly alter the pattern of fractures seen in skeletal remains; however, battle victims are usually removed from the original context and it is usually impossible to say if a person actually used a shield or similar means of defence. Although there seem to be no archaeological finds of Neolithic shields in Central Europe, these appear in highly developed forms in the following Bronze Age. These finds are made of metal, but there is consensus that many more were made of organic materials like wood or leather, which simply did not survive in the ground. The sudden appearance of fully developed metal shields does strongly suggest a tradition of shield use which may well stretch back into the Late Neolithic (Uckelmann, 2004/2005). Unfortunately, as there is a lack of evidence for the effects of prehistoric shields on the distribution of blunt force trauma, any thoughts on this matter have to remain speculative. Although the radius fractures in the male skeletons from Eulau may of course be the results of everyday accidents or falls, there actually is no easy distinction between postcranial fractures due to interpersonal violence or accidents, as forced and unintended movements resulting in bone trauma may be part of hand-to-hand combat. This is especially relevant when securely violence-related injuries, such as cranial trauma or projectile injuries, occur along with fractures usually interpreted as accidental. The presence of both helps to support the interpretation of a given injury as non-accidental. The bioarchaeological interpretation of such injuries has to make use of all available information to arrive at the most plausible explanation. In the case of the Eulau skeletons there is also a healed fracture of a scaphoid bone and an impacted fracture of two manual phalanges. The scaphoid fracture can be explained as a result of the same impact that resulted in the distal radius fracture of Ind. 11, as this bone directly articulates with the radius. The healed trauma to the phalanges of the hand of Ind. 3 may also be related to the same event that caused the other antemortem lesions. Fractures of the hand may be defensive wounds (Milner et al., 1991; Lovell, 1997; Rogers, 2004), or occur while attacking with the unprotected hand (Brink et al., 1998). It is therefore possible that the hand fractures occurred in a combat situation, raising the possibility that the other fractures are non-accidental as well. In conclusion, one cannot be sure about the events that led to the antemortem lesions observed in the two male individuals. Occupational hazards primarily affecting the male part of the population may be one explanation (Lovell, 2008), injuries received during hand-to-hand fighting may be another. What can be stated is that although they survived these accidental or violent events, both men are likely to have suffered some functional impairment resulting from their injuries as indicated by the presence of ankylosis and osteoarthrosis, respectively. Perimortem injuries Projectile points embedded in bone are universally seen as the best and most direct type of evidence of interpersonal violence in ancient times (Lambert, 1997; Lovell, 1997; Walker, 2001; Milner, 2005; Smith et al., 2007). Neolithic arrowheads were either made of flint (or related material) or bone, the former being by far the most common material in Central and Northern Europe. Cases of embedded projectile points in prehistoric skeletal remains from Europe are relatively rare but constant findings, and are known from a wide range of different archaeological cultures (e.g. Wells, 1964; Wahl and König, 1987; Roksandic et al., 2006). Recently, a 420 C. Meyer et al. / Journal of Anthropological Archaeology 28 (2009) 412–423 comprehensive list of known arrow injuries from European sites has been included in a book by Guilaine and Zammit (2005). Although prehistoric projectile wounds sometimes also affected the cranium, depressed or comminuted fractures due to blunt force trauma are more common cranial injuries (Wahl and König, 1987; Walker, 1989; Lambert, 1997). When blows were meant to kill, the head was and often is selected as the primary target. Numerous cranial injuries with lethal outcome from different sites and times clearly indicate this pattern (e.g. Wahl and König, 1987; Weber and Czarnetzki, 2001; Ta’ala et al., 2006). The presence of projectile wounds and extensive cranial trauma in a group of skeletons therefore is a good indicator of larger scale violent events most certainly involving more than one attacker (Walker, 2001). In the case of the Eulau multiple burials, both types of injury are represented. The