Resilience as a multi- directional movement process: A conceptual and empirical exploration

Movement is a recurrent thematic within extant resilience scholarship. Ecological theorizations of resilience emphasize systems that are in constant movement and flux. Terms such as “bouncing back” and “bouncing forward” are frequently used to describe how individuals recover and move on from adversity and trauma. However, integrated individual-systemic explanations of resilience movement dynamics are lacking. Seeking to address this gap, this interdisciplinary ar ticle expressly frames resilience as a multi- directional move ment process. Individuals do not just move forward and backwards. Rather, they move in multiple and varied ways as they seek to navigate their way through adversity— and through systems and social- ecological traps that both shape how they move and require them to keep moving. Drawing on interview data with victims- /survivors of conflict- related sexual violence in Bosnia- Herzegovina (BiH), Colombia, and Uganda, the article empirically explores what “movement” looks like at the microlevel through a focus on everyday forms and expressions of movement, while also examin ing how wider systemic environments shape and influence these movements.


| 3
CLARK their lives. The conclusion articulates and develops the core idea of resilience as a multi-directional movement process.

| RE S ILIEN CE SCHOL AR S HIP AND THE THEME OF MOVEMENT
In their research on resilience and climate instability, Scheffers et al. examine the significance of tree living (arboreality). According to them, "Species that are capable of exploiting the vertical gradient provided by trees have a broader niche space available to them than is available to ground-dwelling species" (Scheffers et al., 2017, p. 788).
They add that "individuals moving through the vertical gradient on a short-term basis experience a large variation in microclimate as they do so," which can make them more physiologically resilient to short-term climatic variation (Scheffers et al., 2017, p. 788).
Addressing the issue of soil water movement and system resilience in the semi-arid Kalahari, Doudill et al., (1998, p. 456-457) found that "the Kalahari soils possess an inherent resilience to changes in profile patterns of soil water and nutrient availability." They linked this to "the dominance of matrix flow 2 in sandy soils," on the basis that flow enhanced opportunities for mineral absorption (Doudill et al., 1998, p. 457). Zhu and Ruth similarly underscore the concept of "flow" in their discussion of industrial ecosystems. Precisely what makes these systems resilient, they argue, is "the ability to maintain their defining feature of eco-efficient material and energy flows under disruptions such as changes in production levels and technologies, and firm closure and relocation" (Zhu & Ruth, 2013, p. 74).
While very different, these examples-drawn from the fields of ecology and industrial ecology respectively -highlight important movement and flow dynamics. Looking at resilience literature more broadly, movement also emerges-directly or indirectly-as a recurrent thematic. Luthar et al., (2000, p. 543), for example, define resilience as "a dynamic process encompassing positive adaptation within the context of significant adversity." Indeed, the adjective "dynamic" is commonly used in relation to resilience (see, for example, Lenette et al., 2013, p. 639;Stainton et al., 2019, p. 726;Windle, 2011, p. 164) and it powerfully conveys ideas of motion, energy, and change. Ungar (2013, p. 256) discusses resilience as individuals engaging in behaviors "that help them to navigate their way to the resources they need to flourish." For their part, Jeans et al., (2017, p. 2) associate resilience with three inter-linked capabilities, namely absorption, adaptation, and transformation. The third is particularly evocative of movement in the sense of change. As will be discussed in the empirical section, transformation in the context of SES means that "a system shifts from one state to another, as defined by a change in system parameters" (Cretney, 2014, p. 630;Folke et al., 2005, p. 457).
Two particular dimensions of movement are central to this research. The first of these is a systemic dimension and the second is an individual dimension. While both are important in their own right, in resilience research they are often discussed separately. This reveals only part of a bigger picture. VanderPlaat (2016, p. 191), for example, reflects that "Sociologists would argue that the focus on adversity as individual, personal, and immediate not only deflects our gaze from the social structures that cause and maintain these social conditions in the first place, but also limits our capacity to deal with these issues on a broad scale basis." This research addresses both the individual and the systemic, through its theorization of resilience as a multi-dimensional movement process that locates individual movement in relation to broader systemic dynamics.
In contrast, the ecologist C.S. Holling was interested in the behavior of systems in response to disturbances, including disease and manmade pressures on resources such as fish stocks. What he particularly underlined is that following disturbances, ecological systems rarely return to their previous state (Holling, 1973, p. 10). In this regard, he made a crucial distinction between engineering and ecological resilience. The former, which he also referred to as the "stability view" (Holling, 1973, p. 21), "concentrates on stability near an equilibrium steady state, where resistance to disturbance and speed of return are used to measure the property" (Holling, 1996, p. 33). In contrast, the latter (ecological resilience) "emphasizes conditions far from any equilibrium steady state, where instabilities can flip a system into another regime of behaviour -that is, to another stability domain" (Holling, 1996, p. 33).
