Reframing historical narratives : two Colombian case studies

(1)

Graduate School of Social Science

MSc Conflict Resolution and Governance

Reframing historical narratives:

Two Colombian case studies

Author: Veera Veijalainen Student no: 11184655 Supervisor: Dr David Laws

Second reader: Dr Martijn Dekker

(2)

THESIS SUMMARY

Over the past two or three decades, there has been a distinct movement towards the analysis of memory within a wide range of academic disciplines, constituting what the prominent historian Jay Winter (2006) has termed, a ‘memory boom’. Memories are socially constructed and, despite their subjectivity, are essential for both individuals and groups in assigning meaning to past events and experiences and as such, developing our frameworks, both personal and collective, through which we understand and navigate the world. In the case of (post)conflict situations, memories are always contested and challenged different ways of remembering and forgetting often support social polarisation and maintain divisions, thereby perpetuating conflict. The reconstruction of memory in the reconciliation process of a conflict is known as memory work. Herein lies the focus of this thesis, seeking to understand how memory work contributes to reframing historical narratives and in turn, what impacts it has on the process of reconciliation.

This thesis’ research is situated within the context of Antioquia, Colombia, or more specifically, the case studies of a community led memorial space called the Salón del Nunca Más (Hall of Never Again) in Granada and a public memorial museum known as the Casa de la Memoria (House of Memory) in Medellín. Medellin is both the former ‘most dangerous city in the world’ and the location of some of the most innovative reconciliation and memory initiatives in Colombia. Grenada is a small town of roughly 20,000 inhabitants that experienced extremely high levels of violence, displacement and trauma in Colombia. Initial research for this thesis would reveal that despite the apparent limited diversity within groups engaging in memory initiatives, they do contribute in a positive way to (and support) the reconciliation process by promoting more complex historical narratives.

To ascertain the impacts that these case studies could potentially have on the broader reconciliation process in Colombia, this thesis employs a normative theoretical framework, understanding memory as a social construct formed through ongoing processes of remembering and forgetting and sought to establish the dynamics of the relationship between memory and conflict through the analysis of historical narratives. This thesis makes a distinction between the case studies according to the social framework in which they are situated and the narratives that they conceptualise through their engagement with identified victim groups and individuals. Following this, the author employs an ethical framework through which to analyse whether the narratives espoused by the case study initiatives contribute to polarisation and escalation or of reconciliation processes. This theoretical framework was

(3)

operationalised through a variety of qualitative research methods including semi-structured and narrative interviews with victims, community leaders and workers within the initiatives, in addition to the semiotic and discursive analysis of various mediums employed by the Casa de la Memoria and Salón del Nunca Más such as photographs, diary excerpts, informative exhibitions, videos and artistic installations. This process included a 6-week data gathering mission to Colombia.

The findings of this research were established through the interpretation of the collected data using the theoretical framework. The impact of the case study initiatives is quantified in this study mainly through analysis the experiences of the visitors, considering their ability to share and address painful memories of the conflict and participate in a collective reconciliation process. The results of the research reveal that, despite the general focus on victims (compromising diversity) and the context of ongoing conflict, the two studied memory initiatives contribute in a positive way to reframing historical conflict narratives. However, the Salón del Nunca Más illustrated the ability to achieve greater levels of participation and ownership of victims through using a more contextual and localised engagement framework whereas the Casa de la Memoria employed a pluralistic discourse and as such undermined its ability to truly engage individuals with the same level of sustainable and meaningful impact.

(4)

ACKNOWLEDGEMENTS

I would like to thank my supervisor, Dr. David Laws, for his insightful comments and seemingly endless availability for discussions despite his busy schedule. I would also like to extend my appreciation to Dr. Martijn Dekker who agreed to be my second reader, and to the whole CRG group for providing endless entertainment and inspiration in this entire thesis process.

To my family – thank you for always being here for me, for proofreading my work, and especially for the moral support over the last 14 hours. To Gwen, thank you for your amazing company and support in all forms during our field research in Colombia – it was super cool. To my baib, I could not have done it without you.

Last but definitely not least, the biggest thank you goes out to all the people in Colombia who made this possible. I hope I have done justice to your incredible stories.

(5)

CHAPTER 1: INTRODUCTION

Over the past two or three decades, there has been a distinct movement towards the analysis of memory within a wide range of academic disciplines, constituting what the prominent historian Jay Winter (2006) has termed, a ‘memory boom’. The proliferation of research dedicated to the study of memory has been particularly evident within the social sciences, in which the central premise is that memory is socially constructed within groups, rather than being one individual’s objective representation of the past. Despite their fluidity and subjectivity, memories have been shown to be essential for both individuals and groups, not just for the purpose of recalling the past but, more importantly, for assigning meaning to past events and thereby forming and shifting models of our experiences that we use to navigate the world (Spreng 2013).

Given its subjective nature and influence on how we make sense of the world, memories are always contested and challenged. In the case of (post)conflict situations, this becomes particularly evident as different ways of remembering and forgetting often support social polarisation and maintain divisions, thereby perpetuating conflict. For example, Patrick Pinkerton (2012) studied the memory struggle manifested in the James McCurrie Robert Neill Memorial Garden (for victims of the IRA) in Belfast to show how those who frequently vandalised the memorial garden remain trapped within the logics of those who erected it, whereby communal belonging is viewed in absolutist terms, and historical memory is separate and not shared. To give another example, Veena Das (2007) analysed the systematic “forgetting” of the previously close relations between Hindus and Sikhs in India, allowing for a conflict narrative of Sikh history viewed as as a series of systematic dualisms separating the Sikh self from the Hindu Other.

With these illustrative examples, it is clear that conflict and memory (or forgetting) are often two sides of the same coin (Wagoner & Brescó 2016). On one hand, conflicts deeply mark the memories of individuals and collectives, thereby hampering future reconciliation as in the example of the memorial garden in Belfast; and on the other, memory is behind many conflicts, insofar as certain ways of remembering and forgetting past episodes imply bringing the past into the present and with it old scars, grievances, resentments, hatreds, and senses of revenge, as in the case of relations between Sikhs and Hindus in India. Although it appears that the majority of previous research on the relation between memory and conflict has focused on how remembering and forgetting support conflict escalation, there

(7)

is a growing body of practice-oriented literature that emphasises the opportunities of steering memory in a way that serves reconciliation and conflict resolution. Much of this has been based on the actions of the Truth and Reconciliation Commission in South Africa (Asmal et al. 1997). It is this opportunity of reconstructing or reframing memory to support reconciliation efforts rather than conflict escalation that serves as the focus of my thesis.

