Social ecologies of health and conflict-related sexual violence: Translating “healthworlds” into transitional justice

Abstract This article discusses the relationship between health and transitional justice through a particular focus on the issue of conflict-related sexual violence. It is not, however, about the individual health needs of victims-/survivors, nor about possible ways that transitional justice processes might address these. Drawing on empirical data from Bosnia-Herzegovina, Colombia and Uganda, it explores some of the health legacies of sexual violence in conflict and their wider significance for transitional justice. Embracing the World Health Organization’s definition of health as “a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity”, the article specifically seeks to demonstrate that conflict-related sexual violence (and its frequent entanglement with other forms of violence) affects not only individual but also social-ecological health. The article’s overall contention, thus, is that in the context of transitional justice, more than individual health matters. The broader “health” of social ecologies themselves is also critically important. Ultimately advocating a social-ecological reframing of transitional justice, the article utilizes Germond & Cochrane’s (2010) concept of “healthworlds” to explore what this reframing might look like in practice.


Introduction
Multiple links exist between transitional justice and health. Following its creation in 1995, for example, the South African Truth and Reconciliation Commission (TRC) conducted special hearings "to explore how decades of systematic 'racial' discrimination had influenced South Africa's health services and how the health sector contributed to the context for widespread abuses of human rights under apartheid" (Baldwin-Ragaven et al., 2000, p. 228; see also Harris et al., 2014, p. 144). Transitional justice processes can also powerfully illuminate some of the long-term health consequences of human rights abuses. As one illustration, the Peruvian TRC, established in 2001, "set the stage for creating a more expansive approach to mental health, recognizing that each rural community, deeply impacted and often divided by the war, faced unique challenges to mental health recovery" (Laplante & Holguin, 2006, p. 137). Moreover, public health issues, like the cholera epidemic in Haiti (officially attributed to the arrival of UN peacekeepers) that began in 2010, can become a catalyst for transitional justice work (Lemay-Hebert & Freedman, 2021). Additionally, factors such as the restoration of the rule of law, or having access to some form of justice, can potentially benefit individual and societal health and well-being (Swenson & Kniess, 2021, p. 1708. This article approaches the relationship between health and transitional justice through a particular focus on the issue of conflict-related sexual violence, drawing on empirical data from Bosnia-Herzegovina (BiH), Colombia, and Uganda. It is not, however, about the individual health needs of victims-/survivors, 1 or about possible ways that transitional justice processes might address these (Gilmore & McEvoy, 2021, p. 7;Lambourne & Rodriguez Carreon, 2016;N ı Aol ain et al., 2015;Pruitt, 2012). Rather, its focus is on some of the health legacies of sexual violence in conflict-which, as the analysis demonstrates, are often linked to other co-occurring forms of violence-and their wider significance for how we frame transitional justice.
According to the World Health Organization (WHO, n.d. [a]), "Health is a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity." Consistent with this definition, this research adopts a social-ecological understanding of health, locating the concept in the intersections between individuals and larger social systems and environments (Golden & Earp, 2012, p. 364). What it specifically aims to demonstrate is that sexual violence in conflict, both directly and indirectly through its frequent entanglement with other types of violence, affects not only individual but also social-ecological health. Social ecologies, or networks of relationality (G omez-Barris, 2017, p. 2)-including family, community, and institutions-can substantially influence the health of individual victims-/survivors. At the same time, and illustrating what Jadhav et al. (2015, p. 12) have termed "ecologies of suffering," conflictrelated sexual violence, and the broader circumstances in which it occurs, can affect social-ecological health.
The article's overall contention, thus, is that in the context of transitional justice, more than individual health matters. The broader "health" of social ecologies themselves is also critically important. In developing this argument, the article invokes Germond and Cochrane's (2010) concept of "healthworlds." In a recent article, Willard (2021) emphasized the limitations of existing transitional justice mechanisms and the need to "imagine possibilities" outside the box of mainstream approaches. The notion of healthworlds, which "requires that we be attentive to the complex whole of human well-being" (Germond & Cochrane, 2010, p. 316), offers a new "possibility" in this regard. Ultimately, this research uses the concept to advocate for a social-ecological reframing of transitional justice.
It is essential to stress from the outset that health-related harms are just one subset of harms that can result from conflict-related sexual violence (or any form of violence). The article's particular focus on health is not in any way intended to detract from other possible harms, including displacement from male and female gender identities (Schulz, 2018(Schulz, , p. 1107, loss of standing/ respect within the community (Josse, 2010, p. 178), and lost opportunities (Aroussi, 2018, p. 292). The crucial point is that health was a significant and recurrent thematic within the qualitative data that underpin this article and therefore merits attention. Moreover, discussions-including at the international policy level-about conflict-related sexual violence and the needs of victims-/survivors often place a strong emphasis on health (see, e.g., UN, 2019a, p. 8;WHO, n.d. [b]). That many victims-/survivors do not have access to (adequate) health care is well documented (Chynoweth et al., 2017, p. 92;Pham et al., 2020, p. 384;Zalesne, 2019, p. 719). From a social-ecological perspective, however, the concept of health care is not simply about the resources available to individual victims-/survivors but, more fundamentally, about the care that is extended to the health of their wider ecologies.
The article's first section provides an overview of extant scholarship on health and transitional justice. It adopts a cross-level analysis, consistent with the empirical section, by exploring some of the ways that transitional justice has endeavored to deal with individual health and also, even if implicitly, social-ecological health. The second section is methodological and introduces the empirical data that underpin and inform the article's theoretical and conceptual arguments. Drawing on the interview data, the third section empirically develops the argument that the health consequences and legacies of conflict-related sexual violence need to be understood in a broader social-ecological context. The conclusion discusses healthworlds and the concept's relevance for advancing the field of transitional justice in a new social-ecological direction.

