INTRODUCTION
In her classic research on emotion in the workplace, Hochschild (Reference Hochschild1983, 7) coined the term “emotional labor” to capture the “management of feeling to create a publicly observable facial and bodily display.” Since then, extensive scholarship has established the pervasiveness of emotional labor in the service sector and the professions (Wharton Reference Wharton2009). A central finding of this research is that employees engage in both intrapersonal and interpersonal emotion management (Thoits Reference Thoits1996). They manage their own and others’ emotions, and the management techniques they rely on are tied to characteristics of the workers as well as the organizations in which they work.
Although the disavowal of emotion is a key feature of judicial authority across contexts (Bergman Blix and Wettergren Reference Bergman Blix and Wettergren2018), judges also engage in intrapersonal and interpersonal emotion management (Roach Anleu and Mack Reference Roach Anleu and Mack2005). This article addresses judicial emotion management in a particular setting: transitional justice courts. Such spaces are rife with emotion, and better understanding the emotional aspects of transitional justice proceedings contributes to knowledge regarding post-conflict reconciliation and stability. Nevertheless, little work has explicitly examined the emotion work of transitional justice judges.
To address this gap, we engage in an in-depth examination of the emotion management techniques employed by judges who served in Rwanda’s post-genocide gacaca courts. Our analysis of interviews with eighty-five gacaca court judges begins by documenting the pronounced need for transitional justice judges to perform emotional labor. We then discuss the intrapersonal and interpersonal emotion management strategies they employed to adjudicate crimes of genocide. Put another way, we assess how judges tried to regulate their own emotions, as well as their self-professed efforts to regulate the emotions of others.
In doing so, we make two contributions. First, we extend work on emotional labor to include transitional justice contexts. To date, researchers have largely focused on the emotions of transitional justice participants, such as people who were victimized (Rimé et al. Reference Rimé, Kanyangara, Yzerbyt and Paez2011; Cilliers, Dube, and Siddiqi Reference Cilliers, Dube and Siddiqi2016). Newer work has emphasized the emotions of those who staff and implement transitional justice mechanisms, such as judges (for example, Hagan and Kay Reference Hagan and Kay2011) and translators (for example, Swain Reference Swain2011), with a particular emphasis on secondary trauma and well-being. We situate our study within this small yet growing literature on the emotions of those who staff and implement transitional justice mechanisms, though we also depart from existing work by addressing emotional labor rather than secondary trauma and well-being.
Second, we take an intersectional approach to understanding emotional labor in a transitional justice context with an emphasis on gender and ethnicity. Most of the intrapersonal and interpersonal emotion management strategies we document are gendered, consistent with previous research on international human rights lawyers (Hagan and Kay Reference Hagan and Kay2011) and studies of emotional labor in other professional settings (for example, Berheide et al. Reference Berheide, Carpenter and Cotter2022). However, we also find that the dimensions of identity most salient during the conflict shaped the judges’ interpersonal emotion management strategies afterward. In particular, judges associated with the targeted group concentrated on lifting the spirits of survivors, while those who were not targeted during the genocide emphasized the need to punish defendants.Footnote 1 We thus consider gender and conflict-related identities in combination, illustrating how social identities linked to the genocide interacted with gender norms to structure judges’ intrapersonal and interpersonal emotion management techniques.
EMOTIONAL LABOR AND EMOTION MANAGEMENT
Brief Theoretical Overview
Hochschild (Reference Hochschild1983) introduced the term “emotional labor” to describe employees’ efforts to manage their emotional states to generate certain feelings in others. Think, for instance, of a cheerful flight attendant or a bill collector who is brusque or rude. These jobs, and other jobs that call for emotional labor, generally entail: (1) face-to-face or voice-to-voice contact with the public, (2) requirements to produce an emotional state in another person, and (3) a degree of employer control over the emotional activities of workers.
In conceptualizing emotional labor, Hochschild emphasized the management of one’s own emotions, known as intrapersonal emotion management. Employees engage in intrapersonal emotion management to shape and direct their own feelings, including efforts to ensure that their emotions conform to the feeling norms associated with their positions.Footnote 2 For instance, flight attendants may need to work to appear cheerful, just as bill collectors may need to work to be unpleasant. As such, they regularly employ specific intrapersonal emotion management strategies. These strategies may be cognitive, such as reinterpreting an event or situation; behavioral, such as controlling emotional displays; or physical, such as taking drugs to reduce arousal (Hochschild Reference Hochschild1983; Pugliesi Reference Pugliesi1999; Wharton Reference Wharton2009).
Social inequality, cultural expectations, and training (among other factors) shape the intrapersonal emotion management strategies that people employ. For example, racial and ethnic minorities promote counternarratives in predominantly white institutional spaces to protect themselves from denigration (Evans and Moore Reference Evans and Leo Moore2015; see also Wingfield Reference Wingfield2010). Similarly, women service sector employees assert control in their interactions with customers to counter ingrained cultural symbols of servitude (Paules Reference Paules1991). In one of many other examples, crisis-response workers rely heavily on formal training to remain calm under emotional duress (Mastracci, Guy, and Newman Reference Mastracci, Guy and Newman2014).
As emotional labor involves producing an emotional state in others, workers also engage in interpersonal emotional management to bring others’ emotions in line with feeling or display rules (Lively and Weed Reference Lively and Weed2014, 203). Early research on interpersonal emotion managementFootnote 3 concentrated on certain types of employees, such as counselors (Francis Reference Francis1997), though many positions involve the management of others’ emotions (Rafaeli and Sutton Reference Rafaeli and Sutton1991; Thoits Reference Thoits1996). During the Covid-19 pandemic, for example, retail workers engaged in exhausting interpersonal emotional management in enforcing safety guidelines (Loustaunau et al. Reference Loustaunau, Lina Stepick, Petrucci and Henifin2021).
Just as with intrapersonal emotion management, factors like gender (Lively Reference Lively2008) and organizational goals (Craciun Reference Craciun2018) also shape interpersonal emotion management techniques. For instance, Pierce (Reference Pierce1999) highlights gender-appropriate ways in which men and women paralegals catered to their bosses’ needs, illustrating that while both were expected to manage their anger, men were given considerably more leeway.