This focus derived from curiosity. While traveling in Colombia in the spring of 2017, I noticed that initiatives around reconstructing memory of the armed conflict1_{could be found almost everywhere.}

This was evident in the numerous memorial spaces, murals, theatre, music, and art centered on the theme of remembering the armed conflict, which continues to cause destruction and displacement in several parts of the country. When I started reading about the reconstruction of memory as part of the reconciliation process, I noticed that Colombia stood out from other contexts where memory work has typically been carried out as a part of peacebuilding in that it remains in the midst of an armed conflict. Quite extraordinarily, even at the height of the conflict in the late 1980s to early 1990s, Colombian victim groups were already engaged in collective remembering as a way to resist oblivion through forced silencing by the conflict actors. Although literature around reconciliation and memory work highlights the importance of engaging diverse voices to foster coexistence and inclusiveness, this appears elusive in practice, as most initiatives in Colombia that engage in historical dialogue have been focused categorically on victims. Be that as it may my initial observations suggested that, despite the apparent limited diversity within groups engaging in memory initiatives, they do contribute in a positive way to (and support) the reconciliation process by promoting more complex historical narratives. Since this would be considered a surprising result given the ongoing conflict context, I was thus interested in exploring the Colombian case further. It was from this apparent gap between theory and practice in Colombia that I derived my general research question: how does memory work contribute to reframing historical narratives around the armed conflict?

I have approached this question by drawing on two case studies from the department of Antioquia, Colombia. The first is a case study of a community led memorial space called the Salón del Nunca Más

1_{The Colombian armed conflict emerged from a combination of economic, political and social factors in the}

country around 60 years ago. The origins are commonly traced back to the ten-year conflict that was fought between the Conservative and Liberal parties known as La Violencia (1948-1958), which caused 200,000 casualties and failed to resolve the bitter political division. From the 1960s, revolutionary guerrilla groups began forming in rural areas from peasant and communist uprisings, soon to be countered by right-wing paramilitary groups. With the lack of state presence in many rural parts, the growing drug economy as a source of financing for guerrilla and paramilitary groups, and the complicity of Colombian state forces in the violence perpetrated by paramilitaries, the conflict expanded and consumed almost every part of the country, leading to the displacement

(8)

(Hall of Never Again) in Granada, a rural town of around 20,000 inhabitants (including those of the surrounding rural areas), 170 km to the north east of Colombia’s second city Medellín and scene of some of the most intense violence in the whole country between 2000 and 2002. The second is a case study of a public memorial museum known as the Casa de la Memoria (House of Memory) in Medellín, capital of the department of Antioquia and a city that, despite the dramatic reduction of violence in the mid 2000s, remains infamous for its recognition in TIME magazine as the world’s most dangerous city in 1988, largely thanks to the fear and destruction brought by the infamous drug lord Pablo Escobar and his Medellín cartel.

This thesis is structured as follows: first, I provide a historical overview of the armed conflict as it played out in Granada and Medellín, as well as a contextual summary of the ongoing Colombian peace process. In chapter three, I lay out the three part theoretical framework underpinning my study. The first part draws a distinction between the two cases under investigation according to the social framework in which they are situated, the second is concerned with conceptualising narratives; how they are developed and communicated, and the third introduces an ethical framework that tracks whether such narrative development moves in the direction of polarisation and escalation or of reconciliation. In chapter four I discuss the chosen extended case study design and methodology comprising of semi-structured and narrative interviews, participant observation and narrative analysis, as well as research limitations and ethical issues. Next, I present my findings and analysis in chapter five. My findings are divided into three distinct sections that correspond with the theoretical framework, with the first two being focused on an analysis of the narratives communicated in each case and their impacts on reconciliation, and the third on a discussion of the two cases in tandem, with reference to the distinct social frameworks in which they are situated. Finally, I offer some conclusions and recommendations for further research in chapter six. The appendix shows all translations of quotes taken from interviews and translated by the author.

(9)

CHAPTER 2: HISTORICAL & CONTEXTUAL

OVERVIEW

2.1 Brief historical overview of the armed conflict in Granada

The origins of the armed conflict in Granada trace back to the 1980s when guerrilla groups, namely the Revolutionary Armed Forces of Colombia (FARC) and the National Liberation Army (ELN), began embedding themselves in local communities in order to gain control of the area, which serves as an important strategic point between Bogotá and Medellín and has a strong agricultural economy (GMH 2016, p. 17). After a decade of guerrilla hegemony with occasional periods of violence, paramilitary groups arrived in the territory in the mid 1990s and, seeking to end the hegemony of the guerrilla, the two sides embarked on an all out war, with devastating impacts on the civilian population.

Between November and December 2000, Granada experienced one of the most intense episodes of recent violence in Colombia (Pinilla 2017). First, the paramilitary umbrella group United Self-Defense Forces of Colombia (AUC) executed a massacre in which 17 people were killed, and just a month later, FARC guerrillas assaulted the town and detonated a car bomb with 400 kg of explosives that left 20 people dead, 124 houses and 82 premises destroyed, and 120 houses damaged (GMH 2016, p. 18). In total, the National Observatory for Memory and Conflict has reported 519 victims of killings and 2,992 victims of forced disappearance from Granada, causing the displacement of almost three quarters of its population (cited in GMH 2016, p. 127). Since 2003 or 2004, the conflict situation has calmed considerably as the AUC paramilitaries entered a demobilisation process with the government, and as the Colombian public forces managed to finally gain control over the territory (GMH 2016, p. 127). Since then, there has been a strong movement, particularly on behalf of the local victim association ASOVIDA, to promote peace and prevent the eruption of more violence through memory work.

2.2 Brief historical overview of the armed conflict in Medellín

The violence of the mid-twentieth century in rural areas such as Granada accelerated the growth of cities and generated mass waves of immigration into urban areas like Medellín, a traditionally conservative, deeply Catholic and productive industrial centre. By the 1970s, the mass arrival of campesinos (rural people) and the subsequent social transformation and inability of the city

(10)

poverty and a proliferation of criminal groups. During the 1980s, this urban conflict entered a phase of open disputes that overlapped with the consolidation of drug trafficking and the end of a period of tolerance that had allowed the drug trade to develop (Ceballos-Melguizo & Cronshaw 2001, p. 114). Although the violence in Medellín is not a direct result of the drug trade, this did have spin off effects such as the rise of gangs and extreme right-wing paramilitary groups promising to bring order and justice (Ruíz-Romero 2018).

