INTRODUCTION
In February of 2013, a young woman named Taringa gave birth to a baby boy in Gorongosa, a rural district in the center of Mozambique. The district was the epicenter of the postcolonial civil war (1976–1992) and periods of severe drought and famine (1988–1992), which were never completely resolved. At the time that Taringa gave birth, she was about twenty years old, according to her close relatives (Taringa's father, Furatira, and her mother, Robina). Her parents remembered that she was born around the end of the Mozambican civil war in 1992, a year of acute drought and famine.Footnote 1 In April 2013, around two months after giving birth, Taringa began inexplicably consuming large amounts of maize meal porridge. When her family's maize stock was exhausted, she began invading the neighbors’ houses to seize theirs. Taringa entered people's houses stealthily, which compelled neighbors to hide their food stocks. She stopped breastfeeding her baby, became careless, and had difficulty sleeping since she often wanted to cook and eat in the middle of the night and at dawn. She stopped working on her farm and during the day she would disappear into the bush; the family followed her to ensure that she was not assaulted for invading people's properties. Her relatives cared for her baby due to her constant wandering, and neighbors complained about her invasions of their homes and unauthorized consumption of their maize meal.
It has been acknowledged that when people face unsettling events, we often use “past experience to render stimuli into a form sufficiently familiar that we can attach some sense to them” (Cohen Reference Cohen1985: 99; Sahlins Reference Sahlins1985: 154). In Sub-Saharan Africa, healers and oracles of all persuasions are known to specialize in negotiating and creating meaning for peculiar events and experiences (Calvão Reference Calvão2017; Geschiere Reference Geschiere1997; Green Reference Green1997; Peek Reference Peek1991; Stroeken Reference Stroeken2010). When social conflicts and bodily afflictions persist, they can be accompanied by disturbing spirits and spirit possession episodes which the people involved claim to be caused by unresolved past violence and violations (Boddy Reference Boddy1989; Honwana Reference Honwana, Meyer and Pels2003; Igreja Reference Igreja, Stan and Nedelsky2013; Lambek Reference Lambek1981; Masquelier Reference Masquelier2001; Stoller Reference Stoller1995). In such cases, local specialists are solicited as part of attempts to identify the alleged culprits through diverse bodily rituals whereby the tripartite interactions of body-spirit and audience play center stage (Igreja Reference Igreja, Wallis, Kent, Forsyth, Dinnen and Bose2018a; Masquelier Reference Masquelier2001; Stoller Reference Stoller1995).
Taringa's family and several members of her community followed a similar path. They searched for the expert support of healers and oracles, known in the Gorongosa vernacular as nahana or nyanga.Footnote 2 These can be madzoca (who use ancestral spirits) and gamba healers (who employ spirits of soldiers who died in the civil war, known as gamba spirits).Footnote 3 Yet the various healers they consulted were unable to provide adequate responses. Taringa continued to roam about searching for and grabbing people's food stocks and ate without satisfying her cravings, and without displaying symptoms of any locally recognized affliction.Footnote 4 In this context, her actions were regarded as singular, partly because they differed from the recent history of practices of dealing with serious bodily afflictions in the region. In the first two decades following the peace accords (1992–2002), the ritual actions of gamba spirits dominated the local and regional ways of conveying disturbing legacies of the civil war (Igreja Reference Igreja2015a; Marlin Reference Marlin2001).
In these rituals, the gamba spirits often acted in a conventional manner by intrusively taking possession of the bodies of the living and disrupting the host's health and relationships. These intrusive acts were locally interpreted as being part of struggles to hold accountable alleged perpetrators of serious violations during the country's civil war (Igreja Reference Igreja, Sriram and Pillay2009; see Marlin Reference Marlin2001). Young women, in particular, were a focal point of the gamba spirits’ intrusions, and these spirits sometimes evolved to use the hosts’ bodies to become gamba healers (Igreja Reference Igreja2014; Marlin Reference Marlin2001). In a state of gamba spirit possession, the spirit's hosts reenacted violent wartime events and indicted individuals allegedly implicated as perpetrators of violations in family and community circles. By “violations” here I refer to neglectful behavior that led to the deaths of people in vulnerable predicaments, and to betrayals, forced disappearances, and violence that led to murders and the abandonment of corpses to rot in public spaces. Community courts reinforced the role of gamba spirits by agreeing to hold hearings involving such spirits and to address the spirits’ charges (Igreja Reference Igreja2010). Such accusations largely placed the burden of proof on the spirit, who spoke through the human host in a state of possession. The ensuing deliberations among the participants in ritual sessions, mediated by a gamba healer, often contributed to further disclosures of violations, admissions of responsibility, and the carrying out of payback (Igreja and Racin Reference Igreja, Racin, Meier and Steinforth2013; Marlin Reference Marlin2001). Recently, with the introduction of electricity and mass media technologies in the region, some oracles, known as paza and specializing in sensing the invisible, introduced television as part of their oracle divination apparatus (Igreja Reference Igreja2015b).Footnote 5
Taringa's actions did not fit into these locally conventional scripts or expected ways of bodily conveying past unresolved conflicts. Her actions were singular also because they lacked conceptual representation and manifested as fragments (Deleuze and Guattari Reference Deleuze and Guattari1987; McMahon Reference McMahon and Stivale2005). Her actions did not signal links to troubling memories or seem to indict anyone, as had occurred during rituals of the gamba spirits and oracle television. Absent was the kind of “momentarily shared reality” that, since the writings of Emile Durkheim and Erving Goffman, has been known to typically occur during rituals (cited in Collins Reference Collins2014: 7). Taringa was conscious all the time and she could recount the number of times she wandered, cooked, and ate. Yet she could not explain her actions. As she told me on one occasion, “I feel hungry and I start moving around in search of maize to cook and eat, but I don't know why I do these things.” Despite her invasions of people's houses, she was not locally regarded as a thief, nor was she considered to be suffering from a disturbed mental state.
In other communities facing aftereffects of civil wars, it has been observed that some individuals, categorized by local audiences as being peculiar and mad, might still be bestowed a voice because of their male and senior status and because the stories they tell are related to the community (Bertelsen Reference Bertelsen, Broch-Due and Bertelsen2016: 204–5). Taringa was still a young woman and without noticeable status other than trying to become a mother who could nurture her first child to survive. It was not apparent that she had social stories or narratives to tell because her parents remembered that she was born around the end of the civil war and so had no “autobiographical memories” of the war or the famine that had ravaged the region (Olick Reference Olick1999: 335). It is known that in famine emergencies various risk factors are always in play (Devereux Reference Devereux and Devereux2007; Newitt Reference Newitt1988). Yet Taringa's parents and the neighbors singled out her actions as the main triggers of a serious risk of famine, or as they expressed it, “Indine ku gopa ndzala”—“We are afraid of famine.”
The perception of risk and insecurity created the context that gave rise to a multiplicity of questions, collective imagination and negotiations, and various scattered testimonies that “transcended” Taringa's bodily experiences (Deleuze Reference Deleuze1997) and became constitutive of and vehicles for contested evidence of past violations perpetrated during the 1980s famine in Gorongosa district. Specifically, in the 1980s and 1990s, Taringa's father, Furatira, allegedly committed violations against some people who were vulnerable and debilitated due to the war and famine. One case was of a young, adopted boy who, under the guardianship of Furatira, disappeared and was never found. Another was a frail woman and her two children who Furatira cared for but then hastily abandoned. Soon after, the woman and one of her children died and were left to rot in the middle of the road. The whereabouts of the second child remains unknown. Locals deemed these actions acute violations of norms of respect (translated in Chi-Gorongose as makodzo), but they had hitherto been silenced and concealed. Using local notions of the body and kinship (dzindza) and following lengthy denials and negotiations, Furatira's violations were considered plausible explanations, at least in part, for Taringa's predicament. Thus, her singular actions, coupled with the various local efforts, contributed to sustaining the struggle to counter the silence and impunity around these acts, and ultimately to cementing memories of collective guilt in the landscape. The people's struggles for justice, as presented here, exceed global accountability and transitional justice practices and highlight the meaningfulness of unexpected bodily actions, collective imagination and negotiations, changing subject positions, and kinship ties as sources of knowledge of violations perpetrated by ordinary individuals in regions beleaguered by war and famine.