Health and transitional justice
McGlynn and Westmarland (2019) use the term "kaleidoscopic justice" to convey the idea that justice is not something fixed. Kaleidoscopic justice, they argue, "is justice as a constantly shifting pattern, justice constantly refracted through new experiences or understandings; an ever-evolving, lived experience" (McGlynn & Westmarland, 2019, p. 180). Transitional justice can similarly be understood as kaleidoscopic in the sense that what it "looks" like and the patterns it creates are highly diverse (and constantly changing), shaped, inter alia, by wider contextual factors and the particular circumstances in which the need for transitional justice processes arises.
Nevertheless, transitional justice can be broadly defined as "the full range of processes and mechanisms associated with a society's attempt to come to terms with a legacy of large-scale past abuses, in order to ensure accountability, serve justice and achieve reconciliation" (UN, 2010). Although this concise definition captures only some of the core aims of transitional justice, the bigger point is that these aims arguably have a latent health-related dimension-linked to the fact that "healthier postconflict communities are less likely to return to violence than less healthy communities" (Davies, 2014, p. 861). Ensuring accountability, for example, is about creating a healthy society that does not tolerate impunity and about fostering institutional trust through the rule of law. Reconciliation is about restoring or building healthy (in the sense of cooperative, respectful, and nonadversarial) relationships within a community or society. In these examples, however, "health" is necessarily a broad and abstract concept.
In contrast, when health is explicitly discussed and addressed in a transitional justice context, it often assumes a more concrete and practical form. The Sierra Leonean TRC (2002)(2003)(2004), for example, recommended a number of reparations projects, including the provision of free physical and mental health care (Williams & Opdam, 2017, p. 1290. Some of its recommendations, moreover, were "directly pertinent to harm suffered due to sexual violence," including the provision of free psychological support and reproductive health care (Williams & Opdam, 2017, p. 1289. In response to the Comprehensive Reparations Plan recommended by the Peruvian TRC (2001)(2002)(2003), Peru's Ministry of Health established a Comprehensive Reparations Programme on Mental Health. Specifically targeting the 10 areas most affected by the armed conflict, the program "included hiring professionals and training staff in what can be seen as a significant improvement in mental health services" (Correa, 2013, pp. 23-24).
These are important examples of some of the ways that transitional justice processes have sought to respond to health-related needs. It is also widely acknowledged within existing scholarship, however, that the very act of participating in a transitional justice process, such as a criminal trial or a TRC, can have health-related consequences (Allan, 2000;Androff, 2012;Broun eus, 2010;Caparos et al., 2020;DeLaet et al., 2020;Doak, 2011;Laplante & Theidon, 2007;Salih & Samarasinghe, 2017). Cilliers et al., for example, studied the impact of a reconciliation processdeveloped and implemented by a local nongovernmental organization (NGO) called Fambul Tok-across 200 villages in Sierra Leone. Drawing on data from 2383 individuals, they concluded that "the negative psychological impacts indicate that truth-telling opened up psychological wounds, pointing to the potency of these war memories when they are evoked suddenly" (Cilliers et al., 2016, p. 794). Focused on the gacaca tribunals in Rwanda, and based on a sample of 373 genocide survivors, Caparos et al. (2020, p. 782) found that individuals "reported lower levels of PTSD [posttraumatic stress disorder] and depression symptoms when they had attended the Gacaca without testifying, compared to when they had not participated in the trials, or when they had attended and testified." Particular efforts have been made to minimize the health risks-and specifically the risk of retraumatization-to victims-/survivors who testify about their experiences of conflict-related sexual violence. In the Sepur Zarco trial in Guatemala, for example, the court allowed the 15 witnesses to give prerecorded testimony after their representatives insisted that requiring them to give live testimony in the courtroom could adversely affect their physical and mental wellbeing. According to Martin and S aCouto (2020, p. 258), "Using the victims' prerecorded testimony allowed the court to admit the evidence it needed in a manner that was consistent with both the rights of the accused and the obligation to ensure the well-being of the victims." The International Criminal Tribunal for the former Yugoslavia (ICTY), which completed its mandate in December 2017, established a Victims and Witnesses Section to provide counseling and support to vulnerable witnesses, "in particular in cases of rape and sexual assaults" (ICTY, 1994, Rule 34). In 2019, discussing the Dominic Ongwen of Uganda trial at the International Criminal Court (ICC)-the first case at the court to deal with charges of forced pregnancy and forced marriagethe then Prosecutor, Fatou Bensouda emphasized some of the measures adopted to help minimize distress to witnesses. In particular, the Office of the Prosecutor "employed a special procedure under the Statute, article 56, to present, record and thus preserve the testimony of sexual and gender-based victims, during the pretrial phase of the case'" (Bensouda, 2019), meaning that these individuals would not be required to repeat their testimonies.
Regardless of the particular form they take, however, and of any special measures adopted, transitional justice processes are necessarily located, taking place within broader socio-cultural contexts which themselves can significantly shape health needs and their expression-and, by extension, moderate the relationship between transitional justice and health. Kammel (2008, p. 384) gave the example of religious or socially shared beliefs. He further accentuated that when working with war-affected populations, important areas of inquiry include "examining local idioms of distress, identifying local mental health concerns and priorities, understanding the effects of organized violence on multiple levels" (Kammel, 2008, p. 385).
