The actual security environment in Mozambique is influenced by the developments starting as far back as the time of the liberation struggle. These bear upon many of the factors shaping this environment, as well as perceptions of and responses to it. A United Nations Development Programme National Human Development Report¹ argues that the security in the country has been characterised by three successive changes that span almost the last three decades: from colonial rule to independence by way of an armed struggle; from centrally planed to a market economy through an International Monetary Fund and World Bank structural adjustment programme; and from war and a single-party state to peace and a multiparty political dispensation.

It is therefore important to look briefly at the way in which these historical developments have shaped the actual security problems as well as its political perception and how security priorities are determined. Irrespective of what the security concerns of the people might have been, security policy priorities have indeed been mediated by political perceptions. As a result, while crime clearly increased since the beginning of the 1990s, the real reforms and reorganisation of security institutions towards crime prevention and combating did not start before 1997.

In order to illustrate these points, this article will first describe the actual security concerns in the country during the 1990s. It attempts to identify the security threats and their causal factors. Secondly, it looks at the political perceptions shaping the security policies throughout the period under consideration by seeking to understand how policy priorities have been accorded to the security environment they are meant to serve. Thirdly, noting the relationship between policies and security environment, the reforms needed by security institutions are considered. It is noted that, despite legal advances and the general political discourse on reforms, the structure and operational culture have remained largely untouched, except that the military have been downplayed and the police have been militarised. The article further discusses the slow start of reforms, trying to reveal the factors which determined the delay, as well as those currently driving them. Finally, a brief conclusion indicates that the major policy challenge is that of closing down the gap between policies and security concerns, where police reform is clearly an imperative.

The end of the 1980s was characterised by unprecedented political, social and economic changes. One of the main changes was the fact that ‘politics through war’ was gradually being pushed to the background. It seemed that once this process was completed, security would be guaranteed. However, war legacies, combined with poverty, environmental degradation and widespread epidemics were to shape the forthcoming security concerns, but not in any better way.

Indeed, by the late 1980s, the war had disrupted fifty per cent of the countryside, nearly one million people had been killed and 4,5 million became refugees and/or displaced.² The public security system deteriorated. The state and society remained apart, creating conditions for lawlessness. Urban areas, which had remained fairly safe during the war, witnessed a crime wave that the police was unable to curb.

The most important factors that contributed to the increased crime rate were:

The end of the armed conflict left large numbers of redundant weapons in the hands of demobilised soldiers and civilians, as well as in the bush. These weapons (previously used for political and ideological reasons) became highly mobile and ignited criminal violence in the country. The causes of the failure of the UN Operation in Mozambique (ONUMOZ) to undertake a comprehensive disarmament process have been extensively reported.³ As Latham commented in 1995, "... there is no longer an ideological war being fought on Mozambican soil, but citizens complain that, if the violence has abated, it has far from disappeared ... No property is safe and horror stories of ambushes, shooting and stabbing abound."⁴



Returning refugees and displaced people, including thousands of migrant workers who had lost their jobs in East and West Germany and thousands of decommissioned soldiers, joined the unemployed army. Given the weakness of the country’s economy, the social and economic reintegration of these groups of people became a nightmare. A Refugee Studies Programme pilot study found no evidence linking former soldiers with armed crime. It recognised, however, that the lack of formal employment coupled with the inability of the small-scale agriculture to guarantee the subsistence of rural families, forced former combatants to look for alternative income earning prospects in the informal commerce of urban centres.⁵ Here, their vulnerability to crime of all kinds was real, not only because the products on offer in the informal economy were stolen, but also because the temptation to use anything at their disposal for survival, including weapons, was very great.⁶ Besides, the simple fact that they joined the unemployed army in the urban areas was potentially a destabilising factor in itself. These redundant soldiers, as well as unsettled refugees, displaced people and the unemployed youth had no other identity but that of vulnerability. In addition to that, the economic liberalisation process had increased consumer goods in the market, while the public’s purchasing power was decreasing.