To capture the dynamics of these sorts of legacies of war and famine and people's attempts to achieve resolution, I employ the concepts of “famine continuum” and “embodied accountability.” I analyze the conditions under which singular events can unexpectedly become catalytic in relationships transitioning from silence, denial, and alienation to voice, social engagement, and acceptance of guilt. While cross-cultural comparisons are important, equally significant is an inward analytic comparison of different embodied accountability practices to grasp their changing qualities over time in specific locations (see Akin Reference Akin2003).
FAMINE CRISIS, BROKERS OF LIFE AND DEATH, AND THE FAMINE CONTINUUM
In conflict-ridden communities in Africa wracked by civil wars, droughts, and famines, people do not die simply from a lack of food, water, or adequate health services, although the global news media often emphasize such causes as “famine newsworthy” to news consumers in the United States (Fair Reference Fair1996) and other industrialized nations (Ebron Reference Ebron2002). Following a famine crisis, the media's interpretations and reporting often reinforce Westerners’ sense of plenty, which justifies their attitudes of benevolence toward the poor in peripheral worlds (Baptista Reference Baptista2017). In this regard, the notion of “famine crimes” has been useful to presenting the ample evidence of how people in famine crises (e.g., in Ethiopia, Eritrea, Nigeria, Niger, Somalia, Uganda, Darfur, Yemen) were depleted and died because brokers of life and death, such as state officials and rebel groups, used famine as a weapon of war and control (Blouin Reference Blouin2008; de Waal Reference de Waal2006; Devereux Reference Devereux and Devereux2007; Keller Reference Keller1992; Macrae and Zwi Reference Macrae and Zwi1992; Marcus Reference Marcus2003).
The problem with the notion of “famine crimes” is that it remains anchored in realist visions of politics by concentrating on the perfidious and fatal actions of government officials and armies and insurgent groups, and on the negotiations and tactics that international humanitarian actors employ to deliver food to suffering populations in conflict zones (Blouin Reference Blouin2008). A number of anthropologists have observed that in civil wars violence is not only undertaken by national and regional rival military groups in conflict, but is also significantly shaped by local social tensions and micropolitical goals, defined and operating at the community and family levels (Lubkemann Reference Lubkemann2005: 495; see Argenti-Pillen Reference Argenti-Pillen2003; Kalyvas Reference Kalyvas2006). That said, anthropologists have not sufficiently analyzed similar dynamics in contexts of famine, while transitional justice specialists (Crawford Reference Crawford2013; Fry Reference Fry2016; Lang Reference Lang2007) have ignored serious violations that also occur during famines. This article addresses these gaps in both the anthropological and transitional justice literatures.
I therefore propose the term “famine continuum” to refer to various actions that diverse people deploy to endure collective famines, the ambivalence of some of these survival actions, violations that occurred among the people directly and indirectly involved, and the consequences that haunt famine survivors for decades afterward. The concept also alludes to the diversity of actors within and interpretations of collective famine events and experiences.
Traces of this diversity of participants and contradictory actions in the famine continuum can be found, for instance, in Maoist China, where in the wake of the Great Leap Forward—a period of reform that led to famine and death—people's survival tactics did not uniformly save lives, since they included coercion, systematic violence, and lethal violence (Dikotter Reference Dikotter2010). Further, in the aftermath of these grisly experiences, their repercussions did not fade away with pacification and the restoration of food production cycles. In the post-Great Leap Forward period, the survivors could find no solace or closure because those who “died of hunger or thirst in the famine became wild ghosts” and disturbed some of the survivors in profound ways (Mueggler Reference Mueggler2001: 194). Comparative analysis of the literature shows that the adoption of peace agreements, and public appeals by political and religious elites for silence, appeasement, and reconciliation do not efface the legacies of violent conflicts since these legacies endure in complex and fragmented ways (Comaroff and Comaroff Reference Comaroff and Comaroff1992; Kleinman and Kleinman Reference Kleinman and Kleinman1994; Perera Reference Perera, Das, Kleinman, Lock, Ramphele and Reynolds2001; Stoller Reference Stoller1995; Wiegink Reference Wiegink2015). These complex fragments can be obscured by conventional procedures, such as formal transitional justice processes in which survivors are allowed to bear witness to violations and eventually hold accountable alleged perpetrators (Krog, Mpolweni, and Ratele Reference Krog, Mpolweni and Ratele2009).
By drawing attention to a complex set of reverberations of famine events and experiences over time, this study underscores the idea of the “violence continuum” (Scheper-Hughes and Bourgois Reference Scheper-Hughes and Bourgois2004; Whitehead Reference Whitehead, Whitehead and Finnstrom2013; Wilson Reference Wilson2001) rather than conceptions of transition founded upon clear endpoints, as evoked by popular metaphors such as “closing the books” (Elster Reference Elster2004), or by certain powerful groups in transitional societies that stress the need to erase the past as a precondition to moving forward (Igreja Reference Igreja and Russell2018b). Along these same lines, it is no surprise that in the post-peace accord (1992 onwards) dispensation in Mozambique many survivors in Gorongosa wished to forget and move on by engaging in food production. Their everyday relationships were shaped by attempts at appeasement and lack of acknowledgment of war and famine-related violations in which some civilian survivors were implicated. The exceptional interruptions of the silence were marked by the intrusive eruptions of gamba spirits. Yet, over time, Taringa's bodily actions, because of their singularity, triggered collective fears, insecurity, and memories of the famine that had never been completely resolved. In turn, some people attributed “many potential meanings” (Whitehead Reference Whitehead, Whitehead and Finnstrom2013: 35) to her unexpected bodily actions.
A focus on the myriad potential meanings here provides us with access points from which we can examine people's multifaceted ways of expressing legacies of conflicts and dealing with deep-seated anxieties, even if they may initially be incomprehensible, as with Taringa's case. While her singular actions might have been dismissed due to the dearth of explanations for them, over time they created a space from within which other triggers and brokers of life and death in the famine continuum could be understood both locally and by external observers.
SOCIAL PRACTICES, SINGULAR EVENTS, AND IMAGINATIVE ACTS
In parallel to transitional justice specialists (Lang Reference Lang2007; Van der Merwe, Baxter, and Chapman Reference Van der Merwe, Baxter and Chapman2009), anthropologists and sociologists have analyzed violence often in contexts of repressive regimes and mass political violence (Argenti-Pillen Reference Argenti-Pillen2003; Lubkemann Reference Lubkemann2005; Nordstrom Reference Nordstrom1997; Taussig Reference Taussig and Dirks1992; Whitehead Reference Whitehead, Whitehead and Finnstrom2013). Yet there is a need in both fields to expand the contexts of analysis to include the dynamics of human involvement in the famine continuum. Unlike transitional justice specialists, anthropologists and sociologists take for granted the idea that “different cultural orders have their own modes of historical action, consciousness, and determination” (Sahlins Reference Sahlins1985: 34; see Sewell Reference Sewell2005). Ethnographic descriptions and analysis of national and local modes of historical consciousness can be as diverse as “visual depictions and material embodiments” (Zubrzycki Reference Zubrzycki2017: 5), routinized bodily actions (Comaroff Reference Comaroff1985; Csordas Reference Csordas and Csordas1994; Farnell Reference Farnell2012; Kidron Reference Kidron2009), rituals and testimonies conveyed by spirits and spirit possession (Argenti-Pillen Reference Argenti-Pillen2003; Boddy Reference Boddy1989; Lambek Reference Lambek1981; Masquelier Reference Masquelier2001; Stoller Reference Stoller1995), and memorials and myths (Basso Reference Basso1996; Levinson Reference Levinson1998). People around the world use these diverse modes to convey local and national political attachments, the legacies of collective violence, secular and religious sentiments, and historical and contemporary experiences.