The conflict intensified in the 1980s with new waves of displaced campesinos entering the city due to increasingly violent combat between guerrilla groups, paramilitaries and Colombian armed forces in the countryside, in addition to the co-optation of the highest spheres of politics by Escobar’s Medellín cartel, which was by now at war with the state (Ceballos-Melguizo & Cronshaw 2001). The growing violence and proliferation of new armed groups had a devastating impact on the residents of Medellín, making it the most dangerous city in the world during the late 80s to early 90s, reaching a staggering homicide rate of 6,809 people in the year 1991 (GMH 2013). Although extreme levels of violence continued well into the 2000s, at least partly as a result of the aggressive policies of Colombian presidents like Álvaro Uribe (2002-2010) towards guerrilla groups, the 1990s were also a time during which the city government started to plan for strategies to reduce violence and civil society groups started to mobilise and participate in that process. At least some of the credit for the ‘rebirth’ of Medellín and the reduction of violence in the 2000s has been attributed to Mayor Sergio Fajardo’s (2003-2007) ambitious experiment in ‘social urbanism’, under which significant investments were made to poorer communities in an attempt turn the city from one of the most violent into the most innovative.

2.3 Steps towards peace

Though Colombia has a long history of violence, Colombians have also demonstrated capacity to resist it. Unusual to societies in violent conflict, efforts aimed at reconstructing historical memory for reconciliation purposes were already underway at the height of the armed conflict. In fact, Colombia has for decades been investigating the causes of violence: over the last 50 years, there have been more than twenty commissions which investigated the political violence and its history (Barkan 2016, p. 19). Nonetheless, it was not until the mid-2000s that there was a sharp rise in memory initiatives. This is at least in part due to the favourable conditions given to paramilitaries in the process of demobilisation, as

(11)

established in a 2005 bill2_{by President Álvaro Uribe. Ruíz-Romero (2012) argues that the responsibility}

to reconstruct the truth about what happened during the participation of paramilitaries in the conflict was thereby allocated to the perpetrators while undermining the participation of victims. Against this backdrop, there has been an apparent growing social mobilisation around memory claims particularly among civil society organisations in Colombia, which have arguably made it an “active response to the ordeals of war, and a rejection of the attempts to impose silence on so many victims” (GMH, 2013, p. 19). With the change of government from the hands of the staunchly right-wing and militaristic President Uribe to President Santos in 2010, Colombia has progressed considerably in its peace process. Decades of peace talks finally resulted in an agreement with the FARC in 2016, while in 2011 Congress passed the Victims and Land Restitution Law3_{, which created a framework for an ambitious and}

comprehensive transitional justice system aiming at restorative justice, truth and reparations for victims. In addition to concrete reparations such as compensation and assistance to displaced people, the Law also includes symbolic reparation measures, such as the creation of a national day of memory and the collection of oral testimonies to preserve historical memory. Although implementation of the Victims law has been extremely slow and fraught with practical issues (Correa 2015), it has to a large extent been inspired by the work of victim groups in recovering or reconstructing historical memory. My study thus includes not only a case of a community-led initiative reconstructing memory, but also a public institution supported by city, regional and national level governments.

The historical and contextual overview above provides the reader with the appropriate background knowledge that is required for the remainder of this thesis.

2_{Law 975, or the ‘Justice and Peace’ Law was aimed at facilitating the reintegration of the members of armed}

groups into civilian life. This law awarded former militants certain legal benefits on the condition that they adhere to the measures geared towards national reconciliation such as truth-seeking, reparations for victims, and an adequate re-socialisation.

(12)

CHAPTER 3: THEORETICAL FRAMEWORK

Research on memory has traditionally been pursued within the domain of psychologists and has focused on individual pathology. Over the past few decades, however, interest in the study of memory has been renewed across disciplines from anthropology and sociology to education, literature, history, and philosophy among others. These academic developments have tended to highlight the social aspect of memory. In this chapter, I lay out a framework to make sense of how social memory is constructed and explore the implications this has on reconciliation after violent conflict and other conflict outcomes.

The first section provides a conceptual overview of memory as a social construct that is affected by individuals, their social environments, and by the wider sociocultural setting and formed through ongoing processes of remembering and forgetting. The second section explores the relationship between memory and conflict. Literature on this topic has tended to focus on how memory is instrumentalised or manipulated and used to fuel and escalate conflict. This has raised the question of whether memory can, under other conditions, also serve to de-escalate conflict or promote reconciliation? Alternative theories on memory and conflict that suggest it can are reviewed in section three. These theories draw on narrative and the capacity to “tell another way” (Ricoeur 1999, p 9). By default, such a perspective requires a normative framework that distinguishes between ‘good’ and ‘bad’ narratives. This is described in section 3.4, in which I detail the operational framework that I have devised to enable me to answer my research question. Finally, I offer some interim conclusions, and sensitising questions arising from the theoretical framework.

3.1 Conceptualising group memory

For the purposes of my research, memory must first and foremost be conceptualised as a collective and social process. The theoretical underpinnings of this understanding of memory trace back to the late French philosopher and sociologist Maurice Halbwachs (1877-1945), who rejected the traditionally individual-psychological approach to memory. In his landmark study ‘On Collective Memory’ originally published in 1925, Halbwachs argued that memory can only function within a collective context through socialisation and communication: “it is in society that people normally acquire their memories [and] it is also in society that they recall, recognize, and localize their memories” (1992, p. 38). More recently, Wang & Brockmeier (2002, p. 47) conceived of individual (or ‘autobiographical’) memory as “an active construction embedded in a social weave of dialogues that are negotiated not only between an individual and his or her immediate social environment (parents, peers and significant others), but also,

(13)

equally important, between the individual and the larger cultural milieu.” Memory, whether individual or social, can thus be characterised as ‘living.’ It is an active past that is “selected, combined, and evaluated in accord with necessities and rules not imposed on the groups that had through time guarded them as a living trust” (Halbwachs 1980, p. 78). A very similar view is also proposed by Pierre Nora, a French historian best known for his seminal work on memorialisation, ‘Les lieux de mémoire’ (1984 [translated into English as Realms of Memory, 1997]). For Nora,

Memory is life, borne by living societies founded in its name. It remains in permanent evolution, open to the dialectic of remembering and forgetting, unconscious of its successive deformations, vulnerable to manipulation and appropriation, susceptible to being long dormant and periodically revived (1989, p. 8).

