Navegação – Mapa do site

Mutual Transformations of State and Traditional Authority. The renewed role of chiefs in policing and justice enforcement in Mozambique

Transformações Recíprocas do Estado e da Autoridade Tradicional. O renovado papel dos chefes na aplicação da justiça e da lei em Moçambique
Helene Maria Kyed
p. 179-201


Este artigo explora o renovado papel dos chefes na aplicação da justiça e da lei a partir da perspectiva das práticas quotidianas e modos de organização das relações entre os chefes e as instituições estatais locais. Baseado em material etnográfico recolhido em Dombe, distrito de Sussundenga, em Moçambique, questiona-se quais as implicações das novas relações quer para o Estado local quer para as autoridades tradicionais. O artigo demonstra que o Decreto 15/2000 foi apropriado pelos membros locais da polícia estatal não como um reconhecimento benigno das práticas já existentes dos chefes mas sim como um meio para controlar os chefes e proteger a autoridade do Estado na antiga zona de guerra em Dombe.

Topo da página

Texto integral

1The Decree 15/2000 revoked more than 25 years of official government ban on traditional leaders or chiefs in Mozambique. Chiefs were from 2002 recognised by the state as community authorities and as assistants of the local tiers of the state in a range of state administrative, developmental and security matters. Amongst these was a return to state recognition of chiefs’ roles in policing and justice enforcement, roles that date back to pre-colonial history and that were officially recognised during late Portuguese colonial rule (Kyed, 2007). Decree 15/2000 obligates chiefs, now community authorities, to collaborate with the local state police in identifying troublemakers and engage in the resolution of civil conflicts in liaison with the local community courts (Boletim da República, 2000a: article 5b-c). Legislation is nonetheless full of ambiguities. It gives few clues to how the future interaction between chiefs and the state police and courts should be regulated and organised. Chiefs’ courts and system of policing are not recognised by law and there is no official incorporation of chiefs within the justice and security sectors.

2This article explores the renewed role of chiefs in policing and justice enforcement from the perspective of every-day practices and modes of organising the relationship between chiefs and local state institutions. Based on ethnographic material from Dombe in Sussundenga District it asks what the newly forged relationship implies for local state and traditional authority. The article shows that the Decree 15/2000 was appropriated by the local tiers of state police not as a benign recognition of already existing chiefly practices, but as a means to regulate chiefs and bolster state authority in the former war-zone of Dombe. This attempt has not been without contradictions, however. In the very process of state police attempts to reshape chiefs, local state practices were reshaped too. Consequently state recognition of traditional authority implies mutual transformations of chiefly and state authority.

3Repeated attempts by local state officials to fix a boundary between state and chieftaincy as distinct domains of authority were circumvented; the chiefs’ own tendency to define themselves in opposition to the state deflected. In fact, multiple practical fusions challenged the distinction between state and chieftaincy. Local police officers began to take decisions on witchcraft accusations using official stamps and procedures, although they claimed that witchcraft exclusively falls under the jurisdiction of traditional authority. Chiefs often referred to state law in dispute settlement, although they just as often flouted the state law. The result is hybrid authorities, rather than strictly separate and distinct forms of state and traditional authority.

4The mutual transformation of traditional and state authority in Dombe questions two major positions on state recognition of traditional authority in Africa. The first holds that recognition of chiefs since colonial rule has resulted in state co-optation and transformation of the chieftaincy in service of state rather than popular interests. This has eroded pre-colonial traditional authority, turning present-day chiefs into little more than state bureaucratic inventions (Costa, 1999; Serra, 1997; Mamdani, 1996; Herbst, 2000; Jordan 1997; Ntsebeza, 1999). The second position by contrast argues that chiefs have resisted co-optation by the state apparatus and retained legitimacy rooted in pre-colonial culture and traditions. Chiefs have only been partially reshaped by years of interaction with state institutions. They have become hybrid authorities, drawing on both modern-state and traditional forms of legitimacy (Ray and van Nieuwaal, 1996; von Trotha, 1996; van Dijk and van Nieuwaal, 1999; Quinlan, 1996; Sklar, 1999). These two positions shed light on the state recognition of traditional authority in Dombe, the first highlighting the state’s attempt to incorporate chiefs, the second chiefs’ capacity to remain distinct from the state. However, none of the positions address whether state practices may also be (re)shaped through interactions with chiefs. The state remains unchanged, either as a distinct domain separated from traditional authority or as a powerful entity that has eroded traditional authority. This article challenges both these propositions. It shows that local state officials’ attempt to co-opt chiefs also reshapes state practices, and that distinct domains of state and traditional authority are expressive of ongoing political processes, rather than inevitable, fixed structures.

5The article begins with a brief background to the state recognition of traditional authority in Mozambique and the ambiguous legal framework that pertains to the formal roles granted to chiefs in policing and justice enforcement. It then moves to Dombe in Sussundenga District, exploring first how the new legal framework was appropriated by the local tiers of the Mozambican Police Force (Polícia da República de Moçambique – PRM) and what immediate consequences it had for chiefly and local state police authority. Secondly it explores the everyday practices of chiefs and police officers in policing and justice enforcement. Finally, the article concludes by discussing what these everyday practices mean for emerging forms of local state and traditional authority.

The ambiguous legal framework

6After the end of the civil war in 1992 the importance of traditional leaders in local governance emerged as a matter of topical interest in diverse national circles in Mozambique. This reflected a general trend across Sub-Saharan Africa in the 1990s in which transitions to liberal-democracy were paralleled by a resurgence of traditional authority, both from below and through top-down state legislation (Kyed and Buur, 2007). Whereas international donors in Mozambique looked to traditional leaders as part of decentralisation and community-based programmes, Mozambican academics saw traditional authority as an important cultural-symbolic value of Mozambican society and as a building bloc of national reconciliation. Conversely, the FRELIMO government became increasingly content on recognising those chiefs it had banned at independence in 1975 for state administrative and political reasons (Kyed, 2007). This change in attitude was intimately related to the former rebel movement, and now opposition party, RENAMO’s effective reinsertion of chiefs in governance in the rural areas it controlled during the war and its use of traditionalist rhetoric as a counter-ideology to the FRELIMO government. RENAMO’s reliance on chiefs was increasingly seen as one reason for RENAMO’s pervasive encroachment of rural territory during the war and its subsequent post-war electoral victory in many rural areas in 1994 (West and Kloeck-Jensen, 1999). One such rural area was Dombe in Sussundenga district, explored in this article.

