Empathy in frontline humanitarian negotiations: a relational approach to engagement

Humanitarian access—people’s ability to reach aid and aid’s ability to reach people—is widely understood to be a central challenge in humanitarian action. One of the most important ways in which humanitarian access is practically secured in conflict settings is through frontline humanitarian negotiations. In this type of negotiation, humanitarians engage in face-to-face interactions with conflict parties to secure safe access to, and protection of, civilian populations in situations of armed conflict. An underdeveloped aspect of such negotiations that is ripe for further exploration is the role of empathy. The purpose of this article is thus to draw on the insights of the empathy literature to explore how empathy shapes humanitarian protection work in the specific domain of frontline humanitarian negotiations. Part one conceptualizes empathy, drawing on the interdisciplinary field of scientific research. Part two introduces the practice of frontline humanitarian negotiation and explains why empathy is critical, particularly in the increasingly fragmented environments that negotiators must operate. Adopting a relational approach, Part three advances a framework for analyzing empathy in frontline humanitarian negotiations. We theorize empathy’s salience across four different axes of negotiation, drawing insights gleaned from scholarship and a systematic review of the grey literature on humanitarian negotiation, including field manuals, training materials, and operational guidance. We do not ultimately argue for ‘more empathy’ in this type of work, but rather a more thoughtful approach to empathy— one that entails the cultivation of core empathy-related skill areas, including: emotion regulation, perspective-taking, social awareness, and strategic conveyance of empathy. We contend that this approach could help to alleviate numerous problems in the humanitarian sector, including aid worker burnout.

"Humanitarians have to abandon their shell of regulations and get under the skin of those who are in front of them. " 1 Humanitarian access-people's ability to reach aid and aid's ability to reach people-is widely understood to be 'the central and overriding problem in humanitarian action' (USAID, 2020). A recent study found that crisis-affected populations in more than 60 countries are not receiving the humanitarian aid they need because of access constraints (ACAPS, 2021). One of the most important ways in which humanitarian access is practically secured in conflict settings is through frontline humanitarian negotiations. 2 In these negotiations, humanitarians engage in face-to-face interactions with various actors who might be able to support their work and help them to reach people in need; this can include members of the affected population, donors, fellow humanitarian organizations, civilians, and so on. Given the fluidity and changing dynamics of conflict contexts, the parameter for these interactions, which are often ongoing, is far from fixed; 'everything is open to negotiation' , as one senior MSF staff member observes (quoted in Allié, 2011, p.5). In this study we focus narrowly on frontline humanitarian negotiations with conflict parties-including state parties and non-state armed groups-which seek to secure safe access to, and protection of, civilian populations in situations of armed conflict.
It is a matter of urgency that scholars, policy-makers and practitioners scrutinize-and devise ways to improve-negotiation processes and outcomes for those in need of humanitarian assistance. Given the highstakes interpersonal nature of these interactions, an aspect of negotiations that is ripe for further exploration is the role of empathy. Empathy's relevance to humanitarian action extends far beyond frontline negotiations, implicating protection work in areas such as: interactions with peacekeepers and members of international military missions (Sutton, 2021); high-level humanitarian advocacy on humanitarian access and civilian protection; the elicitation of empathy, sympathy, or solidarity from outsiders (e.g., donors, the media, and public); and service delivery with, and programming on behalf of, project beneficiaries. While our analysis is scoped to focus narrowly on frontline negotiations, its findings generate important insights for multiple domains of humanitarian protection work.
Although the field of humanitarianism has yet to fully grapple with empathy's role, scholarship on empathy in other disciplines has thrived over the last decade. From neuroscience to psychology and behavioural economics, scholars have analysed how empathy impacts individual feelings and motivations, and how levels of empathy relate to pro-social or anti-social behaviour (McGinley, et al., 2014;Klimecki et al., 2016). Scholars have also shown how parts of the brain involved in empathy, and our capacity to use empathy in reasoning and decision-making, can be impaired or enhanced. Research in a range of professional fields from medicine to education, law, journalism, and social work has illuminated the importance of empathy in shaping workplace outcomes, particularly in so-called "helping" or "caring" professions (E.g., Moudatsou, et al., 2020;Westaby and Jones, 2018). Fields adjacent to humanitarianism, such as conflict resolution and diplomacy, have also established that empathy is critical to face-to-face negotiation practices. Empathy is a precursor to trust and it enables individuals to infer the intentions, motivations, and interests of their counterparts (Holmes and Yarhi-Milo, 2017;Mansbridge and Martin, 2013;Galinsky, Maddux, Gilin, and White, 2008). While myriad accounts in the practitioner literature treat trust as a key building block of humanitarian negotiations-essential for fostering positive relationships and fruitful negotiation encounters-the link between empathy and trust in the humanitarian negotiations context is not yet well fleshed out. We adopt an approach here that views trust and empathy as generally working in tandem to enhance negotiation outcomes, but a caveat is needed. As we explain later in the article, there may be instances where empathy fosters distrust as it raises doubts about another's intentions. Further, it may be possible to secure a good negotiation outcome in the absence of trust or indeed in the presence of deepening suspicions.