In general, ethnographers in their studies have been concerned “less with events than with meaningful practices” (Comaroff and Comaroff Reference Comaroff and Comaroff1992: 37). Yet several anthropologists and sociologists have also grappled with the role of unexpected events (Sahlins Reference Sahlins1985; Sewell Reference Sewell2005) and have argued that they can either trigger dramatic change or reproduce and reinforce the social order (Sahlins Reference Sahlins1985; Sewell Reference Sewell2005). Such events may also generate innovations that do not fundamentally change the social order, as aptly suggested by Barrington Moore (Reference Moore1993 [1966]: 315). In this article, I pursue this question further by analyzing how people in Gorongosa searched for tools to navigate a predicament that defied “definition, measurement, and coherent representation” (Parr Reference Parr2008: 12).
Deleuze and Guattari (Reference Deleuze and Guattari1987) proposed giving attention to singular events as a way to capture the array of knowledge sources people draw upon to categorize specific problems. In this approach, the lack of conceptual representations available to label a certain problem or condition shifts the analytical focus to the various pieces of the problem. The shift is an important step, given that such pieces can be imbued with potential meanings that transcend them individually (Deleuze Reference Deleuze1997: 57). Deleuzians such as McMahon have added that a singular event that exceeds our representational capacities can also provoke “the exercise of all our powers, creating a relay between sense, memory, imagination and thought” (Reference McMahon and Stivale2005: 47).
The spark of the imagination in this context is different from fictions or fantasies as defined in psychological studies. Following Severi (Reference Severi and Lloyd2015: 22), imagination here forms the context that guides inference. In this case, imaginative acts can be regarded as “to some degree unfinished, teasingly schematic in such a way as to, almost perforce, mobilize the imaginative powers of the recipient to fill in the blanks” (Graeber Reference Graeber, Severi and Lloyd2015: xv). The outcome that the relay between sense, memory, imagination, and thought might generate—that is, transformation or reproduction of the social order (Sahlins Reference Sahlins1985; Sewell Reference Sewell2005), or innovation without change (Moore Reference Moore1993 [1966])—cannot be predicted. Considerable time and intellectual effort are required to grasp the diverse ways in which people struggle and negotiate to live with an enigmatic problem (Lefebvre Reference Lefebvre2005: 111).
When considering how analysts should best invest their time and intellectual efforts, Deleuze argued that the individual, and the solitude of philosophical contemplation, are the primary site and mode of initiating analyses of singular events, writing, “We exist only in contemplating” (Reference Deleuze and Patton1994: 74). However, new ideas, analyses, and theories of the social field need to be situated within specific historical conjunctures and geographical terrains (Highmore Reference Highmore2002: 18). Thus, unlike Deleuze's claim for the centrality of the individual philosopher, instances of vita activa (Arendt Reference Arendt1958) (or doing with one's body and hands), as part of collective modes of production and reproduction, are also crucial for comprehending the potential of singular events in specific locations. This article builds upon and expands these theoretical reflections on the ways people respond to singular and adverse circumstances, their evocations of fragments of past traumas, and mobilizations of collective imagination. I begin by presenting the cultural and historical contexts that inform family organization, relationships, and processes of conflict, resolution, and change in Gorongosa.
CULTURAL AND HISTORICAL CONTEXTS
Like many rural districts in Sub-Saharan Africa, Gorongosa is politically, legally, and religiously diverse. Everyday life is shaped by community and state institutions and actors in tandem with a self-sustaining form of agriculture. The sociocultural organization revolves around patrilineal kinship, polygyny, and monogamy. Most people speak “Chi-Gorongose,” which is a mixture of regional vernaculars, and there is a minority of Portuguese speakers. Since the signing of the Mozambican peace agreement in October 1992, the Gorongosa region has enjoyed some economic progress (Igreja Reference Igreja2015b).Footnote 6 Yet long-term studies of the region indicate that, for large segments of the population, unbridled poverty, unemployment, and intractable strife related to past unresolved violent conflicts have remained grave challenges (Bertelsen Reference Bertelsen, Broch-Due and Bertelsen2016; Igreja Reference Igreja2015a; Marlin Reference Marlin2001; Wiegink Reference Wiegink2013). Local survivors and their offspring have faced intractable conflicts that have encroached upon their social relationships, heightened collective insecurity, and reinforced the perception that the transition from the civil war to peace is incomplete (Igreja, Dias-Lambranca, and Richters Reference Igreja, Dias-Lambranca and Richters2008).
My analysis draws broadly on community research in Gorongosa that I initiated in 1997 and which has evolved over time. My research has been anchored in the idea of an “engaged and self-reflexive” process (Whitehead and Finnstrom Reference Whitehead, Finnstrom, Whitehead and Finnstrom2013: 11). I have taken seriously my interlocutors’ practices of knowledge production as well as their subjective positions and doubts, and responses to singular events. I have tried to make sense of my experiences there in light of what I came to know of the people over several years of fieldwork. The specific data analyzed in this article was gathered in the context of Taringa's actions, which emerged in 2013.
I first met the main protagonists in Taringa's case, particularly Furatira and his family, in 1997, prior to the public flare-up over it. I interviewed Furatira several times, and others in his area, about their experiences with the civil war and famine. I also interviewed his wife, his sister, and his mother Bandemba. Over time, I chatted with Furatira about diverse issues. Prior to Taringa's case, I had witnessed the resolution in community courts of intractable conflicts that Furatira had with his family and some of his neighbors.Footnote 7 Those encounters taught me also about several other disputes within his family that later became meaningful for my analysis of Taringa's case, particularly in relation to issues of forgetting and concealment (see Ricoeur Reference Ricoeur, Blamey and Pellauer2006).
As that case evolved, I interviewed Taringa, then in her early twenties, her husband Buraco (twenty-five), her father Furatira (fifty-five), who was a village chief, her paternal grandmother Bandemba (seventy), her paternal uncle Tovumba (also seventy), Taringa's mother Robina, and her sister. At the community level, my study involved an oracle healer and three community judges. The latter are lay individuals, men and women, chosen by the community to adjudicate their disputes (Igreja Reference Igreja2010; see Kyed Reference Kyed2009). I also talked with various community members who tried to categorize Taringa's singular actions.
I gathered oral life histories and conducted in-depth interviews and participant observation in multiple sites of accountability, namely people's homesteads, community courts, and the yard of an oracle healer. It has been rightly pointed out that during fieldwork “the ethnographer only ‘sees’ fragments of a cultural field” (Comaroff and Comaroff Reference Comaroff and Comaroff1992: 16), and sometimes it is not only ethnographers who are confronted with perplexing acts that “seem totally unconnected to the social context in which they are occurring” (Basso Reference Basso1996: 113). In the wake of Taringa's case, around thirty neighboring families were equally puzzled and encroached upon by a sense of insecurity, and they continually lamented Taringa's uncontrolled consumption of their food stocks. I recorded the arguments and counterarguments of the various participants. I followed up the various accountability processes by re-interviewing the protagonists in late 2013, June 2014, and January 2015. In the two latter periods my contact was limited due to the escalation of military confrontations in the region between, once again, the Frelimo government military forces and the armed faction of the Renamo party (Igreja Reference Igreja2015c; Wiegink Reference Wiegink2015).
Mapping Local Moral Bodies, Violations, and the Burden of Relationships
The social order in Gorongosa, as in many Sub-Saharan African communities, is shaped by visible and invisible forces, as well as the loose presence and logics of the modern state (Comaroff Reference Comaroff1985; Geschiere Reference Geschiere1997; Igreja Reference Igreja2014). Secularized notions of the body dominant in Western cultures consider bodies as physical entities stripped of agency after death. That said, sometimes even in the Western taxonomy notions of the body, life and death, persons, and objects become diffuse and processual through disputes over “quarries for transplantable organs” (Palmié Reference Palmié2006: 874). In Gorongosa, the body is a relational entity and mediator between past, present, and future temporalities and the visible and invisible forces that impinge on their social world. These local notions are consistent with the idea that the body encapsulates processes that are at once secular, political, and religious (Csordas Reference Csordas and Csordas1994). The body is a site of struggle and contestation in that the destruction of corporeal life opens spaces for the politicization of death (Clarke Reference Clarke2009: 28). Politics or disputes after corporeal life are possible because the spirits of the dead are social actors capable of meddling in bodily and community functioning through spirit possession (Lambek Reference Lambek1981).
