Better researchers, better people? The dangers of empathetic research on the extreme right

ABSTRACT The difficulties of empathising with violent perpetrators has preoccupied a small number of social scientists, particularly criminologists and critical security and terrorism scholars. Most of this work examines how researchers can mitigate the impact of engaging with violent perpetrators and minimize any distress this may cause. This article examines whether a deeper empathetic exchange with violent actors - one in which researchers open themselves cognitively and emotionally and that accepts the dialogical nature of empathy - is the next step that needs to be undertaken by critical scholars. It then examines the costs of such a move. The discomfort and danger of empathizing with violent actors is examined through the particularly thorny example of research with violent actors from the extreme-right.

New Zealand Prime Minister Jacinda Ardern asked the world not to use Brenton Tarrant's name when referring to his live-streamed attack in Christchurch, in which he killed 51 people in two mosques on 15 March 2019. She told the New Zealand parliament four days after the attack: He sought many things from his act of terror, but one was notoriety. That is why you will never hear me mention his name. He is a terrorist. He is a criminal. He is an extremist. But he will, when I speak, be nameless. And to others, I implore you, speak the names of those who were lost rather than the name of the man who took them. He may have sought notoriety but we in New Zealand will give him nothing, not even his name. 1 She says he deserves nothing, not even a name, i.e. not even subjectivity or indeed humanity. And if he is to remain nameless, undeserving of anything, he is certainly not deserving of empathy.
As a human being and as someone who tries hard to live by anti-racist feminist principles, I understand what Ardern wants to do. She wants to tell New Zealanders and then the whole world that Tarrant is not "one of us." He is a terrorist, a criminal, an extremist, and we are better than that. As a researcher, however, I have spent most of my academic career arguing that terrorists, criminals, and extremists cannot be reduced to single identities. They are complex human beings with multiple identities. To understand them and their violence, we need to recognise their humanity -how they are "like us." We must speak and write their names.
Recognising that someone is "like us" enables us -indeed it may require us -to empathise with them. Empathy -understood in the person-centred approach as "to learn what it feels like to be in [another's] skin and to perceive the world as [they] perceives it" (Mearns, Thorne and McLeod 2013, 13) -is both an emotional and cognitive process (Head 2016). It is also an aspiration. I do not believe we can ever completely be in another person's shoes, but we can and should try as much as possible. Crucially for the argument being developed here, empathy does not necessarily lead to sympathy -to think well of another or feel for the other. Indeed, Wheeler (2013) in his work on empathy and trust in International Relations, notes that social actors can use their capacity to empathise with their opponent to better defeat them. This still requires researchers to be able to put themselves in the other's skin or shoes, another commonly used expression. This in turn requires of researchers to see and accept the humanity of their coparticipants. Empathy demands that researchers start any interaction with another by saying: "They are like me, human." 2 Despite emerging from different disciplines, several similarities exist in the engagement with this question and the general approach can be summarised as one of "managing ambivalence" (Knight, Phillips, and Chapman 2016, 6). This is largely carried out by attempting to minimise the distress and impact of such engagement on researchers. Three pathways can be identified from the literature on how to manage the ambivalence.
The first pathway is to point out that the perpetrators of violence are often themselves victims of violence, either of direct violence or of other less obvious forms such as structural or cultural violence (Galtung 1969). As noted by Cottee (2013, 7: emphasis original): Many criminological theories, especially those of a strongly sociological bent, situate the offender within a structural context of radically limited choices and possibilities, and, hence, serve to improve the moral image of the offender. Implicit in these theories is the suggestion that criminal conduct, far from being a reflection of the depraved and deep-seated personal character traits of the criminal offender, is in fact a consequence of extraneous societal factors, which caused or 'pushed' them into criminality.
Cottee believes that these approaches to criminology do not aim to excuse or justify the violence perpetrated, but argues that it is important to recognise that research in violence does not only do scientific work, but also does "moral work" (Cottee 2013, 3 emphasis original). This moral work is done onto others -students, fellow academics, policymakers -but also onto the researchers themselves. "Having been enlightened about the sources of their criminality, are we likely to feel more or less indignant towards [offenders]?" he asks (Cottee 2013, 2 emphasis original). As is noted above, he concludes that contextualising the violence and seeking to understand it through an analysis of the structural conditions surrounding the offender, are likely to make us less indignant towards the offender. Aside from making one possibly less indignant/angry, this contextualisation also is likely to make empathising with offender/perpetrators easier and less disturbing for researchers. The perpetrators become fellow human beings who were placed in circumstances that -to greater and lesser degrees -contributed to their use of violence. I would argue that a similar argument can be made for critical approaches to security and terrorism, where the field has focused particularly on the structural violence that has framed, contributed to, or caused -depending on the epistemological standpoint of the author -the direct violence of those labelled as terrorists or extremists. Jackson, Smyth, and Gunning (2009, 218), in one of the founding texts of CTS, explicitly name structural violence as "a reason for oppositional violence." My work has also adopted this position by taking on a grievance-based understanding of terrorist violence and examining how underlying conditions of structural violence can form part of negotiated agreements with non-state armed groups (Toros 2012). In no way do I mean to imply that CTS is guilty of any form of justification of terrorist violence -an accusation that has been pointed at too many of us far too often. On the contrary, a considerable amount of work in CTS is based either on a pacifist grounding (see Jackson 2018) or on the aim of reducing violence in all its forms (Toros and Gunning 2009). 3 However, I want to tentatively argue here that part of the "moral work" that this position undertakes eases the emotional and ethical dilemmas that may emerge from empathising with those labelled as "terrorists." By linking direct "terrorist" violence to other forms of direct violence (often carried out legally and illegally by states) and widespread forms of structural and cultural violence, CTS generally has argued that violent actors need to be seen as having multiple identities, that of perpetrator of terrorist violence but also that of victim of direct and structural violence. This recognition of the multiple identities of violent actors -including a victim identity -serves to allow researchers (and indeed practitioners who engage with violent perpetrators such as probation officers or lawyers) to empathise with the victim identity and not have to confront the moral work of empathising with perpetrator identity. With the argument that "superordinates have superiors too and each will argue that the world is as it is because those above them determine that it is so" (Liebling 2001, 478), researchers can skirt the discomfort of empathising with perpetrators.