arrowhead lodged in the L4 vertebral body of Ind. 5 can be seen as the best evidence for violent death among the Eulau individuals (Fig. 8). To arrive at its present location, the projectile had to probably penetrate clothing as well as the skin, abdominal wall, small intestines, and most probably at least one of the large blood vessels (abdominal aorta, vena cava) running in front of the vertebral column. The heavy blood loss due to this injury would have been almost instantly fatal, and no signs of bony remodelling could be found. A virtually identical case affecting a lumbar vertebra has been presented by Wells (1964). As the trunk of the human body presents the largest target area for projectile weapons, thorax and abdomen are in fact the most frequent sites for arrow injuries (Wells, 1964; Lambert, 1997; Smith et al., 2007). As has been mentioned above, not all arrows hit bone. This leads to the inevitable conclusion that the number of embedded arrowheads in the abdominal area known from archaeological skeletons truly are only a very small proportion of all the wounds received. Referring again to the analysis of documented historic arrow injuries, only about 2% of arrows hitting the abdominal area embed themselves in bone (Milner, 2005). The unambiguously fatal arrow wound of the woman from grave 6-0090 is important when considering the other Eulau skeletons. The closely associated remains of the subadult Ind. 6 (Fig. 1b) show no signs of violent injury, but the bones are badly deteriorated and the skull could not be removed for detailed analysis. What can be said, though, is that both individuals in this grave died about the same time and were buried simultaneously in a true double inhumation, raising the possibility of a violent death for Ind. 6. Regarding the arrowheads themselves which are associated with the female Ind. 5, both are of a typical Corded Ware design and therefore are not unusual for this time or region. Cranial vault injuries which clearly occurred perimortem and most likely had a lethal outcome could be identified in two skeletons from two of the graves. The parietal bone as a classic site for penetrating blunt or sharp force trauma (Bach, 1963; Willey, 1990; Weber and Czarnetzki, 2001; Paine et al., 2007) is affected in Ind. 7. Two separate blows were inflicted on this woman (Fig. 9). The similar size of both defects suggests the same or an almost identical weapon was used in both cases. Damage to the brain and associated soft tissues apparently resulted in death, as again, no signs of healing are evident. As both wounds are located on the right parietal, the interpretation of this injury, as commonly done in palaeopathology, would include a probable right-handed attacker and therefore a situation where the victim turned her back to the assailant. A constellation like this is usually reconstructed as an unsuccessful attempt to flee from an attack (e.g. Wahl and König, 1987; Walker, 1989). Similarly to grave 6-0090 the adult female in grave 6-0098 is also associated with a subadult. Again, the skeletal remains of this child are too badly preserved to allow any interpretation of possible trauma, but the close proximity of woman and child is highly indicative of simultaneous burial and hence simultaneous death, possibly by similar causes. The remaining two subadult skeletons in grave 6-0098 bear no clear signs of perimortem trauma. They have been placed to the northwest of the female Ind. 7, who has her back to them. The DNA analyses carried out by Haak et al. (2008) revealed that the two children are most likely siblings but are actually not maternally related to the woman. Nevertheless, all four individuals have been buried simultaneously, again indicating death within a short period of time. Another perimortem cranial vault wound is found on the occipital of Ind. 4 (Fig. 7). This location may be interpreted as a wound received while fleeing, but could also be the result of an execution. As this boy of 8–9 years reached an estimated body height of about 109 cm while alive, and the attackers were most likely adult individuals, it would be expected that a wound inflicted during a combat situation would have hit the upper parts of the cranial vault, most likely the parietal bones. As this is not the case, the attacker probably either delivered an underhand-blow or the victim was hit with the head held or hanging down or the individual was actually lying on the ground, maybe protecting the face. Since we do not know the relative positions of attacker and victim when the blow hit the child’s skull, this interpretation has to remain speculative. That fractures of