An important feature of social memory is that it always takes place within a defined group. All persons are a part, successively or simultaneously, of multiple groups. Some examples of such groups are a family, a class of students, a neighbourhood, a village or a city, and a nation. From this perspective, it is erroneous to refer to a singular ‘collective’ memory because each group constructs and reconstructs its own memory. Some of these groups are abstract communities, such as nations, while others are more immediate. Halbwachs refers to these as “distant frameworks” and “nearby milieus” respectively:

Ordinarily [...] the nation is too remote from the individual for him to consider the history of his country as anything else than a very large framework with which his own history makes contact at only a few points [but] the smaller the group, the greater the interest members have in these events. Whereas one may easily be lost in the city, village inhabitants continually observe one another. The group memory faithfully registers everything that it can about each member, because these facts react on this small society and help change it. In such milieus all persons think and remember in common (1980, p. 76).

The distinction that Halbwachs draws between nearby milieus and distant frameworks is analytically useful to consider how the social milieu influences the development of collective memory. In the Colombian context it helps to differentiate between a community-level memory initiative that plays out in a nearby milieu and a distant framework like the Casa de la Memoria. The size of the framework or group within which memories are formed bears relevance to the process of memory construction. Larger, distant, frameworks can be conceptualised as what Benedict Anderson (2006) famously described as ‘imagined communities’. Writing about the nation, Anderson argued that it "it is imagined

(14)

because the members of even the smallest nation will never know most of their fellow-members, meet them, or even hear of them, yet in the minds of each lives the image of their communion” (2006, p. 6). For Halbwachs, the idea of a nation as a distant framework means that it “can’t be said to be interested in the destiny of each of its members,” and so national history “retain[s] only facts of interest to the citizens as a whole” (1980, p. 77). While the nation remains an abstract community, Halbwachs admits that “between individual and nation lie many other, more restricted groups. Each of these has its own memory. Changes in such a group more directly affect the life and thought of members” (1980, p. 77). Building on such a perspective, it is evident that in larger groups the process of constructing memory is more distant from the memories of individual citizens than in smaller groups. In other words, the identities produced or reinforced by the construction of memory in smaller groups are expected to be deeper and more engaging than those produced or reinforced in bigger group. The effect that the size of the group has on the dynamics of memory construction is an important pillar of my theoretical framework, allowing me to envisage a distinction between the Casa de la Memoria as a distant framework and the Salón del Nunca Más as a nearby milieu. This distinction is useful in that it enables a more situated understanding of the practical impact of the narratives on visitors and participants of the two memory spaces.

3.2 Memory & conflict: general expectations

During times of mass violence and conflict, sectarian groups from different ‘sides’ of the conflict often exploit memories to foment further conflict (Barkan 2016). In the case of Colombia in 2018, we see some of the effects of this playing out in the presidential election. Here, polarisation between the political left and right remains arguably as sharp and severe as it was at the height of the armed conflict. The two candidates in the recently held second round of the election were Ivan Duque, widely seen as a puppet of former president Álvaro Uribe, who has been investigated and prosecuted by the Colombian Supreme Court for ties to paramilitary groups and death squads during his political career, and Gustavo Petro, one of the former leaders of the M-19 guerrilla group (demobilised in 1990).

In times of conflict and particularly in transitional phases, groups tend to compete with each other about their victimhood. This is because there is always a selective process of accounting for history and the need to negotiate victimhood reflects a need to look through conflicting accounts of history in order to clarify the roots of violence and determine who might bear the greater responsibility for them. On the other hand, the tendency toward competitive victimhood could also be due to the possibility that a sense of victimhood can legitimise further violence and aggression as a way to reclaim justice (Gill 2014). In any case, it is recognised that the tendency towards competitive victimhood is deeply

(15)

problematic and can greatly impede the prospect of future peaceful coexistence between conflicting groups. Keynan found that in conflict and transitional settings, group narratives tend to enhance the sense of being right, but also evoke fears of the Other’s intentions, and of the reoccurrence of past traumas (2014, p. 20). Memories of suffering and trauma thus tend to lead groups to identify more with their in-group at the expense of with the wider society, promoting a narrative of ‘us’ versus ‘them’ (Gaertner et al. 1993). Memories are therefore both shaped by conflict and they can fuel conflict. Wagoner & Brescó express this eloquently:

Conflict and memory are often two sides of the same coin, which seamlessly feed into each other. On one hand, conflicts deeply mark the memories of both individuals and collectives, thereby hampering future reconciliation. On the other hand, memory is behind many conflicts, insofar as certain ways of remembering dramatic episodes (whether recent or remote) imply bringing the past into the present and with it the old scars, grievances, resentments, hatreds, and senses of revenge (2016, p. 3)

Although much research has focused on how memory has served to foment further conflict, there is also a growing interest in how memory might be used for collective healing and reconciliation. In the following section I explore some alternative theories that reveal the transformative potential of memory work.

3.3 Memory & conflict: alternative approaches

As I have already mentioned, there are a growing number of initiatives aimed at reconstructing historical memory of the conflict in Colombia. In order to do that, these initiatives engage in what Michael Schudson calls ‘narritivisation’ (1997, p. 355), a process through which intangible memories are translated into discursive reality by concretising them in a story about the memories. Narratives are arranged in temporal order so that they have a beginning, a middle and an end, making them “the principal way in which our species organizes its understanding of time” (Abbott 2002, p.3). Only relevant events are included in the story, and the causal links between events are based not on formal logic or probability, but on plot. Since we cannot tell a story without neglecting or dropping some important events according to the kind of plot we want to build, narratives are “at the same time the occasion for manipulation through reading and directing narratives, but also the place where a certain healing of memory may begin” (Ricoeur 1999, p. 9). The potential for healing for Ricoeur comes from the fact that “it is always possible to tell in another way” (ibid.).