Increasingly weakened state institutions were unable to maintain law and order, in general, and the police, in particular, were unable to provide effective answers to increasing crime figures.⁷ Along with this inability to combat crime, the police force was generally seen as highly corrupt, and citizens sometimes took the law into their own hands. The newly formed army was characterised as desperate and undisciplined. With impunity, crime thus increasingly became an "occupational hazard" as Vines ironically puts it.⁸

The conditions were in place for the period of the cease-fire to be characterised by increasing crime and violence. Not the least, landmines were in place to kill innocent civilians in areas where armed gunmen were not interested to operate.⁹

Statistical data collection in Mozambican state institutions is at the best disorganised, mainly because information systems are very poor. In addition, popular trust in the police is very weak. Many people prefer not to report crime, because it does not make sense to waste time when it is well-known that nothing will happen to the criminals. Even with regard to some of the reported cases, the information on cases may ‘disappear’ before the Department of Statistics takes note.

Despite qualifications that the table above may be a gross under-estimation of what the crime incidence might be, the figures in it show an increase. The alarm was rung in 1993 when, between May and September, the number of reported crimes in Mozambique included 167 homicides, 726 armed robberies and hundreds of cases of physical assault, rape, and other violent crimes.¹⁰ Since then, the trend seems to have remained consistent. Police statistics show an increase of 22 per cent between 1994 and 1996.¹¹ Another important trend worth mentioning is that, since about 1993, property crime has consistently accounted for more than fifty per cent of all reported crimes. This suggests that crime has become a way of life for many people.

Conversely, state institutions have shown a consistent inability to combat crime. Between 1990 and 1998 the police have only managed to submit an average of 37,2 per cent of all reported crimes in the country to the courts (see Table 1). In turn, the attorney-general reported that, in 1997, 58 per cent of the 13 993 criminal cases brought to court were tried.¹² In other words, less than half of all reported crimes reach the courts of which about one half are tried, meaning that about one quarter of all reported crimes are brought to book.

This state of affairs is a direct reflection of a weak judicial system. The weakness of the criminal system is also illustrated by the fact that the correctional end of the spectrum, the prisons are over-crowded and many of the inmates have not been tried. For instance, reports indicated that, in 1995, over 110 prisoners in Xai-xai, Gaza province, rioted in protest against the long delays in cases being brought to court. Of the 164 prisoners in jail then, only 25 were serving sentences. In reaction to these reports, the Justice minister admitted that, of the 2 572 prison population in 1995, more than a third (1 451) were waiting trial.¹³ The factors behind this include:

a lack of de facto independence of the judiciary (see below);


a lack of capable district and provisional judges;


lack of lawyers in general;


a severe lack of funding; and


widespread corruption.

Although the fact that these developments took place during the transitional period should be comforting, what is disturbing, is the generalised impunity in the country. Above all, the risk that this impunity will entrench a culture of lawlessness in the country, where law breaking becomes the rule, and law abiding is discouraged, remained real.

At independence, power was transferred to the Frente de Libertação de Moçambique (FRELIMO), known as the legitimate representative of the people of Mozambique. The struggle was transformed into a social revolution which could only be challenged by external enemies, albeit with the use of internal ‘traitors’. This, coupled to the regional context, was at the basis of Machel’s security concerns. Until 1980, it was believed that the most serious security threats emanated from South Africa and Rhodesia. Therefore, the emphasis in security-related efforts was placed upon the modernisation and professionalisation of the army in anticipation of a conventional attack from the two neighbouring giants (South Africa and Rhodesia).

However, from 1980 onwards, increasing military pressure from the Resistencia Nacional de Moçambique (RENAMO) throughout the country led the government to undertake a fundamental reorganisation of the army and a massive recruitment drive. Defence expenditure was reported to be above forty per cent of the government budget.¹⁴ In 1982, when about 8 000 RENAMO guerrillas were scattered all over the country, the government created ten provincial commands which were assigned to recruit soldiers to serve at district and local levels, as necessary. It seems that the number of people employed by the military (Forças Armadas de Moçambique — FAM) was not known since then, although rough estimates reported that, by 1990, Mozambique had around 145 000 people on its payroll. Although these estimates may be exaggerated, it is an incontrovertible fact that the government had, on the one hand, a large, uncontrollable, disloyal army. Unable to provide people with security, on the other hand, it had distributed an unknown number of weapons to militia groups and civilians alike, for the protection of towns, suburbs, villages, districts, state-owned companies.¹⁵ The idea in FRELIMO’s mind was that of reorganising the armed forces to counter an external aggression operating as an internal guerrilla movement. Analysts have argued that FRELIMO has consistently denied any political value to RENAMO and instead tried to assimilate it.¹⁶