The Gorongosas acknowledge two general types of contentious death. One is death caused by the failure of expected caregivers to provide adequate care for the victim. The other is the death of a person due to the breaking of taboos, such as taboos of witchcraft and cannibalism (Igreja Reference Igreja2014; see Geschiere Reference Geschiere1997; West Reference West2005), or through betrayal and physical violence leading to the spilling of blood and killing (ku puthua or ku zunguza) without moral legitimacy. Both types of death represent serious offences and, in cases where the culprits are unknown, the process of determining responsibility includes inspection of kinship ties (dzindza). That is because liability for violations is shared by family members of the perpetrator. Bloodshed and homicide pollute the earth and the actions of avenging spirits can endanger the entire community (Douglas Reference Douglas1980: 69). Yet, epistemologically, people cannot tell in advance when and how the spirits of the dead will avenge their deaths or how embodied accountability actions might unfold. These cultural notions of violations and death contrast starkly with the definitions of accountability proposed by the global accountability project, though that project is not monolithic.
ACCOUNTABILITY AND EMBODIED ACCOUNTABILITY
The political science literature on accountability distinguishes at least two types: answerability and enforcement (Brinkerhoff Reference Brinkerhoff2001, cited in Ribot Reference Ribot2002: 29). Here I will emphasize the former due to its parallels with the notion of embodied accountability. “Answerability” consists of “the obligation to provide information and explanations concerning decisions and actions” (Ribot Reference Ribot2002: 29). Central to this idea of accountability is the need to increase levels of scrutiny in state-society relations. This type of accountability has its roots in the Western notion of individual, autonomous moral agency. It is consistent with the indictment and prosecution of individuals regarded as being responsible for, or the masterminds of, large numbers of violent deaths and atrocities, mainly in contexts of dictatorships and modern wars (Crawford Reference Crawford2013: 7). Since the end of the Cold War, considerable attention has been given to this notion of accountability, sometimes expressed as “global legalism,” which is grounded in an excessive faith in international law's effectiveness in resolving conflicts (Posner Reference Posner2009: xii). Under this banner, international justice institutions have received and spent millions of dollars to prosecute a small number of individuals accused of brutal human rights violations and crimes (Clarke Reference Clarke2009; Fry Reference Fry2016; Hagan, Schhoenfeld, and Palloni Reference Hagan, Schoenfeld and Palloni2006; Van der Merwe, Baxter, and Chapman Reference Van der Merwe, Baxter and Chapman2009). This growing emphasis on international law has meant a neglect of the plurality of local expressions of the legacies of such violence, and of varied manifestations of, and ongoing shifts in, community justice mechanisms (Igreja Reference Igreja2010; Honwana Reference Honwana, Meyer and Pels2003; Nordstrom Reference Nordstrom1997: 216).
Embodied accountability refers to bodily predicaments that evolve and become crucial to initiating struggles to right wrongs related to past violations perpetrated in contexts of family and community relations (Igreja Reference Igreja2015a; see Mueggler Reference Mueggler2001). At the root of embodied accountability is a relational conception of the body and personhood wherein “connections to and relationships with others and our membership in groups” play a role in determining someone's responsibility for wrongdoing (Laidlaw Reference Laidlaw2014: 193) and the liability of the kin group of the alleged perpetrator in such harmful acts (Igreja Reference Igreja, Wallis, Kent, Forsyth, Dinnen and Bose2018a; see Akin Reference Akin1999: 44–47; Reference Akin2013: 79–84). Despite these disparate conceptual roots, the idea of answerability provides parallels between accountability and embodied accountability. In Gorongosa, the obligation to provide an explanation in ordinary conflicts is expressed through the notion of ku vundzissa, which translates as “to raise a question” when confronted with a certain ordinary problem.
The differences between accountability and embodied accountability emerge during the resolution of cases, particularly those involving alleged killings and marred by a lack of firm evidence and conceptual representation, as in the case of Taringa. Here the tone of the obligation is elevated and the category of relational personhood, including blood and kin ties, is emphasized by acting it out. It is expressed as mu vundzissene—“ask questions of one another.” This questioning in the context of embodied accountability occurs among the participants in the conflict and during it the imagination is an important tool (see below, for example, regarding Bandemba's questioning of her son, Furatira). Yet questioning and answerability can evolve through a comprehensive and dialectic process that includes spirts, animals, places, and objects (see Calvão Reference Calvão2017; Geschiere Reference Geschiere1997; Igreja Reference Igreja, Collins, Igreja and Danaher2019; Peek Reference Peek1991; Stroeken Reference Stroeken2010).
Another difference between accountability and embodied accountability lies in how this obligation to provide answers is implemented and met. Demands for answers are triggered by enduring and complex bodily afflictions that torment and terrify the kin of the alleged perpetrator. Thus, embodied accountability is an apt term given that in this cultural milieu, as in many around the world, most people take for granted that bodily torments can be “agentive” (Asad Reference Asad2003: 79; Kleinman and Kleinman Reference Kleinman and Kleinman1994: 718). Based on “education” and “circumstances of life in common” (Mauss Reference Mauss1973 [1935]: 86), embodied accountability is a stable practice expressed in various conventional bodily rituals and routine bodily actions (Boddy Reference Boddy1989; Csordas Reference Csordas and Csordas1994; Kidron Reference Kidron2009; Lambek Reference Lambek1981; Mauss Reference Mauss1973; Stoller Reference Stoller1995). But, the body also generates unpredictable actions and conditions (Cohen Reference Cohen1985; Farnell Reference Farnell2012).Footnote 8 This latter quality adds another valence to embodied accountability, making it continuously emergent and enigmatic. In contexts of the resolution of violations, this emergent and enigmatic quality offers possibilities for exploring how people continuously struggle and negotiate to bridge gaps between past and ongoing conflicts and actions.
Here I focus on the second dimension of embodied accountability, answerability, to suggest that bodily acts which seem incomprehensible can nonetheless be potential “channels of meaning” (Douglas Reference Douglas and Douglas1975: 85; Deleuze and Guattari Reference Deleuze and Guattari1987; McMahon Reference McMahon and Stivale2005). The meanings in such bodily actions may not be immediately mastered by either locals or outsiders (Highmore Reference Highmore2002: 4). Yet, we should not dismiss them as devoid of any sense, or evidence, of something substantial, as mainstream transitional justice specialists often mistakenly do (see Krog, Mpolweni, and Ratele Reference Krog, Mpolweni and Ratele2009). Instead, we must gain knowledge of local conceptions of the body, kinships norms, and the dynamics of violence in specific places as a prelude to determining the diversity of actors, contexts, and conceptions of violations, ways of conveying evidence of disturbing pasts, and attempts at achieving a resolution.
Bodies of Evidence: Accountability and Embodied Accountability
Transitional justice specialists and global accountability experts often acquire evidence of violations by narrowly relying on factual verbal accounts of witnesses, material vestiges of mass killings, archived documents, and visual imagery in contexts of repressive dictatorships and warfare, and more recently of electoral violence (Hagan, Schoenfeld, and Palloni Reference Hagan, Schoenfeld and Palloni2006). Accountability here is triggered only in cases of “widespread or systematic attack” directed against civilian populations (International Criminal Court 2002: 4–10). To assign legal responsibility, “either bodies must be uncovered from mass graves and identified” or “the number of deaths must be otherwise convincingly established” (Hagan, Schhoenfeld, and Palloni Reference Hagan, Schoenfeld and Palloni2006: 334).Footnote 9
Several authors have criticized the global application of accountability because it fails to acknowledge that massive human rights abuses are perpetrated by collectives, be they governments or rebel armies (Clarke Reference Clarke2009; Crawford Reference Crawford2013). Therefore, they argue, we must determine the responsibility of states and rebel groups, writ large, not just that of their commanders (Lang Reference Lang2007; Radzik Reference Radzik2001).Footnote 10 Other criticisms have focused on the failure of global accountability regimens to prosecute government and rebel groups that use famine as a weapon of war and control, which may cause more deaths than actual fighting (de Waal Reference de Waal2006; Keller Reference Keller1992; Macrae and Zwi Reference Macrae and Zwi1992; Marcus Reference Marcus2003). These criticisms, though meaningful, remain anchored in state-centric perspectives. To effectively eliminate the impunity such actions currently enjoy, any global strategy must consider other dynamics of violations, spheres of collective responsibility, and modes of gathering evidence.