the occipital bone can actually form evidence for an execution with simple weaponry has recently been shown by Ta’ala et al. (2006). All recognized perimortem cranial wounds are compatible with typical Corded Ware stone axes, three of which have been found as grave goods associated with the three oldest male individuals (Fig. 13). The two adults and the remaining boy from grave 6-0099 show no clear evidence of perimortem cranial trauma; however, this does not necessarily indicate that none was present. Because the skull bones could not be removed for examination, it was not possible to exclude violence as a reason for the observed fragmentation of the skulls. In the case of the adult individuals violent injuries of the cranial vault or face are certainly possible. Fig. 13. Corded Ware stone axes found with three of the male individuals (Inds. 3, 9 and 11). C. Meyer et al. / Journal of Anthropological Archaeology 28 (2009) 412–423 While most cranial vault injuries in prehistory were probably sustained during violent assaults and projectile points embedded in bone are diagnostic of traumatic events with implied lethal intent (Lambert, 1997; Lovell, 1997), perimortem fractures of the forearm and hands are often indicative of attempted defense from hand-to-hand attacks (Wahl and König, 1987; Milner et al., 1991; Rogers, 2004). One example of a well-known defensive wound is the parry-fracture, which has been discussed above. Trauma to the metacarpals and phalanges of the hand can also be taken as good evidence for fights involving bare hands or predominantly blunt force weaponry (Brink et al., 1998). The observed perimortem fractures of the hands of both male skeletons (Inds. 3 and 11), involving both metacarpals and phalanges (Fig. 6) are in accordance with defense wounds described in the cited literature. Alternative explanations for these fractures, like post-burial disturbance or soil pressure, can be excluded. There is no evidence for any disturbance of the graves and most of the numerous postmortem bone breaks can easily be distinguished from the perimortem ones on the basis of their morphology, colour and location. The fractures of the right ulna and radius in Ind. 11 with clear displacement of the distal part of the forearm including the hand are another instance of perimortem trauma (Fig. 12). As discussed above, an injury like this might result from a fall, but in the light of all other perimortem trauma noted on these skeletons, it is also likely that the forearm fracture was the result of interpersonal violence. It possibly represents a defense wound as well. The presence of perimortem violent injuries as well as perimortem defense wounds in this skeletal series is a very strong indicator of a violent event which most probably led to the demise of all the people buried in the multiple graves at Eulau. The presence of old and healed fractures, which are quite similar to some of the perimortem lesions and possibly combat-related further supports a violence-related interpretation, as hostile encounters appear to have been a relatively regular occurrence in the Central European Neolithic (e.g. Wahl and König, 1987; Christensen, 2004). Reconstruction of the most probable scenario It has been shown that the context in which the deaths of the Eulau people occurred clearly was a violent one. Still, a number of different scenarios can be imagined that may have resulted in of the death of 13 people of different ages and of both sexes. Here, we try to reconstruct the cultural context of the observed injuries by comparison with other, similar sites from the European Neolithic and elsewhere. The scenario which fits best with all the known osteological and archaeological facts is deemed to be the most probable one (Walker, 2001). The people under study here were related to one another by social ties, which can be reasonably deduced from the fact that they were buried closely grouped according to Corded Ware customs (Fig. 1). Biological kinship has been proven for several individuals by Haak et al. (2008), including a nuclear family (Inds. 1–4). Combined with the archaeological and isotopic evidence there is almost no doubt that these biologically related people lived and died at the same time in the cultural context of the Corded Ware of the third millennium BC. This points logically to a local settlement group, connected by biological and social kinship. The presence of women and children also supports the view that the individuals formed part of a residence group. Based on the strontium isotope evidence, this group most probably was organized patrilocally, as the children and male