  • 1 On reasons why it took almost ten years from the first policy studies of traditional authority to t (...)

7During the war RENAMO managed to take control of the entire territory of Dombe administrative post and establish a system of partly military and partly civilian rule based on chieftaincy (Alexander, 1997). Consequently, official state administrative and police institutions were absent in Dombe at the end of the war. This continued until the end of 1995, when state police presence was re-established after several failed attempts since 1994. Despite the fact that FRELIMO had won the 1994 elections, RENAMO and local chiefs refused to recognise the result. This led to a number of violent clashes between the state police and disgruntled chiefs (Notícias, 1995; Savana, 1995). The ‘Dombe case’ was held out as paradigmatic of the problems facing post-war local state administration and policing vis-à-vis unrecognised chiefs, and was paralleled by similar incidents in other rural areas. Thus many local state officials saw state recognition of chiefs as counterparts of the state as a means to ease the cumbersome process of rebuilding state presence and legitimacy in the rural areas (Kyed, 2007). This view underpinned the state administrative reasons behind the FRELIMO government’s increased positive outlook on a legislation that would include traditional leaders in local governance. It was gradually incorporated in the policy-making process that led to Decree 15/2000, the first piece of post-colonial legislation to officially recognize traditional authority in Mozambique1.

  • 2 The Decree includes three categories of community authorities that can be legally recognised by the (...)

8The Decree 15/2000 envisages chiefs as community legitimised authorities and as key assistants of local state institutions (Buur and Kyed, 2005). As community authorities, chiefs are delegated a range of key state-administrative and security duties that inter alia include policing, taxation, population registration, justice enforcement, land allocation and rural development. They are also obligated to perform various elements of civic-education in their communities, e.g. preventing crime, informing about the law, fostering a patriotic spirit, supporting the celebration of national days, and preventing epidemics, HIV/Aids, and premature pregnancy and marriage (Boletim da República, 2000a: art. 5)2.

9What emerges from the Decree and its regulation’s list of duties is a rather multifaceted cocktail of tasks, which even the most advanced bureaucratic machinery, would struggle to carry out. One would therefore expect a clear delineation of the concrete steps to be taken. However, the Decree is unclear on how the relationship between the state institutions and chiefs should be organised in practice. Lack of clarity pertains in particular to the policing and justice enforcement duties. The regulation of the Decree states in few words that the community authorities are obligated to “articulate with the community courts, where they exist, in the resolution of small conflicts of a civil nature and in accordance with the local uses and customs, all within the limits of the law” (Boletim da República, 2000a: art. 5b). For policing they are obligated to “participate with the administrative and police authorities in attending to the commitments of violations and in the existence and localization of troublemakers, hidden arms and mined areas” (ibidem, art. 5d).

10These articles leave several operational questions unanswered: What precise types of cases should be solved by state institutions (police, the formal court system and the administration) and by traditional or community authorities? What procedures should be used for dividing and passing on cases between state and non-state institutions? How should community authorities concretely assist the police in localizing and dealing with troublemakers and violators of the law? And finally, while the regulation makes clear that conflict resolution should take place within the confines of state law, it is still pertinent to ask: who has the authority to define the rules and procedures for solving non-criminal or civil cases (e.g. forms of punishment/sentences)? With regard to this question it is significant to add that there is no universally encoded customary law in Mozambique as opposed to South Africa and Zimbabwe. What is referred to as such, is therefore highly localized, negotiable and varies between the abundance of chieftaincies. In addition, while Decree 15/2000 implicitly recognizes chiefs as authorities who enforce justice in respect to cases of a civil nature, their courts are not officially acknowledged by the formal justice system. So far this only covers the community courts (tribunais comunitários), which with Law 4 of 1992 were established to enforce justice of a non-official character at community level (Trindade and Santos, 2003: 72). Moreover, chiefs’ role in assisting the local tiers of the state police (from hereon PRM) is not officially attached to the Ministry of Interior, under which the PRM falls. Rather the regulation of chiefs and community authorities more generally falls under the Ministry of State Administration.

11Lack of operational clarity in the Decree and the separation of administrative, justice and security sectors presented legal grey-zones to its implementers, i.e. local state officials. In practice the legal grey-zones gave way to the dominance of localised strategies, interpretations and huge room for manipulation and negotiation in the implementation of Decree 15/2000 in the area of Dombe (Buur and Kyed, 2005: 14-5). Importantly, lack of operational clarity also reflected a core pretension of legislation: state recognition of traditional authority was presented by policy-makers, the FRELIMO government and higher ranking state officials as a simple recognition of what already existed, chiefs and their communities. In official discourse it was widely held that state and traditional forms of authority could peacefully collaborate to the advantage of both, without disturbing either of the two domains of authority. Ultimately this pretension was based on the assumption of state and chieftaincy as representing generically different types of authority. As I address next, this assumption did not reflect local reality in Dombe. Neither did the pretension of non-disturbance mirror how the local state police went about translating the Decree into practice. The renewed relationship between chiefs and the state police reshaped the authority of each, rather than preserving presumably separate domains of authority.

Local state appropriation of decree 15/2000 in Dombe

  • 3 On the process of legitimisation and recognition in Dombe see Kyed (2007) and Buur and Kyed (2006).

12By August 2002 eight chiefs had been recognised as community authorities in Dombe Administrative post, and an additional fourteen sub-chiefs had been registered by the local state administration3. The formal recognition of chiefs was met with enthusiasm from local state administrative and police officials. Alliances with chiefs facilitated state re-establishment in the former hostile war zones of Dombe and the delegation of tasks to chiefs was seen as a means to make state governance more effective. As held by a state official: “The Decree (15/2000) is a blessing to us. It now means that we can formally work with the chiefs and reach the populations out there”. Similarly, the chief of police asserted: “Alone the police can do nothing. We need the collaboration of the chiefs to secure law and order. Because it is the chiefs who know what goes on out there with the people”.

