Violence is a murky issue to research and to theorize: this introduction suggests that it has also often been approached differently by anthropologists and historians. In the pages that follow, we reflect on the ways in which both disciplines have worked to interpret violent events in Africa, whether in the deep past, during the colonial era or in more recent periods. To better contextualize these disciplinary advances, we intersperse them with brief reviews of general theories on violence. The three articles featured in this special section, while dealing with very dissimilar case studies, provide common insights on three main themes. The first engages with the paradox of the contingency and continuity of violence, and with the unevenness of perpetrators, victims and targets. The second deals with the refractive meanings attached to violent events. The third probes, underneath the apparent turmoil of violent acts, the deep moral and cultural frameworks of action that underwrite them. We have composed this introduction around these main questions.
Although historians and anthropologists have devoted considerable work to violence, it seems to us that the former have primarily focused on collective forms of instrumental violence, or, to follow Charles Tilly's formula, on ‘violence as politics’. This includes the uprisings and actions of subaltern groups and individuals, usually deemed ‘illegitimate’ by established regimes, and the legal ‘force’ exercised by institutions and state officials to control such actions (Tilly Reference Tilly2003). This disproportionate attention is explained by the nature of historical sources. In Africa, archival series inform mostly on institutional violence and on collective resistance and uprisings: they include prison statistics, judicial and administrative records, complaints and petitions to international organizations (Terretta Reference Terretta2013). It is much more difficult to find the traces of individual acts and temporary outbursts. Although court hearings, life stories and private correspondence offer rich stories of domestic and interpersonal violence (Van Onselen Reference Van Onselen1984; Jewsiewicki Reference Jewsiewicki1993), many are too tenuous or isolated to construct a more general interpretative narrative. Moreover, historians of Africa have been late in dealing with the history of emotions and their works have focused primarily on productive affects rather than anger and hatred (Cole and Thomas Reference Cole and Thomas2009).
Yet historians of violence have not worked in isolation: most have been influenced by theories outside their field. Interpretations of institutional and popular violence, for instance, have been inspired by the concept of hegemony (Gramsci and Boothman Reference Gramsci and Boothman1995) and the notion of ‘symbolic violence’ crafted by sociologists such as Pierre Bourdieu (for a critical discussion, see Addi Reference Addi2002: 156–79). One of the most accomplished examples is Jonathan Glassman's study of urban riots on the Swahili coast in the late nineteenth century, in which he argues that the violence of the crowd reflected intense struggles about the moral economy of various stakeholders (Reference Glassman1995). In turn, the concept of the moral economy signals how historians have affected other fields. The work of E. P. Thompson has proved widely influential among social scientists, including anthropologist James Scott, who reworked it in his study of peasant rebellions and resistance in Malaysia (Thompson Reference Thompson1971; Scott Reference Scott1976; Reference Scott1985). For historians, empirical research has been guided by two main agendas: the need to interpret the causes and factors of violent acts, and to identify long-term patterns while paying attention to contingent convulsions and crises. Recently, for instance, in a landmark study of a German police brigade operating in the ghettos of Poland in 1942 and manned by middle-aged men from Hamburg who were ineligible for regular military duty, Christopher Browning engaged with Hannah Arendt's work on the banality of evil (Reference Arendt1970). Browning argued that the men terrorized and massacred Jews out of basic obedience to the authorities and the fear of ostracism rather than primal hatred or ideological commitment (Reference Browning1992a). A moderate functionalist, he highlights the importance of the cumulative radicalizing of the Nazi state over time, a theory that critiques ‘intentionalist’ historians who rely on Hitler's grand plan and totalitarian politics to explain the development of the final solution (Browning Reference Browning1992b).