individuals grew up locally, while the women originally came from somewhere else (Haak et al., 2008). Some of the observed fractures could be interpreted as defense wounds and were not necessarily deadly by themselves, whereas other injuries consisted of lethal perimortem cranial and postcra- 421 nial wounds. The combination of both is a strong indicator of homicide (Lovell, 1997; Rogers, 2004). This pattern most probably rules out any interpretation of ritual killings or practises like ritual suicide. Although there have been some theoretical thoughts on this matter regarding multiple burials (Hein, 1987; Dresely, 2004) there is no conclusive evidence for any of the two in the Corded Ware archaeological record. If ritualised killing had been practised as a cultural phenomenon, for example as a consequence of the death of an important individual, defense wounds would be rather unexpected. Ritual suicide is also highly unlikely, as this is incompatible with the injury patterns of the skeletons and their demographic profile. The most probable scenario which fits the demography, injury patterns and subsequent burial rites is a raid, carried out by a group of attackers on a small resident community. One characteristic evidence of a raid is a high proportion of subadult and female victims (Christensen, 2004; Guilaine and Zammit, 2005), which is definitely the case at Eulau with 85% of individuals in this category, accepting that all were killed in a single event, as argued here. As shown above, both males had suffered previous wounds which resulted in physical impairment in the form of osteoarthrosis of the wrist and interphalangeal ankylosis of the hand (Figs. 5 and 10). The three adult women were each closely associated with at least one child, which in fact is also true for both men. The proven genetic relationships between some of the individuals, especially in grave 6-0099, suggest that even more individuals in this series are most probably direct kin. Therefore it is likely that all three women were indeed the mothers of the children that were most closely associated with them in the burial record. Positive evidence is available for one pair (Inds. 1 and 2), the children from the other two pairs did not allow a DNA analysis because of their bad preservation (Haak et al., 2008). Maternal responsibilities of the women might have presented a limiting feature in terms of survival of a raid. Thus, all of the adult individuals represented in the multiple graves were most probably either burdened by old injuries or child care and would have been rather easy prey for the unknown attackers (Milner et al., 1991). The absence of adolescents and healthy men may also be interpreted in this manner, as these would have had the best chances to either flee or fight back and therefore not to die. The one site from Central Europe with the best parallels to the Eulau skeletons is the Early Neolithic mass grave from Talheim, Germany (Wahl and König, 1987). At Talheim, several dozen people were murdered in a raid that most probably annihilated a whole village. Due to broadly comparable weaponry used, the injuries of the Talheim skeletons also consist of penetrating cranial trauma and arrow wounds, much like those observed in the Eulau skeletons. The 14th century Crow Creek massacre site from South Dakota also has many parallels to the Talheim mass grave, although it resulted in the deaths of several hundred individuals (Willey, 1990). Many crania from Crow Creek showed perimortem traumatic injuries, and men, women, and children were among the victims, which is not surprising when a settlement is overrun (Schulting and Wysocki, 2005). The physical remains of the people were buried in a common mass grave. While the Crow Creek victims had at least been partially disarticulated upon burial, the Talheim victims were interred as complete and probably fresh bodies. If indeed the whole village population had been killed at Talheim, the question of who actually buried the dead still remains. While the exact answer will probably never be known, it is possible that the attackers or some unrelated inhabitants of neighbouring settlements threw the bodies into the grave pit. The bizarre and haphazard body positions at Talheim have nothing in common with the usual burial customs of that time, indicating lack of care and religious rites. The same is evident for the victims of the Crow Creek raid, as the remains have been thrown into a defensive ditch which 422 C. Meyer et al. / Journal of Anthropological Archaeology 28 (2009) 412–423 had not been intended as a grave site when constructed (Willey, 1990). Another site from