13Despite these positive views of chiefs, the actual organisation of the collaboration between chiefs and local tiers of the state that followed the formal recognition ceremonies in 2002 underpinned inherently ambiguous attempts by local state officials to reorder and regulate chiefs. Central to this endeavour were attempts to reclaim superior state authority by incorporating chiefs under the command hierarchy of the state and prohibiting them from performing particular functions and taking decisions on issues that they had hitherto be accustomed to during the war years of state absence in Dombe. Thus state recognition and collaboration came at a price for chiefly authority. Local state appropriation of Decree 15/2000 did not reflect, as officially asserted, a benign recognition of already existing forms of chiefly practices and claims to authority, but rather a reshaping of these to fit state demands. This was not least reflected within policing and justice enforcement where the local tiers of the PRM took charge of decisions on chiefs’ future roles.

  • 4 The chiefs’ courts in Dombe as elsewhere in Manica province, consisted of the chief and his/her cou (...)

14In light of the legal grey-zones described earlier, the PRM in Dombe invested enormous energy in organising and formulating a set of rules for how the chiefs could perform, what their mandates were and how the labour division between the state police and chiefs should be. This de facto gave way to what can best be described as a localised, secondary body of law that filled the gaps left open in formal legislation. But they also expanded formal legislation with a range of prohibitions and obligations on chiefs that were attached to a set of extra-legal sanctions. Central to the secondary body of law were attempts by the PRM to reclaim state monopoly on settling criminal offences, use of force and issuing of punishments of expulsion. By the time of implementing the Decree in Dombe chiefly practices and claims to authority considerably overlapped and competed with these claims to state monopoly. Chiefs had regularly settled what in the state Penal Code is referred to as criminal transgressions such as theft, homicide, arson and physical assaults, as well as issued punishments that covered expulsion and physical disciplining. This co-existed with areas of conflict resolution outside official law, such as witchcraft and transgressions of traditional norms (i.e. adultery, intercourse in the bush, violation of sacred places, land disputes and marriage payment) and the issuing of penalties that differed from the official courts, i.e. material compensation and restorative justice. However, in chiefs’ courts (banja in local Chi-Ndau dialect)4 no conceptual distinction was made between criminal cases covered by the Penal Code and other types of delinquency outside official law. Rather the common concept of kushaisha (literally evil-doing) was applied to all kinds of transgressions of the mutemo yo passe chigare (the rules or traditions of the ancestral spirits), which both included and went beyond categories of crime in the Penal Code.

15The crux of the matter is that the chiefs’ self-proclaimed jurisdictions of order and justice enforcement were not entirely distinguishable from, but considerably overlapped with the mandates of official courts and the role of the state police in order enforcement. The difference lay in conceptualisations of delinquency, not in the actual acts of delinquency that were dealt with. This reality on the ground challenged the official assumption of the Decree that the state institutions and the chieftaincy represented entirely separate and distinct forms of authority that could co-exist and collaborate without competition. Rather than being a fact from the outset, the separation of state and chiefly jurisdictions of authority, was a central aim of the secondary body of law of the local tiers of the PRM. This body of law reflected a direct response to the overlap of state and chiefly jurisdictions.

Local state police law and regulation

16The secondary body of law of the local police covered three sets of rules to reverse the situation of chiefly competition with state institutions. First, it covered a set of rules that separated chiefly and state jurisdictions by drawing a boundary between the types of transgressions that each were allowed to settle. This boundary was defined according to the PRM’s classification of two categories of cases: criminal and traditional. Only the official state institutions were permitted to settle criminal cases. The chiefs’ courts had the exclusive authority to settle the traditional cases. The PRM’s definition of criminal cases covered those acts that violated the land, insulted authorities and that inflicted violence on human bodies. In short, acts that were physically destructive, covering inter alia homicide, fights in which blood is spilt, rape, stabbings, larger thefts involving the use of weapons and violence, the use and production of drugs and arson. All these acts were defined by the PRM officers as crimes against the state and thus as punishable by the state and the state alone. Chiefs were told by the PRM, that they were strictly prohibited from settling such transgressions. To do so would be treated as criminal offences – i.e. as law-breaking. This aspect reflected how the secondary body of law, although never written down, was communicated by the PRM as having the status of codified state law, i.e. as attached to sanctions. But it also had other consequences.

17If the PRM’s category of criminal cases corresponded to the Penal Code, they also overlapped considerably with those elements of the mutemo yo passe chigare, which underpinned the chiefs’ authority to make decisions on the taking of life, insulting of chiefly authority, the spilling of blood and violations of the land. Consequently the claim to a state monopoly on settling crimes against the state, implied that the PRM re-defined what counted as traditional cases: the category covered, according to the PRM, those kinds of conduct that chiefs considered to be against the tradition (mutemo), but excluded those acts defined as a crime against the state. This considerably delimited the self-proclaimed jurisdictions of chiefs, and ultimately their authority within important spheres of order and justice enforcement. If this can be seen as a general affirmation of codified state-law, then the PRM also expanded official law by recognising witchcraft as part of the category of traditional cases. Hence, while the PRM criminalised the authority of the chiefs to enforce significant elements of mutemo yo passe chigare, it de facto recognised witchcraft which is officially outside of or unrecognised by state law. The same applied to other cases that chiefs were permitted to settle such as adultery, divorce, marriage payments, debt, and land disputes between neighbours. The recognition of these types of cases reflected another key characteristic of the PRM’s extra-legal rules: as an element of adapting the Decree to the local context the police recognised institutions and practices outside the codified law, but at the same time prohibited chiefs from entering the domain of what the PRM defined as inside the law.

18Consequently, the PRM’s rules marked a boundary between the domains of traditional and state authority, which hitherto had been blurred. This re-assertion of state authority also applied to the second set of rules that outlined the kinds of punishments that chiefs were allowed to issue. Whereas chiefs were allowed to enforce monetary compensation, they were strictly prohibited from using any kind of corporal punishment, physical discipline and expulsion. These were claimed the monopoly of the state. The prohibition had particular consequences for the chiefs, whose use of such punishments had been a significant marker of a chiefs’ superior authority. Although expulsion and corporal punishment was rarely issued by chiefs in Dombe, it was an important option (often in the form of a threat) to regulate those members of the chieftaincy that threatened the authority of the chiefs or who repeatedly violated the mutemo yo passe chigare. Thus the prohibition can be seen as another element of criminalising those aspects of chiefly authority enforcement that not only challenged state monopoly on physical violations, but also underpinned chiefly authority to enforce order. The state recognition of chiefs’ role in justice enforcement was therefore highly restricted by the PRM’s rules. The same applied to the third and final set of extra-legal rules covering prohibitions and obligations for how chiefs should assist the PRM in policing activities.