(16)

In my understanding, narratives thus serve as the potential link between remembering and reconciliation, however, as seen in the preceding section, they may also serve as the link between remembering and conflict escalation. Actors who work on historical memory reconstruction face both an opportunity and a risk. The opportunity is that memories can be reinterpreted to narrow differences between conflicting memories and narratives and to acknowledge the details that overlap and are shared across different narratives. The risk is that engaging in historical dialogue will lead to destructive outcomes if the stories are told in a way that bolsters the historical constructs that serve to maintain or deepen existing divisions, as is often the case in post conflict settings (Edlmann 2017; Barkan 2016; Lategan 2007). The key in determining the likely implications of remembering on conflict is narrative complexity. For Sara Cobb, “conflict escalation can be understood as itself a process of narrative simplification” (2013, p. 86). Implications of such simplification are that it “intensifies certainty and reduces complexity, two trends that combine explosively to generate and perpetuate violence” (ibid., p. 197).

For Ricoeur, the manipulation of memory (a bad use of memory in his normative model) can be seen in the excesses of certain commemorations that attempt to fix it in a reverential relationship to the past, while the ethics of memory is in the ‘good’ use of commemorative acts that “revert from past to future, and this revision from past to future is by way of drawing out the exemplary significance of past events” (1999, p. 9). A similar view is also proposed by Bernard Lategan, reflecting on his work and research in peacebuilding in South Africa:

In times of war […] and social transformation, historical memory often divides groups of people as each seeks to establish its particular ‘memory’ […]. Reconciliation in such conditions requires the conflicting memories to be articulated, recognized, respected – and also brought into a process of interaction with each other, which can lead to reassessments of the past and adjustments in understanding (2007, p. 26)

Such reassessments of the past have the potential to enable an understanding of individual and communal stories within a broader context. When this is successful, “the need for both individuals and a society or context to heal can be understood within dynamic and unfolding ways, thus feeding into and enabling broader systemic work” (Edlmann 2017, p. 232). A key element here is to create images of the past that will work to change or enlarge the identity of individuals and groups to being more inclusive (van Beek 2007, p. 207). The idea is to move beyond dichotomous and simplistic categories of victim/perpetrator and move towards a more complex understanding of the conflict and its dynamics.

(17)

3.4 Towards a normative framework of memory

Given my research focus on how memory work contributes to reframing historical narratives of the armed conflict, I am interested primarily in the practical level. This calls for further elaboration of a framework that can guide my analysis. In this section I describe the practical theoretical basis through which I analyse historical narratives in my research. There are two components to this: first, I stipulate how narratives are constructed in order to focus my analysis. Second, I lay out a normative framework that bridges these narratives with implications on reconciliation outcomes.

A useful framework for understanding how narratives are constructed in memorial spaces or museums like Casa de la Memoria and the Salón del Nunca Más is offered by Brockmeier (2002). Accordingly, narratives are presented using three orders: the linguistic (the plot), the semiotic (the wider symbolic space around the narrative), and the the discursive (the ‘telling’ of the story) (Brockmeier 2002). Narrative as a linguistic order comes into play when the story is told to someone through oral or written form. As mentioned before, this involves a process of ‘emplotment’ to foster a meaningful storyline. Second, since narratives cannot be considered without taking into account the sociocultural ‘language’ of the narrator, attention must be paid to the wider semiotic order. From this viewpoint, all meaningful organisations of signs, including material installations, constitute narrative texts (Brockmeier 2002). Finally, the discursive order, or the telling of the story, is seen as carried out by narrators who occupy a situated, sociocultural position - i.e. narratives are always performed in a specific local, cultural context which give them meaning. According to Brockmeier (2002, p. 35), “it is not the narrated event but the narrative event that makes a plot. It is not the historical ‘facts’ as such but the discursive practices of their presentation that symbolically activate [a memory] installation and turn it into an agent in a cultural system.” These three orders of narrative together integrate several elements into a specific structure of meaning. Here, single elements acquire meaning only because they are integrated into the narrative plot of the constructed memory. Brockmeier’s framework is useful to apply in my research because it draws attention to the different ways in which meaning is constructed, looking beyond just narrative texts. Together, these three orders constitute a mnemonic system, “a system of past experience (that we usually call memory) and present experience (that includes the process of remembering that past experience)” (Brockmeier 2002, p. 37). It is thus through these three orders of linguistic, semiotic and performative, that memory and remembering (and forgetting) are constructed.

Though Brockmeier’s three orders are useful for clarifying the different dimensions that are included in narratives, beyond just narrative texts, they do not constitute a normative framework that bridges

(18)

enables me to make sense of the narratives with regard to reconciliation. A suitable model to apply to my cases is Sara Cobb’s (2013) theory about “bad” and “better formed” conflict narratives. This builds on Ricoeur’s normative framework, which distinguishes between (good) uses and (bad) abuses of memory, but with an explicit perspective towards conflict/reconciliation outcomes. A “bad story” according to Cobb is one in which “they tell a narrative of wrongdoing and provide an account of violation, but this narrative perpetuates and deepens the kind of discourse that contributes to destroy rather than open public debate and deliberation” (2013, p. 37). By contrast, and in line with Ricoeur, a “better formed story” is “one that does contribute to “healing” social illnesses” (2013, p. 226). The better formed story should thus be understood as a critical narrative practice in that it seeks to destabilise subject positions and create a more complex subjectivity. It also provides distinctions that can be used to analyse practices that aim at narrative development. We might track whether and how stories move from simple to more complex storylines.

Cobb (2013) elaborates this move further in an analytically useful way. Bad and better formed stories can be recognised, and the move from one to the other can be tracked, by the way that they respond to four criteria: time, characters, causality and themes/values. Bad stories focus on the past, while the present and future remain absent/underdeveloped. Bad stories involve few roles/characters and focus on victim-victimiser relations. They treat causality as linear (and the other as responsible); and they feature themes of hopelessness, suffering, vengeance, and in-group loyalty. In contrast, better formed stories offer a rich account past, present and future; they include a diverse array of characters who are seen as human and fallible and whose relations and action is seen as mutual. Victim/victimiser boundaries are blurred or transformed and care is exhibited. Causality is seen as circular - expressing interdependence, connecting the narrator to others; and they feature themes of hope, charity, justice, learning development, participation. All in all, “the moral frameworks emerged through the creation of the better-formed story ensure that the value system is more diverse — good and evil can be clearly identified, but those judgments are complicated by the multiple ways in which people/things can be good or bad” (Cobb 2013, p. 223).