Extensive scholarship has documented the importance of such emotion management techniques across diverse service sector settings (Brotheridge and Grandey Reference Brotheridge and Grandey2002; Erickson and Ritter Reference Erickson and Ritter2001; Gimlin Reference Gimlin1996) and the professions, including law (Bandes and Blumenthal Reference Bandes and Blumenthal2012; Bergman Blix and Wettergren Reference Bergman Blix and Wettergren2016; Pierce Reference Pierce1995). Perhaps most notably, Roach Anleu and Mack (Reference Roach Anleu and Mack2005, 606) find that Australian magistrates’ intrapersonal and interpersonal emotion management involves (1) direct engagement with defendants’ unmediated emotions, (2) concern for the impression a person will take from the court, (3) management of their own emotion, and (4) management of the effects of emotional labor on themselves.
Others have likewise examined how judges in disparate contexts regulate their and others’ emotions. Maroney and Gross (Reference Maroney and Gross2014) illustrate how judges must manage their own emotions while meeting professional obligations; for instance, if a judge cannot change or ignore a case, they may alter how they think about it, hence utilizing a cognitive strategy. In therapeutic jurisprudence and problem-solving courts, judges often attempt to apply the law in ways that can “improve the psychological functioning and emotional well-being of those affected” (Winick and Wexler Reference Winick and Wexler2015, 479). In fact, judges frequently engage in managing their and others’ emotions simultaneously, such as by expressing patience “to calm a nervous witness” while keeping their own impatience in check (Bergman Blix and Wettergren Reference Bergman Blix and Wettergren2016, 34).
Taken together, this body of research establishes that emotional labor occurs across diverse positions, including the judiciary, and that individuals develop various strategies to manage their emotions, as well as the emotions of others. Further, these intrapersonal and interpersonal emotion management strategies are impacted by societal norms and structures as well as by organizational factors. In this article, we extend these findings to a new organizational realm: transitional justice courts.
Emotional Labor, Transitional Justice, and Rwanda’s Gacaca Courts
Emotions are social, as research has demonstrated across innumerable contexts, including that of transitional justice (Thoits Reference Thoits1996; Karstedt Reference Karstedt2016). Researchers have established the emotional impact of transitional justice endeavors (Brounéus Reference Brounéus2010; Cilliers, Dube, and Siddiqi Reference Cilliers, Dube and Siddiqi2016) and called for greater attention to emotion in these processes (Mendeloff Reference Mendeloff2009; Swain Reference Swain, Sköld and Swain2015). To date, however, few studies have systematically examined the emotional labor of frontline workers, officials, and decision makers in transitional justice mechanisms.
Much research in this vein has emphasized the emotional impacts of serving in justice-related roles. For instance, Hagan and Kay’s (Reference Hagan and Kay2011) study of legal professionals in the Office of the Prosecutor at the International Criminal Tribunal for the former Yugoslavia found that the shocking content of work at the tribunal led to depressive symptoms among women legal professionals (see also Hagan Reference Hagan2010; Nah Reference Nah2020). To a greater extent than in other court proceedings, emotional testimonies are often a cornerstone of transitional justice proceedings (Karstedt Reference Karstedt2016). Trials in post-atrocity settings typically include detailed survivor testimonies of brutal violence, frequently accompanied by extensive crying and emotional outrage. This places court actors, like translators and other staff, into close emotional proximity to the traumatic narratives they must interpret and absorb (Swain Reference Swain2011). Hayner (Reference Hayner2010, 159) accordingly warns of the secondary traumatization of statement takers, data entry staff, and journalists who may become “sponges of trauma” in absorbing such narratives.
Though judges across many courts are tasked with emotional duties, those serving in transitional justice settings engage in emotion work in a uniquely challenging situation. Hagan and Kay (Reference Hagan and Kay2011) consequently conclude that additional work is needed to understand the emotional lives of lawyers, as well as nonlawyers, working in human rights settings. Here, we take up this call and extend research on emotional tolls of legal service in human rights–related fields to emotional labor in such fields. As no scholarship has explicitly examined emotional labor among judges who serve in transitional justice courts, we bring together research on emotional labor (Hochschild Reference Hochschild1983; Thoits Reference Thoits1996), judges (Roach Anleu and Mack Reference Roach Anleu and Mack2005), and transitional justice (Cilliers, Dube, and Siddiqi Reference Cilliers, Dube and Siddiqi2016; Karstedt Reference Karstedt2016) to productively extend this literature. To do so, we examine the case of Rwanda’s post-genocide gacaca courts. Since extant literature has documented that occupational roles and broader social settings shape emotional labor (Lively and Weed Reference Lively and Weed2014), we briefly review the social setting in which gacaca trials took place, as well as the roles associated with serving as a judge.
Turning first to the social setting, the 1994 genocide in Rwanda targeted Rwanda’s TutsiFootnote 4 minority alongside Hutu moderates. The genocide occurred after decades of strife between Tutsi, who had dominated Rwanda’s precolonial and colonial monarchy despite making up no more than 15 percent of the population, and Hutu, who had served in positions of power since independence. An economic downtown and a civil war between a mostly-Tutsi army (known as the Rwandan Patriotic Army) and the Hutu-led government resulted in countrywide fear and mistrust during the early 1990s. Following the assassination of Rwanda’s Hutu president in 1994, political elites encouraged Hutu civilians to protect themselves and their families from Tutsi (Des Forges Reference Des Forges1999; Straus Reference Straus2006; McDoom Reference McDoom2021).