It merits mention that empathy's relevance is not limited to interactions with non-humanitarian actors. Empathy matters internally for the sector as well, as it is inextricably linked to the functioning and well-being of humanitarian workers themselves. These actors deliver life-saving services in situations shaped by violence and conflict, often in direct contact or close proximity to those who are suffering. That many humanitarian staff are nationals and may have complex relationships and histories in the areas in which they work, including in some instances as former combatants, makes the dynamics surrounding empathy all the more 1 West African humanitarian respondent, cited in Deborah Mancini-Griffoli and Andre Picot, 'Humanitarian negotiation: a handbook for securing access, assistance and protection for civilians in armed conflict' (Humanitarian Dialogue, 2004), p. 62. 2 In examining frontline negotiations, the present article focuses on fieldlevel interactions that take place between humanitarian actors and conflict parties. It might be expected that studies focusing on different levels of, or forums for, negotiation-such as virtual interactions or higher-level planning and strategic decision-making-would produce different findings about empathy's role and relevance. Differences could also be anticipated were the analysis to focus on negotiations with the other types of actors listed, such as members of the affected population and donors. For a discussion of the different levels of humanitarian negotiations, see Deborah Mancini-Griffoli and Andre Picot, 'Humanitarian negotiation: a handbook for securing access, assistance and protection for civilians in armed conflict' (Humanitarian Dialogue, 2004), p. 22. complex (James 2022;Lombard 2018). This remit raises questions about whether better humanitarian outcomes will be secured if humanitarians feel more empathy-for the populations they serve and for other interlocutors such as armed groups. While empathy can indeed be critical to effective engagement, the literature also cautions that empathy creates vulnerabilities to stress, emotional exhaustion, and burnout (Decety, 2020;Wagner, et al., 2019;Ferri et al., 2015;Rothschild, 2006;Samra, 2018). These types of vulnerabilities have long plagued the humanitarian sector, a dynamic which is increasingly being recognized and given due attention (Young, et. al., 2021;Cockcroft-McKay and Eiroa-Orosa, 2021). Experts warn of a 'mental health crisis' amongst humanitarians, with some studies indicating that 70 to 89% of humanitarian aid workers have experienced mental health issues related to their job (Jachens, 2019;Cardozo et al., 2012;Welton-Mitchell, 2013). Mental health issues in the humanitarian sector have been linked to increased absenteeism, low commitment, high turnover, and poor decision-making (Antares Foundation, 2012;Professionals in Humanitarian Assistance and Protection, 2015;Loquercio, Hammersley, and Emmens, 2006). These problems inevitably implicate those receiving humanitarian services, a point of concern given the access constraints already mentioned. Perversely, the risk strategies that humanitarian agencies deploy to ensure their safety in challenging contexts-such as physical distancing and remote management-reduce opportunities for face-to-face interaction, exacerbate humanitarian burnout, and may hinder empathy (Andersson and Weigand, 2015). The broader drive towards professionalization of the sector, with its emphasis on results-based management, may also pit efficiency against altruism (Carbonnier, 2015).
The above observations establish that to neglect the role of empathy is to ignore crucial dimensions of humanitarian practice. The purpose of this article is to draw on the insights of the empathy literature to explore how empathy shapes humanitarian protection work in the specific domain of frontline humanitarian negotiations. The article proceeds in three parts. Part one conceptualizes empathy, drawing on the interdisciplinary field of scientific research. Empathy, as numerous studies suggest, is the ability to adopt another person's perspective and understand their cognitive and affective states. Like emotion, it is at once individual, rooted in human biology and experienced within one's body, and something that 'emerges from, and exists within, social relations' (Holmes and Yarhi-Milo, 2017, p. 110). Part two introduces the practice of frontline humanitarian negotiation and explains why empathy is critical, particularly in the increasingly fragmented environments that negotiators must operate. Adopting a relational approach, in Part three we advance a framework for analyzing empathy in frontline humanitarian negotiations. We theorize empathy's salience across four different axes of negotiation, drawing insights gleaned from scholarship and a systematic review of the grey literature on humanitarian negotiation, including field manuals, training materials, and operational guidance. 3 By bringing the scientific literature to bear on this field of practice, our analysis reveals that empathy, while essential, is not an unalloyed good. As such, we do not argue for more empathy, but rather a more thoughtful approach to empathy-one that entails the cultivation of core empathy-related skill areas, including: emotion regulation, perspective-taking, social awareness, and strategic conveyance of empathy. The conclusion discusses the implications for humanitarianism generally and for the training and capacity development of frontline humanitarian negotiators.

What is empathy?
The term 'empathy, first coined at the turn of the twentieth century, is a translation of the German word Einfühlung to 'feel into' . While empathy has been the subject of extensive research over the last century, consensus on a precise definition of empathy remains elusive (Engelen and Rottger-Rossler, 2012). Nonetheless, it is generally accepted that empathy encompasses both affective and cognitive elements that serve separate functions but that work in concert (Decety and Jackson, 2004;Shamay-Tsoory, 2011).
The first element, affective empathy, refers to the sensations and feelings one has in response to others' emotions (i.e., "I feel what you feel"). This capacity, developed early on in life, has a neural basis. The presence of a mirror neuron system in the human brain helps people feel what another feels simply through observation. When an individual observes the actions, pain, or affect of another, the motor mirror systems in the brain function in such a manner that the very same neural networks are triggered as would be triggered if one had experienced the feelings in question firsthand (Armstrong, 2017).