In a similar vein, Al-Krenawi and Graham's work drew attention to the fact that in some cultures, health has wider meanings that extend beyond individuals. They underlined that "all interventions with Arab clients need to be couched in the context of the family, extended family, community, or tribal background" (Al-Krenawi & Graham, 2000, p. 13). In other words, and highlighting the article's central argument, health has significant social-ecological dimensions (Grzywacz & Fuqua, 2000, p. 102).
Within the field of transitional justice, there is some broad engagement with social-ecological thinking and ideas (upon which this article specifically builds), although it is often more implicit than explicit. As one illustration, health is frequently mentioned in larger discussions about economic and social rights. According to Cahill-Ripley (2014, p. 189), transitional justice has largely glossed over such rights, as they are "not viewed as a legitimate concern for prosecutorial justice or are deemed too difficult an issue for criminal accountability" (see also Arbour, 2007;McAuliffe, 2017;Schmid & Nolan, 2014). Linking health to wider violations of economic and social rights, however, draws important attention to health as a systemic issue and recognizes that "the individual is connected to the larger social ecosystem" (Grzywacz & Fuqua, 2000, p. 102).
In her work on Nepal, for example, Pasipanodoya (2008, p. 351) argues that postconflict reconstruction efforts to provide communities with crucial socio-economic resources, such as health and education, would be more effective "if they fully considered how historical injustices related to health and education manifest themselves in the current situation." In their research with female victims of sexual and reproductive violence (SRV) in Guatemala and Peru, Duggan and colleagues emphasize that broader structural factors, linked to inequality, can exacerbate the secondary effects that women report (including depression, low self-esteem, and shame)-or even create new health-related problems. As one illustration, they note that poverty and economic precariousness force some women into prostitution, in turn exposing them to the risk of HIV and other sexually transmitted infections (Duggan et al., 2008, p. 196).
If complex structural injustices, because of their wide-ranging impact, indirectly highlight some of the social-ecological dimensions of health in transitioning or postconflict societies (McAuliffe, 2017, p. 3;Mullen, 2015, p. 471), some scholars have more explicitly engaged with social-ecological ideas in the context of health and transitional justice. An example of this is Fletcher and Weinstein's (2002, p. 580) development of "an ecological model to understand social breakdown and to identify the critical elements of social repair." For them, the importance of such a model is that it illuminates some of the wider legacies of violence beyond individual harms (Fletcher & Weinstein, 2002, p. 634). Emphasizing six particular interventions they regard as critical for social repair-including individual and/or family psycho-social support, externally driven community interventions, and community-based responses-they maintained that "[u]nder an ecological framework the synergistic effects of interventions at multiple points in a system lead to social reconstruction" (Fletcher & Weinstein, 2002, p. 626). In other words, they effectively adopted a social-ecological approach to health, qua social repair, that seeks to address multiple harms across "a broader community of suffering" (N ı Aol ain, 2000, p. 335) or different layers of the social fabric. Pham and colleagues similarly embraced an expansive conceptualization of health, emphasizing "the relationships between transitional justice and individual, community, and societal health" (Pham et al., 2010, p. 104). This conveys the image of health as forming interlocking nested systems, consistent with Bronfenbrenner's (1977) pioneering work on social ecologies.
Within transitional justice practice, collective reparations are an obvious example-albeit not overtly framed as such-of thinking in more social-ecological ways about health and the wider "spillover" effects of harm. In Morocco, individual and collective reparations were implemented pursuant to the recommendations of the country's Equity and Reconciliation Commission, established by King Mohammed VI in 2013. Individual victims of human rights violations committed during the period of the Commission's temporal mandate  received health care and monetary compensation, and collective reparations-specifically, social and economic projectswere granted to regions most affected by political violence (International Center for Transitional Justice, 2016).
More recently, in the Al Mahdi case, the ICC, for the first time, awarded individual and collective reparations for destruction of cultural heritage in Mali. Illustrating Balasco's (2017, p. 2) argument that "The harm committed to individuals has communal consequences that are equally in need of reparation," the Trial Chamber found, inter alia, that both the victim applicants and the wider Timbuktu community suffered "moral harm"-in the sense of emotional distress (ICC, 2017, para. 88)-and economic harm (ICC, 2017, para. 76). In other words, Al Mahdi's crimes affected interlinked individual and collective health in a broad sense.
A further practical example that illustrates, again implicitly, a social-ecological framing of health is the recognition by the Special Jurisdiction of Peace (JEP) in Colombia of Black and Indigenous territories as victims of the country's armed conflict. Thereby highlighting some of the environmental dimensions and legacies of the conflict, the JEP has also emphasized "the disruption of socio-ecological relations, such as forms of subsistence farming and other cultural practices, as well as disruption of the spiritual world" (Huneeus & S aiz, 2021, p. 211)-and, by extension, disruption of social-ecological health.
These are promising developments and, indeed, the example from the JEP comes closest to the social-ecological reframing of transitional justice that this article advocates. Whereas this first section has broadly examined the relationship between health and transitional justice, in what follows I explore this relationship specifically by drawing on interviews with victims-/survivors of sexual violence in conflict. Thinking in social-ecological terms about the health dynamics and legacies of conflict-related sexual violence is not only an under-researched way of approaching health in this particular context, but it also illuminates some of the limitations of so-called "survivor-centered approaches"-terminology that is now widely invoked at the international policy level (see, e.g., UN, 2019b). The very act of "centering" victims-/survivors arguably deflects from the significance of their social ecologies and from crucial connectivities-or entanglements-between the two (Clark, 2021; for an alternative interpretation, however, see Di Eugenio & Baines, 2021). Barad (2014, p. 175) has argued that, "Entanglements are not unities. They do not erase differences; on the contrary, entanglings entail differentiatings, differentiatings entail entanglings. One move-cutting together-apart" (emphasis in the original). This article's empirical section is effectively a "cutting together-apart." In "one move," it examines some of the individual health consequences that interviewees talked about, as well as some of the wider and interrelated social-ecological legacies.