As a result, internal and external security functions were not discernible. The police and the intelligence agency were literally subordinated to the military to the extent that one of the requirements for recruitment to the police force was to serve in the army¹⁷ and intelligence officers carried a military identity document.

Any unlawful behaviour was interpreted as part of the war. The death penalty,¹⁸ initially introduced for high treason, was soon expanded to include economic crimes such as black marketeering and armed robbery. When the Faculty of Law at Eduardo Mondlane University was closed in 1982, the argument was that lawyers were defending colonial and capitalist ideas.¹⁹ In other words, crime disorder and political opposition were seen as functions of the war and thus deserving the same response. The concepts of defence and security seemed to find their explanation in the conception of security that sought to integrate the regime’s response to external aggression, manifested as crime, political opposition, or the like. Given the confusion between political and criminal security factors, internal safety and security functions were militarised.

These facts have shaped both the government and RENAMO’s behaviour throughout the peace process. Indeed, as Mazula²⁰ has noted, distrust between the parties was a permanent constraint that shaped the entire process from the negotiation phase through to implementation.

RENAMO was suspicious of FRELIMO’s intentions and wanted to avoid post-settlement marginalisation. Thus, while RENAMO tried at the negotiating table to get as many FRELIMO concessions as possible, it did not lose sight of the fact that these could be withdrawn as soon as it had been militarily minimised. It was clear therefore that, until RENAMO could guarantee that it had achieved enough political space to ensure its own survival within a new political dispensation, it had to maintain what Cock has termed the "material base for political bargaining."²¹

Similarly, FRELIMO had to make numerous concessions as the political price for achieving the cease-fire. Unaccustomed to competitive politics, it knew that it had made mistakes in governing, which the constituency would scrutinise for the first time. Moreover, the whole process meant the reallocation of the power it had monopolised. The question was how to guarantee the surrender of specific power, while preserving fundamental political control. To prevent the latter from being jeopardised, every step was taken, including resorting to armed means. All these concerns led to the equivalent of an arms insurance policy.

One of the most important moves in this direction was the rearrangement of the defence and security forces, for which no data is publicly available. What is known, however, is that from about 1990 onwards, there were a large number of military personnel being transferred to the national police. The government was sure that, in the event of disarmament, it would be possible to protect the police, based on the need to maintain law and order. An arsenal was allegedly found in the Ministry of Interior in September 1994. Other reports indicated that:

"Over time, it became clear to UN Civil Police (CIVPOL) officers that government military troops and equipment were being transferred to the police, especially to the Presidential guard."²²

As a matter of fact, police issues were non-negotiable during the negotiation and implementation phases, as the government alleged that some kind of sovereignty should be maintained. There is little doubt that the FRELIMO government were preparing a force like the Ninjas in Angola. Apart from the fact that FRELIMO was not quite sure about RENAMO’s intentions, the idea that the war had always been one of external destabilisation seemed to prevail among FRELIMO’s leadership. Although it is partly a political game, President Chissano’s insinuation that RENAMO was a creation of Portuguese, Rhodesian and South African fascist regimes from which it continues to receive guidance,²³ supports this argument to this day. The point is that the reliance on military means led both sides to keep a certain level of military muscle, not as a response to the people’s security needs, but simply for their own political survival.