Both theorizations of resilience reflect aspects of movement but in different ways, underscoring that the critical difference between the two "lies in assumptions regarding whether multistable states exist" (Holling, 1996, p. 38). In engineering resilience, the accent is on movement back to an equilibrium state. From an ecological resilience perspective, in contrast, there is no equilibrium to return to (Berkes et al., 2003, p. 15). Rather, there is continual movement as an integral part of broader system dynamics that are "subject to cycles of continuous change and renewal" (Berkes et al., 2003, p. 7). Gunderson and Folke (2005) thus refer to resilience as "the capacity for renewal in a dynamic environment." Two core concepts within ecological resilience literature particularly exemplify movement. The first of these are thresholds, usefully illustrated by the ball-in-the-basin model. A ball in a basin will always be moving toward an equilibrium state, which is itself in constant motion due to changes in external conditions (Walker & Salt, 2006, p. 54). Hence, the ball is never still in the basin and if it goes over the edge, it has "crossed a threshold into a new basin of attraction -a new regime" (Walker & Salt, 2006, p. 55). Thresholds, therefore, represent "crossing points that have the potential to alter the future of many of the systems that we depend upon" (Walker & Salt, 2006, p. 53), either positively or negatively. The crucial point is that complex ecological systems-which operate at different organizational levels (Walker et al., 2004)-necessarily need to adapt, and this constant movement can flip them "from one point of equilibrium to another" (Palma-Oliveira & Trump, 2018, p. 116). How much disturbance a system can absorb before crossing a threshold and changing its structure is the quintessence of ecological resilience (Holling & Gunderson, 2002, p. 28). As Holling (1973, p. 17) defined it, "Resilience determines the persistence of relationships within a system and is a measure of the ability of these systems to absorb changes of state variables, driving variables, and parameters, and still persist." The second core concept is the adaptive cycle metaphor. According to Holling and Gunderson (2002), systems move through four key phases. The first of these two phases, namely exploitation (r) and conservation (K), "comprise a slow, cumulative forward loop of the cycle, during which the dynamics of the system are reasonably predictable" (Walker et al., 2004). In the K phase, the system starts to become less flexible as resources are locked up. The resultant collapse leads to a release (Ω) and reorganization (α) phase. These two phases "together comprise an unpredictable backloop," and the α phase leads into a new r phase which can be similar or dissimilar to the first one (Walker et al., 2004). In other words, the α phase is not a bouncing back to a pre-disturbance equilibrium, but, rather, a potentially transformative phase. As Holling (2001, p. 395) emphasized, "The adaptive cycle therefore embraces two opposites: growth and stability on the one hand, change and variety on the other." The concept of the adaptive cycle further underscores how complex adaptive systems repeatedly move and shift. As the adaptive cycle progresses, "a system's ecological resilience expands and contracts," with resilience highest during the back loop of the cycle when resources are freed up (Holling, 2001, p. 395).
Systemic theorizations have hugely enhanced the field of resilience scholarship. Ecological resilience, however, is arguably an abstract concept (Desjardins et al., 2015, p. 156), and concepts such as thresholds and adaptive cycles are not easily translatable to everyday adversities and disturbances, which affect not just systems but also human (and non-human) lives. Resilience scholars thus increasingly talk about social-ecological systems (SES), precisely to capture "the interdependencies and feedbacks between ecosystem development and social dynamics…" (Gunderson & Folke, 2005; see also Adger, 2000, p. 347;Alberti & Marzluff, 2004, p. 242;Berkes et al., 2003;Cretney, 2014, p. 628). Yet, even with this emphasis on the inter-connections between ecological and social systems, exploring movement from a purely systemic perspective gives few insights into what movement actually entails at an individual level.

| Individual movement
Some scholars have focused on individual dimensions of movement, referring to resilience as a process of "bouncing back" (Emlet et al., 2017;Silverman et al., 2017;Smith et al., 2010), "bouncing forward" (Houston, 2018;Sleijpen et al., 2013;Walsh, 2002) and/or "moving forward" (Isaak et al., 2015;Tenhula et al., 2014). The idea of "bouncing back," which reflects Holling's aforementioned concept of engineering resilience (Smith et al., 2008, p. 199), is particularly problematic (Clark, 2020;Lenette et al., 2013;Vale, 2014), not least in its assumption that a return to a pre-disturbance state is possible and/or desirable. Indeed, Walker (2020) maintains that viewing resilience as a process of "bouncing back" is "[p]ossibly the most common misinterpretation of resilience." This is because resilience entails a process of learning-"learning from a disturbance, to be able to better cope with a similar disturbance in the future" (Walker, 2020). Although Walker is discussing the resilience of systems, his argument is also pertinent to individuals. In short, just as a system "does not bounce back to look and behave exactly like it did before" (Walker, 2020), neither does a person.