Hundreds of thousands of Hutu civilians committed violence against Tutsi, resulting in widespread death, displacement, and sexualized violence. The genocide ended after the Rwandan Patriotic Army took power, cementing the associated political party—the Rwandan Patriotic Front—as the new leaders of Rwanda. To try the many people accused of genocide,Footnote 5 this new government created courts known as gacaca courts (see Clark Reference Clark2010; Bornkamm Reference Bornkamm2012; Chakravarty Reference Chakravarty2015; Palmer Reference Palmer2015; Doughty Reference Doughty2016; Ingelaere Reference Ingelaere2016). The courts were loosely based on a precolonial dispute resolution mechanism and had five official goals: identifying the truth, increasing trial speed, fighting impunity, contributing to reconciliation, and demonstrating Rwanda’s problem-solving capacity (Clark Reference Clark2010). They also served to shore up the new government’s power and propagate its narrative of the violence (Chakravarty Reference Chakravarty2015). Indeed, critics have suggested that crimes committed by the Rwandan Patriotic Army in 1994 were not tried, mirroring issues with victor’s justice in many transitional justice contexts (Waldorf Reference Waldorf, Sullivan and Tifft2006).
Per the government’s instruction, every community elected panels of local judges to preside over these courts and, ultimately, to determine their neighbors’ fates.Footnote 6 These judges were called inyangamugayo, which translates to “person of integrity.” The position was unpaid, and no previous legal training or formal education was required. Rather, judges had to be Rwandan residents, at least twenty-one years old, and people of “good character” who had neither participated in the genocide nor spent more than six months incarcerated (see Bornkamm (Reference Bornkamm2012) and Ingelaere (Reference Ingelaere2016) for more on elections).
In total, over 250,000 individuals staffed approximately 12,000 courts. These individuals were Hutu and Tutsi, though there are not data regarding their identities beyond studies of courts in specific areas. While precolonial gacaca was only implemented by men, data from 2005 indicate that at least 34 percent of the judges were women (Final Gacaca Report 2012). This number likely rose as some judges stepped down amid allegations of genocide, thrusting many women into the public sphere at a time when such public positions were often held by men (Mukandori Reference Mukandori2014).Footnote 7
After accepting the position, the inyangamugayo underwent several weeks of training regarding the laws governing the courts, punishments, evidence, witness traumatization, and ethics, among other topics (Bornkamm Reference Bornkamm2012). As Doughty (Reference Doughty2016, 196) explains, they were inquisitorial judges, meaning they “played an active role conducting investigations and questioning, during and outside of trials, much as the prosecution, defense, or even police do in other legal systems.” Since there were rarely lawyers at the courts, the judges were responsible for gathering evidence, scheduling and presiding over trials of anyone whom neighbors accused, and determining guilt and punishments. For crimes against people (e.g., killing or joining a killing group), these punishments typically involved prison sentences with community service options for those who confessed or those who committed less serious crimes. For crimes against property, typical sentences involved fines that were meant to be paid (though were not always paid) to the victims’ families. While there were thus elements of restorative justice present in gacaca, retributive justice was core to the gacaca model (Waldorf Reference Waldorf, Sullivan and Tifft2006; Ingelaere Reference Ingelaere2016).
After three years of pilot trials and several more weeks of training, the courts opened in March 2005. Trials occurred weekly, and all community members were expected to attend. The courts closed in June 2012,Footnote 8 meaning that many judges worked for a decade.
The unpaid nature of this work clearly diverges from the employment and paid work context of much research on emotional labor. Nevertheless, there are compelling reasons to examine the experiences of the inyangamugayo through the lens of emotional labor and, specifically, emotion management. First, the judges engaged in intensive, formalized, and public work over a sustained period of time. In a setting in which most people are self-employed in agricultural pursuits around their homes, the inyangamugayo received formal training and worked regular hours at particular sites, often in full view of the community. Their tasks were also explicitly undertaken for the state, like many paid positions, and judges were continually reminded that they were working for their country. As they began their roles, for instance, the inyangamugayo swore an oath to Rwanda (Doughty Reference Doughty2016). Each time they worked, they donned sashes in the colors of the Rwandan flag, again symbolically connecting their positions to the state itself.
Furthermore, as Chakravarty (Reference Chakravarty2015) has poignantly argued, many of the inyangamugayo stepped into the unpaid position to advance their own careers, meaning that they were expecting a transactional outcome not unlike employment in other contexts. In fact, multiple inyangamugayo were motivated by the possibility of gaining visibility or leadership positions, suggesting that their reputations and future careers were at stake.Footnote 9 Recent research has indeed illustrated that some judges subsequently garnered positions, including positions as mediators or local leaders, as well as perceived social capital in the form of enhanced reputations (Nyseth Brehm et al. Reference Nyseth Brehm, Frizzell, Uggen and Gertz2021).
The inyangamugayo also differ from judges in other contexts in important ways, though many of these differences likewise highlight the significance of emotion management. With respect to their own emotions, inyangamugayo adjudicated extremely serious crimes and had the ability to levy heavy punishments (including life sentences), though they had very little training compared to judges in other contexts. The several weeks of training judges did receive emphasized that they needed to regulate their emotions and follow sentencing guidelines rather than allow themselves to be swayed by particularly emotional or shocking cases (Chakravarty Reference Chakravarty2015; Doughty Reference Doughty2016). As such, the inyangamugayo were told not to make decisions based on emotions, though they were given very few professional tools for handling such emotions and thus likely needed to develop intrapersonal emotion management strategies.
With respect to regulating the emotions of others, the inyangamugayo’s jobs were arguably uniquely complicated compared to judges in many other settings, as they lived in the same communities as defendants and witnesses (Chakravarty Reference Chakravarty2015; Ingelaere Reference Ingelaere2016). Accordingly, their training instructed judges to be impartial, which was broadly construed as not taking personal relationships into account with respect to rendering decisions and stepping down when a case was too personal (Doughty Reference Doughty2016). Training also instructed the judges to assist people with trauma and help resolve conflicts (Clark Reference Clark2010; Bornkamm Reference Bornkamm2012; Doughty Reference Doughty2016). In fact, the judges’ training emphasized keeping trial attendees calm, and one training unit even covered “group entertainment” (Final Gacaca Report 2012). As such, the inyangamugayo regulated the expression of emotion within the public domain as a form of service. Unlike judges in other contexts, however, the inyangamugayo did not benefit from “status shields” (Hochschild Reference Hochschild1983) that would have helped protect them from the displaced feelings of others attending trials that were often brimming with palpable emotion. Interpersonal emotion management strategies may have therefore been of paramount importance, and we consequently explore both intrapersonal and interpersonal emotion management techniques undertaken by Rwanda’s inyangamugayo.