Fieldwork and methodology
The idea for this article developed in the context of an ongoing research project ("the project")now in its final year-about resilience and victims-/survivors of conflict-related sexual violence. To date, extant scholarship on conflict-related sexual violence has largely overlooked resilience. There are some exceptions (Gilmore & Moffett, 2021;Zraly et al., 2013), but even when scholars working on conflict-related sexual violence refer to resilience, they often omit to substantively engage with the concept and its meanings (see, e.g., Edstr€ om et al., 2016;Koos, 2018). This is also true at the international policy level (UN, 2020a(UN, , p. 37, 2020b. As this article is not specifically about resilience, here is not the place for a detailed discussion of the concept. What is important to underline, however, is that the project approaches resilience as a social-ecological concept, to emphasize different relational connectivities within complex networks across multiple systemic levels. More specifically, it embraces a definition of resilience as "the qualities of both the individual and the individual's environment that potentiate positive development" (Ungar & Liebenberg, 2011, p. 127). Against this conceptual backdrop, it explores the different ways that victims-/survivors of conflict-related sexual violence, through interactions with their wider social ecologies, express resilience. It accordingly focuses on three case studies from different continents-BiH, Colombia, and Uganda-that reflect a maximum variation logic across multiple social-ecological dimensions (including conflict dynamics, patterns of conflictrelated sexual violence, history, and cultural context).
During the quantitative phase of the project, 449 research participants (all of them victims-/ survivors of conflict-related sexual violence) across the three countries completed a study questionnaire between May and December 2018. 2 A key aim of the questionnaire, as an entry point into the empirical part of the research, was to ascertain participants' resilience "scores" and to compare and contrast them. To this end, I opted to use the Adult Resilience Measure (ARM), a 28-item scale-consisting of individual, relational, and contextual subscales-that measures an individual's protective resources (Resilience Research Centre, 2016). Answers are scored from one to five, with higher overall ARM scores reflecting higher levels of protective resources and, hence, resilience (for an analysis of ARM results in relation to this research, see Clark et al., 2021). Measuring and quantifying resilience necessarily poses challenges. Scholars have described the concept as "stretchy" (Walklate et al., 2014, p. 410) and "messy" (Ziervogel et al., 2017, p. 123), and Saja et al. (2018. p. 863) underscore that resilience is "a set of diverse and inter-related nonlinear properties within complex and dynamic social systems which most often changes differently over time." Hence, to apply the ARM at a particular moment in time is necessarily also to measure individuals' resilience at a particular moment in time. It is precisely for this reason that the ARM was used in the context of a broader mixed-methods research design, aimed at capturing more of the richness and nuances of resilience.
As preparation for the qualitative stage of the fieldwork, participants in each country were grouped into four quartiles based on their ARM scores. Interviewees were then selected from each set of country quartiles, and particular care was taken to ensure that the choices made reflected the demographic diversity within the quartiles (especially gender, 3 ethnic, and age diversity). In total, 63 participants (21 from each country) took part in semi-structured interviews between January and July 2019. I undertook all of the interviews in BiH, and two postdoctoral researchers employed at the host institution conducted the interviews in Colombia and Uganda, respectively. No interpreters were used and all interviews were recorded using fully encrypted digital voice recorders. Ethical approval for the research was granted by the Humanities and Social Sciences Ethical Review Committee at the University of Birmingham, by the research funder, and by relevant authorities in BiH, Colombia, and Uganda.
Interviews were transcribed verbatim and translated into English. Both the transcripts and the questionnaire data were subsequently uploaded into NVivo. I developed the coding book over a period of approximately 12 months and made multiple adjustments and amendments during the coding process. Most of the interviews were double coded, to ensure consistency of coding, and I proceeded to use thematic analysis (Braun & Clarke, 2006) to identify overarching themes and patterns in the data. Health was a recurrent theme.
Interviewees talked, inter alia, about their own health issues, about injuries they had sustained, and about some of the emotional and psychological consequences of their experiences. However, as the next section explores, they also spoke more broadly about the health of their families and communities, and about the interconnections between their own health and the health of their social ecologies. In drawing on data from all three countries, one of the inevitable tradeoffs is the lack of sufficient space to discuss the wider historical and political contexts surrounding interviewees' experiences. Yet, it is also important to balance this tradeoff against the fact that presenting data from BiH, Colombia, and Uganda together (rather than from just one country) more fully captures some of the complex and multilayered health legacies of conflict-related sexual violence in very different countries. Hence, it also ultimately makes a stronger case for the social-ecological reframing of transitional justice that this article puts forward.
