Drug Violence, War-Crime Distinction, and Hierarchies of Victimhood

Issues related to victimhood are central to transitional justice and international criminal justice. However, processes of transitional justice do not usually include victims of drug-related violence, despite the fact that in several Latin American countries deaths caused by cartel violence easily meet criteria of civil war. This article's central argument is that distinctions between victims of war and victims of what is often termed conventional crime are of great importance to notions of legitimate victimhood in transitional contexts. Taking Colombia's Victims’ Law (2011) as a case study, we argue that the binary distinction between war and crime fails to address the needs of victims of mass drug violence and creates a hierarchy among victims. This has important symbolic, legal and material implications for those who find themselves in the less favoured category. Victims of drug related violence struggle to access justice and to make their voices heard in public discourses about violence. We argue that the current understanding of mass drug violence as ‘conventional crime’ represents a Northern perspective on violence, which can be counter-productive when used uncritically in Southern contexts.


Introduction
On 27 November 1989, the Medellín Cartel bombed the HK-1803 domestic flight from Bogotá to Cali in Colombia. 107 passengers died in the attack. The event was part of a terror campaign orchestrated by Pablo Escobar and his associates to resist U.S. pressure on the Colombian government to make them stand trial in the United States on trafficking charges (Smyth, 1998). The plane bombing was part of a larger, long-standing conflict involving the Colombian state and the Medellín Cartel which, at the height of its power, controlled 60 per cent of the world's cocaine and made the city where it was based among the most violent in the world. During the 1980s, 375 homicides per 100,000 inhabitants were recorded annually in the city, which is more than 35 times the World Health Organisation's definition of epidemic violence (Maclean, 2015: 2).
It, therefore, came as a surprise to many that victims of the Medellín Cartel were excluded from the recognition and benefits granted by the 2011 Colombian Victims' Law 1 . The Colombian Congress declared that the law would only cover 'direct victims of the conflict, i.e. those affected by the guerrillas, paramilitaries or agents of the state' (Unidad para la Atención y Reparación Integral a las Víctimas, 2018: 7). The law specifies that 'those whose rights have been harmed as a consequence of common criminality will not be considered as victims' (art. 3, para. 3; italics added). When interpreting the law, also the Constitutional Court' 2 defined as victims only those affected by violence related to internal armed conflict. Although Federico Arellano-son of the musician Gerardo Arellano, who died in the HK-1803 planeeventually received official recognition as a victim of the Colombian conflict (Unidad para la Atención y Reparación Integral a las Víctimas, 2018), few other victims of drug violence have achieved the same.
Processes of transitional justice do not generally include victims of drug-related violence, despite the fact that, in several Latin American countries, deaths caused by cartel violence easily exceeds the common 1000-battle deaths-deaths-per-year criterion of civil war (Lessing, 2018(Lessing, : 1487. In this article, we show that distinctions between victims of war and victims of what is often termed conventional crime are of great importance for notions of legitimate victimhood in transitional contexts. Drawing on previous scholarship on hierarchies of victimhood (McEvoy & McConnachie, 2012, we argue that the position of victims of drug violence reflects a globally established hierarchy, which distinguishes between victims of war and those of 'conventional crime', and gives the former greater priority. Our data collected in Colombia-the first Latin American country to experience drug violence on a massive scale-show that hierarchies of victimhood have serious legal, symbolic and material implications for those who find themselves in the less favoured category. Victims of drug related violence struggle to access justice and to make their voices heard in public discourses about violence. We argue that current binary distinctions between war and crime do not offer sound conceptual tools for understanding the Colombian conflict or the empirical realities of mass violence and atrocities in many other societies in the global South. They have a distinct Northern bias which can be counter-productive when used uncritically for understanding Southern contexts. The article begins with an overview of the existing scholarship on hierarchies of victimhood. While research on the issue has brought significant advances in the past decade, we point out that there has not yet been an examination of how global epistemological power imbalances contribute to our understanding of victimhood. We then continue to discuss the distinction between war and crime and the relevance of scholarly perspectives on the coloniality of knowledge for understanding its genealogy. Our empirical findings then allow the voices of Medellín's inhabitants and victims of cartel violence to be heard, and provide a sensorial account of what it is like to live in an environment where violence is so widespread that it seems like a war. By making use of the sociology of the senses, we seek to convey the bodily experience of being in a city dominated by fear, uncertainty and mass violence (Herrity et al., 2021).
Our discussion then asks why the experiences of victims of mass cartel violence are not considered important for transitional justice processes and what the implications of this exclusion from legitimate victimhood are for those whose lives are affected by the violence. We argue that the character of drug-related violence in Colombia (and other Latin American countries) is so pervasive and significant that it can often be indistinguishable from mass atrocities. Furthermore, in practice, drug-violence and political violence are intrinsically intertwined and fuel each other, making it difficult to separate them into two isolated phenomena. We sustain that there are two explanations for the existence of the distinction between drug-violence and violence related to the internal armed conflict: first, the impact of Northern perspectives in Southern contexts. Second, the denial of the broader political character of drug-violence reflects efforts to hide a link between such violence and the actions of the powerful. Consequentially, we conclude that policies and legal and institutional frameworks aiming to achieve justice in transitional contexts should transcend binary distinctions between crime and warfare, and become more aware of the hierarchies of victimhood that may result from them.