METHODS
Many studies of emotional labor employ interviews or surveys, ethnography, or a combination of these methods. For instance, Hochschild (Reference Hochschild1983) distributed questionnaires to students in her classes, interviewed bill collectors, interviewed flight attendant trainers, and observed flight attendant recruitment and training. As such, she did not directly observe people at work but emphasized how they were trained and how they felt they had to manage their emotions to produce a certain emotional state in others. Many others have followed suit. For instance, Stenross and Kleinman (Reference Stenross and Kleinman1989) interviewed detectives to learn about aspects of their emotional labor involved in dealing with those who committed crimes and those who were victimized, while Bergman Blix and Wettergren (Reference Bergman Blix and Wettergren2018) relied on both interviews and observations of court actors for their study of emotions in courts.
To assess judges’ emotional labor during Rwanda’s transitional justice processes, we rely on in-depth semi-structured interviews. Because the courts closed in 2012, we were unable to undertake ethnographic observations, which is a limitation of our approach. Nonetheless, interviews are particularly well suited for ascertaining intrapersonal emotional management techniques, and they also shed light on people’s intentions as they pursue interpersonal management strategies. We also triangulate our data with information regarding the inyangamugayo’s training as previously outlined.
Specifically, we interviewed eighty-five former inyangamugayo in Rwanda during 2015 and 2016.Footnote 10 As gacaca courts operated at the sector and cell (subsector) levels of Rwandan geographic administration, we selected participants through a stratified random sampling procedure in four sectors: Gikondo, Gahanga, Masaka, and Mwurire. These sectors were chosen due to their comparatively urban (Gikondo), semi-rural (Gahanga, Masaka), and rural (Mwurire) compositions, though proximity to a central location (the capital city) partially guided this choice.Footnote 11
We randomly selected twenty trials from each of the four sectors using a random number generator and a sampling frame of all gacaca court trials. Next, we identified all inyangamugayo involved in the randomly selected trials, including judges in cell, sector, and appeals courts (which also functioned at the sector level). We then obtained phone numbers and addresses through local contacts, approached these individuals, and asked them to participate in the study. Participation was voluntary, and two judges declined.Footnote 12
Participants were invited to conduct an interview in English, French, or Kinyarwanda. All chose Kinyarwanda, and the interviews were thus conducted with translators or by Kinyarwanda-speaking interviewers. These interviews took place in or near respondents’ homes and typically lasted about one hour. The conversations followed a semi-structured interview guide that had been translated and back-translated into Kinyarwanda.
Our conversations generally involved discussions of their election, training, duties, and experiences as judges, as well as the impacts of serving as a judge. References to emotion management often surfaced when judges spoke about their duties, and we then probed to better understand how they managed their and others’ emotions. If such references did not surface organically, we asked the judges about the emotions they felt and how they managed them.
Fifty of the participants were men and thirty-five were women, ranging in age from thirty-three to eighty-six. Most engaged in farming and had finished at least some primary school, though a minority had completed secondary school and held formal employment. None had received formal legal training prior to their election as judges.
Additionally, forty-five of the interviewees (53 percent) were considered Tutsi during the genocide and thus among the group that was targeted. The thirty-four participants (40 percent) who were considered Hutu during the genocide often included those who had either refused to perpetrate violence or had risked their lives to save Tutsi. The remaining six judges were not born in Rwanda.Footnote 13 Notably, it can be difficult to discuss ethnicity in Rwanda. National laws passed in 2008 and 2013 deem the pregenocide ethnic categories to be part of a broader genocide ideology. We therefore did not directly ask participants about their past ethnicity but gleaned their identities from the information they volunteered or the context of the conversation. Many explicitly referenced being targeted or explained that they “were Hutu” or “were Tutsi” in past tense. As these categories are no longer officially recognized in Rwanda and as respondents did not verbally identify with them at the time of the interview, we refer to “former” ethnicities or otherwise discuss people who were targeted (typically Tutsi) and those who were not (typically Hutu). If we do not mention an identity when introducing someone in our results section, it means the respondent was not targeted during the genocide. Note also that while we do not suggest that individual victimization equates to group membership, in this case, all respondents who were considered Tutsi were in the country in 1994 and targeted during the genocide.
Numerous scholars have noted the difficulties of conducting research in Rwanda. For instance, many highlight how the state tightly regulates narratives of the genocide, in part to address concerns about divisiveness, denial, and genocide ideologies, and in part to extend and consolidate state power (see Straus and Waldorf Reference Straus and Waldorf2011; Chakravarty Reference Chakravarty2015). Some may thus worry that respondents would have been reluctant to discuss the gacaca courts. To mitigate such concerns, we stressed that the results would be confidential and that no names or personally identifiable information would be included in any publications. We also spent time talking with participants prior to the interviews to develop rapport.
It quickly became clear to us that respondents were not reluctant to discuss their role in the courts. Although some were hesitant to criticize the courts, most openly spoke about weaknesses and challenges, the negative effects the positions had on their lives (e.g., grudges from neighbors whose children they sent to prison), and their desire to be fairly compensated. This provided some reassurance that participants felt comfortable sharing critical views and experiences with us. Additionally, as this article examines emotional labor and related occupational demands, rather than opinions of the courts, the content we rely on is perhaps less affected by worries about critiquing the government. Of course, our positionality as Western scholars likely influenced our interviews, though two Rwandans conducted interviews as well. We compared themes that surfaced across foreign (and white, Western) interviewers as well as Rwandan interviewers,Footnote 14 and we did not observe notable differences in the findings presented here.