The health legacies of conflict-related sexual violence As King and Meernik (2019, p. 349) have pointed out, "One piece of the transitional justice puzzle centers on whether testifying produces healing and closure for witnesses or whether it does harm to those who recount wartime events." Although this article is not about the possible health consequences of testifying or otherwise participating in transitional justice processes, the interview guide used in the underpinning research did include a set of questions about transitional justice. 4 What some interviewees strongly articulated was a deep sense of frustration and disappointment regarding their own experiences of transitional justice-feelings that could have implications for health, particularly emotional health and well-being. Reflecting on her experiences of testifying in a local court, for example, a Bosnian interviewee recounted: To tell you the truth, it was awful for me, after, at the court and when … when I think that as a witness, I did not contribute much and that my testimony, the way I gave my observations would not be much help in convicting someone or something. I was not happy with my testimony. … Dear God, I even regretted going. And I said I will never go again, not for this or anything else. No one, whatever happens-no! At the end, they [referring to how she was treated in the courtroom] almost declared me a culprit. (Interview, BiH, July 3, 2019) A Colombian interviewee spoke about having to tell her story over and over again to different officials. In her words, "It really isn't easy, and what's worse is when the person who is supposed to look after you doesn't have the right kind of outlook, way of talking or the words to help you." She further explained, "I went into so many offices where they asked me questions that made me feel even more dirty than I already did" (interview, Colombia, March 29, 2019). These are just two examples from an extremely rich dataset that provides some important insights into the fundamental issue of how victims-/survivors would like to be-and deserve to be-treated within transitional justice processes.
Interviewees themselves made few direct connections between health and transitional justice, generally discussing the two separately. The crucial point, however, is that how they talked about the health legacies of conflict-related sexual violence (entangled with other conflict-related experiences such as physical detention and beatings)-not only at the individual level but also at the level of their social ecologies-has wider implications for transitional justice and how it approaches the basic concept of health.

Individual health legacies
That interviewees spoke so frequently about health underscores the argument that "Insofar as the body tends to disappear when functioning unproblematically, it seizes our attention most strongly at times of dysfunction" (Williams, 1998, p. 61). Interviewees attributed numerous health problems to their particular experiences connected to war and armed conflict (including but not limited to sexual violence). In BiH, for example, a male interviewee who spent nine-and-a-half months in detention after being captured as a wounded soldier told me: "I have more and more health problems due to my time in the camp. My back hurts, my hands, my … From the beatings and those things [sexual violence] that I experienced there" (interview, BiH, February 11, 2019). A Colombian interviewee explained that eight years after the sexual violence and torture that she suffered from members of the guerrilla Revolutionary Armed Forces of Colombia (FARC), she started to feel very unwell and needed an operation to remove her womb and one of her ovaries. Reflecting on everything that had happened, she confided, "I lost my internal organs because of the beatings they gave me-my insides are a mess. … I try to look after myself, although whichever way you look at it, things aren't the same" (interview, Colombia, March 29, 2019).
Many of the Ugandan interviewees had been abducted at a young age by members of the Lord's Resistance Army (LRA; Pham et al., 2008). One interviewee described how she was 13 years old at the time and was subsequently captured by a government soldier who locked her in a hut and sexually abused her. She had been left with several long-term health issues, including an injury to her left arm ("Even a basinful of clothes, two basins, I can't wash them") and problems with her pelvis, which had become dislocated several times. She linked these ongoing issues to the overall fact of "'being' with a man forcefully" when, as a child, she was "not ready for it" (interview, Uganda, March 4, 2019).
Relatedly, Ugandan interviewees spoke particularly about physical pain as a lasting legacy of their war experiences. One of them described how she was "given" to a man in the LRA when she was just 12 years old and how she subsequently sustained a bullet injury to her leg. She explained, "the bullet was acidic, so the pain comes periodically when the acid is eating up the muscles and so making it hard for me to walk on bare feet when the sun is hot" (interview, Uganda, February 1, 2019).
According to Humphrey (2002, p. 1), "Pain is the bodily feeling produced by violence, political power is the source of violence, and suffering is the legacy of violence remaining as a memory in individual bodies." Although transitional justice processes purport to "deal with" the past (Bell, 2004, p. 304) in order to enable societies to move on, this neat temporal compartmentalization (Doughty, 2017, p. 125;Nagy, 2008, p. 280) neglects the myriad ways that the past can continue to live in and through bodies. Fundamentally, the interviewees' discussions about and reflections on health-related issues highlight the "corporeal embodiment of the traumas" (Vassallo, 2008, p. 92) they had experienced. In the words of one of the Ugandan interviewees, who, after her abduction by LRA rebels when she was 10 year old, had to start "sitting" (a euphemism for having sex) with a man: "That thing [referring to sexual violence] sits in the life of that person" (interview, Uganda, March 19, 2019). One of the male Ugandan interviewees, for his part, talked about chronic pain in his abdomen area and revealed that "there is no power [sexually] to be with a woman" (interview, Uganda, February 22, 2019).
Pain was also a prominent thematic within the data in a broader sense. Interviewees across all three countries frequently articulated a deep sense of emotional pain and hurt (which the Ugandan interviewees commonly expressed through references to a "bleeding heart"). In some cases, the particular circumstances in which they were raped, abused, and mistreated had enhanced these feelings of pain and hurt.
A Bosnian interviewee, for example, stressed her young age when she suffered rape in 1992 (the first year of the Bosnian war) and reflected, "My soul is hurting. I mean, my childhood was destroyed with it, because I was very young. … I was not even 15. I was a child. I find this horrible. I can never … win this fight and come to terms with why this had to happen" (interview, BiH, February 19, 2019).
One of the Colombian interviewees described how the man who kidnapped and raped her (she identified him as being from the Popular Liberation Army [EPL], a guerrilla organization) had urinated on her, told her that she was his wife, and threatened to sodomize her father. In her words, "That really hurt me. It was something that really ate me up from a psychological point of view and it ate me up physically. That was really difficult for me" (interview, Colombia, March 30, 2019). Indeed, violence (or threatened violence) against family members, including children, had further helped to entrench interviewees' deep-seated feelings of pain and hurt. If, as Stolorow (2011, p. 143) has argued, "Trauma is constituted in an intersubjective context in which severe emotional pain cannot find a relational home in which it can be held," this "relational home" can be especially elusive when loved ones themselves are suffering pain.