Between 1990 and 1994, the emphasis of the political debate on security was on the need to guarantee the respect for political rights of the emerging political parties, in particular RENAMO. The fact that half of the members of the National Commission for Police Affairs (COMPOL) — formed as part of the implementation of the peace process — sat at the political rights subcommission illustrates this point. What is witnessed, in other words, is the gap between the country’s security concerns and its security policies. Security policies were very much a reflection of the narrow political interests of the former warring parties. Efforts to curb crime were postponed or crime was used as the reason for militarising the police.²⁴

Thus, looking at the actual security challenges of the early 1990s and the way in which political concerns shaped the policy responses, a simple conclusion emerges: despite the internal security environment, characterised by criminal violence, the need for political survival kept security policies apart from security needs. This fact has shaped the reform of security institutions as the discussion below attempts to show.

If ‘the police’ is replaced by ‘security institutions’ in general, Shearing’s statement that, in democratising countries, "... the police ... are to be transformed from authoritarian, partisan instruments of government into modern police institutions that are non-partisan and democratically accountable,"²⁵ will remain valid and applicable to Mozambique. Generally, security sector reforms are meant to change the institutional structures within which the security forces in non-democratic states operate, in particular "... their military organisation, and the occupational culture of the police that guides and shapes police practice."²⁶ This involves a clear legal and operational definition of roles, command and control, and the mission of each security branch. Of course, it is assumed that, in a non-democratic country, for the virtue of primarily defending the regime’s interests as opposed to state interests, the mission, institutional structure, and the command and operational culture of the different security branches are hardly differentiated.

As far as Mozambique is concerned, the assumption above is very true. Indeed, as has been pointed out above, internal and external security factors were seen as the same thing. Crime, violence, public order, the civil war, and more were all functions of the war. Thus, internal and external functions of security could not be differentiated. Institutionally, the reflection was a unified, politicised and militarised concept of defence and security.

Security sector reforms in Mozambique therefore mean, among others, separating the internal and external components of security, changing the operational culture, including command and control structures, redefining the roles of the different security agencies, and involving civil society in the security debate. It also implies a mechanism to conduct a formal and realistic security needs assessment, where the size and structure of the security forces are determined by the country’s security environment, the available resources and the possibility of using non-military means to ensure external security through regional security agreements.

Attempts to reform security agencies started with the 1990 constitution. While this constitution has laid the legal foundation for the political transition, it largely failed to provide a comprehensive basis for security sector reforms.

From a legal point of view, it considered command and control, and accountability issues by providing that the national defence and security forces should be subordinated to a national defence and security policy and to the constitution. The president was to remain the commander-in-chief of the defence and security forces, but was now to be advised by a national defence and security council whose composition was not made clear. Parliament had to ratify the president’s authority to declare war, a state of emergency or siege.²⁷ The need for both an apolitical stance by the security forces, and the separation between the internal and external components of the security function remained largely unclear.

Following the adoption of the democratic constitution, specific acts pertaining to security matters were also passed to comply with the new constitutional provisions. In early July 1991, the newly renamed parliament, the Assembly of the Republic, passed a law that abolished the state secret police, the People’s National Security Service (SNASP), replacing it with State Information and Security Services (SISE). The new SISE was meant to remain solely an intelligence-gathering agency with no police powers, as was the case in the past. Law 19/92 of 31 December 1992 created the Police of the Republic of Mozambique (PRM) to replace the existing Mozambique People’s Police (PPM). According to this law, the police service would have a general command that is equivalent to a national police commissioner. In 1993, the police statutes were approved by cabinet decree 22/93, while national defence force reforms were left to be part of the peace agreement.

If these moves were to signal political will and commitment to peace and the democratic transition, the actual implementation has lagged far behind. In other words, while one would tend to believe that the government felt the responsibility to maintain law and order and to establish an environment conducive to political pluralism, the gap between the legal provisions and the general political discourse, on the one hand and operational practice, on the other, seemed to indicate a different direction. The PRM and SISE remained very much the same as the PPM and SNASP throughout the peace process and beyond.

In fact, defence and security budgets remained a covert matter, making it difficult for civil society to understand. Differentiation between the defence, police and intelligence budgets started to be made public only in the 1999 government budget.