Upon completion of fieldwork, we transcribed and analyzed the interviews. The first and second author initially coded inductively for statements pertaining to emotion, resulting in our identification of emotional labor. We then developed and followed a coding scheme in NVivo to classify statements pertaining to emotion management. This included codes for emotional labor and emotion work more broadly, as well as codes for duties, training, impartiality, unemotional nature, controls and demands of their jobs, successes, and hardships. Additionally, we coded discussions about how judges viewed themselves and their interactions with others. While we did not originally distinguish between intrapersonal and interpersonal emotion management strategies, these techniques inductively emerged as we conducted the first wave of coding, and we subsequently coded for these strategies as well. The second author conducted the first wave of coding, separating the respondents by whether they were targeted during the genocide and gender based on initial themes. The first author conducted a second wave of coding, which also involved checking the second author’s coding and adjusting or augmenting coding as needed.
EMOTIONAL LABOR AT THE GACACA COURTS
We begin by demonstrating that the inyangamugayo engaged in emotional labor. We then discuss the intrapersonal and interpersonal emotion management techniques the judges employed. In doing so, we illustrate that gender and organizational dimensions shaped intrapersonal emotion management techniques but that interpersonal emotion management techniques were also guided by the social statuses most relevant during the genocide.
The Need for Emotional Labor
Laurent, a fifty-three-year-old man, provided an emblematic description of “an exhausting day” in the life of a gacaca judge. In his words:
We would go to start the trials at 6:00 AM. We had to go there … put benches and chairs where people were going to sit; and we would go back at like 6:00 PM, so it was an exhausting day … For example … sometimes we could face a difficult case whereby someone would have many people to testify in favor and many people to testify against. And then committee members contribute, and you had to give everyone time to make sure that they go home content.
One of Laurent’s primary concerns was ensuring that everyone involved in a trial, including those providing testimony and fellow judges, left feeling content with their experience. In his view, this entailed providing witnesses the time and support needed to express their perspectives and ensuring that those perspectives were carefully considered by the judges. It also required actively listening to fellow inyangamugayo and giving due attention to their opinions. Beyond Laurent’s codified duties to listen to testimonies and render judgments, he believed he needed to facilitate a cathartic environment.
Laurent’s duties also illustrate two criteria of emotional labor: face-to-face contact with the public and producing an emotional state in other people (Hochschild Reference Hochschild1983). First, much of the inyangamugayo’s work involved presenting themselves to an audience, with little time for private reflection or space to express authentic emotion. Trials took place in public spaces such as classrooms or outdoor courtyards. Thus, for the majority of a gacaca workday, their every expression and gesture were on stark public display before an often-divided community deeply invested in court outcomes.
Second, producing an emotional state in other people was a fundamental requirement of the inyangamugayo’s job, as again, training had emphasized their role in aiding reconciliation and calming crowds. Just as Laurent saw it as his duty to ensure that all felt “content,” Esperance, a seventy-two-year-old woman who was targeted during the genocide, sought to raise victims’ spirits when they became distraught over sentences they perceived as overly lenient:
We convinced the defendant to confess, and we convinced the witness and the victim to drop evidence that was … baseless. We know victims were not happy. They were sad, but we could show them that we wanted to rebuild our country. We could show them that what we were doing was about reconciliation. I know that there were reservations, but we could show them that we wanted to rebuild a future for Rwanda.
As Esperance illustrated, the inyangamugayo strove to honor grief and to nurture reconciliatory attitudes—at least reconciliation as defined by the state, as we address shortly. Attending to both emotional processes were vital aspects of their work. Mathias, a forty-six-year-old man, explained these imperatives sequentially, saying, “We could sit as judges and community members, and those killers could come and confess. They tell the stories of how they killed people, then family members could forgive them. The punishment was not heavy … because the objective was to reconcile Rwandans.” Mathias and other judges thus aimed to generate feelings that could facilitate peace.
The inyangamugayo’s work also embodied Hochschild’s (Reference Hochschild1983) third component of emotional labor—their emotions were controlled by a supervising entity: the Rwandan state. In creating the positions and coordinating training and supervision, the state exercised a degree of control over the judges’ emotions. And while inyangamugayo retained some discretion over court proceedings (Doughty Reference Doughty2016), they were bound by the laws the government created and were also required to submit all court records to the state, which again had a vested interest in the outcomes of the trials (Chakravarty Reference Chakravarty2015).
Ferdinand, a sixty-year-old man, explained that the state forbade judges from displaying emotion during court proceedings out of explicit concern that they might be accused of bias or amplify trauma. He told us, “It was not acceptable to show your emotions before the public or the participant because your emotions could affect others.” He continued, “As a judge, I could try to hide my emotions because we were trained that we will face problems because people will get trauma.” Many others used similar language, noting that they were not allowed to show emotions and citing their state-run training, which did indeed emphasize such themes.
In sum, the inyangamugayo spent much of their time performing emotional labor and, specifically, emotion management. Trained, supervised, and monitored by the state, they felt compelled to induce feelings of reconciliation—again, a goal of gacaca and an emphasis of their training—among court attendees while also managing their own emotions. Thus, although their positions were unpaid, the concept of emotional labor appropriately captures an essential dimension of the inyangamugayo’s duties. Next, we turn toward documenting some of their specific intrapersonal and interpersonal emotion management strategies.
Intrapersonal Emotion Management Strategies
The inyangamugayo we interviewed told us that they called on three main cognitive strategies to manage their own emotions. These involved suggesting that (1) they were personally unemotional by nature; (2) their training prepared them for the emotional nature of the position; and/or (3) it was their duty to bear the emotion for the good of the community.
Beginning with the first strategy, inyangamugayo of all genders and former ethnicities said that they did not react emotionally to traumatic situations, with some proposing that this quality was why they were elected to serve. For example, Marc, a fifty-year-old man who was targeted during the genocide, explained that it was natural for him to remain cool and unemotional. “Nothing like that could raise emotions,” he said, “That’s my nature, and it is so hard for me to show my emotions.” Josephine, a forty-six-year-old woman who was also targeted, agreed that nothing could “excite” her, as she put it, even when she was reminded of the murder of her own family:
They [defendants] would tell those stories, and right there I saw that’s how my relatives were killed … Maybe he’s the one who even killed my family members. Again, the woman that they killed, we had a relationship, but a distant one. It was from my family but not close, not that close of a relative. You could be strong. You could be strong and listen to those stories. Nothing excited us.