More broadly, ideas of "bouncing back," "bouncing forward," and "moving forward" do not sufficiently convey what resilience as a movement process actually entails in a quotidian sense. They neglect, for example, potential frictions between desired movements and those that are possible. Relatedly, just as systemic theorizations of resilience are limited in terms of elucidating movement at the microlevel, a strong emphasis on individual movements can decontextualize them from wider systemic dynamics. In their research on resilience in Thai rural older people, for example, Pathike et al., (2019, p. 317) found that "Moving on in the rural sociocultural and economic context describes the Thai elders' appreciation for living, earning an income and being courageous. Moving on for these participants involved expressing their feelings and connecting with their people, beliefs and customs." In other words, the research participants were moving on in connection with other elements (i.e., sub-systems) within their social ecologies. The key point, thus, is that individuals move in relation to, and in interaction with, broader systems (Adger, 2000, p. 347, 350) whose non-linearity-juxtaposed to the highly linear ideas of bouncing/moving forward-"leads to multiple possible outcomes of dynamics" (Alberti & Marzluff, 2004, p. 243).

| ME THODOLOGY AND FIELDWORK
The idea for this article developed out of a larger research project-led by the author who is Principal Investigator -about resilience and victims-/survivors of conflict-related sexual violence. Notwithstanding a plethora of resilience literature across multiple disciplines, resilience is a concept that remains under-examined and overlooked in the context of conflict-related sexual violence (see, however, Edström et al., 2016;Koos, 2018;Zraly et al., 2013).
Addressing this gap, and focused on three countries 3 that have experienced high levels of such violence over different temporal periods-namely, BiH, Colombia, and Uganda-the project is exploring some of the ways that victims-/survivors in these countries demonstrate resilience in interaction with their wider social ecologies. The choice of three maximum diversity case studies is important in this regard for elucidating cross-contextual and cross-cultural similarities (and differences) across the data set, and thus for building an overall resilience narrative. Consistent with its social-ecological approach, the project theorizes resilience as "a dynamic and contextual process in dialogue with local worlds and environments" (Hatala et al., 2020, p. 8).
In the quantitative phase of the project, a questionnaire was designed to measure participants' levels of resilience and to assess which independent variables had the biggest impact on the dependent variable (individual resilience scores) and accounted for the greatest variance. A crucial part of the questionnaire was the Adult Resilience Measure or ARM (Resilience Research Centre, 2016). The ARM is a 28-item scale that measures a person's protective resources across individual, relational, and contextual sub-scales. Answers are scored from 1 to 5, and a higher overall ARM score indicates more protective resources and, hence, greater resilience. The questionnaire additionally included a Traumatic Events Checklist, a Centrality of Event Scale (Berntsen & Rubin, 2006) and questions about current problems. In total, 449 victims-/survivors of conflict-related sexual violence across the three countries completed the final version of the questionnaire between May and December 2018. Although the number of male respondents was small (n = 27), this reflects the immense challenges of locating and establishing contact with male victims-/survivors (see, for example, Ferrales et al., 2016, p. 571). The author (based in BiH) and two postdoctoral researchers (based in Colombia and Uganda, respectively) administered a portion of the questionnaires, and several non-governmental organizations (NGOs) applied the remainder (after receiving training).
In the qualitative phase of the research, ARM scores were used to divide respondents in each country into four quartiles, from those with low ARM scores to those with high ARM scores. Interviewees were chosen from across the different quartiles, and every effort was made to ensure that these selections respected demographic diversity-and in particular gender, age, and ethnic/racial diversity-within the quartiles. Sixty-three people in total (21 from each country) were interviewed. The interviews, which took place between January and July 2019, were conducted in the relevant local languages, and the author undertook all of the interviews in BiH. With the interviewees' informed consent, all of the interviews were recorded using encrypted digital voice recorders. The average length of an interview was 1 hr, although some lasted longer. The Humanities and Social Sciences Ethical Review Committee at the author's host institution, the research funder, and relevant authorities in BiH, Colombia, and Uganda granted ethics approval for the research.
The interview guide was designed to elicit important information about the interviewees' war/conflict experiences (which included multiple traumas and not only sexual violence), their current lives, and their support networks. A key aim was to mitigate issues of research fatigue (Boesten & Henry, 2018) by asking some questions that interviewees were unlikely to have previously been asked. These included the following: If you were to tell the story of your life, what title would you give it? Are there parts of your war story that are important to you and that you are never asked about? After everything that you have gone through, what are the factors that have been most important in helping you to rebuild/start to rebuild your life?