General human rights violations of detainees in police custody have been reported. The 1998 Human Rights Watch report points out that, although human rights records are improving, police behaviour and prison conditions remained a serious concern. RENAMO alleged that its officials have been harassed by local government officials and FRELIMO members in some areas, in contradiction to the need for political impartiality. What is more worrying, the report adds, is that no serious prosecution of the police involved has taken place to date. Police brutality more commonly went unpunished, and the police seemed to be increasingly militarised.

While claims that a lack of resources in the government, in general, and the police, in particular, has led to poor police force performance,²⁸ the lack of political willingness cannot be dismissed. In fact, following both the 1990 constitution and the 1994 general elections, it was expected that a defence and security white paper would be formulated, setting the parameters of policy in the "...areas of civil-military or civilian-police relations, the roles and functions of the relevant security forces ..."²⁹ This did not happen until October 1997. What was evident, was a process of changing names and legal provisions with little if any operational changes taking place.

Another important step in the reform of the security agencies in the country was the General Peace Agreement (GPA). Protocol IV set objectives for the ‘depoliticisation and restructuring of the security forces’.

The GPA provided that a new Mozambican Armed Defence Forces (FADM) should be formed to "... defend and safeguard the Sovereignty, independence and territorial integrity of the country in strict respect of professionalism, non-partisan and the rule of law."³⁰

The formation of the FADM was meant to start after the cease-fire and the Control and Supervision Commission (CSC) were in place. It would take place simultaneously with the warring forces being dismantled. After disagreements between the parties, the strength of the FADM was put at 30 000 (24 000 army, 4 000 air force and 2 000 navy) drawn equally from the warring forces. As Young argued, the formation of the FADM "... was not as much a practice of integration, but one of disintegration."³¹ A transitional structure, the High Command, was jointly led by two generals, one from each party. In reality, this dual structure trickled down to all levels of the forces and laid the foundation for the post-election force structure.

Again, notwithstanding the emphasis on forging democratic civil-military relations, both the structure, as well as the strength of the new force had little to do with the prevailing security needs of the country. It was, after all, a result of distrust between the former warring parties, their internal dynamics and external pressures.

The government, with a far larger force, predicted few employment opportunities in the post-election period and wanted to retain the soldiers in a bigger army. RENAMO, on the other hand, with fewer forces, was concerned with the equilibrium within the new armed forces and therefore found the number unacceptable. To the advantage of RENAMO, which wanted a smaller defence force, the international community insisted that the regional and internal strategic environment was conducive for a general downsizing. On the other hand, the numbers were simply incorrect. "Those who negotiated force levels for the FADM proved to be wildly optimistic about its anticipated popularity. By February 1995, the FADM had only 12 195 members (8 533 from FAM and 3 662 from RENAMO) and this number was decreasing."³² Both government and RENAMO troops were not interested in joining the new armed forces.³³

It seems that the government’s defence budget cut of 36,7 per cent in 1995, was more in response to increasing external pressure, rather than its own security needs calculations.³⁴ This explains the contradictory remarks made by a senior officer shortly after the announcement of the budget cut that, if Mozambique was to remain as a nation, it would need to sacrifice its scarce resources to consolidate the armed forces.³⁵

Although the GPA emphasised the need for a smaller and democratic defence force, the format of the emerging defence force in practice was therefore meant to accommodate the political security interests of the former warring parties. When a special police unit was deployed along the highway between Maputo and Ressano Garcia and national Highway number 1, Afonso Dlakhama did not question the need and appropriateness of the deployment of a well-equipped and trained force. However, he did question the fact that the special police unit had military characteristics. To use his own words, "... what is beyond my understanding is ... why FRELIMO is using its former forces to build a parallel army ... if there is a need to create special police force, let us then do so openly."³⁶

As far as other branches of the security establishment are concerned, the GPA provided that the SISE’s activities should be restricted to information-gathering and analysis as required by the president and within the limits allowed by law. Under no circumstances would the SISE exercise policing activities. To ensure that these provisions were complied with, a national commission for information was envisaged.