Like historians, anthropologists have had to confront the ambivalent and contradictory aspects of violence. As Nancy Scheper-Hughes and Philippe Bourgois argue in their influential collection Violence in War and Peace, violence can be destructive or productive, necessary or useless, legitimate or illegitimate, senseless or gratuitous, or utterly rational and strategic (Reference Scheper-Hughes, Bourgois, Scheper-Hughes and Bourgois2004: 2). Anthropologists have paid better attention to its interpersonal manifestations (violent rituals, brawls, individual aggression and scattered attacks), and a number of them have argued that interpersonal violence is primarily expressive and utilitarian. In 1976, for instance, Emanuel Marx looked at the assaults and threats perpetrated by Moroccan immigrants in Israel as trying to provoke a response from town officials (Reference Marx1976). Following Marx, anthropologists such as David Riches (Reference Riches and Riches1986) and Anton Blok (Reference Blok, Aijmer and Abbink2000) argue that violence can improve one's status, change one's social environment, or bar the victim's claims.
Recently, Isak Niehaus has published a critical review of utilitarian theories (Reference Niehaus2009). With Sahlins (Reference Sahlins2004), Niehaus thinks that systemic explanations preclude the possibility of unintended and unanticipated incidents carried out through rage and anger. Looking at homicides in a rural municipality of South Africa, he finds that the public is often struck by the senselessness of the crimes, and the fact that, in the long term, violent actions almost always undermine or destroy the assaulters’ position. While the economic and social context informs the meanings of violent acts, the emotional bursts of perpetrators, rather than their intentions, better explain their violent assaults. Other studies show that, by using violence, the perpetrators project themselves outside the bonds of normal social and cultural processes. Sometimes this is a conscious project, as in the Balkan wars, where, Mattijs Van de Port demonstrates, Serbian soldiers used torture, cruelty and murder in order to transform themselves into ‘wild men’ possessed by unreason and unruliness (Reference Van de Port1998). Sometimes this happens in ways unknown to social agents, like the savagery exacted by rubber company agents on Putumayo Indians in the Amazonian forest; their brutality partly enacted and mirrored the putative savagery that Europeans feared from local people (Taussig Reference Taussig1987).
The contributions presented in this special issue have been written by young historians of Africa who look at collective and interpersonal violence as a form of political action pregnant with productive and destructive aspects, hence the focus on ‘control and excess’ conveyed by our introduction title. Moreover, by highlighting the singularity, diversity and contingency of violent acts in each of their respective case studies, the authors avoid reducing violence to interpretative meta-narratives. Instead, they rely on fine-grained analyses of specific events to explore the possibility of writing histories that preserve the irreducible contingency and inexplicability of violent crises. Nicole Eggers examines the millenarian revolt led by Bushiri Lungunda that erupted in 1944 in the easternmost part of what is today the Democratic Republic of the Congo. Bushiri and his followers arrested several Europeans and agents of the state, and tortured and executed numerous African men and women accused of witchcraft. While the rebels sought primarily to restore a perilously imbalanced economy of power, Eggers shows that violence still remained contingent on certain social and political circumstances.
Aidan Russell focuses on the selective technologies that, in Burundi in 1972, allowed state agents to slaughter approximately 10 per cent of the Hutu adult male population and many women and children by triaging victims with official lists and road barriers. The key feature of the violence, Russell argues, is that it was genocidal in scale while being selective in method and process. The bureaucratic devices central to the unfolding of the genocide endured, in the years that followed, as the central implements through which the population of all ethnicities and all places in society learned to reconsider their relationships with each other and with the state.
Finally, Franziska Rueedi looks at black youth violence in the townships in the Vaal Triangle of South Africa in the mid-1980s. Contrary to the opinion that these actions were senseless and merely the result of high levels of unemployment (Sitas Reference Sitas1992: 629), Rueedi argues that violence was part of a concerted political strategy against the apartheid regime. Attacks against the police and the intimidation of collaborators sought to free the townships from fear while promoting new forms of political loyalty and belonging. Yet the incidents could also aggravate dissent in the townships and weaken the liberation movement.
