Challenges and Possibilities of Memory and Reconciliation: Empirical Evidence for Colombia

| The signing of the peace agreement with the Revolutionary Armed Forces of Colombia-People’s Army (FARC-EP) deepened the debate on the reconstruction of memory as a right of the victims and a way to advance in the reconciliation of Colombian society. This article uses data from the 2019 Colombian Reconciliation Barometer of the Program of Alliances for Reconciliation of the United States Agency for International Development and ACDI/VOCA to analyze the individual and contextual factors that affect how individuals perceive the reconstruction of memory as a tool that contributes to reconciliation. We estimated a probabilistic model for which the dependent variable is the predisposition toward memory reconstruction as a predictor of reconciliation based on the context of the Colombian armed conflict, and a set of individual factors that capture the effects of community initiatives and state institutions and programs arising within the framework of transitional justice. We sought to verify whether the characteristics of individuals and their way of relating to each other in a community and municipal context shape the perception of the contribution of reconstructed memory to reconciliation. The results show that being a female victim of the armed conflict, developing empathy, residing in municipalities with the presence of Places of Memory, and trusting in the processes of transitional justice —such as clarifying the truth— increase an individual’s predisposition to consider that the reconstruction of memory contributes to reconciliation. In contrast, other policies and programs that emerged after the agreement, such as residing in municipalities with Territorially Focused Development Plans (PDET in Spanish) and the presence of Casas de Verdad (Truth Houses), operate in the opposite direction. This work suggests challenges for transitional justice by identifying the factors that condition positive outlooks toward the reconstruction of memory in reconciliation processes and opens the possibility of formulating actions with greater citizen acceptance.

la sociedad colombiana. Este artículo utiliza los datos del Barómetro de la Reconciliación 2019 del Programa Alianzas para la Reconciliación de la Agencia de los Estados Unidos para el Desarrollo y ACDI/VOCA, para analizar los factores individuales y contextuales que inciden en la percepción individual sobre la reconstrucción de la memoria como una herramienta que aporta a la reconciliación. Se estima un modelo probabilístico cuya variable dependiente es la predisposición hacia la reconstrucción de la memoria como predictor de reconciliación en función de un conjunto de factores individuales y del contexto del conflicto armado colombiano que capturan los efectos de las iniciativas comunitarias y de las instituciones y programas estatales surgidos en el marco de la justicia transicional. Se busca comprobar si las características de los individuos y su forma de relacionarse en un contexto comunitario y municipal moldean la percepción sobre el aporte de la reconstrucción de la memoria a la reconciliación. Los resultados muestran que ser mujer víctima del conflicto armado, desarrollar empatía, residir en municipios con presencia de Lugares de Memoria y la confianza en los procesos de justicia transicional -como el esclarecimiento de la verdad-aumentan la predisposición individual a considerar que la reconstrucción de la memoria aporta a la reconciliación. En contraste, otras políticas y programas surgidos tras el acuerdo, como residir en municipios con Programas de Desarrollo con Enfoque Territorial (PDET) y la presencia de Casas de Verdad, operan en sentido contrario. Este trabajo sugiere desafíos para la justicia transicional al identificar los factores que condicionan las actitudes positivas hacia la reconstrucción de la memoria en procesos de reconciliación y abre la posibilidad de formular acciones con mayor aceptación ciudadana.

Introduction
Societies that are in transitional justice processes face the challenge of overcoming massive human rights violations. In this sense, the Colombian conflict has some particularities that make it an attractive case in the international context: on the one hand, because of its long duration, the severity and magnitude of the human, economic and environmental damage caused, and its coexistence with a formal democratic regime; on the other hand, because of the establishment of innovative institutions such as those agreed upon in the Peace Agreement signed in 2016 with the former Revolutionary Armed Forces of Colombia-People's Army (FARC-EP) guerrilla.