The second mechanism commonly used to avoid or reduce discomfort is to see empathy as a tool that can be used strategically to improve research. This is the "empathy as method" approach. Indeed, there is considerable literature on the importance of empathy in research on violence -whether political or criminal (see the work of Head 2016). This leads to terms such as "emotional intelligence" or "emotional literacy" being used. This language of skill, I want to argue, implies that the researcher is able to open themselves up to feeling "like" the perpetrator without giving much of themselves. Indeed, here the position is still one where "the researcher extracts from the research participant and is unaffected by and uninvolved in the process . . . the research relationship is understood as monological" (Cowburn 2010, 70 emphasis original). If empathy is a deployable skill, thus not engaging the researcher's being, there is little danger of the latter being transformed personally or politically by such an engagement. The researcher remains at a safe distance.
In the third approach, difficulty is acknowledged, but this difficulty can be dealt withmanaged -through an effective support and/or debriefing mechanism. Here, the literature points at a researcher's need to unload the disturbing information they have received from research participants or the stress related to having to hold back their rejection or even disgust at what they have heard. This work can be "draining, difficult, stressful and anxiety increasing" and several scholars note that those exposed to violent perpetrators often bring their work into their family lives or have to rely on colleagues and friends (Knight, Phillips, and Chapman 2016, 3). The assumption however seems to remain that if researchers empathising with violent actors can unload their burdens effectively (without putting undue stress on their families, friends or colleagues), they can return to their daily lives largely unchanged by the encounter. The assumption is that there are no profound and/or long-term implications.
A dialogical relationship however, cannot be a one-way encounter. If, as researchers, we open ourselves up to empathise with another -to be able to better understand the other even simply to carry out "better research" -we leave ourselves open to other unplanned for and most likely unwanted connections with the other. Indeed, conflict transformation scholars -often cited in critical approaches to security and terrorismremind us that dialogue is essential and transformative for all parties. Without the possibility of transformation, it is not dialogue. And this requires the "full use of your self" (Liebling 2001, 475

emphasis original).
This is not only the sole way one can enter into a dialogical relationship based on empathy but it is also ethically, I would argue, the right way of engaging if one adheres to an "ethic of care." Indeed, Cowburn (2010, 67) notes that an "ethics of care is concerned with the reciprocal and dialogical nature of human relationships." If reciprocal and dialogical, the relationship requires that the researcher be open to transformation, that is, to receiving more than simply data from violent perpetrators but also a human connection that is potentially transformative. This requires more than "managing ambivalence." It requires abiding by Rogers' (1990) three principles of empathy, congruence, and unconditional positive regard.
For the first, we need to open ourselves up cognitively and emotionally. Indeed, as argued by Naomi Head, empathy is a relational process "whereby cognition and affect play out in interpersonal exchanges through which participants can generate shared new meanings" (Head 2011, 19). Meanings are not transferred from one person to another in research but are being renegotiated in all interaction between social actors, even when actors are not "explicitly seeking to renegotiate their relations" (King 2000, 428). This requires researchers to open themselves up to this renegotiation; they need to be willing to negotiate meanings with their co-participants to be able to create shared meanings. Furthermore, congruence means being honest with our research participants and therefore acknowledging the difficulties, concerns and fears that may be triggered by our association with them. Finally, to have unconditional positive regard for a person does not require us to unconditionally "think well" of them; but it does require that we accept them in their entirety, including their positive and negative attributes, their consistencies and inconsistencies (Toros 2016). It requires recognising and accepting their perpetrator as well as their victim identities.
I have met more people who have killed -legally and illegally -than most researchers I know. I have interviewed them, carried out participant observation with them, run workshops and training courses with them. But I have not only been a researcher with them. I have also been a fellow human being. So, I have sat with them, laughed and drank, introduced them to my children. Some became friends. Some of them, often without me noticing, have changed who I am.