CONTINUUM OF VIOLENCE, UNEVENNESS OF POWER
In Violence in War and Peace, Scheper-Hughes and Bourgois argue that violence reproduces violence ‘in a continuum’ (Reference Scheper-Hughes, Bourgois, Scheper-Hughes and Bourgois2004: 131). The kind of continuum examined by this volume makes sense of another paradox: the fact that violent actions break down existing norms and structures and reconfigure them in original ways. These patterns are better thought of as contiguous rather than opposed. Moreover, they are continuously shaped by specific historical circumstances. Franziska Rueedi, for instance, argues that black youth's violence against local agents of the apartheid regime constituted a multi-layered web in which certain expressions and forms overlapped and others were discrete. Activists used contingent actions such as stoning, petrol bombs, trenches and barricades of burning tyres against the police and armoured vehicles. They also used premeditated ones such as arson, public flogging, beating and ‘necklacing’ against anyone who was rumoured to be associated with the police or the state. For Rueedi, moreover, township violence always emerged within a particular political, moral and social context; hence it could never be said to simply ‘erupt’, nor could it be reduced to a mere by-product of broader political dynamics. Violent acts aimed to forge a new political order and reinforce the moral boundaries of township communities. By promoting general ungovernability, the youth wanted to protect residents from the assaults of the state and the infiltration of informers, while at the same time they sought to overcome fear and internalized repression. As a consequence, violence blurred the line between the political and the subjective and encouraged the emergence of new identities among township residents.
Yet it did not always succeed: in Soweto, for instance, people argued that township violence alienated residents rather than encouraging them to join the struggle. Moreover, the percentage of the youth who participated in violent acts was comparatively low, and did not represent the experiences of youth in general. As William Beinart perceptively argued in an introduction to a special issue of the Journal of Southern African Studies on political violence in South Africa: ‘Violence requires particular forms of agency, and the actions and predispositions of those agents were not simply created by broader lines of conflict’ (Reference Beinart1992: 473).
In turn, the inconsistencies and patchiness displayed by the dominant institutions of power that work as either the targets or the engineers of violent actions can be essential factors in shaping the nature and efficacy of violence. Aidan Russell's study of the 1972 genocide in Burundi provides a fascinating example of bureaucratic terror in a weak, technologically underdeveloped state. In contrast to the mass killings performed by the Nazi regime, the Burundi government utilized deeply hybridized means of persecution that hardly worked as an unstoppable killing machine. State agents, many of them illiterate, selected the victims through road barriers and haphazardly created lists of suspects’ names. The communicative function of these theatrical tactics helped to rationalize the genocide and construct violence as a fully institutional process, although it remained uneven and relatively anarchic. Paradoxically, by facilitating denunciations and collaboration, and sometimes the compliance of unsuspecting victims, the weakness of state power worked to the advantage of the genocidal enterprise.
REFRACTIVE MEANINGS
Outside Africa, historians have been turning their attention to the ways in which past violent events are being remembered and memorialized, uncovering the many issues and contradictions that riddle such processes (Terdiman Reference Terdiman1993; Rothberg Reference Rothberg2009; Silberman and Vatan Reference Silberman and Vatan2013). Selective re-evaluations of the past, for instance, can lead to counter-narratives and competitive forms of remembering. Repetitive violence and multi-layered stories can sometimes blur the distinction between perpetrators and victims. For both groups, contingent violence is often difficult to read, creating the urge to create convincing explanations and retrospective interpretations that can obscure the original meaning of past actions. In contemporary Moscow, for instance, the attempt to memorialize violence in twentieth-century Russia has to wrestle with two World Wars, two revolutions, civil wars, forced collectivization and resettlement, the Great Terror of the 1930s, and the victims of the German occupation of the Soviet Union (Schlöger Reference Schlöger, Vatan and Silberman2013). Last but not least, commemorating violence while seeking reconciliation and inclusivity remains a daunting task. In South Africa, the recent events surrounding the removal of the statue of Cecil Rhodes at Cape Town University demonstrate the difficulty in reconciling divergent memories.
For the victims whose core identity and collective relations have been deeply impaired by violence, representing the past is an almost impossible task (Silberman and Vatan Reference Silberman and Vatan2013: 1–3). In a recent study of the Pollsmoor Prison and Valkenberg Asylum in the Cape region (South Africa), Natacha Filippi argues that researchers should always remember that violence is, above all, an intimate experience of disruption, partial annihilation and power branded on a person's body and mind, whether that person was subjected to or responsible for it. In the process of seeking information, researchers might cause a re-emergence of violence as they can mirror the trauma of past events. As an often white, occidental, middle-class individual in an assumed position of knowledge, the researcher can unconsciously reproduce the very categories and world views that contributed to the emergence of violence in the first place (Filippi Reference Filippi2013).