Additionally, Colombia has sustained an armed confrontation simultaneously with the development of a transitional justice process as a result of the actions of the dissidents of demobilized armed groups, other guerrilla groups such as the National Liberation Army (ELN) and paramilitary groups closely linked to the drug trade (Delgado Barón 2012). Added to this situation are phenomena that have intensified after the signing of the Agreement, which put the progress of the reconciliation processes at risk: massacres (162 in 2020), homicides of social leaders (140 in the same year) (Garzón Vergara 2021), and murders of ex-combatants of the former FARC-EP (approximately 250) (Politics and EFE 2020). For this reason, strategies that strengthen reconciliation in Colombian society with the purpose of consolidating a stable and lasting peace are important.
This article focuses on the reconstruction of memory as a tool that contributes to the processes of reconciliation in post-conflict societies. When analyzing the Colombian case, we seek to answer whether there are individual factors-such as socioeconomic characteristics and victimization status-as well as factors of the municipal context, and others implemented from transitional justice, that shape the perceptions of people about the reconstruction of memory as a generator of reconciliation. Likewise, we aim to show whether individual perceptions about memory as a tool that contributes to reconciliation are shaped only by factors associated with the way in which the armed conflict was experienced, for example, being a victim, or the level of exposure to violence, or if they are also forged by other aspects of the local and municipal context, such as empathy, community initiatives of memory, and the presence of institutions and programs arising from the peace agreement.
To verify this, data from the United States Agency for International Development (USAID) Reconciliation Barometer and the Program of Alliances for Reconciliation and ACDI/VOCA are used to estimate a discrete probit choice model, which allows us to evaluate, in terms of marginal effects, the probability or predisposition that people have to believe that the reconstruction of memory contributes to reconciliation or if, on the contrary, it hinders it by opening the wounds of the past.
The empirical literature has indicated that the Colombian case "shares with other conflicts of long duration -such as Angola, Guatemala, Northern Ireland, and Sri Lanka-a history of spurs of violence as well as fatigue, multiple actors involved in fighting in addition to profound social divisions, and a legacy of conflict-related institutional atrophies" (Rettberg and Ugarriza 2016, 519). Hence, the need to identify which elements potentially affect reconciliation, in particular, those factors that shape the perceptions of people about memory as a predictive tool of reconciliation. Although the literature reflects dissent about the contribution of the reconstruction of memory to reconciliation, it is appropriate to delve into the individual and contextual factors that shape such perceptions; with this, inputs would be available to formulate strategies that contribute to fostering empathy toward the victims and improving citizen attention toward the reconstruction of memory.
The article is organized as follows: the next section after this introduction is divided into two parts. In the first, the literature on the relationship between memory reconstruction and reconciliation is reviewed, while the theoretical approach addresses how memory contributes to advancing reconciliation in transitional and post-conflict contexts. In the second section, the state of the art on the factors that affect reconciliation processes in transitional contexts is presented as a reference to analyze their effect on the formation of perception on the reconstruction of memory. The third section presents the data used and the method applied to the empirical analysis, and the fourth presents the results. Finally, these findings are discussed in light of the proposed hypothesis and the results of other studies.

State of the Art on Memory and Reconciliation
Within the framework of the formulation of transitional justice policies, it became common to understand that "historical memory and the pursuit of truth and justice writ large will bring long-term healing to society" (Rettberg and Ugarriza 2016, 533). However, this idea is based on a nonexistent consensus in the literature about the contribution of the reconstruction of memory to reconciliation and in the absence of sufficient empirical evidence to support it (Mendeloff 2009  . Some authors state that "negative memories weigh more than positive ones, since they drag higher emotional content" (Badruzaman 2018, 8) and generate long-term effects, giving rise to a "ceaseless cycle of repetition and re-enactment of the original trauma" (Stockwell 2019, 11).