The second element, cognitive empathy, involves some form of role or perspective-taking in which the individual recognizes and understands another individual's mental state (i.e., "I understand what you feel"). The degree of cognition involved in this second element of empathy can vary, as psychologist Davis explains, from classic conditioning-'where the observer has previously perceived 3 Our review of the grey literature involved a systematic and forensic analysis of existing works. Based on search criteria that emphasized recency, thematic relevance, geographical diversity, and number of citations, over 15 manuals and guidance documents on humanitarian negotiation and humanitarian action were reviewed and coded for references to law, negotiation strategy, and emotions. The HD Handbook was included despite its somewhat dated 2004 publication, given its engagement with empathy. affective cues in others while experiencing the same affect'-to more sophisticated language-mediated association, in which the observer's reaction to the observed is produced by 'activating language-based cognition networks that triggers associations with the observer's own feelings or experiences' (Davis, 2015, p. 284). The cognitive outcomes produced by these processes include 'interpersonal accuracy' , the effective appraisal of another individual's feelings, motivations, and interests (e.g., Dymond, 1950;Ickes, 1993). Further, perspective-taking has been tied to changes in the cognitive representations that observers form of the observed (Davis et al., 1996;Galinsky and Moskowitz, 2000).
While humans are, in a sense, hardwired for empathy, it is neither a given nor a fixed dispositional trait or capacity (Kirman and Teschl, 2010). Empathy is relational and inextricably linked to social context. The development of empathic capacity and expression of empathy are shaped by a range of factors including genetics, personality, experience, and pre-existing social relationships. There are variations in how empathy is experienced and for whom, as well as the ways in which it impacts individual behavior. Studies in neuroscience, for example, suggests that neural mirroring will be more activated when one is observing an individual one identifies with or with whom one is closely connected. Further, studies show that perception of closeness is tied to empathic concern and serves as a predictor of helping behavior (Decety and Lamm, 2009).
Focusing on the social dimensions of empathy, scholars underscore the importance of expressions of empathy and conveyance of empathic capacity, not just actual possession of it (Pentland, 2008;Zhou, et al., 2003). This is particularly important in contexts where empathy serves as a prerequisite for trust, defined as the '[belief] that the other(s) now and in the future, can be relied upon to desist from acting in ways that will be injurious to their interests and values' (Booth and Wheeler, 2008, p.230; see also Wheeler, 2018). For example, in their work on international peace summits, Holmes and Yarhi-Milo illuminate the ways in which 'individuals in negotiations act on beliefs about empathic capacity' (2017, p. 110). They note the range of behaviours and dispositions through which empathic capacity is conveyed and inferred including 'facial cues, emotions, interpersonal mimicry, gestures and language' (p. 110). Further, Holmes and Yarhi-Milo's research reveals that in circumstances where expressive empathy fails between two individuals, where they cannot convey empathy to each other, third party actors such as mediators can build relational empathy between such individuals.
While empathy is often portrayed as a normatively positive ability, it does not necessarily lead to more trusting, "helping" or prosocial behaviour. For example, understanding another's intentions may result in the realization that the other is not trustworthy, as Holmes and Yarhi-Milo explain (p.109). Further, empathy, sympathy, and compassion are not synonymous. Empathy may give rise to feelings of sympathy and compassion, but the three are conceptually distinct (Decety, 2011;Burton, 2015). As Keen (2010, p.5) describes-focusing here on the affective dimension rather than the cognitive-empathy is akin to feeling with (i.e., "I feel your pain") whereas sympathy is feeling for (i.e., "I feel pity for your pain"). Compassion, by contrast, is an extension of sympathetic consciousness but is coupled with a desire to help or alleviate another's distress.
As such, in contrast to both sympathy and compassion, empathy can be conceived as ethically or value neutral. Nussbaum explains that a 'good sadist or torturer has to be highly empathetic, to understand what would cause his or her victim maximal pain' (2006, p.320). What is more, research indicates that too much empathy, without the ability to self-regulate, can be harmful. Individuals may feel overwhelmed and distraught, and distance themselves from the individual they "feel with" (Decety, 2011). Moreover, the inability to distinguish one's own feelings from another's can result in what psychologists refer to as 'emotional contagion' , the loss of self and agency (Todorov, et. al., 2011).
Two of the leading criticisms of empathy come from Prinz, a philosopher, and Bloom, a psychologist. A field-specific critique of humanitarian empathy has also been advanced by Breithaupt (2015); it will be discussed further below. Both Prinz and Bloom challenge the assertion that empathy has an overall beneficial influence on human behavior. Prinz argues against empathy, noting that it may 'lead to preferential treatment' , 'may be subject to unfortunate biases, including the cuteness effects' , 'is prone to in-group biases' , and 'is subject to proximity effects ' (2011, pp.227-30.). For his part, Bloom problematizes empathy's 'spotlight' and 'innumeracy' features-namely, the tendency to focus one's attention on certain people in the here and now, and the difficulties of empathizing with more than one or two people at the same time (Bloom, 2016). As a result, Bloom argues that empathy can mislead, especially in settings where there is a need to account for statistically relevant information when tackling a moral or social problem.
That Prinz and Bloom are explicit in conceptualizing empathy as an affective phenomenon is significant. Studies indicate that simply sharing another person's emotion-feeling what they feel-does not increase motivation for pro-social action. The same, however, is not true for the aforementioned cognitive dimensions of empathy. Evidence suggests that perspective-taking and empathic concern reduce implicit biases, increase cooperation and charitable giving, and enhance sensitivity to injustices done to others (Galinsky and Moskowitz, 2000;Todd, Bodenhausen, Richeson, and Galinsky, 2011).