Dorothee, a fifty-year-old woman, expressed parallel sentiments, emphasizing her impartiality, lack of sentimentality, and commitment to justice when listening to confessions:
There were sad stories. They could tell stories about how they killed the people you knew … They actually pronounced names of people that they killed … There were sad stories to listen to, but we had to. Again, concerning people that I knew or my friends who participated, we had to be impartial. It was all about justice. We had no emotions. We had no sentiments on deciding the cases.
Marc’s, Josephine’s, and Dorothee’s statements exemplify inyangamugayo’s claims that they could control their emotions. Their insistence also speaks to their continuing efforts to embody and perform the role of judge, and to exemplify the qualities that they believed led to their election. In other words, judges tried to bring their emotions in line with state and public expectations to remain calm and unemotional. Indeed, one of the core aspects of the judges’ training involved maintaining “integrity and impartiality, avoiding making decisions based on emotions or prior relationships” (Doughty Reference Doughty2016, 201). Telling themselves they were unemotional may have thus helped judges align their feelings with this job requirement.
Although this strategy was widely employed, nearly all men respondents insisted that they repressed emotion during the trials. Many women judges said that they, too, remained impartial, but men were most apt to deny any emotional influence. This likely reflects the expression of gender norms in Rwanda, where the masculine ideal encourages men to be a constant symbol of strength and protection (Burnet Reference Burnet2012; Hogg Reference Hogg2010). Mathias, a forty-six-year-old man, reported experiencing nightmares from the narratives of torture and slaughter. Nevertheless, he made clear that he “stood strong” at the trials:
Killers could come and confess that … “we put the whole family in the house and then we lit a fire and burned them in the house… After killing them we used the same machetes to slaughter cows that we ate … ” We could not understand those stories … After going home, you could have nightmares because of the stories, but at the bench, I could stand strong because we were there to listen to those stories and then try cases.
Second, many inyangamugayo maintained that their training, typically only a few weeks in duration, sufficiently prepared them to hear stories of genocide. For example, Leonidas, a forty-six-year-old man who was targeted during the genocide, remarked, “we were given training to calm our emotions so that … we shall manage the situation, not become too emotional or sentimental.” Colette, a fifty-three-year-old woman who was also targeted, similarly noted, “We were trained so that we were not allowed to show emotion.” Martin, a fifty-three-year-old man, elaborated on how this training helped him manage his anger:
it is like we were forced to hide our sentiments or our emotions. That [training] we had before helped us very much because someone could come testify how they killed someone, how he hunted someone, how he speared him or how he shot him with an arrow. That is something that actually made you angry but because of the trainings that you had before, we could behave … We try to manage the situation.
Similarly, Josee, a thirty-eight-year-old woman who was targeted during the genocide, explained, “by the fact that we had trainings on trauma, I could handle it, I could manage my emotions.” Like the notion that they were unemotional by nature, the training seemingly provided a cognitive mechanism to cope with emotional burdens, irrespective of its substantive content. As such, state organizational demands almost certainly shaped this strategy, perhaps due to the absence of status shields. To be clear, however, we emphasize how the judges told themselves they could handle the emotions invoked by the trials, rather than their actual ability to attend to their psychosocial well-being in the face of disturbing content.
Women were also more likely than men to invoke training during our interviews. This may reflect their anticipation of gendered public assumptions about their eligibility or capacity to serve as a judge. Many women respondents shared that they were initially hesitant to step into a role that had traditionally been held by men, as precolonial gacaca courts were only convened by men and as men were more likely to work as legal professionals (Clark Reference Clark2010; Mukandori Reference Mukandori2014). Women inyangamugayo may have consequently cited their training to manage others’ impressions of them as strong and capable.
Finally, a third major cognitive strategy involved bearing emotion for the community’s greater good, hence reframing a hard situation. For example, Innocent, a forty-seven-year-old man who was targeted during the genocide, explained that hearing testimony about graphic violence “was very difficult,” but “because of its importance, because it was useful to the Rwandan society, we committed ourselves and we did it.” Similarly, Grace, a fifty-six-year-old woman who was also targeted, stated “as days went on, [hearing] those stories, we had to stand firm and continue with gacaca to make sure that we give, we help in healing other people.” Both Innocent and Grace said they found strength in the importance and meaning of the task, mirroring the statements of many others.
This strategy was likely shaped by state-provided training and other state messaging, which emphasized how serving as a judge served the country by aiding reconciliation (Chakravarty Reference Chakravarty2015; Doughty Reference Doughty2016). While men and women judges relayed using this strategy, men were especially likely to reference a patriotic sense of duty, consistent with gendered expectations in Rwanda (Williamson Reference Williamson2016). For instance, Mathias expressed his sense of duty by noting, “We had to help our country get peace; we had to help our country unite Rwandans.”
Taken together, these three cognitive intrapersonal emotion management techniques—claiming an unemotional nature, drawing on formal training, and believing in a superseding mission to reconcile their communities—were pervasive in inyangamugayo’s accounts of managing their emotional states. These gendered strategies reflect, at least in part, state-driven narratives that many inyangamugayo internalized through their election and training, which they then embraced as a means to maintain emotional stability in the face of potentially debilitating emotional labor.
Interpersonal Emotion Management Strategies
As the judges managed their own emotions, their positions also entailed managing the emotions of others. Although intrapersonal emotion management appears to have been largely guided by organizational imperatives and gender norms, we find that interpersonal emotion management strategies were additionally guided by former ethnic identities. Because the genocide unfolded along ethnic lines, former ethnic boundaries were particularly rigid (Alba Reference Alba2005). In the aftermath of the violence, this boundary manifested when the judges attempted to manage the emotions of others during the trials. As gender also seems to have shaped this form of emotional labor, we address the interpersonal management strategies by gender and ethnicity.