The National Commission for Police Affairs (COMPOL) was also established and with 21 members, had the mandate to oversee PRM activities in order to guarantee respect for the rule of law, justice and human rights by the police during the peace process, laying the basis for a post-conflict policing culture.³⁷ Despite the limited resources and authority, the commission has been able to identify some of the most challenging problems facing the police force, such as low standards of education and training, a lack of objective criteria in the selection of police candidates, a predisposition among police officers to take bribes, and a lack of resources. It regretted the fact that, of the 51 cases it investigated and reported to the relevant authorities, no corrective measures were taken. The COMPOL³⁸ final report recommended the establishment of an institution which could take over its watchdog activities in the post-election period. But the government turned its recommendations down.

In the post-election period, however, legal definitions of defence and policing tasks and operational structure reforms were deliberately delayed. This suggests that the government deliberately maintained an ambiguous legal environment to allow it to keep a highly militarised police force to fill the perceived security gap resulting from a weak defence force with dubious loyalty. To be sure, although the ruling party had the power to approve the white paper on defence and security drafted in 1995, the situation of policy uncertainty was much more desirable than policy clarity. Thus, it allowed the delay to continue until late 1997.

This reasoning is supported by the fact that the apparent confusion regarding the responsibility for border protection seems to indicate a de facto deliberate, benign neglect by the FADM. Indeed, the notion of territorial integrity assigned to the defence force has been used to deploy the defence force in some instances, and not in other similar cases. FADM units were deployed in Maravia District in January 1997, allegedly to ensure the inviolability of the borders.³⁹ The reason for the deployment was an unconfirmed report that foreigners were moving in the area — Zambians who often crossed the borders on illegal business. In contrast, there were instances when the armed forces were precluded and the police deployed, such as the cracking down on the so-called chimwenjes⁴⁰ known to be armed, alleging that the issue were of a public order nature and that border protection is the police force’s task. This example suggests that militarised special police units would be deployed in any serious trouble spot regardless whether it falls within their legal mandate, as long as the need was felt to keep RENAMO away.

For the government, the formula of a joint defence force has been just a price to be paid for peace — on the one hand, to ease RENAMO’s fears and, on the other, to please the international donor community tied to its ‘butter or guns’ dilemma. As a result, the perceived weakness of the new army was matched by moving core army units to the police force. Imparcial has noted that, in 1991, military border protection and other special units were transferred to the police, because the government was suspicious of the commitment of RENAMO.⁴¹

This claim seems to explain the fact that, although there has not been any public recruitment since 1994 — on the contrary, there were expulsions for misconduct⁴² — the police have grown from an estimated 18 000 in 1994 to approximately 21 666 in 1998. And the police budget has been consistently above that of the defence force since then. (As a result of demobilisation, defence and security expenditure has decreased in real terms from about twenty per cent of the government budget in 1990, to nine per cent in 1998 — see Table 2.) However, the police’s share has increased by 47,4 per cent in three years (1996-1998).

The contention therefore is that, despite the legal instruments — the 1990 constitution and the GPA — it was apparent by early 1997, that beyond the name change, little had been done to change internal controls or the orientation and ethos of the police. This is precisely because, as Bayley argues, policing cannot be separated from politics and reform can only go as far as the local political situation will allow. The political environment remained characterised by high levels of distrust among the major political players. Reform of the police or other governmental institutions could not realistically take place, as "... unstable governments rarely create democratic police forces."⁴³

As confidence in the peace process grew and crime established itself as the most challenging security threat, a slow process of reform started.

The defence and security white paper was passed in 1997. It was an effort to adapt the defence and security agencies to the new political dispensation. The white paper calls for citizens to accept responsibility for the security of the state and the public order as the governing principle of defence and security. It provides that the defence and security institutions shall abide exclusively by the constitution and obey the president of the Republic as its commander-in-chief. The overall political control lies with parliament.

The goals set for the defence and security agencies are as follows:

to guarantee the independence, sovereignty and territorial integrity of the country;

to defend and consolidate national unity,

to defend and ensure the normal functioning of democratic institutions;

to safeguard the internal and external security of the state;

to ensure the fundamental rights and liberties of citizens; and

to maintain law and public order.