All of the interviews were transcribed verbatim and translated into English, and the transcripts were subsequently uploaded into NVivo. The author developed a detailed codebook over 12 months and all of the interviews underwent a first and second stage of coding over a period of several months. To ensure coding consistency and rigor, almost all of the interviews were double-coded. The author used thematic analysis-"a method for systemically identifying, organizing and offering insight into patterns of meaning (themes) across a dataset" (Braun & Clarke, 2012, p. 57)-to identify and develop core themes.
The eight themes-which include "'I am all that I've lived': Connectivities of violence" and "'We have to live': Reconnecting with life"-have a common connectivity thread, which inductively emerged from the qualitative data. The project thus seeks to tell a novel "connectivity story" about resilience through a focus on the dynamic connectivities between individuals and their social ecologies. Connectivity is a widely discussed concept within the literature on ecology and is frequently linked to movement. Sheaves (2009, p. 108), for example, offers "a broad ecological definition of connectivity that refers to physical or ecological events that allow materials or organisms to move between or influence habitats, populations or assemblages that are intermittently isolated in space or time." Movement, in turn, can contribute to fostering resilience. As one illustration, Olds et al., (2012, p. 1,195) argue that "Mobile organisms enhance ecological resilience by linking ecosystem functions across landscapes." In other words, this article's focus on movement and resilience developed in the context of the project's broader connectivity frame.
The following section discusses some of the ways that ideas of movement emerged during the analysis of the interview data. In so doing, and consistent with the approach taken in the underpinning research project introduced at the start of this section, it adopts an explicitly comparative approach. The use of comparison is an important part of exploring how individuals and wider systems "move" and interact with each other in different social-ecological contexts-and how these contexts can variously shape, necessitate, and frustrate movement.

| E VERYDAY MOVEMENT AND RE S ILIEN CE
According to Ogden, "Those who learn to operate in a vastly changing global environment, those who can walk on quicksand and dance with electrons…those who see connections where others see chaos will flourish and find opportunity in every disturbance" (Frank Ogden, cited in Wass, 2020, p. 231). The reference to "connections" underscores crucial individual-systemic synergies. The ideas of walking on quicksand and dancing with electrons reinforce the significance of movement. Walking on quicksand, moreover, can be contrasted to walking in quicksand. Some people with health conditions have talked about walking in quicksand. Poindexter (2017, p. 349), for example, cites a patient with fibromyalgia who states: "I feel like I am walking through quicksand on a daily basis." Walking on quicksand has more positive connotations-of walking without sinking-but it is no less challenging.
Underlining this, Bauman reflects: "…the art of walking on quicksand is still beyond me. What I've learned is only how difficult this art is to master and how hard people need to struggle to learn it" (Zygmunt Bauman, cited in Dawes, 2011, p. 131). Notwithstanding these difficulties, it is important to keep moving and to avoid being sucked into the quicksand.
This idea also strongly emerged from the interview data in various ways. Interviewees-and in particular Bosnian and Colombian interviewees-explicitly spoke about myriad forms of movement; these included falling, sinking, carrying, climbing, and flying. While none of them used the analogy of walking in/on quicksand, they had a clear sense of how they should be "moving." A male interviewee who had spent more than a year detained in various camps during the Bosnian war, for example, underlined that following his release, he was determined to "get down to business and move on, forward, just forward…." While now retired on health grounds, he was a builder by trade and stressed how much his work-and generally keeping himself moving in the sense of not looking backwards-had helped him. In his words, "I have not allowed myself to, to, to fall into depression and to return to what was. I immediately jumped over it. I talked about it and pushed it out the door. Getting down to some work or something" ( interview, BiH, 2 July 2019).
From her very first sentence-"My life is very busy at the moment"-a Colombian interviewee conveyed a strong sense of movement throughout her interview. Reflecting on the significance of her Indigenous roots, for example, she explained: Accentuating the aforementioned challenges of "walking on quicksand," interviewees also conveyed some of the difficulties involved in moving forward-which notions of "bouncing" do not sufficiently capture. Memory was particularly significant in this regard. In the aforementioned adaptive cycle metaphor, memory is positively associated with resilience. More specifically, the K (conservation) phase has been termed "remember." Memory herein refers to "the accumulated experience and history of the system, and it provides context and sources for renewal, recombination, innovation, novelty and self-organization following disturbance" (Folke, 2006, p. 259).