While the articles in this issue do not look specifically at memory and commemoration, they interrogate the continuity between contemporary acts of violence in Africa and the colonial past. All demonstrate that the boundaries between the two periods are markedly fluid when it comes to the dispensation of violence. Aidan Russell traces Burundi's selective genocide in 1961 and 1972 to the bureaucratic methods, road blocks and nominative lists deployed by colonial state agents against ordinary citizens. Nicole Eggers unveils how Bushiri's use of the chicotte during the 1944 Kitawalist movement in Eastern Congo both mirrored and critiqued colonial violence. In today's Congo, she suggests, the repetition of violent acts shows that they are not tethered to the past only by causality and continuity, but by more elusive and less studied processes of ‘reproduction’ and ‘somatization’ (Hunt Reference Hunt2008). First choreographed and implemented by the colonial state, then re-enacted by Bushiri and his followers, the same forms of violence are now used by the militias roaming Eastern Congo.
The case of colonial violence itself, long recognized as central to white domination, has experienced a considerable reappraisal. Recently historians have tended to downplay its scope and reach in Africa, arguing that it applied only to limited moments and places. Yet, as Frederick Cooper has forcefully reminded us: ‘the violence of colonizers was no less violent for the narrowness of its range and the limits of its transformative efficacy’ (Reference Cooper1994: 1545). Michael Watts’ study of famine in Hausaland, for instance, uncovers how seemingly ‘silent’ economic violence in colonial Northern Nigeria destroyed the lives and economic assets of many (Reference Watts1983). Some authors restrict the term merely to acts of physical violence perpetrated by colonial agents. From this perspective, however, it is not at all clear whether colonial violence primarily refers to exceptional events, such as the wholesale slaughter of colonial subjects – particularly in the early colonial period and after the Second World War, during the struggles for independence – or to more quotidian forms of violence, such as the routine use of verbal abuse, slapping, whipping and flogging that often occurred in the workplace and in penal institutions (Anderson and Killingray Reference Anderson and Killingray1991; Bernault Reference Bernault2003; Anderson Reference Anderson2005; Branch and Cheeseman Reference Branch and Cheeseman2006; Hynd Reference Hynd2008; Carotenuto and Shadle Reference Carotenuto and Shadle2012). In hindsight, the repressive and reformative institutions built by colonialists have proved remarkably enduring post-independence (Pierce and Rao Reference Pierce and Rao2006; Bernault Reference Bernault and Dikötter2007). However, while the history of state violence has become a mature field, only a handful of authors have ventured into the terrain of interpersonal violence (Crowder Reference Crowder1988; Burrill et al. Reference Burrill, Roberts and Thornberry2010). Some of the most gripping accounts of these outbursts, otherwise difficult to trace in conventional archives, can be glimpsed in the life stories of Congolese informants collected in the 1980s by Bogumil Jewsiewicki (Reference Jewsiewicki1993), the vampire rumours recovered in Central Africa by Luise White (Reference White2000) and the leopard-men killings in colonial south-eastern Nigeria studied by David Pratten (Reference Pratten2007).