The white paper distinguishes three main components of security: external, internal and state security (intelligence). External security is explicitly defined as armed aggression exclusively assigned to the army. Article 10 of the white paper defines the mission of the national armed defence forces as follows:

to defend the vital interests of the country against all forms of aggression, including terrorism;



to ensure the territorial integrity, sovereignty and freedom of citizens;



to ensure the normal functioning of state institutions;



to participate in protecting agencies and property, as well as to adopt preventive and rescue measures required in certain circumstances as decided by the competent authorities; and



to participate in peacekeeping missions.

Simultaneously, the National Defence and Armed Forces Act was promulgated. While incorporating most of the provisions of the defence and security white paper, the Act seeks to regulate the tasks and conditions under which the armed forces are deployed. It provides that the mission of the defence force consists of military defence against all forms of aggression. Any extraordinary employment of the defence forces is defined by the president and ratified by parliament. This is meant to regulate the use of the defence force in internal tasks, such as rescue missions in emergency circumstances. Given the poor conditions in the armed forces, the Act also reintroduces compulsory military service as a result of fears that a voluntary system would not attract enough personnel to staff the army. On the basis of these legal instruments, the minister of Defence has announced that the recruitment of new personnel into the defence force will correct the current inverted pyramid where there are more officers than soldiers.

Internal security (Article 11), on the other hand, is defined as the activities undertaken by the state in order to ensure law and order, public safety, security and tranquillity, to protect the people and their property and to prevent crime, as well as to contribute in enhancing the normal functioning of state institutions. These tasks are assigned to the PRM.

Finally, state security is an activity undertaken by the state to ensure respect for the constitution and the gathering and production of the necessary information to ensure national independence. For that purpose, the responsibility of gathering, analysing and assessing the information required to ensure such security, is assigned to the SISE.

If things have gone this far at the legal end of the spectrum, much remains to be done operationally. Nevertheless, given the crime wave and the powerful pressure it has ignited since 1996, the police have been in the spotlight. Thus, of the three security agencies, it is in the police service where reforms have started.

Indeed, it was the government that had resisted any involvement of external actors, including the United Nations in police affairs and largely jeopardised the ability of the civilian police (CIVPOL)⁴⁴ to contribute to police reforms, which started looking for external partners to assist with such reforms.⁴⁵

Negotiations that had started in 1996 had to be sped up. An agreement for assistance with police reforms was signed with the UNDP in 1997 and became operational early in 1998. With technical assistance from the Spanish Guardia Civil, the project was focused on three main components: training, reorganisation and equipment.

In order to enhance the PRM structures to enable it to maintain law, order and peace within the current political framework, the internal structure of the PRM general command is being analysed. Under this project, whenever partial reforms of the current legal framework are required, they will be proposed. As police officers are retrained, or new recruits trained, and current structures and operational procedures reorganised, appropriate equipment will be provided, including substantial salary increases, so that the territorial infrastructure of the new police force is accordingly adjusted.

By the end of 1998, 580 police officers were retrained by the UNDP-led project. A project team report has raised concerns, however, that there is a need to retrain all police officers to counter the risk of having two policing standards.⁴⁶ The problem lies in the fact that many of the existing police officers will not meet the conditions required for the retraining courses. According to the same report, this is creating uncertainty among police officers about their possible future in the police force. The question remains whether the government is prepared to retrench those who are found unfit for the police standards needed.

However, it is too early to evaluate the impact of both the legal provisions and the police reform project. For the latter, it suffices to point out at this stage that, if all the objectives stated in the project document are met, then a new police force will have been created. The challenge is that a thorough implementation of the project has a political, as well as a financial price to be paid. While the International community seems to be willing to provide core funding,⁴⁷ it remains to be seen whether the government will have the necessary political courage to sacrifice some of its political clients in the name of impartiality, professionalism and performance in case they do not meet the academic requirements for retraining. For the former, the most optimistic attitude is to wait and see, as it is commonplace that sound legal provisions have never solved problems until a determined finger is lifted to make them work.