As Nykvist and Heland (2014) point out, however, this social-ecological memory can also foster undesirable resilience-including inertia and rigidity-in the sense that "memory is undoubtedly a process that 'stores' and 'forgets' all types of information." What strongly emerged from the interview data was a tension-or friction-between interviewees' desire to forget, as part of the process of moving on, and the intrusiveness of traumatic memories (Ehlers et al., 2002)-including pain, hurt, fear, and distrust-that were "stored" in multiple locations, including in participants' own bodies. A Ugandan interviewee, for example, spoke about some of the injuries that she sustained following her abduction by Lord's Resistance Army (LRA) rebels when she was 10 years old. Asked how she would describe herself, she answered: "A person who has passed through that problem of sleeping [a euphemism for rape], because sometimes it can spoil even your private parts. That thing [rape] sits in the life of that person" ( interview, Uganda, 19 March 2019). In some cases, moreover, memories were "stored" in broader systems, including in everyday structural violence and gendered inequalities that fostered feelings of injustice, thereby impeding a sense of closure. In Colombia, an interviewee who did not identify with any particular ethnic group but strongly emphasized her campesina identity as a woman who grew up in a rural area insisted that there is no justice. In her words: For me, no, justice doesn't exist. Not for me it doesn't because nobody cares about any of the things that have happened to us [meaning other women with similar socio-economic profiles]. At least, I think…if I were a magistrate, or if I were a president and somebody went and raped one of my children. Well! What would happen? But if it's a poor countrywoman, then that has no importance; it's not worth anything. A countrywoman isn't valued anywhere. You're just poor and they don't look at you as a human being; as a person and the value you might have. Really, they don't look at that.

( interview, Colombia, 3 April 2019)
Memories themselves thus evince movement dynamics; they "possess an energy of flow that leads to their movement from whatever locked places, whatever traumatized neural pathways they inhabit, into the interstices of the everyday" (Culbertson, 1995, p. 175). This movement of memories can significantly affect how people move in the sense of getting on with their lives. Speaking about her memories related not only to the Bosnian war, but also to the abuses that she suffered as a child and, subsequently, in her marriage, one of the interviewees explained: "…you always carry it [the past] with you, like a bag, and sometimes you forget that it is on your back, you get used to the load. And sometimes, it is heavy and you put it down for a bit, and then, again, what is it? Simply, you go on… [long pause]. You go on" ( interview, BiH, 22 February 2019).
If memories of the past affected movement, so too did the adversities and stressors that many of the participants continued to face. In all three countries, for example, almost all of the interviewees expressed financial worries and spoke about their everyday struggles to make ends meet. Significant in this regard is the concept of social-ecological traps. According to Haider et al., (2018, p. 318), "Trap dynamics are an emergent outcome of complex social-ecological interactions." As one illustration, most of the Ugandan interviewees were subsistence farmers who relied for their existence on natural resources that are highly sensitive to broader climate dynamics.
Drought, in particular, had affected people's livelihoods. Emphasizing the economic challenges that he faced in sending his three children to school, one of the Ugandan interviewees explained: "It is hard to get money. We could also bait fish in the river there. But these days, the water has dried up and you can't get any. How to get money is not easy now" ( interview, Uganda, 22 February 2019). Another interviewee, who repeatedly described her life as hard, lamented: Even the [financial] strength…to send the children to school is no longer there. This is because I rely only on cultivating land to what? To enable me to get some money whenever I sell the produce.
With the proceeds, I can send the children to school. But this year, it did not happen. I have three children who stayed home without studying. There has been no rain, so my life is now very hard indeed.

( interview, Uganda, 15 April 2019)
Relatedly, trap dynamics emerged from the data in the sense of the limited opportunities that some interviewees' wider systemic environments afforded. A Colombian interviewee, for example, spoke about her dreams for her children, while also painting a picture of unfulfilled dreams in the context of a state which, she maintained, "forgets you." This interviewee, who was a victims' leader, revealed: "My daughter couldn't finish her degree. Now she is a mother-she's got two babies-and worst of all she isn't married. She couldn't achieve her dreams of being a doctor because she couldn't stay at the university and here there are no opportunities for a poor person..." ( interview, Colombia, 29 March 2019). These different examples from Uganda and Colombia show how "Poverty traps work through self-reinforcing (vicious) cycles that over time accentuate the disadvantages experienced by poor people" (Rudel et al., 2013, p. 168).
The existence of trap situations, however, is only part of a larger story. How people respond to traps is also significant. In this regard, Boonstra et al., (2016, p. 879) identify five types of possible responses that reflect different interplays between desires and opportunities: "thick conformity, thin conformity, resignation, innovation, and rebellion." For them, the crucial point is that "…individual responses have a potential to resolve traps" (Boonstra et al., 2016, p. 886). As an illustration of this, the final part of this section provides examples of "innovation." The bigger issue, however, is that interviewees' relationship with traps extended beyond just desires and opportunities. Fundamentally, they were moving in relation to social-ecological traps, which themselves necessitated movement.