the European Neolithic which has to be mentioned in this context is Asparn/Schletz, which is contemporaneous to Talheim and rather similar to the Crow Creek site (Teschler-Nicola et al., 1996). At Asparn, the partly disarticulated skeletons of at least several dozen victims of a violent attack have been found at the bottom of a defensive ditch. Many show evidence of multiple cranial trauma, including females and children, as well as carnivore gnawing marks. This has been interpreted as evidence that the bodies were lying exposed for some time before the then partly scattered remains were dumped into the ditch. Again, similar to Crow Creek, this ditch was not intended for burial upon its construction (Teschler-Nicola et al., 1999). Other archaeological sites with human remains which have been interpreted as those of raid victims show a similar pattern of lacking care for the dead (e.g. Owsley and Bass, 1979; Melbye and Fairgrieve, 1994). The care expended for the burial of the dead is the most obvious difference between the Eulau graves and the other sites mentioned. At Eulau, regular graves were dug and the bodies arranged with care and equipped with few, but common Corded Ware grave goods (Figs. 1 and 13). This indicates that somebody who cared for the dead remained behind. It is therefore highly probable that only part of the original settlement group had been killed in the attack. The survivors apparently had detailed knowledge of the familial relationships between the victims and enough time and motivation for the construction of the multiple graves, which were partly equipped with burial mounds and encircling ditches. That all people died in a single attack, rather than during repeated violent encounters, cannot be proven without doubt but is in fact the more likely interpretation. The multiple graves were found as closely neighbouring features and not spread out randomly among the other typically scattered graves of the Corded Ware burial site. In a prehistoric North Amercian cemetery with a high number of violent injuries the latter distribution of burials has been found and interpreted in such a way that people were killed in ambushes on a rather regular basis and not in a single event (Milner et al., 1991). Putting the anthropological, archaeological and biomolecular information together, the following scenario emerges as the most probable: the healthy and physically fit members of a group of Corded Ware individuals left their settlement for unknown reasons. At least two men with previous, healed injuries as well as three women and at least eight children between 0.5 and 9 years of age remained behind. They became victims of a murderous raid, which was carried out by unknown attackers who seized the chance to assault the now largely unprotected settlement. The assailants used common Late Neolithic weaponry for dispatching their victims (Figs. 8 and 13). After the attackers left, the rest of the residence group returned to find members of their group murdered. With evident care for the deceased they constructed at least four multiple graves and buried the dead according to their biological and social relationships. Implications for Corded Ware society and burial customs The detailed analysis of human remains can give important insights into past events (Jurmain, 2001). While the cause of death remains unknown for most prehistoric humans, including many multiple graves, it was possible to reconstruct a violent end for the people under study here. Lethal interpersonal violence was part of Corded Ware life and in this case the underlying cause for the construction of the four graves. As true multiple burials are known from a large number of Corded Ware sites (e.g. Buchvaldek, 1967; Wahl et al., 1990; Dresely, 2004), this raises the question if lethal violence and multiple graves are correlated in this culture. A complete survey is beyond the scope of this paper as many of the hundreds of Corded Ware skeletal remains have not been studied in detail (Bruchhaus and Neubert, 1999). There is some evidence for violence in Corded Ware or culturally related graves (e.g. Wahl et al., 1990; Schutkowski et al., 1996; Dresely, 2004), but to date no comprehensive palaeoepidemiological survey for Neolithic skeletal remains has been carried out. Only through such a survey would it be possible to meaningfully compare trauma frequencies (Walker, 2001; Schulting and Wysocki, 2005). Due to the lack of reliable data, we therefore cannot assess if the period of the Late Neolithic was more violent than other times, although our own review of the literature suggests that this might have indeed been