19In accordance with the regulation of the Decree chiefs were obliged to locate criminals and suspects and forward this information to the local tiers of the PRM. However, the local PRM’s third set of rules also went beyond this by obliging chiefs to arrest law-breakers and suspects and bring them to the police station. Refraining from collaborating with the police in these ways would be regarded as a crime, the chiefs were told. The same applied to concealing information about criminals. Chiefs were also allowed to tie up criminals or suspects if they resisted arrest, but to use force that resulted in physical injuries would be regarded a crime. Only the PRM officers, chiefs were told, were allowed to use force. Thus chiefs were, as a matter of obligation, drawn into the PRM’s domain of crime control and law-enforcement, i.e. as the extended arm of the state police in the rural hinterlands. At the same time chiefs were set apart from the state police, as marked, for example, by the PRM’s claim to a monopoly of force. Outsourcing of functions to chiefs was accompanied by criminalisation of those chiefly practices that challenged the authority of the state police. This was backed by the threat of a set of local state-enforced sanctions that were equally outside the law: i.e. neither Decree 15/2000 nor any other law included a list of sanctions for disobedient chiefs. For example during fieldwork in 2004 and 2005 the PRM in Dombe punished two chiefs and seven sub-chiefs for failing to abide by the police’s rules and thereby challenge state police authority. Punishments were issued at the local police stations, using force, compelling chiefs to work for the police and/or detentions of chiefs in the local cell. The cases thereby never went to the official court, but were strictly enforced outside the law.

20In sum, the PRM’s secondary body of law represented a particular localised appropriation of Decree 15/2000. It marked an attempt to reclaim state police authority in law and order enforcement by congealing a strict separation between the jurisdictions of chiefs and the formal state institutions that did not exist prior to the Decree. If this congealment reflected state police adaptation to the local reality of overlapping forms of state and chiefly authority, then the intriguing part was that reclaiming state police authority relied on rules and sanctions that partly lay outside state law and official mandates. This, I suggest, reflected both the legal grey-zones in codified law and a particular dilemma facing the local tiers of the PRM: local state police officers depended on chiefs’ assistance in dealing with criminals and thereby in bolstering their own authority, but at the same time felt threatened by chiefs’ self-proclaimed authority to deal with transgressions and make decisions that competed with state police authority. This underpinned the need for both collaboration and strict regulation of chiefs. Consequently chiefs were simultaneously incorporated within the state system and set apart from it as a separate traditional domain of authority enforcement.

21The immediate implication of the PRM’s secondary body of law for the authority of chiefs and the local state police was precarious. Chiefs were positioned in an anomalous position as state assistants, but prohibited from enjoying the same authority as the local state police. In addition, their authority to enforce particular punishments and settle important transgressions of mutemo yo passe chigare were criminalised, and backed by severe (local) state-enforced sanctions. Conversely, the local PRM’s own rules for regulating chiefs’ roles also instigated a particular form of local state police authority that relied on extra-legal rules, exemplified by recognising chiefly practices outside the codified law (e.g. witchcraft and policing tasks) and by enforcing sanctions and punishing chiefs outside the official justice system. Thus the local police’s attempt to regulate and collaborate with chiefs in order to reclaim state authority, underpinned a mutual transformation of chiefly and local state authority. If this was the immediate implication of the PRM’s extra-legal rules then it was even more so in the everyday practices of policing and justice enforcement.

Everyday practices of policing and justice enforcement

22The local chief of police in Dombe and his inferiors spent a lot of energy on communicating the secondary body of law at larger public gatherings in the chieftaincies and at closed meetings with chiefs and community court personnel. The punishments of chiefs also served as pervasive examples of the force behind the PRM’s rules and interviews with rural residents and the chiefs revealed widespread knowledge of them. Irrespectively, the classificatory boundaries drawn between criminal and traditional cases and between which authorities should deal with these were continuously blurred in practice. In fact my analysis of 243 cases, including criminal and non-criminal transgressions, clearly shows that adherence to the PRM’s secondary body of law was less the rule than the exception (Kyed 2007: 261-3). While chiefs increasingly collaborated with the police, they continued to settle a very large number of what the PRM defined as crimes against the state, and rural residents frequently took their cases to the wrong authorities. Most intriguingly the local police officers, including the central police station in Dombe, also engaged in breaching their own rules. They did this by receiving and hearing an increasing number of witchcraft cases and other so-called traditional cases that according to the police’s own rules were supposed to be exclusively dealt with by chiefs. In doing this the PRM entered the domain of chiefly authority enforcement, whereas chiefs continued to challenge the state’s exclusive authority to settle criminal cases. The boundaries of distinct domains of authority remained blurred, despite attempts to the contrary.

23The question is why the PRM’s secondary body of law was constantly breached, and what this implied for the forms of local state and traditional authority that emerged in everyday practice. Based on interviews and participant observation of case settlement within the police stations and at the chiefs’ courts, I suggest that the answer to the first part of the question lies with local state police officers and chiefs’ adjustment to rural residents’ understandings of transgressions and preferred forms of justice as an element of claiming and maintaining authority. These local understandings fared uneasily with the PRM’s categories of transgressions and the kinds of justice enforced by the official courts. Let me briefly elaborate on these two points, and their implications for everyday practices of policing and justice enforcement.

Blurred boundaries – local conceptions of justice and evil-doing

  • 5 This perception of misconduct or transgressions is intimately linked to the worldview of the Chi-Nd (...)