Highlighting this, the Ugandan interviewees spoke most about economic worries and their efforts to find ways to support themselves and their families. These efforts, moreover, had a strong movement dimension; many of the interviewees were engaged in physical land-based activities and, more broadly, they strongly conveyed the idea of moving between different resources in the sense of making use of whatever was available to them. A widowed interviewee with eight children, five of whom were living at home, did not have her own land 4 and was living on Catholic Mission land. However, she described how she farmed, planted, and harvested cassava and sorghum, worked on other people's land when opportunities arose, brewed and sold various types of alcohol (including a local gin which she called ting ni ling, meaning "lift it whole") and baked bread for sale ( interview, Uganda, 12 June 2019). For his part, the aforementioned male interviewee who discussed the local river drying up talked about making charcoal for people, making bricks, climbing rock outcrops, and processing coarse aggregate for sale, as a result of which he had been able to purchase a bull to help him work in his field ( interview, Uganda, 22 February 2019).
Although to lesser extent, some of the Bosnian and Colombian interviewees also spoke about managing economic and existential problems in ways that further demonstrated a movement thematic. One of the Bosnian interviewees, for example, was internally displaced in the BiH Federation, having previously lived in Republika Srpska 5 prior to the war, and she underlined the financial pressures of paying off a loan that she had taken out in order to provide her and her children with a home of their own. Complaining that her husband was a drinker who contributed nothing to the family's finances, she described herself as "capable" (in addition to "smart" and "strong"). 6 In her words: "Well, capable because, well, I have three children in school, I feed them, I cook for them, I go to one job, then to the other job [both cleaning jobs]. This house, here, it would not be standing if it were not for me, err, if I did not take a loan and do two jobs" (interview, BiH, 3 February 2019). The fact of being internally displaced, however, and of physically moving from one location to another (typically from rural areas to cities), had affected how some interviewees "moved" in the sense of trying to deal with their economic issues. Referring to her former life in the countryside, a Colombian interviewee explained: There I had all the skills to earn money, get what I needed, to do so many things… Here it's a lot more difficult. For me especially since I have no qualifications… You go to get work and they say: "Do you have a degree?" and I have to say: "No señor." You understand me?...I want to get a farm and go to the country to work the land, to produce something, grow tomatoes, all those things that I know how to do. Grow yucca [cassava], plantain-just imagine! ( interview, Colombia, 3 April 2019) According to Ungar (2008, p. 225), "resilience is both the capacity of individuals to navigate their way to the psychological, social, cultural, and physical resources that sustain their well-being, and their capacity individually and collectively to negotiate for these resources to be provided and experienced in culturally meaningful ways." The interview data strongly speak to both themes, but interviewees were not simply navigating their way to, and negotiating for, resources. Significant in this regard is the distinction between specified and general resilience.
Specified resilience is "the resilience of what, to what" (Walker et al., 2009). In other words, it is about resilience to a specific disturbance or shock. For the interviewees, these stressors spanned a broad temporal period, relating to particular events and experiences both during war and armed conflict and in the context of their present lives.
They included loss of loved-ones, physical displacement from land, illness, and aforementioned environmental disturbances such as drought. Specified resilience, however, can exist at the expense of general resilience, thus highlighting resilience trade-offs (Walker & Westley, 2011). Walker et al., (2009), for example, argue that in feedback systems, which include SES, "increasing robustness to disturbances at a particular frequency range may reduce robustness to disturbances at another range…." Looking at the interview data, it was particularly striking that as well as dealing with specific stressors, interviewees in all three countries were living in environments where they faced longer-term, systemic stressors that called for more general resilience-which itself could be interpreted as a trade-off in the sense of wider systemic failures. In Colombia, for example, notwithstanding the signing of a peace agreement in November 2016 between the government and the Revolutionary Armed Forces of Colombia (FARC), high levels of insecurity persist-a legacy of more than five decades of armed conflict and an illustration of how social problems themselves can be highly resilient (Cinner & Barnes, 2019, p. 51). One example of general resilience in this context is knowing how to "move" in particular areas. An interviewee whose husband was killed by paramilitaries talked about some of the challenges of living in the city where she is now internally displaced. As she explained: …where I live now is an area very…you have to know how to behave; you have to know how to talk because of the [armed] groups and organized gangs. So, well, I have my son and everyone else just gets on with their own thing. My daughter-in-law goes to her job and I stay at home with my son.
They go to work and I stay at home-on my own. The other Saturday, some people came and they wanted me to pay protection money. Protection costs 1,000 Pesos, so I say: "I don't have that kind of money," 7 ; "Ok then". Then at 8:00, they come again and well, if you don't hand it over, and if you refuse to pay, then you're marked-they have their eye on you. That's how it is here in XXX [name of her city]. In some of the neighbourhoods in XXX, you get marked out like that. So, you have to know how to get along with people….