For example, for the Gorongosas, the numbers of victims, the time of their violent deaths, and the existence of material vestiges of the victims, are not central to their understanding of violations. Their notion of serious violations is linked to beliefs that the failure to right past wrongs fosters relationships in transition which feature enduring bodily afflictions and intractable conflicts: they believe these afflictions and social conflicts are prolonged by invisible agents (aggrieved spirits of dead victims) who interfere in people's daily affairs. They also believe that persistent bodily afflictions and conflicts are maintained through silence, concealment, and obstinate denial (upoca). Without resolution through local cultural mechanisms, the consequences of the violation cannot fade away. This is expressed through the saying “a conflict does not rot” (micero ai vundi).
In this evidential regime, the kin of the alleged perpetrator are not immune from the retributions of aggrieved spirits. Thus, when aggrieved spirits strike, they experience these spirit-assaults as injustices, and family relationships can become a burden. They stridently argue that they are suffering as payback for a wrongdoing in which they had no part. Yet the kin of the alleged perpetrator suffer because of their blood ties, and in this case, their crucial role is to pressure the kin that allegedly committed the offence to disclose knowledge of the violent events. Such knowledge is crucial for the reparative acts that are undertaken by the perpetrator's kin, even if they, as individuals, did not commit the offence. “To pay back” is termed ku lapiwa (Chi-Gorongose vernacular), and it refers to repair in cases of violent death. If the perpetrator denies responsibility and fails to atone, then their extended kin members are doomed to suffer (Igreja Reference Igreja2014). In this way, the burden of relationships through sharing of guilt, bodily suffering, and intractable social conflicts due to kin membership becomes a form of power employed to pressure suspected family members to deal with wrongs that they were presumably responsible for in the past (see Douglas Reference Douglas1980: 69).
Given the patrilineal feature of the Gorongosa region, healers and oracles often inspect the genealogy of concealed offences in the body of a plaintiff and his or her patrikin. The hearings and production of evidence occur when the morally competent person in the family, who is often an elder among the patrikin, is present to assume responsibility (ku himirira) (Igreja Reference Igreja, Wallis, Kent, Forsyth, Dinnen and Bose2018a). The exception to this patrilineal rule is in cases of children who belong to the family genealogy of the mother (known as makuio children). Makuio children are those whose father and patrilineal kin did not effectuate the cultural rituals of respect required to access full rights over the offspring. Thus, the offspring remain under the custodianship of their mother and her kin. In cases of conflicts that require a de jure representative (ku himirira), that representative is chosen from the mother's lineage. This kinship contingency reiterates the idea that “the consequences of a given act are not intrinsic in the act but rather will depend on the nature of the social world within which it takes place” (Sewell Reference Sewell2005: 10). From this perspective, the combination of Taringa's actions, the audiences’ responses, and kinship norms helped articulate a notion of culture “in terms of a capacity for innovation” (Keane Reference Keane2008: 113). I now turn to how the rule of makuio unexpectedly played a significant role in the process of imagining probable causes and explanations for Taringa's conundrum.
POSTCOLONIAL CIVIL WAR AND FAMINE
The origins of the Mozambican civil war have been explained by equating factors such as repercussions of the Portuguese colonial rule, geopolitics of the Cold War, the repressive nature of the Frelimo's postcolonial Marxist-Leninist revolution, and interventions by now-defunct regimes of Rhodesia (now Zimbabwe) and apartheid South Africa (Cahen Reference Cahen, Chesterman, Ignatieff and Thakur2005; Honwana Reference Honwana, Meyer and Pels2003; Lubkemann Reference Lubkemann2008; Nordstrom Reference Nordstrom1997; Pitcher Reference Pitcher2006; West Reference West2005). Numerous academics have also analyzed various state and community mechanisms aimed at coming to terms with the civil war legacies (Bertelsen Reference Bertelsen, Broch-Due and Bertelsen2016; Igreja Reference Igreja2008; Kyed Reference Kyed2009; Lubkemann Reference Lubkemann2008; Nordstrom Reference Nordstrom1997; Obarrio Reference Obarrio2010; Pitcher Reference Pitcher2006; West Reference West2005; Wiegink Reference Wiegink2013). Yet, the dynamics of the famine that ravaged the country in the 1980s and 1990s and its enduring aftereffects have been little studied.
During the civil war, people in regions such as Gorongosa were divided between government- and rebel-controlled areas. Populations sometimes shifted from one area to another. Order in rebel-controlled areas was enforced by paramilitary forces composed mostly of men known as mujibas and village chiefs. For instance, Furatira was a village chief. Populations living under control of the rebel group Renamo participated in maize crop production, while the mujibas compelled the people to contribute part of their crop to the rebels and gather the collected crop in the mujiba's house. Through forced labor (gandira), groups of men and women were assembled to carry the crops for Renamo soldiers, and sexual violence against girls and adult women became rampant in the context of gandira.
Severe droughts and famines occurred in Gorongosa in 1987–1988 and 1990–1992.Footnote 11 Many survivors told me of the hardships: “There was no food or water”; “We relied on dangerous roots of wild trees and these required considerable work to remove the poisonous substances” (see de Waal Reference de Waal2006; Dikotter Reference Dikotter2010). During my previous research in Gorongosa, I had also heard scattered stories of how, when confronted by soldiers, some ordinary individuals lied to and betrayed one another. Alliances with soldiers meant that some people had access to food while others were left to starve and died (Igreja Reference Igreja2007). Yet survivors generally gave vernacular labels to these periods of famine, which blamed nature for the suffering and downplayed the role human actions or inactions played in aggravating conditions. These labels were tchaca ia ndzala (year of famine), crime ia ndzala (crime of famine), or crime ia dzua (crime of the sun) (ibid.). In turn, researchers who employed the notion of “famine crimes” did not fully understand the dynamics of the collective famine because they only analyzed how government and rebel troops used famine as weapons of war and control (Devereux Reference Devereux and Devereux2007; de Waal Reference de Waal2006; Keller Reference Keller1992; Macrae and Zwi Reference Macrae and Zwi1992; Marcus Reference Marcus2003). Taringa's actions ignited local debates that widened the scope of the famine continuum and raised questions among the citizens (mu vundzissene) regarding the responsibility of ordinary civilians for exacerbating the situation.
In government-controlled areas the Red Cross sometimes provided food aid (maize) to local populations, whereas in rebel-controlled areas reports of insecurity deterred them from doing so. This unbalanced Red Cross assistance contributed to people having uneven access to food. The local politicization of the Red Cross raises serious ethical questions regarding the humanitarian claims they made in Mozambique.Footnote 12 Various studies have analyzed how humanitarian organizations in conflict zones have, willingly or unwillingly, taken sides so as to be able to deliver food supplies (Blouin Reference Blouin2008). Clandestine suppliers of food (known locally as fontes) often sold the Red Cross supplies in black markets located along the floating borders between government- and rebel-controlled areas. The food supplies were exchanged for chickens and wild animals smuggled from the rebel-controlled areas. These exchanges were known as candonga (clandestine actions). Expeditions to participate in candonga required walking long distances for days, which posed considerable risks. Those in charge of the candonga, mainly men, exerted strict controls. For example, there are testimonies about neighbors betraying young men who smuggled salt, who were then publicly executed (Igreja Reference Igreja2007).