Hierarchies of Victimhood
A large body of scholarship has shown the importance of the construction of victimhood in transitional societies and those that have experienced mass victimisation. Questions of victimhood are, as Jankowitz (2018: 236) points out, central 'to better equip societies to address violent pasts and secure a more peaceful future'. One of the most salient findings in previous studies within victimology, transitional justice and criminology has been that victims are accorded unequal value and legitimacy. Examples range from von Henting's classic study (1948), in which victims were classified according to the nature of their involvement in the criminal act, to Christie's (2018Christie's ( [1986) influential concept of the ideal victim. In transitional contexts, the category of victimhood has proved to be highly contentious. As Ferguson et al. (2010: 860) point out, 'legitimate identification as the victim may be used to attract resources and international support to the ingroup while strengthening ingroup solidarity and moral superiority over the outgroup.' The label of victim has been contested particularly fiercely because it is closely associated with notions of blameworthiness. This, as McConnachie (2012, 2013) show in their seminal study, leads to the creation of hierarchies of victimhood, which are based on a distinction between deserving and undeserving victims. Such hierarchies may be problematic in that they are based on standards and criteria that reflect the narrative or ideology of the group making the ranking and may lead to further conflict and re-victimisation by privileging certain victims' needs and experiences over others' (Jankowitz, 2018).
The growing scholarly engagement with the politics of victimhood found in contexts of internal armed conflict shows that the notion of innocence is central to the social understanding of victimhood. 'Innocent' victims are 'at the apex of a hierarchy of victimhood' and become 'a symbol around which contested notions of past violence and suffering are constructed and reproduced ' (McEvoy and McConnachie, 2012: 532). van Dijk (1999) and Schwobel-Patel's (2018) studies show that notions of innocence and ideal victimhood also inform how international crimes are understood. Conflicts where distinctions between 'good guys' and 'bad guys' (van Dijk, 1999: 169) are clear and simple attract greater media attention and enable victims to publicize their fate and thus 'benefit' from their status as victims.
Although the innocence paradigm has been by far the most influential approach to the understanding of legitimacy in transitional contexts, some commentators identify further elements to explain and predict differential ranking of legitimacy. Brewer and Hayes (2015), for example, find that ideological and religious convictions affect perceptions of innocence. Moving beyond the 'innocence paradigm', Jankowitz's study of the conflict in Northern Ireland distinguishes between two meta-types of hierarchies: normative, which 'communicate beliefs about how victim experiences should be prioritised or distinguished against one another', and descriptive, which reflect 'beliefs that certain victim experiences are treated as more profound or deserving than others ' (2018: 224). More specialised studies have investigated the creation of hierarchies of victimhood in national legislation. Hearty (2016: 334) points out that the meta-conflict positioning of legislators-i.e. the 'continuing political disagreement over the causes and consequences of conflict'-determines who is recognised as a victim and which victims are prioritised. Consequently, even regulations intended to end the conflict end up being 'war by other means via restrictively ascribing blame for suffering in accordance with meta-conflict positions on the (il)legitimacy of certain violence' (ibid.: 335). Similarly, Killean (2018: 227) shows how processes leading to inclusion in the category of victimhood are permeated by political interests, particularly the desire to 'limit the extent of this [the state's] accountability'. Prosecutorial and judicial selectivity and visibility in the courtroom may also shape hierarchies of victimhood (ibid.).
In the global domain, scholars have pointed out that the hierarchy of victimhood is also shaped by international politics (Carrabine et al., 2020) and have stressed the importance of victims' ability to attract the attention of politicians, international non-governmental organizations and, above all, international media. van Dijk's (1999) study, mentioned above, thus argues that Christie's model of the ideal victim seems largely appropriate in the context of international crimes, although in crucial ways it is shaped by media attention.
Here we also encounter the paradox described by Christie (2018Christie ( /1986: victims have to be vulnerable in order to seem deserving of victim status, yet strong and powerful enough to gain it and fend off counter-claims (see also Schwöbel-Patel, 2018).
Most existing scholarship on hierarchies of victimhood focuses on victims' characteristics, particularly their perceived innocence, relationship to perpetrators, and ability to attract international support and media attention. For the purposes of our discussion, however, van Dijk's (1999) study is important because he also draws attention to the characteristics of conflicts. Not all conflicts are able to endow victims with legitimate victimhood. This had already been pointed out by Knox (2001: 181) who noticed that the innocence of victims can be determined by accidental geographical factors, such as the communities in which they live.
In this article we argue that an important factor determining legitimate victimhood, is whether a conflict is defined as war or as a conventional crime. As exemplified by the Colombia's Victims' Law, distinction between the two has important legal and symbolic consequences for victims. It helps create structural conditions in which victims of mass drug violence are placed at the bottom of the hierarchy of victimhood, although they have other qualities usually associated with 'ideal victims'. In what follows, we shall first explore the origins and nature of the distinction made between war and crime and examine how it may be seen as a reflection of deeper epistemological inequalities between the global North and global South. We will ask to what extent the conceptual placing of mass drug violence in the category of 'ordinary crime' represents a Northern perspective on violence.

Mass Drug Violence: A Civil War or an 'Ordinary Crime'?
In many Latin American countries today the casualties of drugs-related violence far exceed accepted definitions of civil war. According to Lessig (2018Lessig ( : 1487 militarised conflict between cartels and states has particularly afflicted Latin America's three largest countries (Brazil, Mexico and Colombia) 'arguably supplanting revolutionary insurgency as the hemisphere's predominant form of conflict'. Next to the war in Syria, Mexico's cartel-state conflict has been so far the most violent subnational conflict of the twenty-first century (ibid.). According to official government count, 70.000 people were killed between 2006 and 2012 in inter-cartel and state-cartel conflicts. As Trejo and Ley (2020: 2) point out, this is 'more than four times greater than the median death toll of all civil wars in the second half of the twentieth century'.