Men and Women Who Were Not Targeted During the Genocide
Many of the twenty-three men we interviewed who were not targeted during the genocide said they felt an obligation to persuade defendants to confess their crimes and, relatedly, to publicly condemn their actions, share information about what they did, and apologize. Confession was not required, though it was highly incentivized with lighter punishments. Government training also painted confession as a cornerstone of the gacaca process that aided reconciliation by airing the truth and providing information about the remains of loved ones (Clark Reference Clark2010; Bornkamm Reference Bornkamm2012).
In line with this, Gaspard, a sixty-nine-year-old man, described how he would “challenge” defendants to take accountability:
I could sometimes get annoyed, and I even try to teach them, show them how they should tell the truth. Then after challenging them, that person … could speak, could open up and tell you everything that happened. Managing those stories was not an easy thing.
Gaspard publicly displayed annoyance toward the defendants as a way to manage their emotions and, hopefully, to get them to confess and feel remorse. Laurent, fifty-three, similarly encouraged defendants to confess, stating, “We could listen to those stories, but again, we had a responsibility of first meeting the defendant. We tell him in isolation how confession would be helpful and important to that person and explain to that person that failure to confess will lead to problems.”
Like Gaspard and Laurent, men judges who were not targeted presented themselves as being principally concerned with those who committed violence. Many discussed how they displayed anger toward defendants, which in turn might have impacted the defendants’ emotions and actions during the trial. Such efforts were often tied to encouraging confession, which the government painted as key to reconciliation in what Fox (Reference Fox2021) terms the reconciliation formula of perpetrators confessing and then asking for and receiving forgiveness for their wrongdoing. Such displays may also have reflected efforts to manage the emotions of the community members in attendance by demonstrating awareness of their group’s positionality during the genocide—especially since the majority of people who committed violence were Hutu men.
Like the men, the eleven women judges who were not targeted also emphasized managing defendants’ emotions tied to confession. For instance, Belancilla, a seventy-two-year-old woman, shared, “the important thing was sensitizing people to confess.” Unlike the men, however, they often focused on the defendants’ feelings and their views of the trials. When Flora, forty-seven, suspected witnesses were bringing false or unsubstantiated testimony, she refused to punish defendants, noting that the witnesses “did not even know these defendants” and sharing a lengthy story about how it was important for the defendants to know they would have a fair trial.
This emphasis on defendant security may stem in part from displays of empathy, as many judges who were not targeted had family ties to defendants. For example, Consolee’s father and father-in-law were each found guilty of genocide. She spoke about encouraging them, and other defendants, to “be fair and try to confess” to obtain a lighter sentence—an appeal to both reason and emotion. Consolee and other women judges who were not targeted displayed empathy in expressing concern for potential injustice, and they worked to manage the emotions of defendants (and, likely, their own families and communities).
In sum, the emotion management of men and women judges who were not targeted often focused on defendants. Whereas the men routinely described displaying anger and annoyance against those who committed violence toward obtaining a confession and remorse, women were more likely to display empathy as they centered the emotions of defendants and their families. These differing strategies may reflect institutionalized gender norms that provide different scripts to neutralize “in-group” responsibility for the genocide. For these women, it was acceptable to express concern for defendants, and likely their relatives, to induce feelings that they were fairly treated. For men, the most appropriate emotional display was indignation and anger toward defendants, to induce feelings of shame and guilt alongside a confession.
Men and Women Who Were Targeted During the Genocide
We spoke with a total of twenty-three men judges and twenty-two women judges who were targeted during the genocide. In contrast to the judges who were not targeted, these inyangamugayo emphasized “standing strong” and providing relief to victims’ families. They consequently prioritized managing the emotions of fellow Rwandans who were targeted. For example, Nathanael, a forty-four-year-old man who was targeted during the genocide, described how he calmed a victim’s family after the defendant confessed. He explained that it was not easy “listening to someone confessing the killings he did. First of all, we could calm down the person who lost his family,” later emphasizing how he wanted the family to feel “some relief.” Fifty-four-year-old Papias similarly said he needed to “stand strong” for those who had borne the brunt of the violence. As he shared, “there were sad stories. They could take us back to 1994. Someone could stand there and tell stories of how he killed people…. we had to stand strong for those who were emotional, those who could not control their emotions.”
Consistent with Papias’s phrasing, standing strong for others who were distraught was the most salient theme in our conversations with men judges who were targeted during the genocide. In this vein, Innocent, forty-seven, focused on comforting survivors and helping them “deal with” defendants and the process of reconciliation:
Of course, some of them [survivors] were not happy with … the light punishments that we could give. But we could take our time and teach them that we are in the process of reconciliation. Putting people in prison will not help bring back their people, so we had a task of teaching those people there. Some of us were even victims who could tell them that even some of these ones killed our people, but we are now teaching you to be with them so you have to take this, you have to digest this. We have to deal with them, we have to reconcile.
Like the men judges who were not targeted, men judges who were targeted typically drew on notions of ideal Rwandan masculinity to convey strength. They emphasized how they attended to the emotions of survivors, however, rather than those of defendants. As such, they sought to engender feelings of forgiveness, reconciliation, and emotional release. In doing so, they also extended notions of masculinity given that traditional conceptions of “strength” in the Rwandan context had been tied to physical strength. In fact, the word for man (umugabo) is also a synonym for strength, though a more subtle, emotional strength tied to vulnerability has slowly been emerging since the genocide (Williamson Reference Williamson2016).
Women judges who were targeted during the genocide likewise attended to distraught survivors. For example, Esperance noted the urgency of “showing” the overriding goal of reconciliation to survivors who were upset by light sentences. Likewise, Grace, a fifty-six-year-old woman who was targeted, repeatedly told us she “stood firm” during sad stories so other survivors could learn what happened to their families, which she saw as key to healing:
We had to stand firm … because we had friends and relatives who we did not know how they [victims] died. In those stories, you could know how they died … We had to stand firm and … help in healing other people.