24In Dombe the PRM’s separate categories of transgressions (criminal and traditional) did not fit local understandings of evil-doing captured under the common term for all transgressions, kushaisha. People in Dombe did not strictly separate acts regarded as crime by the Penal Code and for example adultery or death caused by witchcraft. All visible acts of misconduct were seen as attached to invisible evil-forces5. In addition, actual cases of conflicts between two parties often involved both a so-called criminal and traditional transgression, which made the separation porous. Whereas adultery could lead to the crimes of homicide or physical assault, witchcraft cases could both emanate from and lead to crimes such as theft, arson, homicide and physical assaults. To understand why this was the case, one needs to understand that witchcraft is locally used to explain manifestations of physical harm in the form of sickness, bad luck or even death. Although the actual act of witchcraft is invisible to ordinary people, it is believed to inflict one person by another person due to some prior conflict between the persons or their family members. Thus witchcraft was used to explain invisible acts of revenge emanating from for example a quarrel or crimes such as theft and homicide. This merger of witchcraft, considered a traditional case by the PRM, with criminal acts, made it difficult to neatly separate those cases that the chiefs should settle and those that should be dealt with by state institutions. Consequently, in everyday practice both chiefs and the PRM ended up being addressed with cases by rural residents that both involved witchcraft and crime. This considerably challenged the PRM’s secondary body of law.

25A second reason why the PRM’s law was precarious in practice was because local understandings of evil-doing underpinned a tension between local notions of appropriate justice enforcement and the kinds of justice most commonly dispensed by the official state institutions. To Dombians, material compensation enforced by chiefs’ courts was by far the most preferred form of justice in criminal and other cases, because it was seen as a means of both compensating losses (in light of no private insurance) and of removing the invisible evil-forces of the perpetrator that had caused him or her to inflict harm. It was also seen as a way to avoid future inflictions of witchcraft, i.e. in the form of revenge. Conversely, imprisonment or fines to the state, enforced by the official justice system, was seen not only as inhibiting material compensation to the victims, but also as worsening the evil-forces of the perpetrator (i.e. giving way to future transgressions). In everyday practice this meant that many rural residents (including victims and the family of perpetrators) preferred to have a criminal case settled by the chiefs’ courts, namely because they wanted to avoid imprisonment and potential future witchcraft inflictions. Taking a crime to the police was seen as risky, because it could mean ending up in the official court at district level and thereby risking lack of appropriate restoration of order.

26These local perceptions of justice and restoration of order help explain why chiefs’ continued to settle what the PRM defined as criminal cases. Despite the risk of punishment by the police that it could imply, chiefs explained that they still at times settled criminal cases as a result of adjusting to the preferences of victims. However, this continuity of old practices also merged with changes of chiefly authority enforcement, which was paralleled by increased incidences of the state police receiving and hearing traditional cases. I address these two aspects next.

Settlement of criminal cases by chiefs

27Continuity of old practices should be seen in light of the precarious position chiefs found themselves in after state recognition, i.e. between the conflicting demands of rural residents and the PRM. Whereas chiefs risked loosing credibility with and punishment by the PRM when settling crimes, they could undermine their own authority with the rural residents if they adhered strictly to the PRM’s rules. However, as the illustrative case below brings to light, chiefs modes of enforcing authority also changed as a direct consequence of their renewed relationship to the state police.

Case 1: In 2004 João accused another man for having slept with his wife and the wife for having committed adultery. The case was taken to the chiefs’ court, because, as is the custom, the husband wanted material compensation from the man who had stolen his wife. Compensation is believed to make the husband avoid becoming sick or doing evil. However on the day of the court session the accused man had disappeared. The chief asked João what he wanted him to do now that the perpetrator was absent. He answered that he wanted the chief to “educate the woman so that she will not repeat what she has done”. The chief first refused, saying “Ahh, but we [chiefs] are not allowed to do that anymore…to punish people with force…this is the law of the police”. However, after thinking a moment, he agreed and ordered one of his police assistants to give the woman five strokes with a whip made of a tree branch. The case was seemingly settled after this. However, at the court session two weeks later, the couple turned up again, this time with the presence of the wife’s parents. João was accused of having severely beaten his wife when he had found out that she was pregnant, believing that it was with the other man. The chief firmly explained that “this is a crime that should be taken to the police…this is the law of the state, because it is these things that can end with murder”. Upon hearing this, João begged not to be sent to the police. The chief then asked the wife and her parents what they wanted to happen. The wife said she did not want her husband to go to the police because then he could end up in prison. She was afraid that if he went to prison, his family would blame everything on her, because she had slept with another man. Her father supported this view and added: “If he goes to prison he will not be able to support my daughter and her two children”. The father also asked the court to make João pay material compensation to him for physically injuring his daughter. After reiterating that “this case is a crime that should go to the police”, the chief agreed to the requests of the victim’s party. He closed the case after João promised never to beat his wife again and to pay compensation to his father-in-law: “If you do not do this”, the chief promised him, “I will personally take you to Dombe” [the police].

28This case is illustrative of how and why chiefs continued to breach the rules of the PRM. It illustrates why chiefs settled what they themselves defined as criminal cases according to the law of the police, namely as an element of adjusting to people’s preferences for justice. The same applied to the issuing of corporal punishment. But this is not all. The case also illustrates a general evolving pattern of how chiefs settled criminal cases by increasingly referring to the law of the police and their renewed role as state police partners. Paradoxically, chiefs’ ability to enforce sanctions, against the law of the police, relied on their capacity to refer to the law and threaten perpetrators with sending them to the police. In other words, people’s knowledge of the chiefs’ formal connection with the police became an asset when chiefs’ courts settled crimes. It bolstered the authority of the chief to enforce decisions. At the same time chiefs formal connection to the police gave way to significant transformations of how chiefly authority was enforced, namely by increasingly drawing on references to the state police and its rules.

29Having said this, it should be noted that many people in Dombe did not exclusively see chiefs as the only route to achieve justice. Some chiefs were less capable of ensuring that compensation was paid, and when this was the case the police could be seen as a last resort to deal with a criminal offence as well as with a traditional case. This brings me to the question of why and how the state police began increasingly received and heard traditional cases.

Police resolution of traditional cases

30As held for chiefs, the police’s involvement in settling traditional cases partly owed to police officers’ adjustment to rural residents’ preferences for justice as an element of claiming and demonstrating state police authority. Police officers were aware that if they simply enforced the law, such as sending people to the district court, or sending people away to chiefs with traditional cases, they would risk loosing popular authority vis-à-vis chiefs. However, it also had to do with issues of crime prevention: local police officers to a large extent shared local conceptions of a potential link between witchcraft and crime. Thus to deal with witchcraft was seen as a means of adverting crime.