This trend also points out that the reconstruction of memory fuels the desire for vengeance while making violent solutions to conflicts seem "more attractive, and opportunities for cooperation and reconciliation […] more limited" (Mendeloff 2009, 600) and hinders the coexistence of former adversaries (Robben 2012) and can produce an "excess of memory" that prevents society from moving forward or even silencing or making some stories invisible, while others are promoted hegemonically (Clark 2013, 119;Pham et al. 2019). Similarly, memory can operate as a "source of division and even social conflict if there is disagreement about its shape and if different narratives about the past are directed against each other," since it prevents a rational debate that in turn generates distrust among actors (Rekść 2021, 48).
In contrast, a second group understands memory as a necessary condition to enable processes of reconciliation (Rettberg and Ugarriza 2016; Oettler and Rettberg 2019) and highlights their role in the generation of conditions for forgiveness, healing, empathy and tolerance, a sense of justice and the recognition of past atrocities to prevent their recurrence (Badruzaman 2018). Likewise, it appeases the desire for revenge and anger, and has therapeutic value as a precondition for restoring relationships between former adversaries (Mendeloff 2009;Molina 2010). Following this line, we propose that the reconstruction of memory is a collective and plural process that evokes facts from group life, which are raised at the moment in which they are remembered, from a particular point of view (Halbwachs 1950). For this reason, the reconstruction of memory is shown as a development of knowledge about past events that depends on "the interpretations from the present and discussions of the cultural and political contexts of memory" (Schwartz 2000, 15-17 cited by Carranza 2018) and makes it possible to cope with and make sense of the events that occurred; in other words, memory as a practice of re-existence (Parrado and Jaramillo 2020).
Considering that the reconstruction of memory is not only focused on recalling historical knowledge but also on the representation of the past that promotes and gives meaning to the identity of a group (Misztal 2010;Stockwell 2019), these practices of memory reconstruction generate cohesion to a particular social and political order. In democratic governments, this is fundamental since it constitutes a way of enriching institutions through reflection and criticism of state actions that make up the stories of memory (Misztal 2010).
Similarly, the reconstruction of memory in transitional contexts operates as a tool that makes reparation for victims possible, given that "it allows to resignify the past and overcome experiences of pain produced in the context of violence" (Arboleda-Ariza, Piper-Shafir, and Vélez-Maya 2020, 129). Thus, "the damage produced by violence will be repaired when the truth of what happened is assumed and the memories of past events are reconstructed," as a result, "the narrative elaboration of memories of violence by victims becomes an indispensable condition for the achievement of the reparation of a nation" (Arboleda-Ariza, Piper-Shafir and Prosser Bravo 2020, 2; Rettberg and Ugarriza 2016; Molina 2010).
Although the literature on the contribution of memory reconstruction in reconciliation processes is extensive, it focuses on theoretical analyses or case studies and rarely uses data derived from instruments such as the Reconciliation Barometer. This work takes as a reference the literature on the mechanisms that generate the necessary conditions for reconciliation in societies in transition and proposes that these factors also act on the individual perception of the role of memory as a tool that contributes to reconciliation. That is, it proposes that the construction of memory is a collective process, as in fact classic works such as those of Maurice Halbwachs or Michael Pollak have already argued, but also that the perception of memory as a tool that contributes to reconciliation is conditioned by individual factors and the community and institutional context.
In this measure, reconciliation is understood as a complex social process of encounter between opposites (Lederach 1998) that "involves the mutual recognition of past suffering and the change of destructive attitudes and behaviors for constructive relationships for a sustainable peace" (Brounéus 2003 cited by Ugarriza 2013, 150), and consequently, we understand that the reconstruction of memory acts as a device that activates the processes of reconciliation.
The literature review on reconciliation in post-conflict societies mainly shows three aspects of analysis, which will be taken as a basis for the analysis of the factors that explain the perception that people have about memory as a tool that contributes to reconciliation. The first associates reconciliation with the degree of exposure to violence and the status of victimization; the second links it with levels of empathy and interpersonal trust, and the last directs attention to institutional trust. . This means that having been impacted allows individuals to be more willing to believe that it is possible to live in peace with former enemies than those who have not experienced it, or in other words, victimization improves the capacities of "social engagement" and "altruism" (Hazlett 2019, 4) and increases empathy.