In this regard, we need to consider two seemingly contradictory realities. On one hand, culture shapes people's violent and non-violent actions and embodied accountability mechanisms. On the other, “Individuals are not constant as situations change” (Collins Reference Collins2014: 4). Microanalyses reveal that, “Incidents shape their incumbents, however momentary they may be” (ibid.: 5). Further, changes do not uniformly impact incumbents in specific situations. As I will describe presently, the famine situation generated a competition for food, to which people responded differently. Some, such as Bandemba, used the situation to express their compassion for the weak, while others, like Furatira, used violence to dominate others and seemingly win the competition. Additionally, various survivors told me, and I heard them reminding one another, about how famine experiences divided families; many couples divorced, and children were abandoned. Some people controlled and punished those they deemed feeble or reckless. Others engaged in smuggling, hid food, or abandoned people to die, their corpses left to rot in communal spaces. People were murdered and many simply disappeared. Stealing became rampant, with accounts settled violently for those allegedly caught red-handed. I was told of how one village member tortured a young woman to death because, he claimed, she had stolen six corn cobs from his field (Igreja Reference Igreja2010). Young people had to hide in the bush and sometimes survived on their own, or sometimes with support from charitable families (Igreja Reference Igreja2015b). Taringa's family did this, hosting people in despair in the midst of the famine, but they nonetheless ended up implicated in unspoken violations, revelations of which were sparked by Taringa's actions.
Bearing Witness to Experiences of War and Famine
It has been observed that political violence, genocide, terror, and famine can all exacerbate complex predicaments, which make it challenging to talk or write about them and overcome memories of guilt and victimization (Taussig Reference Taussig and Dirks1992: 163, see also Argenti-Pillen Reference Argenti-Pillen2003; Kidron Reference Kidron2009; Pitcher Reference Pitcher2006; von Kellenbach Reference von Kellenbach2013). In some post-civil war contexts, alleged perpetrators roam the inner circles of victims or state institutions, and this prevents survivors from opening up about their past terrors and ongoing fears. In other communities in conflict, disturbing spirits and spirit possession have contributed to creating spaces that survivors have seized upon to talk about complex and frightening legacies of bloody proxy-wars linked to the Cold War, problematic state-reform plans, and political persecutions and repression (Argenti-Pillen Reference Argenti-Pillen2003; Mueggler Reference Mueggler2001; Perera Reference Perera, Das, Kleinman, Lock, Ramphele and Reynolds2001). In still other socio-political contexts official truth commissions have been established, but there are indications that these mechanisms can fail to secure from witnesses disclosures of all of their past violent experiences and subject positions (Sriram and Pillay Reference Sriram and Pillay2009). Truth commissions can be limited by the strict forms of testimony-giving and notions of truth and evidence that they impose (Krog, Mpolweni, and Ratele Reference Krog, Mpolweni and Ratele2009; Wilson Reference Wilson2001), or because they allow witnesses to refer only to events that occurred within narrowly conceived temporal spans (Igreja Reference Igreja, Mueller-Hirth and Oyola2018c; Mueller-Hirth and Oyola Reference Mueller-Hirth and Oyola2018).
In Gorongosa, as I have said, gamba spirits interrupted the post-peace agreement silence and compelled survivors and the postwar generations to actively deal with unresolved violations of the civil war (Igreja, Dias-Lambranca, and Richters Reference Igreja, Dias-Lambranca and Richters2008). Taringa's case was different from these established embodied accountability practices in that there were no spirits or altered states of consciousness (spirit possessions) to voice indictments, and the local healers failed in their attempts to address the problem. Taringa's predicament therefore placed greater responsibility on audiences, which engaged their imaginative abilities and triggered confrontations and negotiations over the meanings of her bodily actions.
In their loose attempts at explanation, the people first described Taringa's actions as mberu, mberu kuienda (meaning, in local vernacular, “to wander”). The expression echoed descriptions of the famine in which people trekked slowly in search of food and water, perhaps never to return home alive. People also referred to Taringa in Chi-Gorongose as munhu, or ndio, or iyo ia ntsima, which translate as “the person of porridge.” This usually means simply a person who eats porridge, but in her case, she ate vast quantities and did not share. This notion temporarily stripped Taringa's actions of gender and age identities, since the descriptor “person of porridge” does not specify either. These imaginative initial descriptions provided a common language, albeit a fragmented one, with which people could describe these enigmatic events without necessary explaining the problem. Taringa's father, at the very time that her actions were increasing community unrest, slackened his search for explanations on the rationale that he had exhausted all possibilities, but unhappy kin group members reported the case to a community court so as to pressure the family to continue searching for answers.
Community Courts, Moral Bodies, and Sensing the Invisible
The judges of the community courts were unable to pronounce a judgment given the impenetrable facts they were asked to deliberate on. They referred Taringa's case to a paza oracle. Furatira and his family, which included Taringa's mother and sisters, attended the oracle. Three community judges (two female and one male, who was also the president of the community court) and I joined the oracle as observers. Initially the oracle searched for the de jure person in the family to begin the proceedings. Since it was revealed that Furatira was a makuio child, the oracle inquest had to initiate the proceedings based on the makuio rule. This required the direct involvement of Furatira's mother, Bandemba, which created a “conjectural agency” (Sahlins Reference Sahlins2004: 155). Bandemba shifted her subject position from that of a mother to that of a key witness and was therefore empowered to take a lead in the initial gathering of fragments of evidence under the oracle's mediation. Rather than Taringa's father, Bandemba was transformed into a moral body and morally competent person because the oracle initiates his examination by touching the body of the person in the leading position or the principal witness. Her position as principal witness was instrumental for the subsequent revelations that were made public for the first time.
At the outset of the divination, the oracle gradually presented a set of hypotheses, and in response kin members were expected to confirm (iomba), deny (by remaining silent), or demonstrate uncertainty (also by keeping quiet). These responses aimed to transform the oracle's hypothesis into material clues. In Taringa's case, the participants remained mostly silent while the oracle indicated that Taringa's bodily actions were related to an unsettled debt, which he then described.Footnote 13 Both Bandemba and Furatira confirmed that the debt had existed during the civil war but claimed it had been settled long ago. However, the oracle's allusion to a past debt, the link that he made to wartime events, and the loose parallels community people had drawn between Taringa's actions and the memories of the famine, created a momentum that stirred Bandemba to ask the oracle to check whether Taringa's actions intersected with horrific events in which Furatira had participated, but which had hitherto remained shrouded in secrecy. The oracle was receptive and Bandemba recounted the gruesome events.
Event One: The Famine and Interpersonal Plots that Killed Breda
Bandemba informed the oracle and the audience that during the 1980s she had adopted a wandering, orphaned boy whose family had abandoned him due to the ongoing famine. The boy, named Breda, about eleven years old, had arrived at their house starving and in precarious health. Bandemba was a war widow, and she adopted Breda for companionship. She fed him for almost a year, and when the candonga intensified, her son Furatira participated in candonga in Npintanhanga, a government-controlled village where the Red Cross was distributing food and candonga had flourished.
On one of these treks to the clandestine food markets, Bandemba asked Furatira to take Breda with him since she had some bush meat to exchange with the fontes for maize. Furatira and Breda left together but Breda did not return. When Bandemba and other family members asked Furatira of Breda's whereabouts he initially said that the boy had stayed in Npintanhanga to complete the exchange. After weeks without seeing Breda, Furatira was pressed to return to Npintanhanga to search for him. When he returned, he said that he had not found him. Bandemba and other relatives kept asking Furatira about the boy's whereabouts but got no satisfactory answers.
Furatira agreed with Bandemba's narrative of events up to the disappearance of Breda. They only disagreed on whether Furatira had knowledge of or was responsible for his fate. While Breda's tragic vanishing was not at first linked to the predicament of Taringa, some in the audience were dismayed that Furatira refused to accept moral responsibility even though he had agreed to care for the boy. One female judge said, “Maybe Furatira could say that the child disappeared while in my care and when I arrived home I tried to conceal the fact.” The paza, too, suggested that Furatira was morally responsible. The male judge and community court president added, “We know that here people killed one another to get bush meat. Why are you trying to pass this conflict on to the next generation? You could save your daughter and let us all live in peace by telling what you have done.”
The child's disappearance was a tragic event for Bandemba, but she had never spoken about it, at least in public. During the meeting with the oracle she was visibly distressed as she insisted, “I asked you to care for the child and then you left him. This conflict, I have it: why did you abandon this child?” During these exchanges, Bandemba never called her son Furatira a murderer. She mostly opted to intervene carefully in her relentless search for the truth because plaintiffs should not allege a serious violation without evidence. If they do, the accused may charge the plaintiff with defamation (ku pambazira). Bandemba simply tried to place the burden of proof on Furatira regarding his degree of involvement in Breda's disappearance.