the case. What points in this direction is the very high number of known Corded Ware trepanations. In many cases these might have been carried out as a therapeutic measure after cranial trauma occurred (Piek et al., 1999; Smith et al., 2007), and several Corded Ware skeletons with both trepanations and evidence of cranial trauma have been published (e.g. Bach, 1963; Grimm and Asamoa, 1964). While interpersonal violence surely was not the only reason for burying several people in the same grave at the same time (Meyer et al., 2004), increasing violence may well have led to the apparently increasing number of multiple burials in the Corded Ware of Central Europe. The knowledge about biological kinship, which has been proven for at least four of the individuals (in grave 6-0099) opens a new window into the Late Neolithic, as we now have for the first time definite evidence for the presence of nuclear families in the Corded Ware. This corresponds to the known settlement structures of this culture (Hecht, 2007). Until now, it was only possible to speculate about the significance of multiple graves and who was included in them. With the Eulau graves, we now cannot only glimpse the genetic relationships within a part of the larger population but also one of the possible reasons for the construction of true multiple graves. Both of the questions raised earlier could be answered to a large extent: (1) Why did the individuals buried together die at roughly the same time? They fell victim to a violent raid and were killed by bow, arrow, and axe. (2) Are they biologically related? Yes, in the case of the most intricate burial 60099 the two children are the offspring of the adult individuals. In another grave, two children are probable siblings (Haak et al., 2008). In the case of the Eulau people the analysis and interpretation of all available bioarchaeological data allowed us insight into a short and violent episode of human history. Although certainly the result of a tragic event, the surviving skeletal remains of these 13 individuals reveal important aspects of past human interactions, treatment of the dead, and underlying biological relationships (Walker, 2001). In conclusion, it becomes increasingly evident that studies which combine multiple avenues of research within a bioarchaeological framework emerge as the method of choice for the analysis and interpretation of complex burial situations. Although each scientific study concerned with one type of evidence undoubtedly has its own merits, the ultimate aim of bioarchaeological reconstruction can only be achieved in the highest possible resolution when results from all relevant subdisciplines are combined (Bentley et al., 2008; Meyer et al., 2008; Haak et al., 2008). Apart from the important new insights into the mortuary practises of the Corded Ware, the signs of interpersonal violence encountered in these skeletons add to the growing evidence for prehistoric warfare in Europe. Although the available data at present still does not allow a thorough and valid scientific treatment of this topic, the detailed documentation and interpretation of antemortem and perimortem violence in human skeletal remains appears to be the best and most direct way to approach this matter. Acknowledgments We thank the Staff at the State Museum of Prehistory in Halle for their help during the osteological analysis, Silvio Brandt of the University Hospital in Halle for the CT-scans of selected bones, our colleagues in Bristol for the isotope analyses of these skeletons C. Meyer et al. / Journal of Anthropological Archaeology 28 (2009) 412–423 and Linda Fibiger, Oxford for discussion and improving the language of the paper. We also thank all reviewers for their valuable comments. This is Contribution No. 622 from the Geocycles Cluster funded by the state of Rheinland-Pfalz. References Bach, H., 1962/1963. Ein schnurkeramisches Skelett mit zweifacher Schädeltrepanation aus Wechmar, Kr. Gotha. Alt-Thüringen 6, 202–211. Bach, H., 1963. Die menschlichen Skelettreste. Prähistorische Zeitschrift 41, 134–149. Bennike, P., 2008. Trauma. In: Pinhasi, R., Mays, S. (Eds.), Advances in Human Palaeopathology. Wiley, Chichester, pp. 309–328. Bentley, R.A., Wahl, J., Price, T.D., Atkinson, T.C., 2008. Isotopic signatures and hereditary traits: snapshot of a Neolithic community in Germany. Antiquity 82, 290–304. Brink, O., Vesterby, A., Jensen, J., 1998. Pattern of injuries due to interpersonal violence. Injury 29, 705–709. Bruchhaus, H., Neubert, A., 1999. Zur Rekonstruktion endneolithischer und frühbronzezeitlicher Bevölkerungen im Mittelelbe-Saale-Gebiet – Ergebnisse einer