In sum, rather than focus on either affective or cognitive empathy, scientific research on empathy underscores the necessity of an integrative model, reflective of the ways in which both elements are entwined (Decety, 2011). Our ability to feel what others feel, as Pfeifer and Dapretto assert, figures as a 'precursor to more explicit processes of reasoning through what others feel ' (2009, p. 185). The science of empathy also attests to the importance of understanding empathy as a social phenomenon, both in terms of how it is expressed as well as its relational dimensions, including the potential for building empathy between individuals and groups.

Conceptualizing frontline humanitarian negotiation
Despite a tendency to separate emotion from cognition, the two are linked with important consequences for decision-making related to humanitarian action. In conflict settings shaped by violence, human suffering and the provision of succor animate humanitarian work and create a situation in which the stakes are high. Many would argue that it is precisely because humanitarians are able to feel the suffering of others that they act. And yet, surprisingly, empathy as it relates humanitarian professionals or agencies is mostly overlooked in the policy literature as well as scholarship on humanitarianism. 4 In the two sections that follow, we analyze empathy in frontline humanitarian negotiations. Humanitarian negotiation is not a new field of endeavour. Humanitarian organizations have long had to negotiate access to civilians by brokering agreements with conflict parties (Clements, 2020;Magone et al., 2011;McHugh, 2006;Pease, 2016). Today, most access negotiations are conducted bilaterally, and at subnational levels, 5 with humanitarian organizations preferring to operate independentlyeither in direct dialogue with parties or through their own intermediaries (Harmer and Stoddard, 2018, p. 11). Face-to-face negotiations rely heavily on the interpersonal skills and expertise of individual negotiators at the frontlines. Different humanitarian organizations have developed training and operational guidance in this area, which we draw heavily on in our analysis. However, as scholars have argued, existing guidance and training modalities pay insufficient attention to skills relating to human emotions.
International humanitarian law (IHL), complemented by the traditional humanitarian principles of humanity, independence, impartiality (and sometimes neutrality), provides the legal basis for humanitarian access and negotiations with conflict parties. For decades humanitarians have advocated for recognition of a right of access to vulnerable populations based on this legal framework. Securing such access has become more complex in recent years, as humanitarians have had to adjust to more multilayered and fragmented operational environments; the proliferation of armed groups poses a particular challenge here (Wise et al., 2021) as well as discordant relations between states and humanitarians (Cunningham 2018). A seminal volume on humanitarian negotiation, encompassing case studies from a wide-range of contexts, found that in most instances humanitarian access results from the shifting and shared interests of different actors and the successful building of relationships over time (Magone, et al., 2011; see also Acuto, 2014). In other words, securing access is a process of continuous negotiation where understanding your interlocuter's interests is essential. Similarly, Harmer and Stoddard argue that the organizations that have adapted well and made significant advances to securing access show a preference for 'pragmatic and contextually driven engagement with parties to the conflict; one which focuses on interests and incentives rather than legal requirements, and which reframes IHL and humanitarian principles in the context of local values and norms' (2018, p. 1; see also Cropp 2021). The Centre of Competence on Humanitarian Negotiation (CCHN) manual conceptualizes this change as a shift in mindset from one of "humanitarian entitlement" to one of "humanitarian engagement"; on this approach, one does not claim 'we have the right to be here' , but instead focuses on the degree of predictability, legitimacy, and trust one can derive from the relationships with the parties in particular circumstances (Centre of Competence on Humanitarian Negotiation, 2019, p. 92). While both scholars and practitioners increasingly recognize the importance of building and maintaining relationships, what is missing from existing studies is the role played by empathy. This is surprising given the centrality of the concept in other domains of negotiation.
Before delving into the relational aspects of frontline negotiations, it is important to consider what makes a given negotiation successful from the perspective of humanitarian negotiators. Those interested in legal compliance might opt for a legalized approach focusing on negotiation outcomes that promote adherence to IHL. Here one could consider a metric akin to the indicators that have been developed in the fields of human rights and everyday peace (Satterthwaite, 2012;Mac Ginty, 2014). While IHL is addressed in the analysis that follows, this article pushes beyond a legal compliance paradigm. To be sure, in some instances there may be a clear 'win' for the humanitarian side. Armed groups might agree to open a port to allow aid to reach civilians, for example, or sign a formal 'Deed of Commitment' with Geneva Call which commits them to refrain from recruiting child soldiers (Geneva Call, 2021;Mercy Corps, 2018, pp. 23-24). However, in many contemporary frontline negotiation contexts, goals will be less concrete, immediate, or tangible. As part of an "engagement" approach, humanitarians may simply wish to build networks in an area or make the new representative of their organization known to the parties to the conflict. Moreover, if it can be agreed that the general aim of humanitarian negotiations is to alleviate the suffering of war-affected populations, differentiating an outcome that alleviates suffering from one that has no such impact-or even exacerbates itwill not always be straightforward (Grace, 2020b, p. 37). Some encounters that have clear, if modest, humanitarian 'wins' might also not have been planned as negotiations in the formal sense (Grace, 2020b, p. 17). The scenario of a humanitarian vehicle being permitted to pass through an impromptu checkpoint manned by an armed group is one such example.
In light of the above, we propose that a given humanitarian negotiation be evaluated in accordance with its relational aspects and, more specifically, in terms of empathy. In this frame, a successful negotiation is one in which humanitarian actors are attuned to empathy's role across each of the axes (see below), navigating various empathy points to build a resilient and trusting relationship with their counterparts. While the emphasis here is on process rather than result, this relational engagement drives towards, and is motivated by, a clear ultimate aim: that of securing access to and protection for populations in need. The next section fleshes out this relational approach to frontline humanitarian negotiation.