31Conversely, for rural residents the willingness of the police to adjust to their preferences for justice opened up possibilities for strategically using the police in securing that cases were settled. People commonly used the police as an institution of appeal, when resolutions at the chiefs’ courts had not materialised or when a chief had failed to make the accused turn up for a court session. In fewer cases people took their case to the police as a first option, because they did not believe that their local chief was capable of enforcing a resolution. That it made sense to turn to the police could not be divorced from people’s view of the particular power of the state police vis-à-vis chiefs. This was related to the police’s instruments of force and the general fear that Dombians had of the police in using such instruments. Importantly, turning to the police also made sense because the police in fact responded to the requests of victims and employed a number of practices that were effective in facilitating settlements of cases.

32In fact a little over a quarter of the so-called traditional cases that I followed during fieldwork were taken to the police by rural residents. Instead of the police sending people back to the chiefs as their own rules prescribed, they heard them and overtime developed set routines for dealing with their resolution. In Dombe there was even a special room within the police station reserved for traditional cases. The everyday practices of the police in hearing such cases considerably resembled the resolutions mechanisms of chiefs: the police, like the chiefs’ courts, carefully listened to and supported the requests of the victims in terms of preferred forms of justice, usually material compensation, and they drew on the local conceptual links between witchcraft and crime in pushing for resolutions of conflicts. This blurred the strict boundary between state police and chiefly authority, and also transformed state police authority enforcement. However, the routine procedures of the police in settling traditional cases could also be distinguished from those of chiefs. This included using state bureaucratic artefacts, such as the official stamp of the police in notifying accused persons and in recording verdicts, and it covered explicit references to the state police’s instruments of force. As illustrated in the cases below from the police station of Dombe in 2005, these distinguishing features of the police was significant aspects of why people in fact took their traditional cases to the police.

Case 2 and 3: Five non-criminal cases were heard at the Dombe police station this day in August 2005. The last two cases involved a father and his son, João and Elias, who had travelled around fifty kilometres on foot to each bring a case before the PRM. Elias was the first to speak. He described a case that had begun two months earlier, when his nephew died after severe illness. Elias’ wife Inês was accused of having caused the death through witchcraft. This had been confirmed by a traditional healer (nyanga in local dialect). But Inês refused the accusation. As a result the father of the child took the case to the local police post. The police notified the parties and, after a hearing, the officer ordered Inês to remove the witchcraft from the child, but she refused. The police decided to send the parties to the local chief, because only a chief can send people to a nyanga. At the nyanga Inês was accused again. The chief as a result imposed a fine of Elias and her wife. Elias did not want to pay the fine, arguing that the nyanga had been a liar. Elias insisted that they consult another nyanga before any compensation was paid. But the chief refused his request and threatened with an even higher fine. After this last information the PRM officer intervened, asking “What are you trying to bring forward here? Who are you accusing?” Elias responded: “We are accusing the chief of solving the case badly…that he refuses to send us to the nyanga”. The officer responded by writing a stamped police notification to the chief, stating aloud that “You have to appear here together with all the other persons in the case this coming Friday the 26th of August and solve this case here at the police station”. After this Elias’s father, João, brought another case forward. He explained that his fifteen-year-old daughter was asked a year ago by a man to marry her. But he refused because his daughter was too young. However, one day she ran away to the man and got pregnant. After hearing this, the police officer asked: “Why have you come here with this case?”. João wanted the man to take responsibility for the pregnancy and pay lobolo (marriage payment). The officer asked for the man’s name, wrote another notification and ended by stating: “You can tell him [the accused] that if he does not appear here on the 26th of August, then he will have two cases. One for making a young woman pregnant, and another for abusing the police [failing to turn up at the police station]…if he does not come we will arrest him and educate him [moving his hands to show that he meant using the sjamboko or baton]. That’s all. You can now go”. After the hearing I asked Elias and João why they decided to travel all the way to the police in Dombe with the cases. Elias believed that “the police can help bring the case to an end by telling the chief to solve it well”. With regard to the marriage problems, João said he was convinced that when the accused received the notification he would comply, because “He will see that it [the notification] comes from the state police…and then he will be too afraid not to turn up…you know, as he [the officer] said there the police will sjambokear him [beat him with a baton] if he does not come”.

33These two cases demonstrate how the police was addressed by rural residents as mediators in traditional cases, because of the specific power of police notifications, holding the official stamp of the state. The power of these notifications marked a clear difference from chiefs, because they were attached to an order, or an obligation coming from the state, and backed by the threat of state-police sanctions. Notably this threat was attached to the use of force: failing to abide by police notifications was treated as abusing police authority for which the abuser would be educated, that is treated with force as indicated by the gesture of the officer in the second case presented above. That police notifications were effective was underscored by the fact that in only one of the incidents I came across did the accused completely fail to turn up. There were also concrete examples to draw on. For example, in 2004 I encountered three incidents in which the accused in traditional cases were punished with force by the police because they only turned up at the police station after a second notification.

34What these evolving action patterns of the police suggest more broadly is that, while entering the chiefly domain of case settlement, the police’s capacity to facilitate resolutions of traditional cases was made possible by simultaneously enacting the distinctive authority of the police as state representatives. Thus while the boundaries between chiefly and state police domains of authority enforcement became blurred, the settlement of traditional cases by the police also became part of claiming and demonstrating the distinctive authority of the police vis-à-vis chiefs. The form of local state police authority this gave way to was not a mirror reflection of an ideal-type Weberian form of state-bureaucratic authority, but a particular localised form of authority that was shaped by rural residents’ preferences for justice and the competition with chiefs over satisfying these preferences. As I suggest in the conclusion next, this localisation of state police authority reflected the emergence of hybrid forms of both state and chiefly authority following the implementation of Decree 15/2000.