It is a commonplace in political speeches, media reports and academic writing to describe the situation in Mexico and elsewhere as a 'drug war' (Gibller, 2011;Shirk, 201). The term is also used to convey the level of militarisation found in the fight against illegal narcotics. For example, in 2006, Mexico's President Felipe Calderón famously declared war on the cartels (Trejo and Ley, 2020). Brands (2011: 230) therefore argues that, due to its propensity to attack state institutions, 'illicit activity in Latin America begins to straddle the demarcation between non-political and political violence, between crime and insurgency'. The numbers of drug violence casualties not only match conventional definitions of war but also represent a threat to the central functions of society (Trejo and Ley, 2020).
The blurred line between criminal and military actors has long been acknowledged as a feature of many conflicts. Terms such as civil war (Schedler, 2013), criminal war (Trejo and Ley, 2020), irregular warfare and criminal insurgency (Brands, 2011) have been used by scholars to describe the scale and gravity of violence in cartel-state conflicts. Other observers, however, have been more reluctant to challenge the established distinction between crime and warfare. Lessing (2018) acknowledges the severity of the threat of cartel violence and its magnitude, but argues that it should be distinguished from war because of 'differences in underlying logics of violence'; 'cartels do not seek to topple the government and seize formal power' (ibid.: 1488).
Our main objective in this paper is to examine the implications such conceptual distinctions have for those affected by large-scale drug violence. While studies of transitional justice, international criminal law, victimology and criminology have paid increasing attention to the position of victims of war (McGarry and Walklate, 2015), they have mostly overlooked victims of drug violence. Similarly, the fields of international relations and criminology have been striving to bridge the divide between studies of crime and war, the 'inside' and the 'outside' (see inter alia Loader and Percy, 2012;Jamieson, 2014;Walklate & McGarry, 2015), however, the categorisation of mass drug-related violence as crime has largely remained unchallenged. This means that we have a growing body of knowledge about the nature and causes of drug violence, the damage it causes to state institutions and how it challenges national security (Durán-Martínez, 2018;Lessing, 2018;Trejo and Ley, 2020), but know far less about the impact of mass drug violence on victims and their possibilities of accessing justice.
Nor has a more nuanced understanding of such violence informed international and national legal frameworks. Colombia's Victims' law is not only an example of national legislation, but also an expression of an international legal consensus. The militarisation of Mexico's drug war has sparked debate about whether the conflict should be classified as non-international armed conflict (Bergal, 2011;Sassoli, 2019), but so far drug crimes have not been included in internationally agreed definitions of mass atrocity crimes, even though they may involve widespread, systematic attacks on civilian populations (Robinson, 2015). 3 Interestingly, the negotiation of the Rome Statute of the International Criminal Court was originally initiated by Trinidad and supported by several other small states that wanted an international tribunal to try drug trafficking offences. This argument was later lost in the negotiation process due to the influence of powerful (Northern) actors, particularly the U.S. (Boister, 2012).
In this article we ask why acts such as massacres of civilians, systematic assassination of politicians, judges and other state representatives are classed as 'conventional crime'. Drawing on interviews with survivors of such events, we will try to convey their sensory experience of these forms of violence and of how (to them) they seemed like war. Ultimately, as Gray observes (1989 in McGarry andWalklate, 2011: 900), while 'the magnitude of violence may vary with the object and means of war, the violent essence of war will never change.' Unlike war, however, much crime is non-violent. This applies particularly to countries of the global North, many of which have witnessed a steady decline in recorded crime, including homicide. According to UNODC's Global Study on Homicide (2019), the homicide rate in Europe has declined by 63 per cent since 2002 and is now 3.0 per 100,000. Although drug-related crime continues to feature prominently in crime statistics in Northern societies, it involves few fatalities. By contrast, the homicide rate in the Americas is the highest recorded in the region since reliable records began in 1990. Central America and South America, at 25.9 and 24.2 per 100,000, respectively, were in 2017 the subregions with the highest homicide rates in the world, followed by the Caribbean, with 15.1 per 100,000 (UNODC, 2019). Large-scale loss of life related to drug activities thus takes place predominantly in the countries of the global South. The conceptualisation of drug-related violence as ordinary violence fits Northern realities much better than it does the extreme type found in Latin American countries. To what extent, then, are current understandings of 'ordinary crime' shaped by epistemological inequalities at the global level?
A growing body of scholarship has in the past decades drawn attention to the connection between geopolitical position and knowledge production. Connell's influential critique of northern theory (2006) is particularly relevant here as it points out how vast global inequalities of resources shape all academic disciplines, including the social sciences. These inequalities 'may also be embedded within a discipline, in the way intellectual workers define their problems and carry out their work' (Connell, 2006: 237). Because, overwhelmingly, social theory is produced in the global North, scholars are socialised into what Connell terms 'reading from the centre': metropolitan theorists, building on personal knowledge or local research, 'generalise the specific experience of metropolitan countries' (ibid. 259). The seeming universality of contemporary social theory thus results from the generalisation of Northern experience. Reading from the centre results, amongst others, in what Said (1977: 3) called orientalism, 'a style of thought based upon an ontological and epistemological distinction made between "the Orient" and (most of the time) "the Occident". Orientalist thought is, among other, marked by the belief that extreme violence in the Orient (in this case Latin America) is normal and unremarkable for the societies in question (see also the work of Amin 2009Amin [1988; Hall, 1992).