In some cases, moreover, feelings of pain and hurt intersected with fear and distrust of others, affecting interviewees' relationships in ways that further exposed powerful psychological and emotional health legacies. A Bosnian interviewee who had served in the BiH army and was raped by two of her army colleagues, for example, disclosed: Simply, you don't have faith in anyone. The trust you have, that is in the moment. And you can be with someone once and it's nice, but you don't want to see them a second time. If they touch you a bit, even a bit, somehow it's over. You have fear and this is it. (interview, BiH, June 2, 2019) One of the Colombian interviewees spoke with great enthusiasm about her work as a social leader and the sense of fulfillment that she gets from helping people in her community, but this was also a form of distraction; "I bury myself in all this community work to forget that I exist, that I'm going through what I'm going through." Just one example of how her experiences of sexual violence continued to affect her in the present was her intimate relations. She underlined, "It changes you so much in terms of how you feel about … about other people." Wanting to be with another person is a powerful and intense feeling but one that she no longer experienced: "[W]hen you want to be with a person, it's because it feels right, it comes through their pores, their eyes, their smell, it calls to you. Now, everything is cold" (interview, Colombia, February 4, 2019).
Interviewees did not only speak about health in relation to themselves, however. This is important because in the context of conflict-related sexual violence, where so much of the focus-including at the international policy level-is on individual health care needs (see, e.g. UN, 2019b, WHO et al., 2020, there is a tendency to overlook wider social-ecological health legacies and their significance.

Social-ecological health legacies
Illustrating "the porosity of bodily boundaries" (Scott et al., 2017, p. 331), the interview data provided some powerful examples of how the crimes done to interviewees' bodies had porously entered and affected the bodies of their children. One of the clearest examples of this was a Ugandan interviewee with HIV whose nine-year-old son also had the virus. She explained that after getting his test results, "I myself just wanted to vanish, to lose my life. Why has this thing [HIV] been found on my child and not on myself only? Does it mean that all the trials and tribulations in this world are only targeting me?" (interview, Uganda, June 12, 2019).
To take a very different example, a Colombian interviewee talked about being raped while her children saw what was happening from another room. Her youngest child was very young at the time and she explained that he has an intellectual disability and cannot walk. She took him to see a doctor from Medell ın, who told her that her son's brain had not developed properly; "The child has suffered because he saw you suffer [including metaphorically from within the womb, such as when paramilitaries pointed a gun at the interviewee's pregnant stomach] and that affected him" (interview, Colombia, March 12, 2019).
In some cases, interviewees' health issues had also affected their families' health more broadly, including in an economic sense. In particular, some interviewees complained that due to poor health, they had fewer opportunities to earn a living. An interviewee in Colombia disclosed that she was not physically able to work much; "I have early onset arthritis. I mean, one of my knees doesn't work-I can't walk very well. That's all because of them [referring to the Popular Liberation Army] shooting me in the leg. I've got back problems." In addition to her arthritis, she had undergone a hysterectomy and had been very ill for six months. She recounted how, due to this protracted period of sickness, she had to give up her job and now struggled to find the money to buy food. Three of her four children were living close by, and they faced their own economic challenges. Her grandchild, moreover, was a hemophiliac, revealing a further nexus between physical and socio-economic health in the sense that the interviewee felt unable to seek financial help from her son (interview, Colombia, March 30, 2019).
Ugandan interviewees-many of whom were subsistence farmers-particularly accentuated the economic consequences of their health issues. One of them talked about the struggles she faced to send her children to school and described growing sesame or groundnuts as a way of generating crucial income. The previous season, however, she had suffered painful stomach ulcers and had consequently only been able to harvest some of the groundnuts. The rest went to waste. The millet she had planted was also ruined, and she expressed concerns that she had not been able to prepare the garden for the next planting season. Her late husband's own health issues-after being beaten by cattle rustlers from the Karamoja region in the north east of Uganda-and subsequent death had further affected the family's economic health. When asked what title she would give her life story, she answered: "Being a widow is not an easy thing" (interview, Uganda, April 2, 2019).
In their discussion of disadvantage, Wolff and De-Shalit (2007, p. 120) refer to what they term "dynamic clustering"; they use this to "mean both cases where a person 'accumulates' disadvantages over time, and the reproduction of disadvantage over generations." For a single individual, an example of dynamic clustering "would be a case where one is first unemployed, then becomes homeless, then loses one's friends, and then becomes very ill, and yet this does not all happen immediately but rather accumulates gradually over time" (Wolff & De-Shalit, 2007, p. 120). Ill health can be a cause of, or a significant contributor to, dynamic clustering of disadvantage. In both of the two previous examples however, the interviewees referred to wider health issues (present and past) within their families, which underlines that the dynamic clustering of disadvantage has important social-ecological dimensions. It was not just that interviewees' own health issues were affecting others within their social ecologies but, also, that sickness and ill heath existed within these ecologies in ways that added to the aforementioned clustering.
Indeed, several of the interviewees talked about health issues within their families, which were a significant source of stress and worry in their own right. In some cases, moreover, these familyrelated health issues intersected with other health dimensions of interviewees' environments, such as their living conditions. Exemplifying this, a Bosnian interviewee who worked as a cleaner and was the family's breadwinner described the challenges she faced in looking after her infirm mother. "I have no bathroom," she explained, "and I wish I had a bathroom where I could take her, shower her, put a nappy on her. I mean, I don't, simply, I don't have the living conditions to do this [looking after her mother] more easily" (interview, BiH, June 2, 2019).