Exposure to Violence and Status of Victimization
On the other hand, there are analyses that identify levels of exposure to violence as negative determinants of reconciliation. As it is direct harm, its effects are stronger than when it is an indirect experience (Fergusson et al. 2018; Gaviria, Ávila and García 2019). According to this perspective, the intensity of the psychological traumas produced by the violent experience increases "as a function of the severity and temporal proximity of the victimization" and has "noticeable effects on the ability of the victims to establish social relationships" ( Finally, the same tension is produced when considering the type of violence that occurred. In fact, the coexistence of an elevated risk of violence and violence by members of an armed group against unarmed civilians from another group emphasizes the antagonistic relationship between individual preferences toward justice and reconciliation in post-conflict areas, making positive attitudes toward reconciliation less likely by preferring actions aimed at justice (Penić, Vollhardt, and Reicher 2020).

Empathy and Community Networks
This aspect highlights the need to address the effects of conflict on people more than on infrastructure and institutions. Therefore, the main challenge for social reconstruction in the post-conflict period (Halpern and Weinstein 2004) is to ensure that people coexist in everyday spaces with former aggressors, so it is key to restore interpersonal relationships transformed by conflict that legitimizes violent actions. For this reason, empathy -understood as the ability to share the emotions of another person (Klimecki 2019)-becomes relevant for the reconciliation of former enemies. In this sense, while social reconstruction occurs at the state and community levels, reconciliation involves the individual capacity to regain empathy for others (Halpern and Weinstein 2004).
According to the literature, empathy increases the will to forgive the aggressors and, at the same time, ignites the desire to provide reparations for actions committed (Čehajić, Brown and González 2009). The assumption of this finding is that there is recognition of collective responsibility, since people are less likely to be moved by the suffering of others if responsibilities are not recognized; however, this recognition can generate empathy or spark the defense of actions stemming from the dehumanization of others. Thus, dehumanization prevents feeling compassion and prevents empathy from emerging (Čehajić Brown and González 2009). Bakke, O'Loughlin and Ward (2009) observe that people who distrust others are less likely to forgive than those who do show an elevated level of trust. This interpersonal trust is linked to empathy as a factor that contributes to reconciliation.
Finally, considering that reconciliation is a complex process that depends on both public policies and collective processes of restoring trust and the willingness of the community to overcome the past, the connection between belonging to political and local networks is relevant, as an aspect that favors positive attitudes toward reconciliation (Fergusson et al. 2018).

Institutional Trust
Trust in the State, including the government and its agencies, the armed forces and the judicial system, represents a factor that explains the relative success in conflict resolution and peace-building. Empirically, positive effects on reconciliation are produced by the favorable perception of institutions (Méndez, Casas-Casas and Pino 2020). This variable broadens the traditional field of study, which limits peace-building to formal aspects, in military, economic and infrastructure aspects; in contrast, it focuses on the social change necessary to achieve an effective transition (Méndez, Casas-Casas and Pino 2020).
The Colombian transition is analyzed in a multi-method field study by Méndez, Casas-Casas and Pino (2020) in five subregions of Colombia. According to the study, trust in the "national government or political trust" (2020, 2) is one of the dimensions that most contributes to peace-building; positive attitudes toward the State and the reforms it proposes lead to effective democratic deliberation on the appropriate institutional designs for peace-building (Méndez, Casas-Casas and Pino 2020). In this same context, regarding the peace agreement and the acceptance of concessions in favor of the former FARC-EP in the negotiations, Téllez (2018) observes that those who have favorable views toward the government appear more in favor of approving these concessions and the peace process, than those who lack trust.