The Latin American decolonial movement has also brought attention to the mutually constitutive power relations 'between the international division of labour, the global racial/ethnic hierarchy and the hegemonic Eurocentric epistemologies in the modern/ colonial world-system' (Grosfoguel, 2007: 95). The notion of coloniality of knowledge highlights the role that knowledge systems play in neo-colonial practices and how Northern influence shapes how Southern realities are understood. The Southern embrace of Northern views of reality blinds scholars to dynamics that may contradict or nuance existing academic understandings. Santos (2014) calls this phenomenon epistemological blindness, which, as interpreted by [Author B] means 'the conscious or unconscious preference to accommodate only that which accords with our existing epistemological and methodological configurations, leaving other possibilities or data ignored'. This results in inadequate, decontextualised readings of reality. Moreover, as will be argued in this article, the impact of metropolitan academic worldviews goes beyond the definition of research issues, concepts and methods, also shaping the issues included on states' agendas, the concepts guiding how the issues are understood, and the public policy adopted to confront them [Author B].

Methods
The empirical material for this article was gathered using different qualitative methods at two stages. During the first stage (March-August 2019), we carried out ethnographic observation in the city of Medellín, Colombia, and media analysis. The observation produced handwritten notes and a variety of photographic material; both helped to contextualise and describe the materiality of the city. We did a thematic analysis of a database of 62 news clips relating to violent events connected to drug cartels in Medellín from 1980 to 2001.
In the second stage (September 2019-November 2020), we collected data to explore the themes identified during the first stage. The material consists of: (1) Interviews with individuals involved in memory construction, including the personal secretary of the mayor of Medellín (2016-2020), a journalist working for the programme Medellín abraza su historia (Medellín embraces its history), and the founder of NarcosLab, an independent initiative to preserve memories of the narco-violence in Medellín.
(2) 26 interviews with residents of Medellín, who lived in the city at the time the cartels were highly active (ca. 1976-1993). Some of the 26 interviewees belong to the category of victims (Walklate, 2007), although there are significant differences between them. Eleven of them had witnessed violent events such as car bombings. Fifteen of the interviewees were directly affected by the violence, through the loss a relative (usually a parent or spouse) or were direct victims who had been kidnapped or physically and psychologically injured. In the article, the voices of Medellín's inhabitants and those of victims and survivors provide a sensorial account of what it was like to live in a city permeated by drug violence. We make use of the sociology of the senses to convey the 'bodily experience' of living in such an environment. (3) A database with information from the mayoralty of Medellín (2016Medellín ( -2020 including official statistics that were not publicly available. The project was approved by the Norwegian Centre for Research Data, the national body responsible for privacy and the ethical approval of research activities. Since the data was collected in Colombia, we also complied with the Colombian legal requirements concerning research ethics, set out in Resolution 0843, published in 1993. We obtained the informed consent of all interviewees, after explaining the purpose of the project to them in Spanish. Throughout the interviews, the researchers reminded participants that they did not have to answer any questions they did not want to. We also took steps to avoid re-victimisation, which might have resulted from the sensitive nature of the interviews and the vulnerability of some of the interviewees. These steps included debriefing and follow-up conversations. Interviews were recorded and transcribed. Due to ethical considerations, we anonymised all the interviewees. The names used in the article are therefore pseudonyms, except for Federico Arellano-the first victim of drug violence to eventually receive recognition as a victim under the Victims' Law-whose name became publicly known and who requested that his and his father's names should be used. We analysed the data using the software Atlas.ti, guided by the premises of grounded theory (Strauss & Corbin, 2012).

A War Within a War and a City Brought to its Knees
Violent internal conflict has been one of the distinguishing characteristics of Colombia from its independence in 1810 (Medina Gallego, 2010;Offstein, 2003). War and conflict in Colombia changed in 1948, when la violencia began-a period when guerrilla groups emerged as a significant force and war became a means to change economic, social and political relations within the country (Medina Gallego, 2010;Offstein, 2003). Between 1958 and 2012, 220,000 people were killed as a result of the internal armed conflict in which the most visible actors were the various guerrilla groups, paramilitary groups, and the Colombian Army (Goyes, 2015;Leal Buitrago, 2011); 81.5 per cent of the victims were civilians, the rest combatants. During this period, 5,700,000 people were displaced by force (GMH, 2013). Particularly in the 1980s, the displacement became fuelled by a deadly conflation of political violence and cocaine trafficking. Paramilitary groups simultaneously acted as 'an outgrowth of a complex alliance between narcos (drug traffickers) and agrarian elites, as drug-trafficking private militias and counterinsurgent battalions' (Ballvé, 2012: 603; see also Medina Gallego, 2009). Also guerrilla groups adopted a more confrontational strategy and began to use force for offensive rather than defensive purposes, with different sections pursuing diverging goals and some getting involved in drug trafficking (Centro Nacional de Memoria Histórica, 2013).
In the 1980s violence in Medellín greatly exceeded that found in the rest of the country (Maclean, 2015: 29). This was a result of the growing power of Colombian drug cartels, which was fuelled by the rapidly increasing US demand for cocaine (Lessing, 2018). Building on the enormous profits of the global trade in narcotics, the Medellín cartel established a substantial power base that enabled it to cultivate the disenfranchised youth of Medellín's barrios as sicarios [youth assassins] (Maclean, 2015: 39-40). Although Escobar's cartel initially collaborated with the armed forces, his relationship with the state degenerated into all-out violent conflict, which could be described as a war within a war.