( interview, Colombia, 12 March 2019) The interviewee's words resonate with Marston's research in the city of Medellín. Discussing people's responses to the criminal gangs that patrol local neighborhoods, he found that "With the aim of remaining, many residents develop coping mechanisms to weather violence and avoid unwanted attention from the gang. For instance, they stay indoors, hide wealth, and keep their heads down" (Marston Jr., 2020, p. 2003. The broader point is that interviewees were negotiating ways to move in and navigate their ways through contracted systems, characterized by limited resources and structural adversities; and such systems contribute to fostering a default resilience. It is precisely for this reason that some scholars are highly critical of resilience, arguing that it promotes a self-help ethos that requires individuals "to govern themselves in appropriate ways" (Joseph, 2013, p. 41), in accordance with a broader neo-liberal agenda that diminishes the state's own responsibilities toward its citizens (Chandler, 2012;Duffield, 2016;Evans & Reid, 2013).
The emphasis that neo-liberal arguments place on the purported agenda underpinning resilience, however, gives little insights into what exactly resilience is (Bourbeau, 2015, p. 379) or how it manifests in diverse sociocultural environments. It also detracts from important resources and connectivities within individuals' wider social ecologies, underscoring the limitations of viewing resilience through a narrow top-down lens. Regarding these resources and connectivities, the analysis thus far has demonstrated that interviewees' movements were less about their own desires and more about the particular circumstances in which they found themselves, thereby foregrounding the idea of frictional movement in the context of ongoing challenges and adversities. According to Tsing, however, friction does not have to be something negative. Exploring the concept of friction in relation to global connections, she argues that "Speaking of friction is a reminder of the importance of interaction in defining movement, cultural form, and agency" (Tsing, 2005, p. 6). Supporting this, some of the interviewees' interactions with their wider socialecologies-partly in response to friction-had helped them to move in ways that they themselves wanted to, in the sense of moving on with their lives. Interactions with local NGOs/women's associations were particularly important in this regard (the Ugandan interviewees also spoke about support from international organizations such as World Vision).
A Bosnian interviewee, for example, talked about the lavender that she had received through one NGO's occupational therapy program. Showing the author some photographs of her land, which had been passed down through her family, and the lavender growing on it, she explained: "I got lavender, prepared the soil, seedlings, a lot of, well…[short pause]…I covered the ground, 1,000 seedlings, and there I have…[short pause]. This land is my resource, my 640 square metres and the soil draws out all the negative energy from me" (interview, BiH, 30 January 2019). Speaking about her involvement with the Red de Mujeres Víctimas y Profesionales (Network of Women Victims and Professionals), a Colombian interviewee stressed how much the Network had helped her. In her words, "I'm so happy now because before I was stuck in a rut, I couldn't do anything." While this reference to being stuck in a rut strongly denotes an absence of movement, she emphasized: "I feel freed, I feel free now" ( interview, Colombia, 12 March 2019). In Uganda, the existence of stigma-as an example of inter-personal friction -had resulted in sensitization efforts by some NGOs, which had contributed to "lubricating" interviewees' social movement within their communities. As one interviewee explained: "At X [name of her village], life is changing among people because some NGOs kept going there to sensitize people. Yes, so they kept sensitizing people that the problem that took place at X should not be held in the heart" ( interview, Uganda, 13 June 2019). Haider et al., (2018, p. 319) underline that "…a diversity of ecological and social interactions is critical for the resilience of people, communities, or regions to shock and ongoing socio-environmental changes as they provide sources for adaptation and transformation." Adaptation and transformation are two crucial concepts within resilience scholarship (Walker et al., 2004). Using the example of responses to floods to explicate the distinction between the two, Fedele et al. note that people can adapt to the floods by, inter alia, borrowing money to repair their homes, replanting damaged crops, or building higher dam walls. Alternatively, "people can also respond to floods by transforming their social-ecological system, for example by relocating houses or crop fields to safer areas or restoring previously degraded wetlands upstream" (Fedele et al., 2019, p. 116). What stood out from the interview data was that interviewees' interactions with their social ecologies-including with fellow victims-/survivors-not only fostered adaptation to challenging circumstances. It was also through these interactions that some interviewees sought to bring about broader transformative change-thereby effectively creating "a new stability land-scape…from which to evolve a new way of living" (Walker et al., 2004)-and by extension new ways of moving.
Overall, the Colombian interviews were most striking in this regard, reflecting the fact that some of the interviewees-as a result of both their personal experiences and the support and trainings that they had received from women's associations and NGOs-were themselves social leaders and had their own associations. Giving her life story the title "Warrior Woman" (Mujer Guerrera), for example, one interviewee spoke about her efforts to change how women in her community are treated. In her words, …concerning problems that come up in the municipality, a big one is the mistreatment of women.