Towards a conclusion – hybrid forms of authority

35The vast literature on African chieftaincy has commonly emphasised how state recognition of chiefs either means that chiefs have become completely encapsulated by the state or that chiefs have remained distinct from the state by also drawing on traditional practices. The latter position has led some scholars to speak of chiefs as hybrid authorities (van Dijk, R. and van Nieuwaal 1999; Ray and van Nieuwaal 1996). The concept is applied to describe the mixture of state and traditional sources of legitimacy that chiefs draw on (state law, ancestral spirits, kinship) and the blending of tasks they perform (state-bureaucratic, state law, ceremonial, dispute resolution according to custom, engagement with witchcraft) (Ray and van Nieuwaal 1996: 22). These mixtures mean that chiefs do not fit a single Weberian typology of authority – such as traditional or legal-rational – but a hybrid mixture, which at the same time underpins transformations. Broadly speaking, the concept of hybrid challenges “the belief in invariable and fixed properties which define the ‘whatness’ of a given entity” (Fuss 1991: xi), by contrast highlighting the “interweaving of elements” which create “something familiar but new” (Meredith 1998: 2).

36I suggest that the concept of hybrid authority is useful for capturing the emerging forms of chiefly practices of authority enforcement, discussed in the previous sections, but that we should also extend this concept to local state authority. The dominant literature on chieftaincy fails to address how the local state may also become shaped by interactions with chiefs and rural residents. Rather there is a tendency to view the state as a fixed entity, representing a pure domain of state legal-bureaucratic authority. As demonstrated in the previous sections chiefly practices were indeed reshaped by interactions with and regulation by the state police, but by the same token the police became localised. While the PRM attempted to regulate chiefs and, through its secondary body of law, to fix distinctions, its officers also adjusted their operations to the local context. If not directly flouting the official law, localisation of the police gave way to new routine practices that lay outside the law and which drew partly on the procedures of resolution, sanctions and local ideas about witchcraft that the police officially confined to the chiefly domain of authority. Chiefs, on the other hand, began to refer to the state law and their formalised relationship with the PRM as an effective element in flouting the law (i.e. in continuing to settle criminal cases).

37These developing action patterns point to multiple practical fusions of chiefly and state police authority enforcement. They point to the emergence of both hybrid state authority and hybrid chiefly authority. If this underpinned the inability of the local police to enforce its own secondary body of law prescribing strict distinctions, it also resulted in mutual transformations of the ways in which chiefs and the local police enforced authority. This emerged in a context of competition over areas of jurisdiction, of unclear boundaries from the outset, and because of the rural population’s particular expectations of chiefs and the state police and their preferences for particular forms of justice. The action patterns of rural residents and the perceptions informing them considerably (re)shaped the action patterns of the police and chiefs. Thus the practical fusions were at least partly influenced by each authority’s attempt to ensure popular legitimacy and demonstrate authority, even if this involved taking risks and flouting the law.

38Does the emergence of hybrid forms of local state police and chiefly authority in Dombe mean that the boundary between state and chiefs become completely erased; that chiefs and state authority becomes indistinguishable? Based on the insights from the previous sections, the answer to this question is negative. First, rural residents did not view chiefs and the state police as similar kinds of authorities. Rather their decision to take a case either to the police or the state police was based on what these could do differently in settling cases. To take a criminal case to the chiefs was seen as a means to ensure material compensation and avoid the risk of perpetrators ending up in prison. This necessarily related to chiefs’ particular way of resolving cases and dispensing justice. Conversely, taking a traditional case to the police was based on rural residents’ view of the particular power of the police in making accused persons turn up for hearings and paying compensation. This power was attached to the police’s use of notifications holding the official stamps of the state, which in themselves were understood as an order of the state, backed by the threat of sanctions. Secondly, chiefs and the state police, even as they clearly acted in a hybrid fashion, also articulated distinctions between chiefly and state police authority. In fact, it was through the very blurring of the boundaries between them that the distinctive authority of each was constituted and articulated. The state police usurped the settlement of traditional cases to demonstrate state authority by references to law and monopoly on force and by using state-bureaucratic artefacts. Chiefs drew references to state police law and formal relationship to the police as a means of making effective the enforcement of inherently localised resolutions and thereby manifesting their authority.

39The crux of the matter is that hybrid forms of authority did not erase boundary-marking, but were part of continuously renewing distinctions between chiefs and the state police. This also means that state and traditional authority should not be seen as representing historically fixed types of authority, but as forms of authority that are re-constituted in the relationship between chiefs and local state officials. Having said this, it is also important to take seriously the significance of power relations in discussions of hybrid forms of authority.

40As I have discussed elsewhere (Kyed 2007), not all chiefs were equally capable of acting in a hybrid fashion. Thus some chiefs were more successful in sustaining authority in light of state recognition. Moreover chiefs also depended on maintaining good relationships with the state if they were to retain their position as state recognised authorities or avoid punishments. This underpinned the need to strike a balance between assisting the police and demonstrating allegiance to the state, while also adjusting to the preferences of rural residents. Not all chiefs were equally capable of doing this in Dombe. The position of the local police as representatives of the central state in the rural hinterlands and as de jure imbued with the superior authority to enforce the law and use force also underpinned a clear hierarchy between chiefs and the local state police. This was most eloquently demonstrated in the punishment of chiefs, when the latter were caught challenging the police’s orders.

41The key point however is that to re-claim and maintain local state authority in Dombe it was not enough for the police to follow its official mandate. Faced with a context where chiefly jurisdictions considerably overlapped with the state police’s official mandates and where local conceptions of transgressions and justice conflicted with state law, the police drew up its own extra-legal rules to mark out the boundaries of its own jurisdictions. When these proved inadequate the police gradually adjusted its everyday policing practices to the local context in order to retain authority. More broadly these local level processes demonstrate how any simplistic understanding of the state recognition of traditional authority as either resulting in a complete state capture of chiefs or in the preservation of a pure traditional domain of authority fails to capture how local authority is constituted through ongoing negotiations and contestations partly overlapping state and chiefly jurisdictions and claims. As shown in this article such negotiations and contestations give way to mutual transformations of local state and traditional authority, not the fixing of distinct domains of authority or the preservation of such distinct domains. This becomes clear once we, as I have done in this article, focus empirically on both the everyday practices of chiefs and local state officials, and the interaction between these and subject populations.

Topo da página


Alexander, J. (1997). “The local state in post-war Mozambique: Political practices and ideas about authority”, Africa, 67(1), 1-26.

Boletim da República (2000a), de 25 de Agosto. Regulamento. Diploma Ministerial n.º 107-a/2000, Ministério de Administração Estatal.