As the discussions unfolded, Taringa's mother Robina lamented, “I am tired. Every year we are suffering. I want my children to be healed.” Bandemba's assertion, “This conflict, I have it,” and Robina's lamentation suggested that the memories and anger of these experiences were still with them, although, until then, those involved had kept silent. Furatira then became confrontational and interjected, “We know that so many people died because of famine and no aggrieved spirits were seen to be involved. Why should a small boy create so many problems now?” This question seemed to imply that Furatira had expected that Breda might die and there would then be no consequences. The oracle answered, “Suffering continues when there is blood spilled and killing.” Robina asked the oracle, “Can you tell us if this boy lives or is dead?”
The oracle asked rhetorically: “Don't you know that there were madjuias in the war as well?” Madjuias was the name for groups of men said to have lived in the bush who were known for abducting and killing children during the tribal wars of the nineteenth century. Nowadays, to instill discipline through fear, parents sometimes warn their children not to move out of reach because madjuias might take them. The healer proceeded by consulting his oracle and said, “The body moved a lot, but now the body no longer moves, his blood has dried. The boy is dead.” Furatira's interjection had offered some circumstantial evidence that he was concealing something about Breda's fate and the paza’s answer provided an incremental confirmation that there had been a violation leading to his death.Footnote 14 Finally, Taringa's sister asked the oracle to check if the alleged death of Breda was the cause of Taringa's singular actions. Upon consulting his oracle, he said, “The oracle rejects. The disappearance and death of the boy is unrelated to the actions of Taringa.” It appeared that the oracle's negative answer would end the allegations of a potential link between Taringa's actions and events during the famine. But instead, Bandemba presented another event for the oracle to inspect.
Event Two: A Woman and Her Children Abandoned on the Roads of Death
As the period of famine in 1990–1992 intensified, more people roamed the villages in search of help. Once, Bandemba told the audience, a sick woman carrying her two young children lost her way while wandering on the road looking for shelter and food. People directed her to Furatira's house since he was a village chief. The woman and her children, whose names were unknown, reached his house at sunset. Bandemba explained to the oracle and audience that her son Furatira had helped the family and on the following morning he told the family that he needed to look for food, and that they could not stay in his house in his absence.
Bandemba told the paza that her son had carried the woman on his back since she could hardly walk. He left her and her children on the road with directions to the government-controlled area, so they could try to obtain humanitarian assistance. He went on his way, but some hours later the woman and one of her children died in the road. The other child survived, but no one knows his fate. As they testified in the oracle, many people at the time saw the woman and her child lying dead on the road and, according to Furatira's paternal uncle Tovumba, “Furatira did not even return to bury the corpse of the woman and her son. They both rotted on the spot.” Again, in this case, Furatira corroborated Bandemba's and Tovumba's account, but refused to assume responsibility. He insisted that he did everything he could for the woman and her children. He argued that, given that he was also struggling to get food, he had no choice but to abandon them on the road. Yet Tovumba became upset and contradicted Furatira: “This is not true. You had food, but you did not share; you concealed the food for yourself. How were you doing candonga many times a month if you did not have food to exchange?” In an attempt to soothe tensions, the oracle used the occasion to recount a debilitating experience in which one of his wives had to choose between feeding their baby or herself to survive, and the choice led to the baby's premature death. Then the oracle warned, “If you start accusing one another of hiding food, which killed many people during the crime of the sun, I am certain you will find that no one here is impeccable.”
Negotiations and Refinement of Contested Fragments of Violations
I was dismayed by the revelations that unfolded during the oracle consultation since I thought that I knew many key aspects of the war and famine experiences of Furatira and his family. For instance, in the previous four court cases involving them, none of these revelations of Bandemba's had been spoken of. The display of emotions and distress that accompanied these revelations indicated that some memories of the famine had been silenced but not forgotten (see note 7). The singular actions of Taringa triggered a process of reconfiguring social relations that had hitherto been characterized by what Paul Ricoeur (Reference Ricoeur, Blamey and Pellauer2006: 414) termed “backup forgetting, a sort of forgetting kept in reserve (oubli de réserve).” Such memories in Gorongosa are kept hidden from public access because they are contentious, but they can be retrieved at any moment for purposes of scrutiny and litigation. Open struggles for accountability for violations are often avoided, except when the principal protagonists in a conflict realize that there are meaningful chances to undo the status quo (Igreja Reference Igreja and Russell2018b). Yet it is impossible to predict when and how such occasions will occur. This is an example of why global accountability and transitional justice projects cannot capture all hard facts about violations within a predefined, limited time-frame, and if they are disconnected from local politics of silence, litigation, and the exposure of what constitutes implicating evidence.
Initially, the singularity of Taringa's actions trumped the healers’ efforts to categorize them. When challenged by singular events, the response should not be an enquiry into “what a given sign signifies” within the context of existing sociocultural structures; it should be about “which other signs” a certain sign might refer to (Deleuze and Guattari Reference Deleuze and Guattari1987: 112). During the era following the 1992 Mozambican peace agreement the dominant embodied accountability practice was based on verbal utterances and indictments made by gamba spirits, so the absence of spirit signs and voices in Taringa's case triggered a sense of doubt and despair, but also the popular imagination and the advent of an array of inferences (Severi Reference Severi and Lloyd2015) that linked seemingly disparate signs. Instead of the usual tripartite interactions of body-spirit and audience, Taringa's case granted a significant role to the audiences, who then used their imaginations and memories, and continuously negotiated ways of piecing together the various fragments evoked by her actions. In this regard, the community imaginings, the oracle's actions, and unexpected kinship norms created the conditions for Bandemba to cautiously reveal the existence of unspoken, lingering conflicts that had to be scrutinized to find out if they could explain Taringa's actions.
This process of gathering evidence through fragments, inferences, social pressures, and kinship norms was not free of contestation. During the oracle forum, Furatira challenged his mother's inferences by reverting to culture as the foundation of his counter argument. Besides the doubt that he had expressed about the boy's ability to return to the world of the living as an aggrieved spirit, Furatira also asked how Taringa's actions could provide insight into the gender and age of the victims. He asked, “How is cooking related to the boy given that in our makaliro (ways of living) it is the women who cook?” Furthermore, he asked, “How is it possible that wandering around is related to the woman who died, given that men are the ones who often move around?” The oracle sided with Furatira's doubts by asserting that he could not grasp how Taringa's actions (since cooking was typically female) had any relevance to the missing person who was male.
On the spur of the moment, Bandemba used practical arguments instead of shaping her answers through cultural norms or values. For example, she asserted, “In the midst of oppression, young and adult men do cook for their own survival.” Furatira also raised doubts regarding the link between Taringa's actions and the deaths of the woman and her child. He asked, “How can it be that a powerful spirit affects Taringa and disrupts an entire village without ku buda [meaning, without the usual visible signs of spirit possession]?” Using a similar logic of practical arguments, the second female judge asserted: “It is the same thing with a pregnancy; it can deliver a boy or a girl, or a miscarriage.” This was to imply that the female body can harbor memories that transcend gender divides.
Following three hours of disclosures, presentations of fragments, evidence of unresolved violations, disputed allegations, and metaphorical talk, the allegations and practical reasoning of Bandemba and the rebukes of Furatira were regarded as subjective, and therefore lacking the power to sway the oracle in a specific direction. It is noticeable that throughout this contest to refine evidence, Taringa remained a silent spectator. The oracle kept his initial verdict by insisting that a war-related debt was the origin of her singular actions. Nevertheless, my understanding is that the oracle's decision did not utterly invalidate everything that was revealed in the session. The issue is that those revelations were meaningful in an abstract sense, while it was also necessary to have a correspondence in a practical sense. In the former sense, the testimonies of Bandemba and of those who backed her contributed to expanding knowledge of the famine continuum. Her revelations paved the way to filling a major gap in the current anthropological and historical literature of the war and famine in Mozambique. In a practical sense, the oracle was authoritative since it reflected the “legitimate instruments of expression” (Bourdieu Reference Bourdieu1991: 109), but the “power to pronounce” (ibid.: 111) does not automatically imply a rightful pronouncement. Initially, Furatira accepted the oracle's version and mobilized his family to accept its authorized decision, even if he and his mother had rejected the existence of a wartime debt as the foundation of the guilt predicament.