Finally, Fergusson et al. (2018) warn that people who have more "trust in the judicial system or in the police and the army are more optimistic about reconciliation and feel less discomfort from daily interaction with the FARC." In addition, they point out that "this trust is associated with less reluctance to pay additional taxes to finance the reintegration process" (2018, 50). This is explained by the role of a solid judicial system in transitional justice, since it is necessary for the people to take individual responsibility for the violent acts committed, which facilitates coexistence with ex-combatants and promotes attitudes toward reconciliation (Fergusson et al. 2018).
This work takes as a reference these three aspects that influence the attitudes that make reconciliation possible in post-conflict contexts to capture, through the proposed estimation, its effect on forming the positive individual perception of the reconstruction of memory as a tool to predict reconciliation in the Colombian case, and to contribute to the identification of the factors that shape this perception.

Methodology and Data
The dataset was constructed using the Reconciliation Barometer, an information system of the Program of Alliances for Reconciliation (PAR) of the United States Agency for International Development (USAID) and ACDI/VOCA, whose purpose is to make inferences and analyze changes in the perceptions, attitudes and behaviors of Colombians toward reconciliation. The survey was conducted in 2019 and contains information on 11,942 people over the age of 16 years from 44 municipalities in the country, 27 prioritized by PAR and 17 not prioritized. The survey has national, regional and municipal representation and a confidence level of 95%.
Regarding the empirical strategy, a probit model was chosen for its capacity for structural interpretation in discrete choice models. The results are evaluated in terms of marginal effects or probabilities, which indicate the predisposition of people to believe that memory reconstruction contributes to reconciliation or, on the contrary, hinders it because it opens wounds from the past.
The model is presented as follows: where P is the probability that people feel that reconstructing memory opens the wounds of the past and Xij is the vector of explanatory variables belonging to individual i in municipality j.
We propose a group of variables identified in the literature that describe the mechanisms and conditions that are potentially related to the fact that reconciliation occurs in societies in transition (Rettberg and Ugarriza 2016). These are used to measure its impact on the perception of memory reconstruction as a predictor of reconciliation.
To identify the explanatory effect of the variables of interest, four independent regressions were estimated. The first includes variables related to the risk of exposure to violence and to victimization status; the second, variables that account for empathy and participation in community networks; the third, variables of trust in institutions, the government and organizations for demobilized combatants; and the last regression includes all the variables of interest and control, the latter relating to socioeconomic conditions and to the context of Colombian transitional justice.
The dependent variable is the predisposition to memory reconstruction as a predictor of reconciliation. It is a latent variable, which is partially observed and can express the difference, in terms of utility, of making a choice (Rdz-Navarro and Asún, 2016); it is represented in a dichotomous way, since it investigates the perception of the respondents about whether or not the reconstruction of memory opens the wounds of the past. On the other hand, the independent variables reflect the trends of the empirical literature regarding the factors that influence the emergence of attitudes (positive or negative) toward reconciliation: i) the degree of exposure to violence and the status of victimization, ii) community participation and empathy; and iii) trust between people and between them and institutions. In the study, sociodemographic control variables such as gender, age, education and income were used because they can be associated with the individual predisposition toward reconciliation. According to Téllez (2018), differences in age shape attitudes in relation to reintegration, which is evident in the fact that older adults are more reluctant to reintegrate demobilized combatants. In this same sense, Bakke, O'Loughlin and Ward (2009) use socioeconomic status, arguing that economic difficulties can cause negative perceptions of others that affect the ability to forgive and the belief that reconciliation is possible. Gender was incorporated as a relevant variable. Some studies show that men are more willing to reconcile than women and that, among direct victimizations, women are less open and less willing to reconcile (Téllez 2018; Gaviria, Ávila and García 2019).