After the assassination of justice minister Rodrigo Lara Bonilla the conflict intensified and led to extreme violence on both sides, with the state increasingly co-opting paramilitaries and the competing Cali cartel into their struggle (Lessing, 2015). During a seven-year period, Escobar alone, in addition to causing numerous civilian casualties, 'assassinated a leading presidential candidate, a standing state governor, an attorney general, chiefs of police, countless judges, and the director of a major national newspaper' (ibid.: 125). Car bombs were a common part of his repertoire to prevent his extradition to the United States (Aristizábal, 2018). Between September and December 1991, more than 100 explosive devices went off in the city 4 .
As Lessing (2018) observes, the conflict between the Medellín cartel and the state is unique in the extent to which the cartel openly challenged the supremacy of state institutions. The conflict has been fully documented in both popular culture and in scholarly literature (see inter alia Maclean, 2015;Vasquez, 2017). However, the focus has mainly been on the perpetrators, especially Escobar's violent clash with the state, rather than on the victims and the lives of the civilian population affected by the violence. In this article we turn the focus on the civilian population and argue that it is not only objective elements (such as the number of casualties) that made the conflict into something very like a war, but also the subjective experiences of those affected by it.
Remembering this period, the inhabitants of the city we interviewed describe how they saw car bombs, executions, homicides, kidnappings and many other forms of destruction and violence, and experienced the pervasive fear that affected all segments of the population and the young in particular. As Luciana, a relative of a kidnapped politician, put it: 'There was not one person who felt completely safe in this city.' She describes it thus: It is that fear when you're going out on a Friday, because you are a teenager and you want to go out with your friends. The parents' fear, in a time where there are no cell phones, there is no way to keep an eye on young people. We wanted to go out, but were afraid we might run into a bomb. The brother of a boyfriend I had at one point was killed in the Oporto massacre 5 , so, like it or not, you experience violence through all your friends and family.
José, whose brother was killed by the Cartel, was a youngster at the time and remembers it this way: You had a psychosis (it lasted a long time) about going out into the streets. I would often go to the intersection and walk six or seven kilometers, or more, to get to my friends. When a motorcycle came along, it was like 'Oops! They're not going to shoot me, are they?' The bike carried on and oops! you relaxed. There was a policeman close by, terrifying! At that time, they killed policemen. You didn't want to have a policeman near you, you walked away. It was a collective psychosis that was part of everyone's mentality. /…/ I mean, at ten o'clock at night there was a sort of undeclared curfew, which later became official as a result of all the fighting that broke out. But before that, the curfew had already been declared by the gangsters. It was unspoken; you couldn't give papaya [slang for 'give others an opportunity to take advantage of you'], as they say around here. It was a time of terrible anxiety.
The statement reveals not only the pervasive sense of anxiety and concern with survival and self-preservation, but also the feeling there was direct threat to state institutions, particularly the police. There was a sense that the very fabric of social life and state institutions was being destroyed. Salomé, who worked as a journalist at the time summed it up like this: Pablo Escobar brought this country to its knees. And brought this city to its knees. And that man brought civil society to its knees too. He brought us all to our knees. Some let themselves be bought and others were compelled to save their lives. We had brave people like Luis Carlos Galán 6 , who knew they were going to kill him, and he was so aware of the role he had chosen and they killed him. But the vast majority, what made them bow down? Was it out of fear, out of ambition, out of terror-whatever. But Pablo Escobar brought this city to its knees and that brought this country to its knees.
Several of our interviewees were children at the time. They remember evacuation drills at school, living with bomb threats and, as Isabella, daughter of a prominent politician, expressed it, a life surrounded by weapons: 'My house had guns everywhere. My babysitters were the bodyguards, my friends were the bodyguards. They had me in one hand and a machine gun in the other. That was what life was like for me'. The sense of constant threat was felt particularly intensely by those whose parents and family members were directly targeted by the cartel, but anxiety and the need for constant alertness was felt throughout the population. As Mariana, whose husband was a victim of Escobar, remembers: 'Airplanes exploded, there were bombs in different places. You always felt anxious, thinking "when will it be my turn?"' Several interviewees remember constantly turning on the radio to get updates about the latest developments. For many, these events had a lasting impact. As David, a taxi driver, said: That period left a trauma in me, there was always anxiety because one was 'just living' and suddenly 'boom'. I experienced that and I had many close friends and relatives that died during those bombings.
Traumatic memories often had a sensory association such as the noise of ambulances and bombs exploding, the smell of blood and corpses on the streets (Goyes and Franko, 2021).
While highly traumatic or well-known events, such as the killing of loved ones, the killing of political figures and well-known massacres were major features of the recollections, most also included less dramatic, but equally insidious effects of the period. These sensory memories are very similar to those described in scholarly literature on the trauma of war. Although research on the psychological impact of war on civilian populations reflects cultural and political variations in the war-affected societies, it also shows a remarkable level of similarity (Krippner and MacIntyre, 2003: 4). The term war stress is used 'to refer to the multiple stressors that people are exposed to when they have experienced war, either directly or indirectly' (ibid.: 6). At a collective level, war stress has a number of long-term health, social, economic, cultural and political consequences that can be described as war trauma (ibid.: 7). While many of our interviewees were still deeply psychologically affected by the loss of loved ones (a spouse, a parent or other close family member), the interviews also describe more widespread effects of the violence that can be felt at the collective level and could easily fit into the category of war trauma. Some also expressed concern about its effect on younger generations.
