53 M. Roksandic: Contextualizing the Evidence of Violent Death in the Mesolithic... CONTEXTUALIZING THE EVIDENCE OF VIOLENT DEATH IN THE MESOLITHIC: BURIALS ASSOCIATED WITH VICTIMS OF VIOLENCE IN THE IRON GATES GORGE Mirjana ROKSANDIC Abstract: Interactions of foragers and farmers are at the center of debate on the Lepenski Vir Mesolithic of the Iron Gates Gorge (Serbia- Romania). While typology played a major role in determining Lepenski Vir culture as Epipaleolithic (Boroneanţ, 1973), Mesolithic (Srejović, 1972) or Neolithic (Jovanović, 1984b), it is currently understood that foraging economy and semi-sedentism characterize the sites even after the contact with the farming communities in the region becomes evident, in the second half of the 7th millennium B.C. (Radovanović, 1996b; Roksandić, 2000a). To date, violent interactions in the region have been discussed only briefly, and in the context of possible conflict between foragers and farmers at the left bank site of Schela Cladovei (Boroneanţ et al., 1999). Here I examine the evidence for violent interactions on the right bank of the Danube, and associated burial practices, in an attempt to distinguish individual acts of interpersonal violence from possible warfare. The data are further contrasted with evicence from the site of Schela Cladovei located in the flood plane on the left bank of the Danube, downstream from the Gorges. Key words: Lepenski Vir, violent trauma, interpersonal violence, burial ritual, Mesolithic/Neolithic transition, Iron Gates Mesolithic. 1. INTRODUCTION Violence in any society has to be examined in the context of local history (Simons, 1999). While it is inevitable in archaeology to look for broader patterns and explanatory mechanism on a larger scale, we must not assume a violent or a non-violent pattern of behaviour to be “typical” of prehistoric hunter gatherers (see Roksandic, this volume). Societies – or types of economies and social structures – do not have a “natural propensity” for violence or otherwise. In any specific group, all interactions within a community, including violence, are sanctioned by learning and socialisation (Ember and Ember, 1997). We have to remember that these interactions, both within a group and with other groups, change over time. As prehistorians, we work with long stretches of time collapsed into a single “cultural” unit or its phase; everyday dealings of individuals, even when observable, have to be interpreted on a larger scale. Short-term changes in behavioural pattern remain unrecognized, and specific incidences of violence are interpreted as a pattern of behaviour applicable to the society as a whole, or – even more broadly – to an entire period. Based on sporadic evidence for violence in Mesolithic series, the period itself is portrayed as the beginning of organised inter-group violence and warfare (Frayer, 1997; Keeley, 1997; Thorpe, 2000; Vencl, 1999). This has been accepted as fact even outside the field (De Pauw, 1998). As pointed out by Dennell (1985), contact of colonizing Europeans with indigenous peoples was used as a model for early contact between European Mesolithic peoples and farmers. For these reasons, it is not surprising that the sporadic evidence of violence in the Mesolithic/Neolithic contact period is often regarded as indicative of endemic warfare (e.g. Boroneanţ et al., 1995). While it is not necessarily incorrect to interpret this violence as evidence of warfare, we must not jump to conclusions, especially when defence structures and differentiated implements for hunting versus warfare are lacking. In fact, what we need is a more solid understanding of how warfare can be identified in the osteological and archaeological records. Given the lack of archaeological data representing defence structures and armament, we have to rely on bio- archaeological data derived from skeletal lesions associated with violent trauma and warfare (Walker, 2001). I have already discussed difficulties in inferring warfare, including feuding, (as defined by Kelly, 2000) from skeletal evidence in the Introduction to this volume. Tracy Rogers’ (this volume: 9-22) forensic perspective on the problem further points out the difficulties in building a plausible model for determining warfare, while Mary Jackes has brought forward archaeological and ethno-archaeological examples that illustrate this difficulty (this volume: 23-40). While not always conclusive, I suggest that the most promising approach is provided by rigorous interpretation of mortuary behaviour associated with victims of violence. In such a way, for example, synchronicity of burials that are presumably derived from a massacre (as claimed for Ofnet by Frayer, 1997) can be confirmed or refuted. Furthermore, a more thorough understanding of the social persona, an understanding that goes beyond age and sex to examine individual’s role or position in the society – to the extent it can be inferred from the totality of burial data – can potentially elucidate the problem and give supporting evidence to skeletal information. In this paper, evidence of violent trauma in the Iron Gates Gorge Mesolithic (IGM), a relatively restricted regional manifestation, is discussed in light of the accompanying burial data. Detailed examination of burial context associated with the individuals who engaged in violent behaviour is proposed as a source of information on their status and on the