3.5 Pulling the theoretical pieces together

Although research on memory in (post)conflict settings has usually been linked to conflict escalation, peacebuilding research and practice have demonstrated the potential for memory and processes of remembering to contribute to reconciliation. The point of departure in this chapter was conceptualising memory as a social construct that is always situated in a certain time and space. The malleability of memory opens the door to explore how memory can be used to serve a given purpose. Narrative was

(19)

introduced as a way to understand how memory can be used in different ways, since they are “at the same time the occasion for manipulation through reading and directing narratives, but also the place where a certain healing of memory may begin” (Ricoeur 1999, p. 9). What Ricoeur refers to as the potential for healing comes from the fact that it is always possible to tell the story another way, a way that is more nuanced and complex. In this way, moving towards a complex narrative is a critical practice in that it seeks to destabilise subject positions and create a more complex subjectivity (Cobb 2013).

My analysis draws on these distinctions to make sense of how narratives are constructed in two memory initiatives in contemporary Colombia. In this chapter I have devised a framework to guide my analysis. It has three elements - the first draws a distant framework / nearby milieu distinction between the two case studies under investigation, the second is concerned with conceptualising narratives: how they are developed and communicated, and the third introduces an ethical framework that tracks whether such development moves in the direction of polarization and escalation or of reconciliation. With regard to the second part, I have adopted the conceptual framework of narratives proposed by Brockmeier (2002), which incorporates three orders of narrative: the linguistic, the performative and the semiotic. It is through these three orders that memory and remembering are constructed and it is these dimensions that will be taken into account in my analysis. The third part of my theoretical framework introduces a normative framework through which I make sense of my research data. While it is used with normative intent by practitioners, I use it analytically to make sense of stories and their development in the initiatives under study. Accordingly, I consider how the studied narratives respond to four criteria: time, characters, causality and themes/values and discuss the implications of that on reconciliation outcomes.

As shown in this theoretical discussion, my main research question - how does memory work contribute to reframing historical narratives of the armed conflict? - has two distinct components: describing what kind of historical narratives are presented through the chosen memory initiatives; and applying existing theory to track how these narratives correspond with reconciliation or escalation. I formulated certain sensitising questions that enabled me to gather and analyse my data with regard to these two components. In order to address the first of these, I was guided by questions arising from Brockmeier’s (2002) theory of narratives: what goes on at the linguistic, performative and semiotic levels? With regards to the second component, I was guided by the criteria in Sara Cobb’s conflict narratives theory, as discussed earlier in this chapter. My sensitising questions thus focus on how

(20)

narratives respond to time, characters, causality and constructive or destructive themes. The subsequent chapter lays out the design and methodology that was utilised to carry out my research.

(21)

CHAPTER 4: RESEARCH METHODOLOGY

My thesis investigates how memory work contributes to reframing historical narratives around the armed conflict in Colombia. In the previous chapter, I discussed the three part theoretical framework underpinning my research. The first part draws a distant framework / nearby milieu distinction between the two cases under investigation. The second is concerned with conceptualising how narratives are developed and communicated and the third introduces an ethical framework that will be used to track narrative development on a continuum between polarisation and escalation and reconciliation. The focus on narratives calls for a qualitative research methodology and a case study design. In this chapter, I justify the choice of cases and the methods used to study each case, primarily semi-structured and narrative interviews, participant observation and narrative analysis. Finally, I discuss some limitations of the research and the ethical considerations that the research raises.

4.1 Research design

My choice of research design was an extended case study, as proposed by Michael Burawoy. This particular approach is best suited to my research since it aims to improve theory by “... turning anomalies into exemplars” (Burawoy 1998, p. 10). This was precisely the overall aim of my research - I set out to improve theory on the transformative capacity of memory in relation to conflict by taking as my starting point my initial observations that suggested that, despite the apparent limited diversity within groups engaging in memory initiatives, they do contribute in a positive way to (and support) the reconciliation process by promoting more complex historical narratives.

Given that taking all of Colombia as a case study would have been impractical due to time and resource constraints, the focus was narrowed to two specific cases of memory initiatives: the Casa de la Memoria in Medellín and the Salón del Nunca Más in Granada. These two particular memory initiatives were selected because they met the subsequent criteria, developed to complement the theoretical framework developed in the preceding chapter. They both feature reconciliation as an explicit aim, they were both accessible at the time of my field research (certain parts of the country remain highly volatile), and they both feature memory exhibitions. It is also useful to highlight the contextual distinctions between my cases: the Casa de la Memoria is located in Colombia’s second city and is managed by the city government, while the Salón del Nunca Más is a community-led memorial space in the small town of

(22)

Granada, 170 km from Medellín. The former is thus a distant framework of memory while the latter is a nearby milieu (see chapter three for discussion). Since my research is not directed at establishing a definitive “truth” about the external world, my aim was not to draw generalisable or comparable conclusions from studying just two (in many ways distinct) cases, rather, I sought a contextual and evaluative understanding of these two cases to extend existing theory around the transformative potential of memory as per Burawoy’s reflexive approach (1998).

One of the key features of my approach is that it constituted an iterative process within the study design, whereby my research question was adjusted during field data collection. This is in line with Burawoy’s approach (1998), which regards the reconstruction of theory as a continual process that unfolds during fieldwork itself, rather than being a task of post-fieldwork analysis. Inspired by literature on the importance of plurality in narrative conflict transformation, I had initially set out to find out how organisers of memory initiatives understand and incorporate plural voices in their work. After some interviews with professionals involved in organising memory work revealed that the focus has been overwhelmingly on victims to date (whilst maintaining a focus towards reconciliation), I was pushed to question whether plurality in terms of victims and perpetrators really would be a necessary precondition of reconciliation outcomes in the case of Colombia in 2018. I had learned through interviews that many people involved in victim-centric memory work communicate narratives that feature many constructive themes as per Cobb’s theory (2013). I thus broadened my research question in a way that would allow me to better situate my research in its context, thereby extending existing theory.

4.2 Research methods

My research methods included interviews and participant observation. I used the data collected through these methods in narrative analysis, which constitutes my approach to interpreting to data.