Women judges who were targeted also oriented to the emotional needs of defendants and their families, often by trying to avoid appearing unduly punitive. Like the women judges who were not targeted, some of these women judges also emphasized the need to reassure defendants and their families that the trials were fair—though sometimes in ways that would be uncommon, or at least not openly discussed, in adversarial justice systems. For example, Epiphanie, sixty, explained how she “never had emotions” to imprison all (former) Hutu, citing cases in which she worked to acquit the innocent. To illustrate, she described how she spoke on behalf of her accused landlord:
I never had emotions of wanting all Hutu to be in prison. I even tried to make sure some of them were innocent and acquitted. I never had those sentiments. I can give you an example of one. He was my landlord during the genocide. I know that person served many people … During the gacaca many people wanted him to be convicted. I tried to make sure that person is acquitted.
Likewise, Clementine, fifty, encouraged confession to ease emotional burdens, noting that some defendants “confess and then they cry. They weep in public. They could say that they are now free because they have actually told the people what they did.” Felicite, a fifty-nine-year-old woman who was targeted, similarly emphasized showing leniency, explaining, “When someone confessed from the bottom of his heart,” she said, he need not “die in jail … the country needs his arms … to rebuild again.”
In sum, women who were targeted underscored meeting the emotional needs of those harmed by the violence and those who lost family members. However, they were also concerned with showing that they could manage their emotions, remain impartial, and pursue fair rather than unduly harsh punishments. This work may have countered stereotypes of women as being overly emotional, as Rwandan women are often viewed negatively when they are perceived as excessively emotional (Burnet Reference Burnet2012, 44). Women from the victim group may have also been especially invested in emotional displays that helped convince defendants and their families they would be treated fairly.
DISCUSSION AND CONCLUSION
Given the precarity of societies in the immediate aftermath of mass violence, scholars are keenly interested in the emotional costs and benefits of transitional justice efforts (Brounéus Reference Brounéus2010; Cilliers, Dube, and Siddiqi Reference Cilliers, Dube and Siddiqi2016). In line with this research, this article has considered emotional labor (Hochschild Reference Hochschild1983) within a transitional justice context. Specifically, we examined the intrapersonal and interpersonal emotion management strategies Rwandan judges employed in adjudicating crimes of genocide, extending prior scholarship by emphasizing the emotional labor of those implementing transitional justice mechanisms.
Our case study of Rwanda’s inyangamugayo first documented emotional labor demands and then investigated the judges’ intrapersonal and interpersonal management techniques. With respect to how the inyangamugayo managed their own emotions, the judges discussed three primary cognitive strategies: (1) maintaining that they were (and are) unemotional by nature, (2) reminding themselves and others about their formal training, and (3) ascribing to a superseding mission. These strategies notably parallel three key facets of emotional labor: (1) face-to-face contact with the public, (2) requirements to produce emotional states in others, and (3) a degree of centralized control over their own emotional activities.
The judges’ face-to-face contact with the public clearly shaped the emotion management strategies they shared with us. Their efforts to present themselves and their life-changing decisions in an unemotional manner, in the face of highly disturbing and emotional evidence, was directed toward producing desired emotions in others. Moreover, their repeated references to state training and the broader mission of their work reflect the centrality of state control over their emotion management. Indeed, the state instructed the judges to be unemotional and emphasized the mission of gacaca as bringing justice and reconciliation via a specific “reconciliation formula” (Fox Reference Fox2021).
Intrapersonal emotion management strategies that the judges relayed were also gendered, as men more commonly expressed that they were unemotional and discussed their duties in patriotic terms, likely due to gendered assumptions about masculinity in Rwanda (Hogg Reference Hogg2010). By contrast, women were more likely to reference formal training, likely due (at least in part) to the paucity of women in such roles and the gendered nature of positions of power in Rwanda (Burnet Reference Burnet2012; Mukandori Reference Mukandori2014).
Interpersonal management strategies that the judges discussed varied both by gender and by whether the judges were targeted during the genocide. Inyangamugayo who were not targeted focused on defendants, with the men displaying anger and annoyance to induce remorse and confessions, and women displaying empathy to calm concerns about fairness. In contrast, men and women inyangamugayo who were targeted generally focused on lifting the spirits of survivors and their families, though women who were targeted also strove to avoid appearing unduly punitive as a technique to manage the comfort of defendants and their families.
Other work has illustrated how gender impacts emotion management techniques in legal and other professional settings (for example, Pierce Reference Pierce1999; Berheide et al. Reference Berheide, Carpenter and Cotter2022), though we believe we are the first to highlight how conflict-related identities are associated with interpersonal emotional management during transitional justice. These different strategies indicate that the identities that were salient during the genocide shaped interpersonal emotion management in its aftermath. As such, in other post-conflict settings, we would expect the identities tied most tightly to the conflict (e.g., ethnicity, religion, political affiliation) to shape emotional labor during transitional justice processes.
While we have considered intrapersonal and interpersonal strategies separately, future work could assess their relationship, as well as how community connections may have affected emotion management. Future research could also determine whether and how the techniques we found impacted sentencing, as well as whether judges were surface acting (e.g., faking emotional displays) or deep acting (e.g., actively trying to change internal emotional states). This work could likewise assess the relative frequency of techniques across a broader sample, as well as whether and how techniques shaped opinions of the courts (Roach Anleu and Mack Reference Roach Anleu and Mack2005). Additionally, while our research highlights how conflict-related identities and gender shape emotional labor during transitional justice processes, future research should investigate the impact of other social identities, such as socioeconomic status (Orzechowicz Reference Orzechowicz2008). Finally, further work is needed to understand points of similarity and difference between the volunteer judges who preside in transitional justice settings and the professional judges discussed in extant literature.
Such research might also investigate whether judges presiding over particularly gruesome cases experienced distinct forms of emotional hardship (Hagan and Kay Reference Hagan and Kay2011). Life-course studies tracking the trajectories of other legal professionals identify job satisfaction and workplace control as key determinants of depressive symptoms (Plickert, Kay, and Hagan Reference Plickert, Kay and Hagan2017). Similar longitudinal research on the effects of transitional justice work conditions could yield important insights regarding best (and worst) practices for structuring this important work, especially in situations where the absence of enduring status shields may lead to increased risk. Such research could aid in developing institutional resources to help judges cope with the enormous burden they carry, which may improve the quality of transitional justice as well as the work lives of those who implement it.