For me that's [she makes a fist with her right hand and punches her left palm]…that gets me going… so, I have a lot to do with officials…I go to various institutions. It's always the same…They try to mask the way things are, but I have the law on my side. I use the law and go knocking on doors-the direct approach. That's a resource I have and I have it because of the same organization [referring to a local women's organization with which she is closely involved]. Without that organization, I wouldn't know what to do.
( interview, Colombia, 10 February 2019) Another interviewee, who stressed her passion for the work that she does, implicitly spoke about transformative change in the sense of encouraging other women to speak out about their experiences (as a way of tackling impunity) and fighting to make sure that they are treated correctly when they do so. Giving some of the background to her work, she explained: I went into so many offices where they asked me questions that made me feel even more dirty than I already did, and there are still women who have that experience when I take them to make their statements. But I have… how can I explain this? I feel that I have the power to get them the respect they need because I lived through it and I don't want them to have the same experiences I had and so I demand they are respected.

( interview, Colombia, 29 March 2019)
These examples further illustrate the aforementioned concept of social-ecological traps. One of the five previously noted possible responses to traps, as identified by Boonstra et al., is "innovation," meaning that "Actors have a desire to change SE traps and have the ability to do so" (Boonstra et al., 2016, p. 880). What some of the interviewees demonstrated was a desire to change the wider systems in which they were living, and the particular trap dynamics within these systems. In so doing, they were utilizing whatever opportunities they had-and creating their own opportunities-aimed at broader systemic transformation.

| CON CLUS I ON: RE S ILIEN CE A S A MULTI -D IREC TI ONAL MOVEMENT PRO CE SS
In an article about herding in Mongolia, Xie and Li explore the impact of a Livestock and Rangeland Double-Contract Responsibility System (LDRCRS), designed by the Chinese during the mid-1980s with the aim of stopping the herders from moving freely. This imposed policy presents an example of a contracted environment; the herders were expected to remain within prescribed boundaries. This not only disrupted their ecological connectivity with the land, but it also negatively impacted on their economic livelihoods (Xie & Li, 2008, p. 36). Many of the herders nevertheless continued to move in a very literal sense, in order to access better quality rangeland. As the authors note, "This is a traditional herding strategy called otor in Mongolian…that represents herders' need for flexible physical and social boundaries" (Xie & Li, 2008, p. 35). In other words, otor involved the herders moving in multiple directions-across both short and long distances-beyond their allocated boundaries. This Mongolian example points to an important linkage between stressors/adversity and movement that was also seen in the interview data from BiH, Colombia, and Uganda. The issues that the herders faced in practicing otor, moreover, including increased livestock production costs (Xie & Li, 2008, p. 49), powerfully convey some of the immense challenges that movement can entail-challenges that commonly-used terms within extant resilience scholarship, such as "bouncing back" and "bouncing forward," often overlook (Clark, 2020). The fact that, ultimately, the Mongolian herders had to adapt their herding strategies around the new restrictions imposed by the LRDCRS illuminates a further important dimension of resilience, namely "learning how to change in order not to be changed" (Walker, 2020). In other words, the herders needed to know how to move and how much to move without changing their entire way of life.
This article began by exploring movement and resilience in the context of ecological systems. Thinking about the different ways that systems themselves move in response to shocks and stressors, however, is of limited application at the individual level. Conversely, descriptions of individuals "bouncing back" from adversity-or "bouncing forward" with their lives-disconnect everyday movement from the broader systems that shape it. This article, in contrast, has brought the individual and the systemic together within a comparative SES framework. Theorizing resilience as a multi-directional movement process, it has drawn on interviews with victims-/survivors of conflictrelated sexual violence in BiH, Colombia, and Uganda to analyze some of the different ways that these men and women were "moving" as they sought to get on with their lives and deal with everyday challenges. Yet, in so doing, it has also explored how interviewees' wider systemic environments were influencing and affecting their movements.
In addition to presenting an original way of thinking about resilience, this research also brings new insights to the question "resilience of what to what at what scales" (Walker & Salt, 2006, p. 252). This question-which primarily speaks to the aforementioned concept of specified resilience-also conveys ideas of movement. In particular, the "to what" part of it reflects the fact that even specific adversities and disturbances are shaped by wider systems that are in constant movement and flux. In this sense, the question itself highlights an important individual-systemic movement dialectic, which complexifies the relationship between specified and general resilience.

CO N FLI C T O F I NTE R E S T
No conflict of interest is declared.

E TH I C S S TATEM ENT
Full ethics approval for the underpinning research was granted by the European Research Council, the Humanities and Social Sciences Ethics Committee at the University of Birmingham and by relevant authorities in Bosnia-Herzegovina, Colombia, and Uganda.

DATA AVA I L A B I L I T Y S TAT E M E N T
Research data are not shared.