Empathy along four axes: a relational approach to frontline humanitarian negotiation
The following analysis considers empathy in humanitarian negotiations along four main axes, centering the perspective of the humanitarian practitioner on each axis. First, there is the empathy of the humanitarian actor for their interlocutor-referring here to conflict parties including both state and non-state actors. 6 Second, there is the empathy of the humanitarian actor for the civilian population. Third, there is the interlocutor's empathy (or lack thereof ) for this same civilian population, which the humanitarian actor can appeal to and seek to strengthen. Fourth, there is the interlocutor's empathy for the humanitarian actor, which the humanitarian may seek to cultivate. 7 As we consider each axis and its interaction with the other axes, we attend to empathy's role and articulate some possible implications for humanitarian work, including the relevance of international legal rules and norms in the negotiation context. While the analysis is not centered on legal questions, we include this dimension because the humanitarian negotiations literature routinely juxtaposes an 'objective' (legal) dimension with the more 'subjective' or emotional components.

Axis 1: The humanitarian actor's empathy for their interlocutor
There are numerous reasons why it is desirable for a humanitarian actor to feel empathy for the conflict parties with whom they engage to protect and assist civilian populations. One of Grace's humanitarian respondents describes this form of empathy thus: 'That means you are able to put your feet in the shoes of the other party. If you cannot understand the rationale behind your interlocutor's behavior, it will be very difficult' (Grace, 2020b, p. 27). A 2004 handbook by Humanitarian Dialogue instructs that, to truly understand an interlocutor's point of view, humanitarians must engage in 'a real act of empathy that tries to imagine their ideology, their experiences, their objectives and their feelings' (Mancini- Griffoli and Picot, 2004, p. 122). In this regard, empathy can help humanitarians anticipate the expectations and reactions of their interlocutor (Mancini-Griffoli and Picot, 2004, p. 122;Lepora and Goodin, 2013). The Handbook presents empathy as a tool that can bolster humanitarians' ability to discern an interlocutor's self-perceived needs and interests, building trust and a predictable relationship. Interests are presented as the most important thing to identify, since they are the ultimate motive for negotiating and thus inform the decisions that negotiators make (Mancini-Griffoli and Picot, 2004, p. 62).
Challenges arise because the interlocutor may have a multiplicity of (potentially conflicting) interests, some of which are shaped by organisational and social group belonging (Mancini-Griffoli and Picot, 2004, p. 62). Conflict actors might not even know their own interests, and they could purposely obscure their true interests from humanitarians. Propensity for obfuscation is potentially heightened when a conflict actor deduces that a given interest will be unpalatable to humanitarians (Ibid., p. 62). This suggests that humanitarians may run into problems where they adopt a moral stance on a given issue-and we would inquire whether this applies to situations where humanitarians suggest or imply that international law has been violated-leaving humanitarians with a lack of information as conflict parties withhold. Humanitarians may also confront obstacles where they make interlocutors feel unappreciated or treat them as adversaries. Fisher and Shapiro denote 'appreciation' and 'affiliation' as two core concerns of all negotiators that must be addressed (along with autonomy, status, and role) (2005, p.19). Parties who do not feel appreciated, they warn, will become angry and impatient; this could lead to negative reactions (Ibid). Interlocutors who are treated as adversaries, furthermore, may feel indignant, disgusted and resentful; this makes them prone to 'go it alone' (Ibid).
A point that has already been raised in this analysis is that there are also limits to empathy, and this is especially important to consider with respect to engagement with conflict parties. Here, humanitarians must exercise sound judgment about appropriate levels of proximity and intimacy. If one over-identifies with the worldview of an interlocuter, one might acquiesce with problematic behavior and be coopted or made complicit (Mancini-Griffoli and Picot, 2004, p. 122). In this respect, the HD Handbook differentiates empathy from sympathy. 'Heartfelt sympathy' is presented not only as unnecessary, but also as something which might lead to a sort of humanitarian Stockholm Syndrome in which humanitarians relinquish their own values (Mancini-Griffoli and Picot, 2004, p. 122). Instead, the Handbook states that in many situations the form of empathy that is needed is cognitive, one that permits an 'objective intellectual understanding of a counterpart's reasoning' (Ibid., p. 122). This linking of empathy to a more rational form of assessment points to a possible circularity in this guidance. While the Handbook suggests that 'real empathy' allows humanitarians to discern the interests and needs of their interlocutor, its practical instructions on how to be empathetic circle back to an exercise of identifying their interests. In any event, the mischief being prevented here is the problem of self-loss, a scenario in which humanitarians over-identify with a conflict actor. The HD Handbook deems self-loss to arise when empathy edges towards sympathy (Ibid., p. 122). However, as shown in the examination of the science of empathy in Part one, while empathy may engender feelings of sympathy, the two are distinct. As we discuss below, the task of the humanitarian is thus to be mindful and aware of this distinction, and to develop their emotion regulation skills.