Some interviewees, moreover, spoke about how war and armed conflict had directly affected the health of their loved ones, thereby illustrating multiple and interlaced health legacies across families. A Colombian interviewee who called her life story "A person with no luck" explained that paramilitaries had killed her brother and that her brother-in-law died during the same year. Just 18 months later, her other brother was killed: "[F]rom that point on, my mum has been ill, and it's something like nine years that she has been ill" (interview, Colombia, February 3, 2019).
Accentuating what Westling (2011, p. 127) has referred to as the "dense ecological texture of interdependence," some interviewees also talked more broadly about the "health" of their "ecological environment" (Bronfenbrenner, 1979, p. 3)-and, in particular, their communities. A male Bosnian interviewee who had spent 15 months in detention highlighted the prevalence of ill health among men in his village. Emphasizing that "Everyone has a code [meaning a classification, like PTSD]," he continued: They're not. … Well, half of them are not normal. More than half are not normal. Well, and here we are mostly from the camps [referring to camps, mostly set up early in the Bosnian war, where large-scale human rights violations took place]. … It is not, like … not like it was before, pre-war. … Everything is, like, everyone is on their own, just out for their own interests. (Interview, BiH, April 10, 2019) This last point exemplifies a common lament among the Bosnian interviewees-namely, that their communities were no longer the same; relationships had changed, social dynamics had altered, young people were leaving. Bosnian interviewees were particularly nostalgic about their prewar lives and the strong sense of community that had existed in Tito's Yugoslavia, established after World War II. One interviewee recalled, "We socialized a lot, believe me. No one ever asked who you were, what were you [referring to ethnicity]. But now, … there is no more socializing like that" (interview, BiH, July 3, 2019).
In short, the health of communities-and the healthiness of relationships within those communities-had changed. Some of the Ugandans also talked about the effects of war on the health of the entire community. Every household in the north was affected by the two-decades war between the government and the LRA, one interviewee reflected, and this had created many challenges. Simply, "there is no softness of the heart [no happiness]" (interview, Uganda, February 1, 2019).
Relatedly, some of the Ugandan interviewees spoke more broadly about the altered health of their physical environments, in the sense that they could no longer rely on these environments to sustain themselves. Although not a direct consequence of the war, imbalances in environmental health further affected, and intersected with, individual and family health, including economic health. Interviewees spoke in particular about the problem of drought (see Branch, 2018). In the words of one interviewee who was living with HIV, "The challenges I face in life are too much drought these days. We do farming, but there would still be no food. How to get it is hard as my body is still not light [healthy].
… I do not have the energy to do casual labour" (interview, Uganda, February 19, 2019). In a similar vein, a male interviewee who had suffered beatings and sexual violence from Karamojong cattle rustlers 5 commented, "It is hard to get money. We could … bait fish in the river there. But these days the water has dried up and you can't get any [fish]. How to get money is not easy now" (interview, Uganda, February 22, 2019).
In their work on marine reserve design, Edwards et al. (2010, p. 459) have argued that, "The health of coral reefs is dependent upon the outcomes of complex relationships between corals, macroalgae and grazing fish." Similarly, the health of social ecologies (and, relatedly, their resilience) depends on complex relationships between individuals, families, communities, and wider environments. The crucial point, as this section has sought to demonstrate, is that when interviewees spoke about health, they were not only speaking about their own health as a direct consequence of the sexual violence they had suffered. Their own health, in web-like fashion, was also connected to many other social-ecological dimensions of health-which, in turn, has wider implications for transitional justice.

Conclusion: Transitional justice and "healthworlds"
Within the field of peacebuilding, direct links have been made between health and peace (Arya & Santa Barbara, 2008;Grove & Zwi, 2008, p. 67;Rushton, 2005, p. 449). One example is the concept of "health bridges," which the 51st World Health Assembly formally adopted in May 1998. The WHO (n.d. [c]) has defined Health Bridges for Peace as "a multidimensional policy and planning framework which supports health workers in delivering health programmes in conflict and post-conflict situations and at the same time contributes to peace-building." From a social-ecological perspective, bridges are a useful metaphor for thinking about the myriad connections that inform and constitute health. Fundamentally, health is an issue that exposes deep interrelationships across and within social-ecological systems (Bunch, 2016, p. 615), making it "impossible to think of harm accruing to one being or set of beings in isolation" (Mitchell, 2014, p. 7). Mitchell (2014, p. 5) has thus underlined the need for a "'worldly' conception of harm," which recognizes that security disturbances such as war, human rights abuses, and conflict-related sexual violence threaten not only individuals but also "whole worlds-that is, irreducible, heterogeneous forms of collective being." Linked to this "worldly" idea of harm is a worldly conceptualization of health, and significant in this regard is the notion of healthworlds. Germond and Cochrane (2010, pp. 308-309) developed the concept based on their work in Lesotho and their observation that "separate words for 'health' and 'religion' do not exist in the Sesotho language, nor did their separation make sense to the worldview of the respondents." Looking for a way to combine both words, they used the single term bophelo, which means "life" or "health." For them, the crucial point is that health cannot be reduced to an individual. Rather, it has to exist across six interconnected socio-spatial configurations, including the person, the family and homestead, the village, and the nation.