ersten Bestandsaufnahme. Berichte der Römisch-Germanischen Kommission 80, 122–161. Buchvaldek, M., 1967. Die Schnurkeramik in Böhmen. Universita Karlova, Prag. Bulla, A., 1998. Untersuchungen zur Frage der geschlechtsspezifischen Beigabenausstattung bandkeramischer Gräber Mitteleuropas unter besonderer Berücksichtigung der Inventare anthropologisch bestimmter Frauenbestattungen. Shaker, Aachen. Christensen, J., 2004. Warfare in the European Neolithic. Acta Archaeologica 75, 129–156. Dittmar, K., Teegen, W.R., 2003. The presence of Fasciola hepatica (liver-fluke) in humans and cattle from a 4,500 year old archaeological site in the Saale-Unstrut Valley, Germany. Memórias do Instituto Oswaldo Cruz 98 (S1), 141–143. Dresely, V., 2004. Schnurkeramik und Schnurkeramiker im Taubertal. Theiss, Stuttgart. Feustel, R., Bach, H., Gall, W., Teichert, M., 1966. Beiträge zur Kultur und Anthropologie der mitteldeutschen Schnurkeramiker. Alt-Thüringen 8, 20–170. Grauer, A.L., Roberts, C.A., 1996. Paleoepidemiology, healing, and possible treatment of trauma in the Medieval cemetery population of St. Helen-on-the-Walls, York, England. American Journal of Physical Anthropology 100, 531–544. Grimm, H., Asamoa, A., 1964. Neue schnurkeramische Skelettreste von Schafstädt, Kreis Merseburg. Jahresschrift für mitteldeutsche Vorgeschichte 48, 107–115. Guilaine, J., Zammit, J., 2005. Origins of War. Violence in Prehistory. Blackwell, Oxford. Haak, W., Brandt, G., de Jong, H.N., Meyer, C., Ganslmeier, R., Heyd, V., Hawkesworth, C., Pike, A.W.G., Meller, H., Alt, K.W., 2008. Ancient DNA, strontium isotopes, and osteological analyses shed light on social and kinship organization of the Later Stone Age. Proceedings of the National Academy of Sciences of the United States of America 105, 18226–18231. Heberer, G., 1938. Die mitteldeutschen Schnurkeramiker. Gebauer-Schwetschke, Halle. Hecht, D., 2007. Das schnurkeramische Siedlungswesen im südlichen Mitteleuropa. Eine Studie zu einer vernachlässigten Fundgattung im Übergang vom Neolithikum zur Bronzezeit. Dissertation. <http://www.ub.uni-heidelberg.de/ archiv/7313>. Hein, M., 1987. Untersuchungen zur Kultur der Schnurkeramik in Mitteldeutschland. Habelt, Bonn. Judd, M.A., 2008. The parry problem. Journal of Archaeological Science 35, 1658– 1666. Jurmain, R., 2001. Paleoepidemiological patterns of trauma in a prehistoric population from Central California. American Journal of Physical Anthropology 115, 13–23. Kunter, M., 1974. Über das Vorkommen von Knochenbrüchen im Bereich des Unterarmes bei Skelettfunden. Homo 25, 78–96. Lambach, F., 1993. La necropole rubanee d’Ensisheim ‘‘Les Octrois”. Description des tombes et anthropologie de terrain. Cahiers de l’association pour la promotion de la recherche archeologique en Alsace 9, 17–48. Lambert, P.M., 1997. Patterns of violence in prehistoric hunter–gatherer societies of Coastal Southern California. In: Martin, D.L., Frayer, D.W. (Eds.), Troubled Times. Violence and Warfare in the Past. Gordon and Breach, Amsterdam, pp. 77–109. Lovell, N., 1997. Trauma analysis in paleopathology. Yearbook of Physical Anthropology 40, 139–170. Lovell, N.C., 2008. Analysis and interpretation of skeletal trauma. In: Katzenberg, M.A., Saunders, S.R. (Eds.), Biological Anthropology of the Human Skeleton, second ed. Wiley, Hoboken, pp. 341–386. Mann, R.W., Hunt, D.R., 2005. Photographic Regional Atlas of Bone Disease. Charles C. Thomas, Springfield. Melbye, J., Fairgrieve, S.I., 1994. A massacre and possible cannibalism in the Canadian Arctic: new evidence from the Saunaktuk Site (NgTn-1). Arctic Anthropology 31, 57–77. Menninger, M., 2008. Die schnurkeramischen Bestattungen von Lauda-Königshofen. Steinzeitliche Hirtennomaden im Taubertal? Dissertation. <http://tobiaslib.ub.uni-tuebingen.de/volltexte/2008/3390/>. 423 Meyer, C., 2003. Osteological evidence for the Battles of Zürich, 1799: a glimpse into soldiery of the past. International Journal of Osteoarchaeology 13, 252–257. Meyer, C., Kürbis, O., Alt, K.W., 2004. Das Massengrab von Wiederstedt, Ldkr. Mansfelder Land. Auswertung und Gedanken zur Deutung im Kontext der Linienbandkeramik. Jahresschrift für mitteldeutsche Vorgeschichte 88, 31–66. Meyer, C., Kranzbühler, J., Drings, S., Bramanti, B., Nehlich, O., Richards, M.P., Alt, K.W., 2008. Die menschlichen Skelettfunde aus der neolithischen Totenhütte von Benzingerode. Anthropologische Untersuchungen an den Bestattungen eines Kollektivgrabs der Bernburger Kultur. Archäologie in Sachsen-Anhalt Sonderband 7, 107–151. Milner, G.R., 2005. Nineteenth-century arrow wounds and perceptions of prehistoric warfare. American Antiquity 70, 