Implications for humanitarian work
Rules, norms and laws may be perceived by humanitarians as interfering with empathy. Too much empathy for an interlocutor, particularly in terms of affective empathy, may equally lead humanitarian actors to abandon the humanitarian principles and to look the other way when confronted with IHL violations. One might imagine a scenario where humanitarians are convinced of the righteousness of the struggle of a particular armed group (against an oppressive or authoritarian state that is committing far worse violations, for example) and this leads humanitarians to take armed group assertions at face value or to not inquire into potential IHL violations or criminal activity by the armed group in question. The Humanitarian Dialogue Handbook flags this problem indirectly when it proposes that unbounded empathy could lead to humanitarian complicity or the condoning of problematic behaviors by others (Mancini-Griffoli and Picot, 2004, p. 122). Creative approaches that adapt IHL rules to connect with the cultural beliefs and norms of particular conflict parties may allow empathy and IHL to interact harmoniously, though it should be noted that emerging research on such efforts shows that this can be quite complex work (Grace 2020a;Cismas and Heffes 2020). Amongst other challenges, grounding IHL rules in local cultural norms demands sustained engagement and it requires sophisticated interlocutors or translators who can bridge different worlds.
In order to engage effectively with armed interlocutors and advocate persuasively on behalf of civilian populations, individual humanitarian actors need to strengthen the cognitive elements of empathy, including their perspective-taking faculties. This refers to the process in which an individual views a situation from another's point of view. Doing so can help humanitarians to mitigate biases, overcome prejudices, and center the agency of affected populations. Understanding the emotional state, interests, intentions, and motivations of armed actors with whom humanitarians engage can also build and sustain trust as well as improve communication over time. This has its limits, however, and humanitarians must discern how and when perspective-taking is useful. Further, as discussed in Part one, studies indicate that the ability to express and convey empathic capacity is critical in relationship building. These points will be elaborated upon in Axis 3, which again implicates perspective-taking and empathy conveyance.

Axis 2: The humanitarian actor's empathy for the civilian population
What are the feelings that humanitarians have, or should have, towards populations they assist? How and why is it important for them to understand the affective and cognitive states of populations in need? This would seem simple on the surface, as it is the desire to alleviate human suffering that lies at the heart of the humanitarian endeavor. However, even if all the 'right' feelings are in place and humanitarians are galvanized to act, it may not be altogether clear in a given situation which course of action would meaningfully alleviate human suffering. By accepting restrictions on their movement and limiting the delivery of aid to certain areas so that they can reach civilians in other areas, humanitarians could unwittingly end up harming those who are not helped (Grace, 2020b, p. 37). While impartial humanitarianism is premised upon selecting the 'most vulnerable' individual or groups, it may be that those selected are not perceived to be particularly vulnerable by others in the local setting (Grace, 2020a, p. 87). Even where vulnerable populations are correctly identified, delivering services to them could still be problematic where it disturbs the local power balance (Grace, 2020a, p. 87). A further issue that might arise is whether humanitarian empathy for the affected population influences how humanitarians navigate limitations in their operating environments. That is, humanitarian decision-making about whether to remain in a constrained environment, or to instead cease operations when conditions become too difficult, could be influenced by empathy for certain segments of the affected population. In such situations, empathy for those who appear especially vulnerable might cloud an analysis of actual needs-and of who could most productively be helped or best served by a continued humanitarian presence (and see discussion of impartiality, below).
Building on the above, there are downsides to empathy along this axis that humanitarian actors need to be aware of before they find themselves in settings in which harm to civilians is particularly grievous or shocking. Similar to both Bloom and Prinz's more general critiques of empathy, (Breithaupt 2015, p. 6) identifies several problems with humanitarian empathy for populations in need. He finds that humans tend to be more empathetic towards those with acute problems, rather than chronic ones, and that it is easier to empathize with someone who is undergoing change instead of stagnating. While the former issue may not be so problematic for humanitarian protection work in emergency settings, the possibility that humanitarians would become frustrated with those they are helping when change fails to materialize-and humanitarians thus find themselves trapped in the world of those they help-merits attention (2015, p. 6). Adopting the motif of a 'scene of empathic engagement' , (Breithaupt 2015, p. 8) proposes that in practice humanitarians are not only empathizing with those they are helping. Instead, humanitarians insert themselves (or an imaginary helper) into the equation, introducing an agent whose presence denotes the potential for things to change or improve. Three main risks arise here: self-loss of the humanitarian, associated with Stockholm Syndrome or over-identification and an inability to take a stand; taking sides, which deepens divisions between the person being helped and others and leads to 'victimhood contests'; and the derivation of 'sadistic empathy' , or vicarious pleasure from another's misfortune, which confirms the need for the helper (Breithaupt, 2015, pp. 10-13). This leads Breithaupt to argue for a humanitarianism without empathy, which instead finds its motivation to act in an inclusive concept of 'we' that does not center the empathy of the empathizer (p. 14). As noted, the concept of self-loss also arises in connection with Axis 1, where it is accompanied by a warning that overly-empathic humanitarians risk losing themselves in a conflict party's worldview.
Others champion sympathy as more pro-social than empathy. Responding directly to Breithaupt's argument, Adloff counters that when 'empathy finds expression as sympathy, it can trigger action' , and that 'compassion' is stirred by those who are innocent or especially vulnerable (2015, pp. 17-20). These differing perspectives attest to a lack of conceptual clarity and competing views on empathy's desirability, showcasing the need for more attention to be paid to how these dynamics play out in humanitarian protection work and specifically in the context of frontline humanitarian negotiation.