Enduring Bodily Afflictions and Landscapes of Shared Guilt
Two months after the oracle's resolution, Taringa's actions persisted. Her family and neighbors took this to mean that the oracle's judgment had been incorrect in a practical sense. Faced with mounting pressures because Taringa's bodily predicament and disturbing actions had not improved, along with complaints from the family and community insecurity, and the resulting social trepidations, Furatira surrendered. He organized a forum in his homestead and asked the extended family to attend. He said he would give his previously untold version of the events. I spoke with Bandemba and Tovumba about the prospects for the meeting, and both were skeptical. Yet the next day both elders showed up at Furatira's homestead. He initiated proceedings with a long speech about the difficult circumstances of the famine. But then he admitted, “In a state of despair I betrayed Breda.” Tovumba interjected, “It cannot be despair; it was greed.” Furatira, however, kept focused on what he was about to say next: “I secretly traded him [Breda] in exchange for food to a group of wandering men. The men were looking for young persons to carry luggage to Manica [a neighboring province], but they promised to bring him back.”
This confession did not bring immediate relief to Furatira's mother, and therefore Bandemba insisted unwaveringly on the truth, saying, “But you knew that leaving Breda there was putting the kid in the firing line.” His paternal uncle Tovumba also weighed in: “The security was precarious. Every day there was news about killings because of food; how did you dare to give the child to mufambi [unknown travelers]?” Bandemba added, “As we heard from the oracle, now Breda is dead.” Furatira responded to these remarks by trying to put his mother on the defensive: “But why are you insisting in this way? Is all this to kill me?” Bandemba promptly replied, “If I had wanted you dead, I could have not risked my life to help you during the war.”Footnote 15 Furatira added, “But you adopted the boy when you knew that we were many people in our family and there was not enough food for everyone.” This latter complaint further shows the ambivalence of some of the people's responses to the severe famine. While Furatira's mother acted to save Breda's life through adoption, he saw that act as increasing the competition for food. Yet Tovumba again interjected, “Why do you always try to do ku kulurira [search for scapegoats]?” Furatira's wife Robina appealed, “Leave aside these secrets, be reasonable.” And Taringa's sister, who at the time was ill and lying on a mat, added, “No one kills someone in our land without tainting the whole family with disgrace.”
To these appeals and references to local culture and practical reasoning, Furatira issued no more counter-arguments. Instead he confessed: “Ndida zunguza manhingue,” which translates as, “I did something very bad and dangerous, I sold the life of the kid.” He gave further details that showed that he had never really expected the unknown men would return Breda to his adopted mother alive. Regarding the mother and her child who had died during the famine, he acknowledged: “I failed to help this family.” He further added, “Even if I did not have a lot of food, I agree I had enough to help them survive, and today perhaps all these people could still be alive.” Tovumba added, “Who does not want to live? We all want to live and see our children and grandchildren growing up healthy.” To pay back for the life of Breda and seek forgiveness from the unknown woman and her children, and to try to reconcile with his mother, Furatira's family ordered him to build a solid hut for the spirits of the dead in the middle of his yard. Building such a monument for the spirits is a rare event, and Bandemba, in commenting on this outcome, told me, “This penance is important because the hut will include all Furatira's ancestral spirits, the spirit of Breda, and the spirits of the unknown woman and her children.” According to Bandemba, the three victims became recognized as kindred spirits that Furatira and his extended family were compelled to continuously care for and worship.
Only when Furatira concluded the building of the hut and the whole family initiated the worshipping of the spirits of the dead victims did Taringa begin to lessen her actions. In June 2014 and January 2015 I conducted follow-up interviews in Gorongosa with Furatira, Taringa, and her husband Buraco. They all agreed that Taringa was no longer acting in unpredictable ways, and that her son had survived and was growing up healthy.
CONCLUSIONS
This article has examined how the singularity of Taringa's bodily actions unexpectedly contributed to breaking silences. This, in turn, stimulated broader collective questionings, testimonies, and imaginative responses tied to unresolved violations against hungry and vulnerable people during the 1980s and 1990s famines in Gorongosa. Though I have presented a few cases of them here, the visible and quantifiable dimensions of violations that are the focus of global accountability projects are not the primary concern of the Gorongosas. In the global projects, accountability matters only in cases of mass killings, when there are too many dead bodies to account for, and guilt is individualized. It is assumed that individuals have an autonomous moral agency, are fully cognizant of the implications of their behaviors, and thus act ruthlessly with intent to cause widespread harm. By contrast, when Gorongosas seek answers regarding embodied accountability, their search is not governed by the number and material vestiges of the victims or the times of their deaths. What sets people to questioning one another is persistent bodily suffering coupled with intractable social conflicts and kinship ties. The logic of embodied accountability is informed by a relational conception of personhood. In Taringa's case, it provided the rationale through which a daughter's bodily suffering and frightening behavior could index her father's past violations of relations of respect. In this context, Furatira's enduring concealment and denial left his family relations in a state of unending transition since they coped with continual social unrest and bodily afflictions because they shared in his liability. Moreover, his reckoning entailed not only physical memorialization but also incorporation of the victims into his ancestral kin group. This manner of ascribing collective responsibility differs also from the ongoing realist debates and analyses of collective responsibility following mass political violence (Crawford Reference Crawford2013; Lang Reference Lang2007; Radzik Reference Radzik2001; Stilz Reference Stilz2011).
The acceptance of shared responsibility and Furatira's building of the monument addressed conflicts of the past and present, but also future events. Bandemba said the shrine would help them meet ongoing goals: “We will also worship these spirits to deliver rain during the droughts that from time to time afflict everyone in our community.” The building of this monument in recognition of guilt in the center of the culprit's yard provides important comparative lessons for those analyzing how the memories of people that have perished unjustly are preserved and respected. This model of commemoration differs from most commemorations of victims of mass violence, massacres, ethnocide, and genocide, where monuments have often been built on the sites of the victimization itself (Levinson Reference Levinson1998). The Gorongosa monument built in the perpetrator's yard served as a powerful, permanent moral message regarding the perils of uncontrolled violence and death.
My examination of the famine continuum has captured layers of wrongdoing that are absent in analyses grounded in the concept of famine crimes (de Waal Reference de Waal2006; Keller Reference Keller1992; Macrae and Zwi Reference Macrae and Zwi1992; Marcus Reference Marcus2003). I have shown how the impacts of famine and deaths were exacerbated by violations perpetrated by ordinary individuals. In the aftermath of the civil war in the Gorongosa region, embodied accountability initially coped with the war's legacies through community courts, gamba spirits and spirit possession rituals, oracle television, and Christian religious practices, and sometimes the state police became involved (Igreja Reference Igreja2014). In relation to the grisly events of the famine, at first survivors attributed the suffering to nature and remained silent and defiant regarding the impacts of human actions, particularly within family and community circles. Taringa's case created spaces and opportunities that expanded the ways in which people could speak about the famine and the mayhem it caused.
The singularity of Taringa's actions and the overall responses to it threatened fundamental aspects of the cultural order but did not change it (see Sahlins Reference Sahlins1985; Sewell Reference Sewell2005). Instead, there was innovation without change (Moore Reference Moore1993 [1966]), given that Taringa's unexpected actions circumvented the customary focus on spirit possession and spirits’ voices as sources of knowledge and indictment. They expanded embodied accountability by creating contexts within which the burden of witnessing was transferred to the audience, who responded by testifying to dreadful experiences of past violations. Nevertheless, this process of resolution did not negate the idea that aggrieved spirits could wreak havoc in the post-peace agreement era.
This family and their community had for nearly three decades lived in the shadows of secrecy and denial. We have seen how the case's resolution confronted the sense of collective insecurity and gradually restored the family's sense of moral rectitude. When the aftermath of violations is permeated by silence, denial, and fragmentary evidence, then unexpected bodily actions, collective imagination, and kinship norms can be sources of knowledge about the meanings of the violations and the diversity of actors involved. They can become sites of accountability. By studying them, we can transcend the limited understandings currently provided by global accountability and mainstream transitional justice projects.