To investigate the incidence of some mechanisms of Colombian transitional justice, variables not used in previous studies are provided, such as the existence of Casas de la Verdad (Truth Houses) 2 and Development Plans with a Territorial Approach (PDET) 3 in the municipalities where the respondents reside. Likewise, a dichotomous variable was included that reflects the perception of people about the contribution of truth to reconciliation in transitional societies. The criterion for including this variable derives from the emphasis of the Peace Agreement on the creation of institutions for the clarification of truth, such as the Truth Commission and the Search Unit for Missing Persons in the Context and Due to the Armed Conflict. Table 1 shows the variables used in the empirical analysis, their definition, characteristics and typology.

Results
The profile of the surveyed group includes, mainly, people under 40 years of age (51%), with low educational levels (45% reach secondary education) and low income (more than half not exceeding the legal minimum wage). One in three people declares to be a victim of the armed conflict, 20% of them are women. Of the 11,942 respondents, 30% feel that the reconstruction of memory opens wounds from the past and, therefore, does not contribute to the reconciliation of societies in transition. The majority (92%) believe that after a conflict, it is important to reconstruct and know the truth of what happened. Finally, more than half live in municipalities where there are community initiatives for the reconstruction of memory (68%) or where there are Truth Houses (56%) (see Table 2). To establish the statistical relevance of the differences between those who feel that reconstructing memory contributes to reconciliation and those who do not, we performed a test of mean differences. In this way, we could determine whether these differences are attributable to the set of variables of interest and control considered.
The differences between the victim status of the respondents and the risk of victimization as a consequence of the municipality where they reside are statistically reliable to identify the difference between those who believe in memory as a predictor of reconciliation and those who do not. It can be stated then that experiencing the war and its manifestations do matter in regard to people's perceptions on whether memory contributes to reconciliation in Colombia. The results of individual socioeconomic variables such as multidimensional and material poverty and income, and to a lesser extent education, reinforce this statement.
The variables of individual and collective empathy, and recognition of the importance of clarifying the truth of the events that occurred in the context of the conflict are reliable to understand this difference, which also focuses on those who trust local and national governments. Those who do not believe in the benefits of reconciliation are people who have suffered some harm from the effects of the conflict (see Table 3).

Results of the Estimations
The estimates are aimed at discussing the variables that are empirically associated with positive and negative attitudes toward the reconstruction of memory as a tool to achieve reconciliation (see Table 4).  The results are presented in four estimates; the first three explain the independent effect of variables of interest proposed in the literature; the latter shows the aggregate effect of the variables of interest considered and of the control variables provided in this study.
First, although gender and victim status analyzed separately do not explain the perception of people of the contribution of memory to reconciliation, the interaction of these variables does. Being a victim of the armed conflict increases the probability of believing that the reconstruction of memory opens the wounds of the past by 7.5 percentage points; therefore, it does not help reconciliation. In contrast, being a woman and being a victim of armed conflict reduces the predisposition to believe that memory opens the wounds of the past by 23 percentage points.
Second, the variables of empathy and community initiatives explain with a high level of confidence the predisposition to consider the reconstruction of memory as a tool that contributes to reconciliation. At a higher level of individual and collective empathy, the predisposition to believe that the reconstruction of memory opens wounds from the past is reduced by 4.6 and 6.4 percentage points, respectively. Likewise, actions for the reconstruction of memory by community initiative, such as places of memory, reduce the predisposition to believe that reconstruction does not contribute to reconciliation by 5 percentage points (Model 2).
The results of the third model are striking because they constitute a challenge for transitional justice processes. First, the positive and statistically significant relationship between the variables of trust -in governments and demobilized combatants' organizations-and the perception that the reconstruction of memory contributes little to reconciliation. This is the case for people who trust the central and municipal governments, which increases the probability of believing that memory is not a tool that leads to reconciliation by 33 and 21 percentage points, respectively. In this same sense, fully trusting demobilized combatants' organizations increases the probability of believing that it does not contribute to reconciliation by 25 percentage points.
The estimate shows results that merit being analyzed in depth in relation to the influence of trust in civil and military institutions on people's predisposition to support the reconstruction of memory. For example, relying heavily on the Colombian Army reduces people's tendency to believe that memory reconstruction opens the wounds of the past by 37 percentage points, while trust in the Police is not significant.