144–156. Milner, G.R., Anderson, E., Smith, V.G., 1991. Warfare in late prehistoric WestCentral Illinois. American Antiquity 56, 581–603. Nieszery, N., 1995. Linearbandkeramische Gräberfelder in Bayern. Marie L. Leidorf, Espelkamp. Owsley, D.W., Bass, W.M., 1979. A demographic analysis of skeletons from the Larson Site (39WW2) Walworth County, South Dakota: vital statistics. American Journal of Physical Anthropology 51, 145–154. Paine, R.R., Mancinelli, D., Ruggieri, M., Coppa, A., 2007. Cranial trauma in iron age samnite agriculturists, Alfedena, Italy: implications for biocultural and economic stress. American Journal of Physical Anthropology 132, 48–58. Piek, J., Lidke, G., Terberger, T., von Smekal, U., Gaab, M.R., 1999. Stone age skull surgery in Mecklenburg–Vorpommern: a systematic study. Neurosurgery 45, 147–151. Puppe, G., 1914. Über Priorität der Schädelbrüche. Ärztliche Sachverständigen Zeitung 20, 307–309. Rogers, J., 2004. Recognizing inter-personal violence: a forensic perspective. In: Roksandic, M. (Ed.), Violent Interactions in the Mesolithic, BAR International Series, vol. 1237. Archaeo Press, Oxford, pp. 9–21. Roksandic, M., Djurić, M., Rakočević, Z., Seguin, K., 2006. Interpersonal violence at Lepenski Vir Mesolithic/Neolithic complex of the iron gates Gorge (SerbiaRomania). American Journal of Physical Anthropology 129, 339–348. Ryan, T.M., Milner, G.R., 2006. Osteological applications of high-resolution computed tomography: a prehistoric arrow injury. Journal of Archaeological Science 33, 871–879. Schulting, R.J., Wysocki, M., 2005. ‘In this Chambered Tumulus were Found Cleft Skulls. . .’: an assessment of the evidence for cranial trauma in the British Neolithic. Proceedings of the Prehistoric Society 71, 107–138. Schutkowski, H., Schultz, M., Holzgraefe, M., 1996. Fatal wounds in a Late Neolithic double inhumation – a probable case of meningitis following trauma. International Journal of Osteoarchaeology 6, 179–184. Schwidetzky, I., 1978. Anthropologie der Schnurkeramik- und Streitaxtkulturen. In: Schwabedissen, H. (Ed.), Die Anfänge des Neolithikums vom Orient bis Nordeuropa 8b. Part 2. Böhlau, Köln, pp. 241–264. Smith, M.J., Brickley, M.B., Leach, S.L., 2007. Experimental evidence for lithic projectile injuries: improving identification of an under-recognised phenomenon. Journal of Archaeological Science 34, 540–553. Smith, M.O., 1996. ‘Parry’ fractures and female-directed interpersonal violence: implications from the late archaic period of West Tennessee. International Journal of Osteoarchaeology 6, 84–91. Spatz, H., 1999. Das mittelneolithische Gräberfeld von Trebur, Kreis Groß-Gerau. Landesamt für Denkmalpflege, Wiesbaden. Ta’ala, S.C., Berg, G.E., Haden, K., 2006. Blunt force cranial trauma in the Cambodian killing fields. Journal of Forensic Sciences 51, 996–1001. Teschler-Nicola, M., Gerold, F., Kanz, F., Lindenbauer, K., Spannagl, M., 1996. Anthropologische Spurensicherung. Die traumatischen und postmortalen Veränderungen an den linearbandkeramischen Skelettresten von Asparn/ Schletz. Archäologie Österreichs 7 (1), 4–12. Teschler-Nicola, M., Gerold, F., Bujatti-Narbeshuber, M., Prohaska, T., Latkoczy, C., Stingeder, G., Watkins, M., 1999. Evidence of Genocide 7000 BP – Neolithic paradigm and geo-climatic reality. Collegium Antropologicum 23, 437–450. Uckelmann, M., 2004/2005. Schutz, Prunk und Kult – Zur Funktion bronzezeitlicher Schilde. Anodos 4–5, 243–249. Wahl, J., König, H.G., 1987. Anthropologisch–Traumatologische Untersuchung der menschlichen Skelettreste aus dem bandkeramischen Massengrab bei Talheim, Kreis Heilbronn. Fundberichte aus Baden-Württemberg 12, 65–186. Wahl, J., Dehn, R., Kokabi, M., 1990. Eine Doppelbestattung der Schnurkeramik aus Stetten an der Donau, Lkr. Tuttlingen. Fundberichte aus Baden-Württemberg 15, 175–211. Walker, P.L., 1989. Cranial injuries as evidence of violence in prehistoric Southern California. American Journal of Physical Anthropology 80, 313–323. Walker, P.L., 2001. A bioarchaeological perspective on the history of violence. Annual Review of Anthropology 30, 573–596. Weber, J., Czarnetzki, A., 2001. Neurotraumatological aspects of head injuries resulting from sharp and blunt force in the early Medieval period of Southwestern Germany. American Journal of Physical Anthropology 114, 352–356. Wells, C., 1964. Bones, Bodies, and Disease. Evidence of Disease and Abnormality in Early Man. Praeger, New York. White, T.D., Folkens, P.A., 2005. The Human Bone Manual. Elsevier, Amsterdam. Willey, P., 1990. Prehistoric Warfare on the Great Plains. Skeletal Analysis of the Crow Creek Massacre Victims. Garland Publishing, New York.