Wrong Kind of Victims. Even though they were traumatised by the violence, many of our interviewees did not like the term victim, preferring to describe themselves as survivors. As Salomé, the journalist, observed: I don't know if the word victim is worn out. But it is a label the state imposed, and I don't like to use it to describe human beings. Because I, being a victim, am someone who has lost their dignity, their ability to move on, or their ability to be and that definition impacts the person forever. As we, the paisas [slang for people from Medellín] say, we live according to how we tell our story. I prefer to think that I am a survivor. I identify with survivors, and I believe that we are all survivors. That's what hurts me the most that day, because Escobar died but he left many hurt survivors. I believe that it is totally irreparable damage. For a girl like mine: What consequences of that does she carry? I went through this through my pregnancy, the story that I am telling you. I was diagnosed with anxiety and fear. She was a baby that suffered even at the womb because of what her mother thought and felt.
The statement demonstrates not only a need to define oneself as resilient (i.e. a survivor rather than a victim), but also the level of collective damage felt to have taken place, including the damage done to her unborn child. Similarly, Isabella did not see herself as a victim but thought-in collective terms-that anyone 'who walked through the streets of Medellín every day is a survivor. There was a bomb every metre; going to shopping centres was a risk, an adventure. I believe that anyone who was going about at that time is a survivor'. Also several other interviewees disapproved of victimhood as an individualistic project and stressed the importance of collective memory. Some resented individual victims promoting their stories because, as Isabella observed, 'many people have got a lot of advantage from that position of victim'.
Others, however, considered it appropriate to be classed as a victim, particularly in relation to the aforementioned Victims' Law. Many were perplexed by the lack of official recognition, as Emiliano, the son of a judge killed by Escobar, says: So one does not understand why the victims of drug trafficking are not regarded as victims /…/. So it seems a legal absurdity to say that the victims of drug trafficking are not victims of the armed conflict in Colombia. When we all knew that the cartels financed guerrilla attacks. They financed the taking of the Palace of Justice. A proven fact. They financed the protection of their crops and promoted guerrillas and paramilitaries. It is incredible that they have made a law that does not apply to the victims of drug trafficking.
As in other post-conflict situations (Jankowitz, 2018;McConnachie, 2012, 2013), in Colombia the political issues surrounding victimhood are fiercely contested. The Victim's Law was drafted in 2007 as a response to the intense criticism that was directed at the so-called Justice and Peace Law. In 2005, a former Colombian president, Álvaro Uribe Vélez, had negotiated a peace agreement with the main paramilitary group operating in the country. Uribe's Justice and Peace Law granted substantial benefits to paramilitary combatants who chose to demobilise. While the declared aim of the law championed by Uribe was to replicate the 'truth, justice, reparation and guarantees of non-recurrence' framework, that the international community had developed to deal with war transitions (Uprimny Yepes et al., 2006), commentators critiqued the initiative for imposing insufficient sentences on combatants who had committed appalling atrocities.
The controversial Justice and Peace Law added to the 'parapolítica scandal', which exposed the close links of many high-ranking elected officials to the paramilitaries (Counter, 2018: 127). This then spurred the creation of the Victims' Law, whose proponents wished to put an end to 'the government's undue deference to paramilitaries and subsequent neglect of conflict victims', while its opponents objected that the law would treat 'terrorists [members of guerrilla groups who were also victims] as equals to police and soldiers' (ibid). Despite such opposition, the Colombian Congress ratified the Victims' Law in 2011.
According to Sandvik and Lemaitre (2015: 259), the Victims' Law 'transformed the small-scale reparation mechanisms embodied in the Justice and Peace Law into a large-scale transitional justice process'. The Victims' Law grants victims the right to benefit from a program to protect a dignified life (including economic support until they cease being in a vulnerable situation), family reunification, a safe return to the place they were displaced, and re-possession of their land (should they have lost it). 7 The law is explicitly founded on the principles of justice, promoted by international institutions and by various norms enterpreneurs, and received a symbolic endorsement of the international community through the presence of the UN Secretary-General Ban Ki Moon at the adoption ceremony (Sandvik and Lemaitre, 2015). 8 Our primary point of interest here is not the overall analysis and efficacy of the law, but rather how the binary distinction between war and crime in the law may have contributed to the systematic and structural exclusion of victims of drug violence from transitional justice processes. Although excluding victims of drug related violence from legitimate victimhood may have been motivated by the state's desire to limit its own accountability (Killean, 2018) it is also in line with internationally established traditions of thinking about transitional justice. As Counter (2018: 128) points out, the law specifies with great precision that the category of victim is reserved for individuals who were victimised by officially recognised actors, and who had been subjected to one of twelve, officially recognised, 'victimising acts'. 'These acts, however, must be related to armed conflict (as opposed to "common crime")' (ibid.).
Valentina, the daughter of a judge killed by cartel violence told us about her experiences when trying to obtain legal recognition as a victim: Because, when the Victims' Law began, [a friend] told me suddenly, 'you can be among the victims'. So, I began to go through the whole process and they told me 'No, this is not classified as being a victim because it is not a guerrilla or paramilitary'. But in the end it is almost the same. Because later, it was like in '94, they saw that they were connected: that the guerrillas were connected to many of Escobar's people. So, I say "Yes, I am", if I am a victim. And that part of the victims of drug trafficking, of the guerrillas, paramilitaries, is the same.