The results of the trust variable in religious institutions highlight the discussion on the role of churches in the transition of Colombian society toward reconciliation. According to the estimate, having complete trust in churches reduces the probability of believing that the reconstruction of memory opens the wounds of the past by 28%.
Control variables such as age, gender, income and educational level do not explain whether people will perceive memory reconstruction as a tool for reconciliation. Conversely, some of the variables of the transitional justice context in Colombia do offer significant evidence. For example, people who feel that clarifying the truth of what happened in the conflict is important are less likely (by 42 percentage points) to believe that the reconstruction of memory opens the wounds of the past.
However, institutional initiatives for clarifying the truth arising within the framework of the peace agreement with the former FARC-EP (for example, Casas de la Verdad) increase the predisposition to believe that memory contributes little to reconciliation by 16 percentage points. The residence within PDET territories variable also follows a similar trend.

Discussion
The purpose of this study is to identify the factors that affect the perception of people regarding memory as a tool that helps reconciliation processes. For this, an analysis is proposed that includes both socioeconomic and psycho-social factors, as well as those associated with the context of transitional justice in Colombia and the relationships of people.
The results of the model coincide with the empirical literature reviewed on the lesser disposition toward reconciliation and forgiveness of people who declare themselves victims of the armed conflict (Bayer, Klasen and Adam 2007; Gaviria, Ávila and García 2019). However, when interactions are proposed in the analysis, novel results are found. In fact, being a woman and a victim of conflict reduces the tendency to believe that memory opens the wounds of the past, that is, it contributes to reconciliation.
This suggests that the experience of direct victimization in women implies a post-traumatic growth (Tedeschi and Calhoun 2004)  The spaces for meeting and gathering that are key to this post-traumatic growth were fostered by gendered care practices and usually devalued (Ojeda and Berman-Arévalo 2020). This devaluation allowed these moments to represent spaces of resistance in the daily life of the conflict (Pérez-Bustos 2016). Though these spaces still represent this resistance, they have also become a means to reconstruct memory, heal wounds (Bello and Aranguren 2020), let people understand themselves as victims and mediate processes of collective struggle for their rights (Quiceno and Villamizar 2020). This allows us to explain how the status of a woman who is a victim of conflict reflects a greater predisposition toward memory as a tool for reconciliation.
Exposure to violence appears in the literature as a factor that negatively affects reconciliation (Fergusson et al. 2018), showing that people who live in areas with high rates of violence are less likely to forgive their adversaries (Bakke, O'Loughlin and Ward 2009), are more reluctant to restructure relationships with former combatants and tend to exhibit preferences toward justice over reconciliation (Penić, Vollhardt and Reicher 2020). In this sense, according to the estimate, inhabiting municipalities with medium and medium-low risk of victimization increases the probability of considering the reconstruction of memory as a tool that opens the wounds of the past. That is, people who live in places with greater exposure to violent acts, compared to municipalities that do not have any level of risk, tend to consider memory less useful as a vehicle for reconciliation. This is explained by the negative predisposition toward reconciliation that people who live in conflict areas or with high rates of violence tend to manifest (Fergusson et al. 2018).
On the other hand, empathy -both individual and collective-has a positive effect in reducing people's predisposition to believe that memory reconstruction opens wounds from the past. The result is consistent with the literature that points to it as a pillar in reconciliation processes because it increases the possibility of forgiving the perpetrators for the acts committed and people's desire to provide reparation to the victims (Čehajić Brown and González 2009).