Implications for humanitarian work
An examination of the humanitarian actor's feelings for the populations they seek to assist highlights the importance of cultivating a particular skillset relating to handling human emotions: emotional regulation. Humanitarians and frontline negotiators will experience empathy by virtue of being human and the nature of humanitarian protection work, which entails proximity to suffering. Further, in many instances humanitarian staff are local actors, working in their own communities. As such, they may feel an even greater connection to (some segments of ) affected populations. Humanitarians thus need to be competent in dealing with their own emotions and, in particular, with a surplus of emotion in situations where failure to do so could cloud their judgment and compromise their impartiality. Indeed, the IHL requirement that humanitarian relief be delivered impartially may present a challenge for humanitarian empathy. Impartiality requires humanitarians to set aside who they might have more feelings for, and to deliver aid purely based on need. We might recall here Prinz' argument against empathy, which highlights the risk of preferential treatment and biases that include 'cuteness effects ' (2011, pp.227-30). There is arguably some tension here with the notion of humanity, 8 which galvanizes humanitarians to act, and also with deeply held beliefs in the humanitarian community that the suffering of victims of war underpins the humanitarian presence in the conflict zone. Further, overly emotive appeals based on the distress of the civilian population may impede progress in negotiations with conflict actors. This is elaborated upon in the analysis of the next axis.

Axis 3: The interlocutor's empathy for the civilian population
The third axis concerns the empathy (or lack thereof ) of the armed actor for the civilian population under their control, which humanitarians can appeal to and seek to cultivate. Humanitarians may attempt to gain access to a rebel-controlled area, for example, by engaging the armed actor's humanity for the civilian population that is suffering from violence and lack of assistance. One problem that arises here is that some conflict actors such as autocratic governments may not be bothered by-indeed, they may actively wish for-the suffering of vulnerable populations (Grace, 2020b, p. 26). Not all, however, is lost. Studies have shown that in circumstances where an individual cannot empathize with another, third party actors can build relational empathy between such individuals by: signaling their own empathic capacity; making the parties feel seen and heard; and by leveraging insights from their own relationship with the party whose empathy they wish to appeal to. In this scenario, humanitarians might activate their own empathic capacity for the interlocutor (Axis 1), to assess and appeal to the latter's relevant interests. One could argue, for example, that a lack of health care in a given area could lead to an epidemic, which would in turn harm the conflict actor in charge of the area, and their troops (Mancini-Griffoli and Picot, 2004, p. 106). Even where armed actors are responsive to appeals relating to human suffering, they might still worry that humanitarians granted access will witness military preparations they are making (Mancini-Griffoli and Picot, 2004, p. 62). Related fears could include the concern that humanitarians support an organization who is opposed to them, or that the humanitarian's organization is linked to the state (Mercy Corps, 2018, pp. 21-22). Empathic engagement by the negotiator with the aim of building a trusting relationship with the interlocutor may serve to allay such fears.

Implications for humanitarian work
In certain circumstances, IHL may be brought into the conversation where emotional appeals fail because the empathy of conflict parties for the civilian population is limited or non-existent. Appealing to IHL and legal entitlements may work better where the legal rules align with the rational interests of the conflict party in question. Further, the humanitarian negotiator's empathy for their interlocuter (Axis 1) may be essential in ascertaining those interests. If IHL prohibits the conduct that the interlocutor wishes to engage in, however, appeals to law could undermine the relationship. The CCHN manual makes a useful observation about the potential limitations of humanitarian advocacy around a right of access to populations in need. Asserting that humanitarians have a right to operate in a particular setting could fall on deaf ears, the manual argues, and interfere with opportunities for fostering trust and relationship building with the actors in question (Centre of Competence on Humanitarian Negotiation, 2019, p. 90). As mentioned in Part two, the manual therefore calls for a shift from a humanitarian "entitlement" to "engagement" approach in certain contexts. This redirects attention away from the interlocutor's feelings for the civilian population, to feelings for the humanitarian actor (see also Axis 4).
Axis 3 again requires humanitarian actors to engage their perspective-taking faculties. As noted in Axis 1, part of the skill involved in perspective-taking is to know whether, when and how to use it. There may be situations where emotional appeals to a conflict party about civilian harm fail because the relevant conflict party has limited or non-existent empathy for the civilian population or where the interlocutor is empathetic but lacks authority and the ability to influence others within their organization, be that a state or non-state armed group. In such instances, advocating on the basis of rules and normsincluding standards drawn from IHL and public international law-could make more headway. Accordingly, this axis of relationship requires a delicate balance in terms of using empathy: the humanitarian actor must first ascertain the armed actor's feelings for the civilian population and their motivations in harming them or blocking humanitarian access (Axis 1). Where armed actors are trying to impede humanitarian access to vulnerable populations, part of the emotion work required may be to allay the interlocutor's fears, such as the concern that humanitarians are aligned with someone who has competing interests with them. The negotiator's ability to assuage such anxieties is closely tied to the final axis.

Axis 4: The interlocutor's empathy for the humanitarian actor
The fourth axis relates to the interlocutor's empathy for the humanitarian actor, their position, and the situation they are in. This must be actively cultivated as part of wider trust and relationship building efforts. The Mercy Corps Playbook advises that the team of humanitarian negotiators should be constructed with an awareness of how interlocutors will perceive their intersectional identities, including age, ethnicity, sex, religion, and ability (Mercy Corps, 2018, p. 10). The CCHN manual for frontline humanitarian negotiation recommends appealing to shared human experiences and making appropriate cultural references (Centre of Competence on Humanitarian Negotiation, 2019, p. 70). The manual identifies a common practice of referring to 'agreed facts of convergent norms' that apply in a particular setting, such as sports, food or music, which might have no direct relation to the negotiation topic at hand (Centre of Competence on Humanitarian Negotiation, 2019, p. 70). What is noteworthy about this axis of empathy is that it may generate progress where the second and third axes-which both concern empathy for the affected civilian population-stall. Instead of insisting on a 'right of access' that is grounded in humanitarian principles or IHL, frontline humanitarian negotiators increasingly rely on an engagement approach that prioritizes relationship-building with conflict actors to gain access to populations in need (Centre of Competence on Humanitarian Negotiation, 2019, p. 90).