The distinction made between categories of victim seemed incomprehensible to her because of her awareness of the intertwining of the two forms of conflict. Similarly, Federico Arellano told us: The lawmakers misunderstood the so-called internal armed conflict. For them it was a conflict between the guerrillas and the paramilitaries, which left out of account important elements such as the relationship between the state and the drug cartels. Drug dealing has been the fuel of the conflict.
Due to this strict demarcation of the 'field of recognisability' of victims, as Counter (2018: 129) points out, 'certain lives are erased from consideration for reparations. ' We shall now argue that these structural erasures, also found internationally, help create a hierarchy of victimhood, which has important consequences for victims of mass drug violence and for the societies damaged by it.

'An Ignored Death': Drug Violence and the Global Hierarchies of Victimhood
Much scholarship on the hierarchies of victimhood stresses the importance of moral considerations, particularly perceptions of victims' innocence McConnachie, 2012, 2013). Although victims of drug violence in Medellín meet the criteria of 'innocence' and 'respectability', they were excluded from the legal recognition as victims of war. Two factors played out in the phenomenon: first, at a national level, the 'metaconflict positioning of the legislators' (Hearty, 2016: 334) led them to be careful about making official the connection between drug trafficking and political conflict in the country. As Giraldo Cerón and Velasco (2019) suggest, the debate about the links between drug cartels and politicians has been a heated subjected for more than a decade, and conceivably legislators avoid the risk of acknowledging the connections. Second, the influence of international actors in the Colombian conflict and peace process influenced the embrace of internationally established approaches and concepts. According to Cujabante Villamil (2016: 216) Northern actors influenced the peace process and the Victims' Law on several levels: politically, Northern nations became facilitators of the processes; economically, they funded the peace dialogues; and theoretically, the Colombian peace processes followed Northern models.
Victims of drug violence have undoubtedly experienced grievous harms, and our findings show that they also feel neglected and impotent in terms of having their voices heard and being able to access justice. Their public standing and the legitimacy of their claims are defined by the fact that they are victims of crime rather than of war or political violence. Victims' objective may not necessarily to achieve official recognition of their status or material reparation; many preferred to describe themselves as survivors and wished to convey an impression of resilience. However, despite differences regarding the victim label and status in legal terms, our interviewees had common concerns, to do with truth, impunity and memory. These concerns resonate with victims' needs in other transitional justice and post-conflict settings. As Brants and Klep (2013: 36) observe: 'Truth, collective memory, and history-telling have become buzzwords in the transitional justice debate, conceptual keys to reconciliation, democracy, and peace in conflict-ridden nations.' The importance of finding out the truth was forcefully argued by María, daughter of a judge who was assassinated by the cartel: I feel that there was no one, no one from the justice system, who took my father's case forward. And that hurts a lot, you know. That it is like an ignored death. I don't think he deserves it. In the same way as he gave his best efforts and his own life to carry out a judicial process, they should also do the same for him. So, 35 years have passed. That's what hurts me the most, you know. I believe that impunity is one of the things that affects me the most Look, nobody's going to bring him back, nothing can fix that. In other words, a family was broken. Six lives (of my grandparents and my uncles) were broken, many days were broken, and beyond my personal process, absolutely nothing can be done about that. But the least I hope for, and what I sincerely want, […] is that they will make some effort at least to investigate what happened. Years go by and years go by and the opportunity to find out is increasingly lost The importance of truth telling emerged in several ways. Sometimes it related to the fate of individual victims: a wish that their death should not be ignored and to find out the truth and make it known. Some started blogs or wrote books in the face of the lack of official interest and recognition. Truth-telling was also frequently described as a project that would contribute to collective memory and help repair the damage done to the community. As Isabella argued: I think everyone should study it, but with the aim of ensuring that it doesn't happen again and of remembering a city to which so many things happened, and which was reborn in the midst of violence. Because Medellín is an example of the rebirth of Colombia in general, and so it seems to me that it should be studied-it should not be forgotten and left in the past Those affected by mass drug violence resemble other communities in post-conflict settings. This is the case not only with regard to the severity of traumatic experiences, but also with regard to the need for truth-telling and the shaping of the collective memory as necessary steps for social recognition. Our interviewees stressed the importance of there being an acknowledgement of the seriousness of the crimes committed, and were afraid that their losses would be forgotten. Such feelings are similar to those emerging in the aftermath of other atrocities and in scholarly literature have been presented as an argument for truth commissions (rather than traditional criminal justice trials) as the preferred response to them. As Brants and Klep (2013: 36) point out: The primary concern of truth commissions is to bring justice to victims through publicly establishing what happened to them. But this obvious difference hides a significant similarity: such truth-finding also promotes the development of a collective memory by establishing a version of history that informs, and is informed by, the memories of those involveda shared truth about crime and injustice that allows sense to be made of a traumatic past and is a prerequisite for a stable future.
However, few such avenues to social recognition and collective repair have been available to those affected by drug violence in Medellín. The criminal justice system there was severely damaged by the onslaught of cartel violence, while in the last twenty years truth-telling has to a large extent been taken over by the global entertainment industry. We have written elsewhere about the large-scale commercialisation of the violence in Medellín, to be seen, for example, in the popular Netflix series Narcos, and about the distress and anxiety this has caused for those affected by it (Goyes and Franko, 2021).
