The earliest mass graves in Europe date back just over 7,000 years. They reveal brutal evidence for violence beyond the simple act of killing. The motives for these events are probably diverse but consistently highlight an intention to kill large numbers of enemies across sex and age ranges. Our study of a 2,850-year-old massacre and resulting mass grave at Gomolava (modern day Serbia) shows the nature of mass killings evolving. Comprised mostly of women and girls, the grave suggests a shift in prehistoric violence. Here, women and children were not collateral victims, but deliberate targets. Looking at who was killed and how they were related can tell us about changes in ancient attitudes to killing combatants – but also choices in targeting non-combatants. The site at Gomolava, first excavated in 1971, is the second of two contemporary mass graves. It contained 77 people. Our recent study has uncovered the circumstances surrounding their death.We found that the original theory – that an epidemic had killed people from a single settlement – is not supported by genetic and isotopic evidence. Our data instead showed that these people descended from the wider region but came from different settlements. Except for a mother and her two daughters, there were no close genetic relationships.Archaeology and mass gravesArchaeology sheds light on the deep history of conflict. When killings are targeted, mass graves can tell us which members of a society were displaced and killed. Then, we pose the question – why them?When we study violence in past societies, archaeologists seek to understand how and why things like discord, ambition, belief or rivalry could lead people to plan to kill others. While archaeology cannot provide ready solutions to modern crises, it helps us to better understand the immediate and longer term societal impact of violence, and how more peaceful times were achieved and managed.Our detailed study of the bones of the dead in the Gomolava grave exposed a brutal story – these people were mostly killed through blows to the head. All the bodies had been crammed into a repurposed old pit house. This structure, once part of a small village, consisted of a sunken pit with low walls and a roof. This hole provided a ready space to stow the dead. It is unclear if that was expedient disposal, or putting them in what was once a home. However, these people were buried respectfully with some of their personal possessions and other offerings. Most of the dead were women and children. Studies of the children’s teeth revealed a disproportionate number were girls. In other prehistoric mass graves, children and young women are often underrepresented because of their value as slaves or for reproduction. Gomolava is a clear exception, inverting this pattern.In comparison, the Neolithic sites of Asparn-Scheltz, Austria (which contains approximately 200 bodies), and Potočani, Croatia (41 bodies) also contained people with few genetic connections. However, children were underrepresented and males and females equally present. At Koszyce, Poland, 15 genetically closely related women, young men and children were buried together. These events highlight larger scale conflicts and often relate to periods of exceptional social change, such as the spread of different material conventions and, at times, inward migration of genetically distinct people.Why women and children?Mass graves have long informed us of the scale of past violent events. Combining archaeology with newer genetic and isotopic methods better reveals demographics, which has revolutionised the ways we can access stories of the victims. At Gomolava, this enabled us to explore the nature, social context and strategic purpose of this mass killing event. In our study, we suggest the women and children in the Gomolava grave were not incidental victims on the periphery of conflict or its aftermath, but primary targets. Selectively killing these women and children ended family lines and cut short the future of communities. Possible motives were retribution or establishing dominance in the region. This represents a sea change in the nature of violent conflict, potentially a case where previous taboos on killing were no longer exercised. Though this is only one site in one specific time and place, it demonstrates how the treatment of mass killings can highlight major changes in attitudes to violence as a social strategy.There is a tension in archaeology between those who argue that violence was a commonplace means of exercising social power and those that concentrate more on characterising the many alternatives to violence for asserting control or authority. In our view, both likely coexisted. In the absence of formal legal systems, the ability to fight could itself function as a deterrent. Violent conflict was therefore an extreme – but ever-present – means of resolving inter-group disputes. Gomolava and the Sava river. Barry Molloy, CC BY-SA The targeting of women and children in Gomolava by people from outside their community provides a bridge between perspectives, beyond a view of men enforcing violence over other men on the battlefield. It reveals how those who shaped social trajectories across many fields of social discourse were central to conflict resolution, whether that be peaceful or through dominance and extermination. We argued that different groups were competing over land ownership in this region, and mass killing of enemies was a strategic choice to assert dominance or hegemony over a community spread across many settlements, likely including Gomolava.Whether such graves represent ethnically, culturally or socially different groups in conflict, the brutality of their selective killing remains a common thread. In seeking to understand how peace was brokered and managed, and how different forms of power played roles in this, working to better understand who was targeted and killed in episodes of violence is a crucial first step. Our comparative approach takes mass graves beyond specific times and places and shows how we may have underestimated the scale, ferocity and purpose of short episodes of violence.Barry Molloy receives funding from The European Research CouncilLinda Fibiger does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.