Implementation power may be a crucial thing to consider when identifying appropriate interlocutors for humanitarian negotiation in armed conflict. This issue intersects with questions of empathy in several ways. The HD Handbook points out that no matter how open an interlocutor is-that is, how receptive they are to humanitarian requests-any agreement made with them will be for naught unless they possess the power and authority to implement it (Mancini-Griffoli and Picot, 2004, p. 51). The Handbook offers the example of a high-level actor guaranteeing humanitarians safe passage through checkpoints manned by members of their armed group, while the latter want to block humanitarian access and do not consider themselves governed by the high-level agreement. In this kind of scenario, empathic engagement with higher-level actors may fail to bring about meaningful results, underscoring the importance of a humanitarian's attunement to how they are perceived, by whom and with what implications.

Implications for humanitarian work
The final axis highlights how international laws and norms may need to take a backseat when humanitarians believe they can make more progress by building and maintaining a good relationship. In certain contexts, humanitarians are advised to avoid law in order to ingratiate themselves with their interlocutor, establish their own legitimacy and build trust. This will especially be the case when a given IHL rule is at odds with the interest and needs of the conflict actor. Further, there is untapped potential in getting humanitarians to elicit more empathy from conflict actors on humanitarians' commitment to their own values, principles, and rules. The social dimension of empathy is especially relevant here. In addition to understanding another's beliefs and feelings, empathy provides insight into how one's own actions, including appeals to legal rules or principles, may be understood and felt by others (e.g., affected populations, state actors, armed groups, donors, and the wider public). Here, expressive empathy, the individual's ability to convey their empathic capacity through, for example, facial expression and body language, is of critical importance in the process of relationship building.

Conclusion: implications for training and operational guidance
The analysis has revealed empathy's relevance across multiple dimensions of humanitarian negotiation, a critical area of humanitarian practice. Empathy is activated not only for populations in need but also for other conflict actors, including those who would harm civilians. Furthermore, the empathy of other conflict actors, whether for humanitarians or for the general civilian population, also matters. While this article has analyzed these dynamics in the context of frontline humanitarian negotiations, they are relevant to humanitarian action more broadly. Humanitarians need to know how to cultivate and manage their own empathy as well as that of others, leveraging it for the benefit of populations in need.
Taking heed of relevant critiques in the scientific literature, empathy has not been presented here as an unalloyed good such that increased empathy would lead to better humanitarian outcomes. Consequently, we do not advocate for more empathy, but rather a more thoughtful approach to empathy, one that recognizes first and foremost that empathy is vital to humanitarian work. While empathy is an innate human behavior that shows inter-individual variability, it is malleable and can be enhanced and shaped by interventions (Stepien and Baernstein, 2006;Kelm, et al., 2014). We conclude by suggesting several respects in which humanitarians-and, by implication, the populations they assist-could benefit from intentionally developing empathy-related skills.
To frame the points that follow on skills development, we emphasize that there is also value in adding more theoretical elements to capacity building efforts. Specifically, it would be helpful to better acquaint humanitarians with the main insights of the empathy literature -for example on the difference between empathy and sympathy, and on how biases and identity characteristics could shape one's experiences of empathy. Such lessons could complement efforts to decolonize the sector and could be integrated into broader trainings on diversity, equity and inclusion, which often involve discussion of unconscious bias.
The first area of skills development is emotion regulation and resilience. Humanitarians will experience empathy by virtue of being human and the nature of humanitarian protection work which entails proximity to suffering. As such, individuals require accessible psychosocial support as well as strategies for coping with emotionally challenging situations, avoiding burnout, and averting empathic stress (Cockcroft-McKay and Eiroa-Orosa, 2021). Boundary management and selfcare, including a healthy lifestyle, are widely recognized as vital to emotion regulation, an important "buffer" to empathy overload and stress. Further, recent studies show that mindfulness serves as a mediating factor between empathy and burnout (Surguladze, et. al., 2018) and that to target the complete mental health of humanitarians, interventions should seek to foster meaning and psychological flexibility, including modalities such as acceptance commitment therapy (Young, et al., 2021).
Second, humanitarians should have opportunities to strengthen their perspective-taking abilities both in predeployment training and while on the job. Cognitive empathy-the ability to intentionally adopt another person's perspective to apprehend their subjective experience-is essential to humanitarian work. In addition to shedding light on another's interests and motivations, evidence suggests that perspective-taking can reduce an individual's biases against groups and increase cooperation. Perspective-taking should thus be promoted in humanitarian work through simulations, role play, and exposure to narrative through art, film, and storytelling. Studies show that it is through active and imaginative immersion into others' stories that empathic reasoning is strengthened.
Finally, individuals and organizations should be attuned to the social dimensions of empathy in humanitarian work. Trainings should target the development of individual negotiators' knowledge of when and how to perspective-take, as well as their active listening skills. Attentive listening to narratives of interlocuters is critical in facilitating empathic engagement. Related, humanitarians should be trained in empathy conveyance and signaling, learning how to communicate in both words and expressive behaviors, such as body language, that one has the capacity to empathize. The kind of empathy we are ultimately arguing for here, as the article's title suggests, is empathy that enables and strengthens relational engagement in the humanitarian negotiation context and beyond.