In 2016, Federico Gutiérrez, the newly elected mayor of Medellín, put the issue on the political agenda and made a valiant attempt to change the narrative about the city's violent past His ambition-ultimately unsuccessful-was to make the violence in Medellín part of a global memory of atrocities. We interviewed (ibid., 2021: 14) the mayor's personal secretary who angrily compared Medellín's situation with that of similar sites of atrocities, arguing that, while victims in Northern societies receive sympathy and solidarity, Medellín is treated as a source of entertainment: The United States had their twin towers destroyed, in Europe there have been recent attacks, and in the Second World War there was a brilliantly evil character; [similarly] thirty years ago we had planes sent to the ground, we had bombs, we had a brilliantly evil character, why does the whole world feel solidarity with the Jews, with the Germans, with the USA, with France, Spain and England, but not with us?
His words show a belief that there is a global hierarchy of victimhood, where Northern victims of atrocities receive a greater acknowledgement and solidarity from the international community.
Global responses to, and ideas about drug violence are, therefore, of central importance. Carlos, a judge who had witnessed the assassination of colleagues and had been forced into exile, stressed the importance of international support and truth-telling in international fora: [W]e had the duty of reporting these events, but no longer among ourselves, because we were clear about what was happening and where the bullets were coming from, but to get this information out of the country. And then, in 1991, when we took on the task of making it known in Europe, and the German judges invited us and the Dutch invited us, and Swedes invited us because we said, as Eduardo Galeano says in Las Venas Abiertas de América Latina, we thought that we were keeping a silence that was quite similar to stupidity, right?

Conclusion
Conceptualisations of violence, including the violence of war, are historically and contextually contingent. Conflict-related sexual violence, for example, has received proper recognition as an element of wartime violence only in recent years, and has led to efforts by the international community to fight impunity (Houge and Lohne, 2017). Because the prevailing view is that drug-related violence is conventional crime, it is excluded from discussions of mass atrocities and transitional justice, irrespective of the scale of its casualties. Consequently, when scholars and practitioners refer to 'fighting impunity'-a powerful slogan used by contemporary justice movements (Lohne, 2019) -this does not include victims of drug violence. As we have seen, intense resentment of impunity was expressed by our respondents.
In this article, we have critically engaged with the conceptual distinctions between war and crime, and explored their implications for affected individuals and communities. Drug-related mass violence is mostly found in societies in the global South, and especially in Latin America. We have argued that the prevailing binary distinction between warfare and crime fails to adequately capture on-the-ground realities in Southern societies. This conceptual distinction is unsuitable for meeting the needs of victims of this type of mass violence by creating a structural divide, and thus a hierarchy among victims, which ignores the severity of suffering undergone by the affected communities.
A sensory approach enabled us to describe drug violence in Medellín not from a (detached) analytical perspective, but as it was experienced by the survivors and the city's inhabitants. Our aim was to get closer to Southern realities, rather than accepting existing conceptualisations and analytical perspectives, which mostly define drugviolence as 'ordinary crime', thus exhibiting a distinctly Northern epistemological bias. According to critics, the contrast between the peaceful North and the violent South-where violent loss of life is normalized-has been one of the defining features of colonial imaginaries, which continue to influence contemporary scientific approaches (Morrison, 2006;Said, 1977). However, our objective has not been simply to critique Northern epistemological influence, but rather to show its consequence for the affected communities. A sense of justice has remained permanently elusive to victims of drug violence in Medellín. More than three decades after the events, they expressed that the conflict was still very much alive in their consciousness and shaped their everyday experiences. They have had little redress from the over-burdened national criminal justice system, and do not feel included in the global discourse about mass atrocities. Although their sense of grievance may have been less acute had the criminal justice system been more capable of responding to the crimes in individual cases, the gap between the two conceptualisations of victimhood is relevant also because the crimes in question affected entire communities. This is particularly pertinent to issues of collective memory and cultural trauma. Since truth-telling and collective memory are extremely important to them, our interviewees feel greatly affected by the commercialisation of Medellín's violence by the global entertainment industry (Goyes and Franko, 2021). McEvoy and McConnachie (2013: 496) point out that giving victims a voice is a central element in achieving justice in postconflict settings. The transformation of Medellín's story into entertainment demonstrates how Southern victims of drug violence yet again differ from victims of war and victims in the global North, who increasingly function as the 'representation of the world through pain' (Walklate, 2016: 5).
If we are to advance conditions for access to justice, particularly in countries of the global South, we need a more nuanced conceptual understanding of mass violence. While scholars of international humanitarian law have recently raised questions about whether drug cartels should be recognised as a party to a non-international armed conflict 9 , little attention has so far been paid to how victims of drug violence can best access justice. The massive onslaught of drug violence on criminal justice institutions and the share scale of casualties indicate a need for crime to be lifted out of its individualised criminal justice frame, and for attention to be paid to collective understandings and responses, to match those accorded to victims and survivors of war and mass atrocities. Finally, drug violence is thoroughly intertwined with politics in terms its causes and responses to it (Durán-Martínez, 2018). An 'intimate connection between political change and large-scale criminal violence' has been observed in several Latin American countries (Trejo and Ley, 2020: 3). An a priori placement of this form of violence into the category of 'ordinary crime', rather than that of political violence, obscures the involvement of political actors in it and closes off other, potentially productive, ways of understanding it and addressing its consequences.