4.2.1 Interviews

I carried out a total of fifteen interviews in Granada and Medellín over the course of five weeks in March - April 2018. My interviewees consisted mostly of ordinary people that visited the Casa de la Memoria or the Salón del Nunca Más, with a few notable exceptions. These included Sara García, a researcher working at the Casa de la Memoria; and Gloria Quintero, treasurer of the local victim association in Granada (ASOVIDA), responsible for the Salón del Nunca Más. Most of my interviewees who were visitors at the Casa de la Memoria were selected through random sampling. In Granada, I primarily used the snowballing method of sampling. The key contact that put me in touch with Gloria (my first

(23)

contact in the Salón) and Sara (my first contact at the Casa de la Memoria) was a very helpful professor named Gabriel Ruíz-Romero from the Universidad de Medellín, whose research has focused on the reconstruction of memory. I had initially contacted him a few weeks before leaving for field work, and he had agreed to meet with me in Medellín. All my interviews, with the exception of Sara, were held in Spanish.

I carried out a mix of narrative and semi-structured interviews. The narrative approach to interviewing is unstructured and aims at stimulating the interviewee to tell their life story about some significant event in their life (Edwards & Holland 2013). The semi-structured interview approach falls somewhere between a structured interview, which is more akin to a questionnaire, and the narrative approach. A typical semi-structured interview will be conducted according to several questions or themes whilst exercising flexibility in how and when the questions are asked. This way, the interviewee has more freedom in the way that they respond, creating a new fluid dialogue, guided by predetermined themes within which the interviewer can probe for answers (Jovchelovitch & Bauer 2000).

I conducted a semi-structured interview with Sara (from the Casa de la Memoria) with the aim of learning about the mission of the Casa de la Memoria and how that is being realised. I also conducted semi-structured interviews with visitors in the Casa de la Memoria and Salón del Nunca Más. I had certain questions that I was seeking answers for. I began the interviews by asking some background questions to ‘break the ice’, followed by which I asked these respondents what the Casa de la Memoria/Salón del Nunca Más means for them subjectively. This was so that I could gain a sense of how they understand the reconstruction of memory and what purposes it serves from the point of view of my respondents, which was vital for the research process given my central focus on subjective experiences. The respondents were mostly women, which is unsurprising given the disproportionately high number of men among the scores of victims who were assassinated or disappeared, and the relatively high involvement of women in the peace process in general.

I also conducted two in-depth interviews with Gloria, the treasurer of ASOVIDA and person responsible for organising the Salón del Nunca Más in Granada. My aim in doing so was to understand the work of the Salón through her narrative (she is the main organiser and person responsible for the space). During the course of the first interview, she guided me around the exhibition, explaining to me the meaning of different features and telling the story of how the conflict unfolded in Granada. In the second interview, I asked her to talk about a specific symbolic forgiveness event that was held in Granada between the FARC and victims that she mentioned during the first interview.

(24)

4.2.2 Participant observation

In addition to the interviews described above, I carried out participant observation to explore each of my cases in greater depth, by immersing myself in the culture and context. In the Casa de la Memoria, I attended a tour that introduced and explained the exhibitions of the museum for a group of university students studying Communication. I did this to gain a sense of how the organisers behind the museum want to present their narrative. I paid particular attention to which features of the exhibition were drawn attention to and how, and which features were overlooked. Throughout the guided tour, I remained at a slight distance so as to minimise my own impact on the event itself. For the same reason I did not use an audio recorder during the tour but instead, I wrote down extensive field notes to facilitate analysis. I was also hoping to attend a memory workshop or dialogue session at the Casa de la Memoria, but unfortunately this did not materialise due the lack of such events being held there during my field research period.

As part of my research in Granada, I was invited to attend an event held in honour of the official national victims’ day on April 9th. The event included a march, followed by a panel discussion and question and answer session with the public at the community hall. The panel consisted of three speakers: a religious leader, a historian from the Universidad de Antioquia, and a community leader. I took down extensive field notes and gained a deeper understanding of the process of how work around memory is carried out. Also in Granada, the first of my narrative interviews with Gloria (when she gave me the tour of the Salón) turned into a session of participant observation when an elderly couple from Medellín came into the museum, joined our tour, and engaged in a heated exchange with Gloria about responsibility for the atrocities committed during the armed conflict. Though unexpected and unplanned, this episode proved very interesting for me as a researcher as it highlighted the contested nature of memory and revealed competing historical narratives in the real-world context.

4.2.3 Interpreting data through narrative analysis

In chapter three I developed a normative framework of memory to guide the analysis of my data. This is based on narrative analysis with regard to two components. First, what kind of historical narrative(s) is/are presented through the chosen memory initiatives? And second, what are the implications of these narratives on the reconciliation process?

In order to answer the first question I operationalised Brockmeier’s three orders of narrative: the plot, the discursive and the semiotic. Considering the plot, I looked at the combination of elements that Brockmeier (2002) calls the constituents of a narrative proper: scene, agent, action, intentionality,

(25)

predicament, and solution. I describe these elements in the narratives of my two cases. Following this, I considered the wider semiotic order that is invoked through the narratives. Accordingly, I documented the exhibitions photographically to enable me to incorporate the broader symbolic order of how the narrative interacts with its material surroundings. Finally, I looked at what goes on at the discursive level. Here I focus on the struggle of memory by identifying dominant themes that arise across the studied narratives as components of the overall discourse. With regard to the second question, what are the implications of these narratives on the reconciliation process?, I discussed the implications of my findings from the first part on conflict / resolution outcomes utilising Sara Cobb’s (2013) theory on conflict narratives. My aim is to contribute to theory on conflict resolution by showing how memory can be used to promote reconciliation in the transitional context.

To enable me to gain an in-depth understanding of my case(s), I adopted a broad conception of what constitutes a narrative material or “text”. Material or data that I used in my narrative analysis includes photos and videos that I took of the two exhibitions themselves, transcripts from my interviews, and field notes made during participant observation. I utilised the qualitative data analysis software Atlas.ti to assist me in analysing the data. With this programme, I was able to code my data (including videos and images) into categories and run cross-category correlations to see what categories overlap.

4.3 Limitations and ethics

By focusing on how memory work reframes historical narratives, I am making the assumption that narratives undergo a process of transformation in the course of memory work. This is an assumption that I gained through reviewing theories on memory and conflict. One of the main limitations is that, given that historical narratives are embedded within certain spatial and temporal trajectories, it is very difficult for me as a researcher to gain an accurate idea of other (previous) narratives - i.e. what was the narrative before memory work was carried out? I would have liked to interview citizens who are not involved in memory work in any way as a kind of ‘control group’ that could help me gain an understanding of their narrative (uninfluenced by engagement with memory initiatives), but this proved to be unrealistic. Understandably, it is difficult to identify these people let alone gain their consent to ask questions pertaining to painful memories. Thus, I decided to focus instead on the narratives arising from and communicated through memory work, and situate and evaluate them with an awareness of common narratives around the armed conflict as heard through casual conversations during my field research in Colombia and through interviews with participants, particularly referring to how their understanding (or memory) of the conflict has changed.