Since I do not believe that empathy can be deployed strategically as a skill, over the years, I have primarily resorted to the first and third method to avoid discomfort from such encounters. I have focused on the victim identity of my interlocutors to sooth my unease.
In one case, I preferred to focus on a former paramilitary's "bad luck" in an unwanted teenage pregnancy that prevented him from leaving the neighbourhood marred by sectarian violence rather than his subsequent murders of unarmed civilians. I did this to feel more comfortable with a relationship that increasingly felt like friendship. In another case, I focused on the death of a fiancée in a drone attack, rather than a militant's leadership of a ruthless military intelligence unit. Such a recognition of others' multiple identities can be seen as part of the process of offering unconditional positive regard to our research participants, but it may also be means to avoid the discomfort of opening up to a perpetrator rather than a victim.
I have also used the third means to mitigate discomfort by setting up informal communication channels with friends and colleagues specifically aimed at discussing my fears of transformation when my research led to daily engagement with violent actors over a prolonged period of time. Interestingly, I never used those channels and by the time I felt the need to reach out, I realised that I was already some way down the road of transformation. My framing of questions had already changed, the value I attributed to the "impact" of my research had changed. This transformation does not appear to be short-term either and has outlived the fieldwork period. For example, I find it increasingly difficult to conceive of research in security and terrorism studies that does not appear to have the potential to impact someone outside the academy -not necessarily the state, but someone. This is relevant because I now read through pages of racist and misogynistic texts, in the form of published documents and online messages. How are they changing me? If, as Liebling (2001, 474) argues, the "relevance of our research is its possible cultural, political and moral implications," what are the personal risks and political implications of adopting an open empathetic approach to the authors of these messages? What are the personal and political costs of opening up to the possibility of a long-term transformation? My instinct -much like Ardern's -is to close down dialogue and empathy towards these actors. I do not want to find communality with the authors of such texts.
Indeed, I am at first drawn by the privilege these actors have overall benefitted from. These texts are written most commonly by white people, often by men (Pearson 2020). As such, in the racialised patriarchical society we live in, they belong broadly to the class of "superordinates." They are written by men like Brenton Tarrant, although the vast majority of them will not turn their acts of cultural violence into acts of direct violence. A more in-depth investigation of these social actors however, often reveals socio-economic deprivation and social isolation. Based on my previous experience working with violent actors, I quickly recognised my attempt to avert discomfort by focusing on the social isolation and/or intellectual poverty of these men. I realise that this is self-preservation strategy No. 1 by focusing on their potential victim identity, despite my initial attempt to dismiss them as superordinates.
However, even if one recognises these actors as superordinates, an ethics of care would seemingly reject any form of "balancing of victimhood" between communities, whether based on race, gender, or class. Would not an ethic of care reject the notion that a victim is worthy of more care than a perpetrator? They may be in need of different kinds of care, but are they not equally worthy of care, particularly as research participants? And who am I to judge who is worthy of being recognised and treated as a full complex human being and who is not? Does not everyone have the right to be treated with care? Tarrant must be granted his name and his subjectivity. I must empathise with him to be able to understand something of him. The transformation that follows means that I will know better what brings about such violence. Of course, I have to be on my guard to challenge his words with mine and with the words of those who sustain me and my antiracist feminist positions. But I need to be open and willing to enter a dialogue. However, empathising with racist misogynist actors does more than improve our knowledge; it does moral and political work. The moral work, as seen above, is hard but necessary from the position of an ethics of care. What of the political work?
Recognising such a victim identity has serious political implications. I do not want to fall prey -or worse still, become an advocate -of the "left behind white working class" narrative put forward by right-wing policy makers (and some academics) who argue that too much focus has been placed on women and minorities "at the expense" of working-class men. Aside from the questionable empirical basis for such claims, their manipulation by right-wing political actors to undermine and defund initiatives to support women and ethnic minorities is not something I want to support. Thus, the political work is difficult but also dangerous. It can provide ammunition to the "left behind white communities" argument, a manipulation of public perception aimed at defunding projects that attempt to address centuries of racial of gender discrimination at the heart of institutional power the world over. Nevertheless, the answer is not to avoid empathy, leaving this field to the right and its media and academic pundits. Rather, the only way forward seems to be to claim this space from them with honest empathetic and emancipatory research that recognises everyone's humanity and community.
Emancipatory research and politics cannot be achieved monologically; they require dialogue which in turn requires being open to transformation from the other. They require accepting that we may have something to learn from the other and that the other always has the potential for emancipatory transformation. They also require however, facing head on the moral and political work of such empathy and not managing our discomfort/ambivalence by overemphasising victimhood.
Empathising with Tarrant and with the extreme right thus makes us better researchers and possibly better people. It is a difficult path because empathy requires an openness that leaves us vulnerable to violent and insidious ideas that hurt the soul as well as the mind. This, however, may be the greatest challenge for a critical engagement with terrorism and political violence yet.