According to them, a lack of bophelo in any one of these areas compromises the bophelo of the broader whole (Germond & Cochrane, 2010, p. 309). Looking for a way to translate the idea of bophelo into English, they introduced the concept of healthworld. Linking healthworlds and broadly defined corporeality, they emphasized not only the condition of the physical body of the individual, but also "the body of individuals in community, and the material environment that shapes both individual and social bodies" (Germond & Cochrane, 2010, p. 315). In other words, the notion of healthworld "resists the impulse to atomization" because individual and social-ecological health are inextricably interlinked. Individuals' healthworlds, in short, "are shaped by, and simultaneously affect, the socially shared healthworld constituted by the collective search for health and well-being" (Germond & Cochrane, 2010, p. 309).
Drawing on empirical data from BiH, Colombia, and Uganda, this article has sought to demonstrate that prioritizing the health needs of particular groups-in this case victims-/survivors of conflict-related sexual violence-can detract from (and decontextualize) the wider social ecologies with which individual health is interwoven. It is important to reiterate that the article's particular focus on health-a highly salient theme within the qualitative data-is not about distilling or compressing some of the many potential harms that result from conflict-related sexual violence. Rather, it is about demonstrating the relevance of social-ecological health-and its significance for transitional justice-in this particular context. Ultimately, therefore, the article's argument is that beyond thinking about the effects of sexual violence in conflict (and related forms of violence) on individual health, it is also necessary-as part of a broader social-ecological reframing of transitional justice-to explore their impact on healthworlds.
One way that transitional justice processes could potentially foster social-ecological health is by giving greater attention to and recognizing structural harms and injustices (see, e.g., Balint et al., 2014;Evans, 2016;Rooney & N ı Aol ain 2018). Structural violence potentially leaves significant health-related legacies (see, e.g., Anastario et al., 2020;Hudson, 2021), which, in turn, can have wider impacts on healthworlds, extending across socio-spatial configurations. Canada's National Inquiry into Missing and Murdered Indigenous Women and Girls, for example, which published its final report in June 2019, found that the country's colonial policies-including its residential schools system-"disrupted family and community dynamics which reverberated across generations," evidenced, inter alia, in "social problems among indigenous communities" (Luoma, 2021, pp. 45, 46). Acknowledgment of structural harms in and of itself is not enough. The bigger point, however, is that locating such harms within a synergistic social-ecological framework is a different way of thinking about them and why they matter for transitional justice. Critically, progress towards achieving key transitional justice goals-including peace, reconciliation, and nonrepetition-requires confronting rather than obscuring or glossing over "continuities of violence and exclusion" (Nagy, 2008, p. 280).
There is a second way that transitional processes might address social-ecological health. In the context of their aforementioned discussion of kaleidoscopic justice, McGlynn and Westmarland (2019, p. 194) talk about connectedness as justice. "Connectedness," they argue, "is about belonging in society, being recognized, being treated with dignity, having a voice" (McGlynn & Westmarland, 2019, p. 194). Although conflict-related sexual violence, and other forms of violence more broadly, can damage and rupture this connectedness, including relationships, it is the multilayered intrinsic connectedness between individuals and their wider social ecologies that is crucial to explaining how and why the health of one necessarily affects and is affected by the health of the other. Transitional justice processes could acknowledge and address the significance of healthworlds by locating the very concept of "legacy"-which is surprisingly underdiscussed within transitional theory and practice-within a wider connectivity frame. This means thinking about the legacies of violence in the sense of how they implicate connectivity and connectedness, and what they do to them.
In their work on corporeal citizenship, Scott et al. (2017, p. 331) argue that, "The movement of toxics across bodies, and through our environments and economies, provides a rationale for why we should extend out our spheres of political and ethical responsibility from the level of the individual or family to cover broader ecosystems and communities." The "movement" of legacies of conflict-related sexual violence across and via connectivities provides a rationale for thinking more about what such violence (and other types of conflict-related violence more broadly) does not only to individual health but also to the health of the social ecologies in which victims-/survivors live and move. However, it is also important to stress that the process of building socially shared healthworlds is necessarily a cross-societal endeavor that reflects the aforementioned connectedness. While transitional justice has a crucial role to play in the process, so too do, inter alia, states, NGOs, health institutions, and communities. Notes 1. This article uses the terminology victims-/survivors to reflect the fact that some research participants primarily viewed themselves as victims, some saw themselves as survivors, and some considered themselves as victims and survivors. 2. The study questionnaire was administered by the three members of the research team (one based in each country); two independent psychologists in BiH and Colombia, respectively; and several in-country organizations-namely, Snaga Zene and the Centre for Democracy and Transitional Justice (BiH), Profamilia, Ruta Pacifica de las Mujeres and Colombia Diversa (Colombia), and Facilitation for Peace and Development (FAPAD) and the Justice and Reconciliation Project (JRP in Uganda). These organizations also played a key role in locating research participants, many of whom were already known to them. 3. Notwithstanding the efforts made to reach male victims-/survivors in each country, one of the project's limitations is that of the 449 victims-/survivors who participated in it, only 27 were men (12 in BiH, five in Colombia, and 10 in Uganda). Of these 27, 11 also took part in the qualitative stage of the research. 4. These transitional justice questions included the following: "Transitional justice refers to the process of dealing with past human rights abuses in a society. It can take many forms, including criminal prosecutions, truth commissions and reparations. Have you experienced any form of transitional justice?" "If yes, did it change your life in any way and how?" "What do you need from transitional justice for it to make a meaningful difference to your life and why?" 5. Research on conflict-related sexual violence in Uganda has focused overwhelmingly on LRA crimes. In contrast, acts of sexual violence committed by the Karamojong, in the context of violence and conflict linked to cattle raids, has received little attention (see Stites & Howe, 2019, p. 143).