VIOLENCE AND DEEP CULTURAL FRAMES OF ACTION
Meanwhile, the sanitizing paradigm of the ‘colonial encounters’ tends to shift historians' attention away from conflicting dynamics and destructive patterns and, by the same token, from the symbolic and moral frameworks that underwrote colonial violence. We are talking more specifically about the kind of cultural, psychological and mental constructions that were indispensable for maintaining colonial rule, the oppression of imperial subjects, and the dissolution of colonial power. After the pioneering work of Franz Fanon on colonial alienation and anti-colonial violence, only a few studies have probed the deeper terrain of the moral imaginations at work during the colonial era (Fanon Reference Fanon1952; Reference Fanon1961; Lan Reference Lan1985; Mbembe Reference Mbembe2001; Thomas Reference Thomas2003). In her study of fertility in the Belgian Congo, for example, Hunt reveals the hybrid lexicon of power and reproduction that emerged from the use of semantic and embodied ‘debris’ born from the quotidian violence of colonial life, when African men were deprived of both social wealth and their hold on youth and kinship; African ‘boys’ were feminized; and African women were herded as a result of hygienic control measures (Hunt Reference Hunt1999). Specialists in religion have mapped out important aspects of the mental underpinnings of past violence and vulnerability (Comaroff and Comaroff Reference Comaroff and Comaroff1991; Landau Reference Landau1995; Meyer Reference Meyer1999; Peel Reference Peel2000). In a pioneering study of the 1908–25 Watch Tower (Kitawala) revival in Malawi and Zambia, Karen Fields shows that, although the movement unfolded primarily in violent anti-witchcraft hunts and cleansing among the African population, it appeared, correctly, to the British authorities as a major political threat that exposed the breaking lines of the British order (Fields Reference Fields1985: 21).
David Schoenbrun recalled recently that, with the important exception of slavery and the slave trades, historians’ ‘work on precolonial violence has been scanty and unsystematic’, including research on ‘the theories of action that frame imaginary violence, the ever-present double of “real” violence’ (Reference Schoenbrun2006a; Reference Schoenbrun2006b; but see Reid Reference Reid2011). Without these early moral imaginations, it is hardly possible to write histories of Africa in people's own terms, or to track the enduring relevance of such frameworks of action over time (Schmidt Reference Schmidt2013). People pursue their aspirations with techniques and ideas both of great antiquity and of more recent vintage (Schoenbrun Reference Schoenbrun2006b). In the Buganda kingdom on Lake Victoria, for instance, from the late eighteenth century onwards, a class of female mediums called Nyabingi appeared as healers of individual illnesses and as leaders of bands of rebels fighting royal or colonial conquest. The moral framework associated with the healers thus seamlessly interpreted acts of rebellion against the central court of Buganda and against British oppression (ibid.).
Nicole Eggers’ analysis of Bushiri's uprising in Congo (in this issue) also attempts to recover some of the overarching imaginations at work in the movement. Although a large part of the violence deployed during the revolt was perpetrated by colonial authorities, Bushiri's movement arrested and beat dozens of villagers accused of witchcraft. Eggers analyses the role of what the Congolese called nguvu or puissance (agency, force), a dynamic theory of power that connected the realm of the visible with the invisible. For Bushiri, violence served to cure the dangerous imbalance in puissance that had been wrought by colonialism, and to restore social health. Yet the brutality of Kitawala cannot be reduced to these restorative tactics. Instead, Eggers argues, the relative morality of violent acts was a contentious subject that Bushiri's followers constantly renegotiated along lexicons of power that articulated violence in dualistic terms: destructive/productive, eating/healing.
As the reverse of social health, witchcraft has been one of the most important ways of approaching common frameworks of action and interpretation, a force that ‘both generates and feeds upon violations of the fundamental norms of kinship solidarity, social prosperity, reciprocity’ (Karlström Reference Karlström2004: 595–7). It is also one of the most studied forms of violence in Africa today: anthropologists have yielded a rich harvest of local, regional and comparative studies on witchcraft attacks and reparative violence (Fields Reference Fields1985; Geschiere Reference Geschiere1997; Reference Geschiere2013; De Boeck and Plissart Reference De Boeck and Plissart2004; Ashforth Reference Ashforth2005; West Reference West2005; Niehaus Reference Niehaus2013). In comparison, we know far too little about changes in witchcraft violence over time (Luongo Reference Luongo2011). An exception has been achieved by anthropologist Heike Behrend, with a rich narrative of the Tooro region in Uganda and the modern crisis of ‘resurrecting cannibals’ accused of reviving the dead in order to feed on them (Reference Behrend2011). Behrend performs a ‘counter-ethnography’ that, instead of opposing vernacular ideas with foreign beliefs, explains how the witch-cannibal emerged both in the European and in the African imaginations.
In attempting to contextualize the research on violence in Africa, this introduction has shown that there are still significant gaps in the existing literature. We hope that future historical and anthropological works will enrich our understanding of the changing methods and meanings of violence in Africa.