Furthermore, considering the variables of Colombian transitional justice, we observed that actions from community initiatives (Places of Memory) reflect greater potential for reducing the inclination to think that memory opens wounds from the past, while the programs of institutional origin derived from the signing of the Peace Agreement with the FARC (Casas de la Verdad, PDET) increase the tendency to reject the idea that memory reconstruction will contribute to reconciliation. These results are explained, on the one hand, by the strength that community initiatives contribute to the construction of a collective, contextualized, plural and inclusive memory that is not infrequently exempt from political manipulation (David 2017) and, on the other hand, by the limited capacity of institutional strategies to reach community bases due to the historical distrust of communities (Fergusson et al. 2018) and the short implementation period of these initiatives.
Additionally, people who consider it valuable to reconstruct memory and to know the truth are 42 percentage points less likely to feel that remembering opens the wounds of the past. This is consistent with the literature on peace-building and conflict resolution, which indicates that the search and telling of the truth are vital elements for healing and reconciliation Contrary to what some of the literature proposes (Méndez, Casas-Casas and Pino, 2020), trust in the national and local government considerably increases the likelihood that people believe that wounds of the past will be opened through memory reconstruction and contributes little to reconciliation. These results could be explained by the "fierce opposition to the Peace Agreement and its implementation, and a rejection of the members of the former FARC guerrilla" (Gaviria, Ávila and García 2019, 73) by the Democratic Center political party. However, a long-term analysis is required to soften the effect of the political situation on this perception.
While trust in the Army reduces people's tendency to believe that the reconstruction of memory opens wounds from the past, trust in the Police is not significant. This finding requires studying in-depth whether the results are explained in part by the counterinsurgency role of the Army and the events where it was involved in the armed conflict -for example, extrajudicial executions-and in part by the civilian nature of the Police. As stated by Trejos (2013, 107-137), "in states of war, violence becomes the means used for the resolution of social tensions and conflicts, that is, violence and those who administer or exercise it become the dynamic element of the development of a shared life" (2013, 109); therefore, the reconstruction of memory and the acceptance of these facts would contribute to ensuring that they are not repeated (Barbosa and Ciro 2020).
Additionally, the results show that trust in churches considerably reduces the probability of believing that rebuilding memory opens the wounds of the past, which makes it "a potentially valuable but critically underused peacebuilding tool" (Clark 2010, 3). Therefore, the discussion on including churches in the processes of memory reconstruction and in reconciliation initiatives is a pending task for institutions and organizations that work towards preserving memory (Goldberg and Blancke 2011). Guatemala, Chile, Rwanda, Kosovo and Algeria are examples that illustrate this relationship, since the presence of religious organizations converges in generating empathy (Clark 2010), which promotes positive attitudes toward memory as a predictor of reconciliation.

Conclusions
The results of this study allow us to accept the proposed hypothesis to the extent that it identifies individual factors, such as being a female victim of conflict, and other factors of the community context, such as the memory construction initiatives that these promote and that explain the perception of memory as a tool that contributes to reconciliation. With these, the importance of the context of individuals and their relationship with the people surrounding them is evident when forming their perception of whether memory reconstruction can be an adequate tool for generating conditions for reconciliation in Colombia. Thus, the perception that is created comes not only from an internal process of the subject, but is also nourished by the collective practice that is fed through collective construction, the community, encountering and participating in networks, and humanization toward other people.
The study raises important challenges for the transitional justice process and the institutions that promote it. First, in the absence of homogeneous thinking about the contribution of the reconstruction of memory to reconciliation, identifying and understanding the individual and contextual factors that condition positive attitudes toward memory reconstruction can contribute to designing strategies with greater acceptance and better outcomes. Second, in Colombian society it is necessary to continue generating feelings of empathy and humanization toward the other, based on awareness and pedagogy strategies that promote the recognition of the harm toward the victims of the armed conflict, by reflecting upon the identification of the conditions that allowed violence to happen and generating strategies to prevent its recurrence.
Finally, it is necessary to extend the narratives on the reconstruction of memory and the clarification of the truth to broad sectors of society as generators of trust; in other words, it is needed that the story about the consequences of the armed conflict is inclusive, not segmented, and has no hegemonic or retaliation purposes.