Boletim da República (2000b). Decreto n.º 15/2000, de 20 de Junho.

Buur, L.; Kyed, H. M. (2006). “Contested sources of authority: Re-claiming state sovereignty and formalizing traditional authority in post-conflict Mozambique”, Development and Change, 37 (4), 847-869.

Buur, L.; Kyed, H. M. (2005). State recognition of traditional authority in Mozambique. The nexus of community representation and state assistance. Discussion Paper 28. Nordic Africa Institute.

Costa, A. (1999). Segregation, customary law and governance of Africans in South Africa. Ph.D Dissertation, University of Cambridge.

Dijk, R. van; Nieuwaal, E. A. B van Rouveroy van (1999). “Introduction: The domestication of chieftaincy: The imposed and the imagined”, in E. A. B. van Niuwaal and R. van Dijk, R. (eds.), African chieftaincy in a new socio-political landscape. Leiden, African Studies Centre, 1-20.

Florêncio, F. (2005). Ao encontro dos Mambos. Autoridades tradicionais vaNdau e Estado em Moçambique. Lisboa, Imprensa de Ciências Sociais da Universidade de Lisboa.

Fuss, D. (1991). Essentially speaking. Feminism, nature and difference. New York, Routledge.

Herbst, Jeffrey (2000). States and power in Africa: Comparative lessons in authority and control. Princeton, N. J., Princeton University Press.

Jordan, P. (1997). “The evolution of so-called African customary law”. Paper presented at ANC 50th National Conference, Mafeking.

Kyed, H. M. (2007). State recognition of traditional authority: Authority, citizenship and state formation in rural post-war Mozambique. Ph.D Dissertation, Roskilde University.

Kyed, H. M.; Buur, L. (2007). “Introduction: Traditional authority and democratization in Africa”, in L. Buur and H. M. Kyed (eds.), A new dawn for traditional authorities. State recognition and democratisation in Sub-Saharan Africa. New York, Palgrave, 1-30.

Mamdani, M. (1996). Citizen and subject. Contemporary Africa and the legacy of late colonialism. Princeton, Princeton University Press.

Meredith, P. (1998). “Hybridity in the third space: Rethinking bi-cultural politics in Aotearoa/New Zealand”. Paper presented to Development Conference, Massey University, 7-9 July 1998.

Notícias (1995). Daily newspaper, 17/07. “Atitude irreversível dos régulos deriva do tribalismo e posse de armas”.

Ntsebeza, L. (1999). Land tenure reform, traditional authorities and rural local government in post-apartheid South Africa: Case studies from the Eastern Cape. Cape Town, School of Government, University of Western Cape, Programme for Land and Agrarian Studies.

Quinlan, T. (1996). “The state and national identity in Lesotho”, Journal of Legal Pluralism and Unofficial Law, 37/38, 377-406.

Ray, D. I.; Nieuwaal, E. A. B van Rouveroy van (1996). “Introduction. The new relevance of traditional authorities in Africa”, Journal of Legal Pluralism and Unofficial Law, 37/38, 1-38.

Savana (1995). Weekly newspaper, 28/07. “Os Régulos: Estão de volta ao poder”.

Serra, C. (1997). Novos combates pela mentalidade sociológica. Maputo, Universidade Eduardo Mondlane.

Sklar, R. L. (1999). “The significance of mixed government in Southern African Studies: A preliminary assessment”, in J. Hyslop (ed.), African democracy in the era of globalization. Witwatersrand, Witwatersrand University Press, 115-21.

Trindade, J. C.; Santos, B. de Sousa (2003). Conflito e transformação social. Uma paisagem das justiças em Moçambique. Porto, Reino e Neves, Santa Maria da Feira.

Trotha, T. von (1996). “From administrative to civil chieftaincy. Some problems and prospects of African chieftaincy”, Journal of Legal Pluralism and Unofficial Law, 37/38, 79-102.

West, H. G.; Kloeck-Jenson, S. (1999). “Betwixt and between: ‘Traditional authority’ and democratic decentralization in post-war Mozambique”, African Affairs, 98, 455-84.

Topo da página


1 On reasons why it took almost ten years from the first policy studies of traditional authority to the passing of legislation see Kyed (2007).

2 The Decree includes three categories of community authorities that can be legally recognised by the state on the basis of local community legitimization: traditional chiefs, former FRELIMO secretaries of suburban-quarters or villages, and other leaders legitimized as such by the respective local communities (Boletim da República, 2000b: art. 1). In this chapter I focus exclusively on the category of traditional chiefs. On secretaries see Kyed (2007).

3 On the process of legitimisation and recognition in Dombe see Kyed (2007) and Buur and Kyed (2006).

4 The chiefs’ courts in Dombe as elsewhere in Manica province, consisted of the chief and his/her council of elders (Madodas) who heard and deliberated hearings, a secretary who wrote notifications, and a group of young police assistants (ma-auxiliares) whose job is to notify contenders in a case and fetch or arrest persons who are unwilling to turn up at the court (Kyed 2007: 234).

5 This perception of misconduct or transgressions is intimately linked to the worldview of the Chi-Ndau people of Dombe in which visible and invisible dimensions of social life are seen as interlinked (see also Florêncio, 2005).

Topo da página

Para citar este artigo

Referência do documento impresso

Helene Maria Kyed, « Mutual Transformations of State and Traditional Authority. The renewed role of chiefs in policing and justice enforcement in Mozambique », Cadernos de Estudos Africanos, 16/17 | 2009, 179-201.

Referência eletrónica

Helene Maria Kyed, « Mutual Transformations of State and Traditional Authority. The renewed role of chiefs in policing and justice enforcement in Mozambique », Cadernos de Estudos Africanos [Online], 16/17 | 2009, posto online no dia 22 Julho 2012, consultado o 23 Março 2017. URL : ; DOI : 10.4000/cea.192

Topo da página


Helene Maria Kyed

Danish Institute for International Studies

Topo da página

Direitos de autor

Licença Creative Commons
O trabalho Cadernos de Estudos Africanos está licenciado com uma Licença Creative Commons - Atribuição-NãoComercial-CompartilhaIgual 4.0 Internacional.

Topo da página