(26)

As mentioned previously, since my study is reflexive rather than positive in its approach, I am more concerned about exploring a particular case (or two cases) in detail to extend theory than to claim representativeness or replicability in my results. Even in positive science, representativeness remains elusive since there are field effects, which dictate that interviews cannot be isolated from the political, social, and economic contexts within which they take place. Replicability is thus undermined by external factors we do not control (Burawoy 1998). Similarly situation effects threaten the principle of representativeness: “insofar as meaning, attitudes, and even knowledge do not reside with individuals but are constituted in social situations, then we should be sampling from a population of social situations and not a population of individuals” (Stinchcombe 1980, cited in ibid.).

Nevertheless, there are certain context effects that the qualitative researcher must be aware of. These include interview effects (interviewer characteristics affect responses) and respondent effects (ambiguity of questions/risk of misunderstanding). With regard to interview effects, I am aware that my presence as a non-native Spanish speaking white female researcher had an effect on responses. My ethnicity stands out in Colombia, and many Colombians I spoke with assumed I was from the United States based on my Spanish accent and how I look. I believe this influences my work negatively because of the perceived power imbalance that it invokes. Colombia has a long history of European colonialism and U.S. imperialism, and the political and economic elites remain overwhelmingly white, while the most marginalised groups are usually indigenous groups and Afro-Colombians. Further, it is important to recognise the possible effects of my gender given the traditionally patriarchal Colombian context. Nevertheless, given that nearly all of my respondents were women, I believe being a woman as well generates trust between myself as a researcher and my respondents. On the topic of respondent effects, given my Spanish language skills and the numerous cultural differences between myself and my respondents, I envisaged that there would inevitably be misunderstandings during the course of my interviews. I thus focused on open-ended questions, and used my non-native Spanish skills to my advantage by often asking for clarification during interviews in an attempt to gain an understanding that does justice to that of my respondent as closely as possible.

Finally, I must draw attention to the primary ethical consideration raised by my research project, which corresponds to the context of Colombia having endured violent armed conflict. Most (if not all) of my respondents had suffered severely during the armed conflict and were very open to talking about it. It was thus difficult at times to focus the discussion on the questions I wanted to ask. Several times the focus of my respondents’ story would turn to the details of how they lost a close family member in the conflict, leading to outbursts of emotion. In these situations, I did not attempt to impose the interview

(27)

on the respondents, rather, I attempted to provide some comfort through active listening. Further, the volatile context affected my choice of case studies. Because conflict is still ongoing in many regions of Colombia, I had to practically limit my research to areas that were, at the time of my field research and given the allocated time for research, accessible and safe.

(28)

CHAPTER 5: FINDINGS

In this chapter, I present findings arising from my the interpretation of the collected data using the theoretical framework as detailed in chapter three. To recall, this framework constitutes three parts: the first is concerned with identifying the narratives communicated through the Salón del Nunca Más and the Casa de la Memoria in accordance with Brockmeier’s three orders of narrative: the linguistic, semiotic and discursive (2002); the second introduces an ethical framework based on Cobb’s theory of conflict narratives (2013) that tracks whether such narratives move in the direction of polarisation and conflict escalation or of reconciliation; and finally, the third draws on Halbwachs’ distinction of nearby milieus and distant frameworks, which enables a more situated understanding of the practical impact of the narratives on visitors and participants of the two memory spaces. This chapter is divided into three sections. The first is concerned with the Salón del Nunca Más and it is split into two parts. In the first part, I describe the narratives communicated by the Salón structured according to the three orders of narrative (Brockmeier 2002). In the second, I analyse the identified narratives against criteria proposed by Cobb (2013) in an attempt to make sense of the ways in which these narratives support reconciliation or conflict escalation. Following this, the second section of this chapter presents the narrative of the Casa de la Memoria, and it is structured in the same way as section one (the Salón). The third and final section of this chapter brings the findings from the first two sections into an interaction with each other, through a reflective discussion of the social framework within which the collective memories of the Salón and Casa de la Memoria are localised.

(29)

5.1 NEARBY MILIEU: Salón del Nunca Más, Granada

The Salón del Nunca Más (Hall of Never Again) is a community-led memory space in Granada, a rural town of around 20,000 inhabitants (including those of the surrounding rural areas) lying 170 km to the north east of Colombia’s second city Medellín. As shown in chapter two, Granada has been one of the most affected areas by the armed conflict and its inhabitants have been the target of horrific acts of violence including forced disappearance, killings and kidnappings, among others. The Salón thus reflects an “active response [on behalf of the local community] to the ordeals of war, and a rejection of the attempts to impose silence on so many victims” (GMH, 2013, p. 19). I have chosen to analyse the narratives of the Salón as an example of a nearby milieu of memory construction. I trace the narratives in the first part, and discuss their implications on reconciliation in the second part.

Part I: Identifying narratives

5.1.1 Linguistic order

To enable me to identify and analyse the linguistic order of narrative of the Salón del Nunca Más, I was guided by the general question: how did this memory initiative come into being? In this sub-section I conduct an analysis of the plot as described in Gloria’s story. It begins with an introduction to the context of violence and destruction in Granada, which demonstrates the apparent need for organised community resistance.

Gloria traces the beginnings of the armed conflict in the region of Granada back to the early 1980s, when the first armed groups entered rural parts of the territory. The ELN entered in 1981, followed by FARC in 1986. Though the violence had not yet been felt by the majority of the population at this time, the two guerrilla groups began a dispute over control of the strategic territory, which includes the main road between Bogotá and Medellín and a strong agricultural economy. For Gloria, the 1980s were a period during which the guerrillas embedded themselves in the rural communities of Granada: “farmers say that [the guerrillas] became involved in their lives as they helped on the farms, so it was a situation of arriving and fitting in to the community.”4

During the 1990s, the conflict escalated considerably. By the start of the decade, both ELN and FARC had consolidated their power in the territory and moved into the town of Granada in search of new

Reframing historical narratives : two Colombian case studies

Graduate School of Social Science

MSc Conflict Resolution and Governance