Domestically, the threat to the stability of Mozambique is increasing crime, made more dangerous by the widespread availability of weapons left from the war and the impunity with which law enforcement agencies are watching. Socio-economic conditions have made returning refugees, displaced persons and demobilised soldiers a ready reserve of criminals if there is no other alternative source of livelihood.

On top of this social foundation for instability, is the fact that security policies throughout the 1990s have failed to respond to the actual security needs of the people. Regime security needs or the former warring parties’ political security concerns have shaped the security policies at the expense of the security of the people.

With all these political and socio-economic complexities, the policy challenge is clearly to close the gap between policy and the actual security needs of the people. This will require a major reform of the agency assigned to combat crime and maintain law and order — the police. There is a need to restore the trust and morale of the FADM to take up some of the militarised functions currently performed by the police. This will release police efforts that can consequently be used exclusively for crime combating. The political price, however, challenges the government’s readiness and courage to sacrifice political clients within the police force for a democratic and effective police force to emerge.

This is a version of an earlier paper prepared as part of the Demilitarisation and Peacebuilding in Southern Africa Project, jointly co-ordinated by the Centre for Conflict Resolution (Cape Town) and Bonn International Center for Conversion (BICC), and kindly funded by the International Development and Research Council (Canada).

UNDP, Mozambique, peace and economic growth: Opportunity for human development, UNDP National Human Development Report 1998, Maputo, pp 30-46.



I Msabah, Negotiating an end to Mozambique’s murderous rebellion, I W Zartman (ed), Elusive peace: Negotiating an end to civil wars, Brookings Institution, Washington DC, pp 204-230.



For a discussion on the ONUMOZ failure to disarm Mozambique, see, among others, E Berman, Managing arms in peace processes — Mozambique, UNIDIR, Geneva, 1996; A Vines, The struggle continues: Light weapons destruction in Mozambique, Occasional Papers on International Security Issues, 25, 1998.



B Latham, Mozambique: Illegal weapons trade threatens security, 6 March 1995, <www.AIAZIM@mango.zw>.



Study conducted by the Refugee Study Programme, cited in Vines, op cit, pp 5-6.



Ibid, p 5.



Latham, op cit.



Vines, op cit, p 7.



The National Commission for Demining estimates that, until 1998, about 20 000 people have been victims of landmines. See UNDP, op cit, p 34.



J L Woods, Mozambique: The CIVPOL operation, <www.ndu.edu/inss/books/policing/chapter5.html>, 10 February 1999.



M Chachiua, The status of arms flows in Mozambique, in T Nkiwane, M Chachiua & S Meek, Weapons flows in Zimbabwe, Mozambique and Swaziland, ISS Monograph, 34, Institute for Security Studies, Halfway House, 1999, p 34.



Procuradoria Geral da República, Informação Annual à Assembleia da República, Maputo, 20 Abril 1998, p 3.



Reuters News Service quoting Radio Mozambique, 11 July 1995.



Although official figures indicated that Mozambican defence expenditure represented 25 per cent of the government budget, analysts believed that it was between forty and fifty per cent of the budget during the 1980s. See Berman, op cit, p 43.



This is based on earlier research.



B Mazula (ed), Moçambique: Eleiçoes, democracia e desenvolvimento, InterAfrica Group, Maputo, 1995, pp 25-30.



Interview with a senior army officer, Maputo, January 1999.



Lei 2/79 de 1 de Março de 1979.



Human Rights Watch, Conspicuous destruction: War, famine and the reform process in Mozambique, Human Rights Watch, New York,1992 pp 139-154.



Mazula, op cit.



J Cock, A sociological account of light weapons proliferation in Southern Africa, in J Singh (ed), Light weapons and international security, Pugwash Conference on Science and World Affairs, British American Security Information Council, Delhi, 1995, p 101.



Woods, op cit.



A Saúde, Homenagem de Eduardo Mondlane nada tem que ver com Eleiçoes, Domingo, 21 Fevereiro 1999, p 5.



Highly armed police special units were deployed to curb armed attacks along the country’s main roads. Conventionally, it would be expected that the army will disband a group of Zimbabwean dissidents (chimwenje), allegedly armed and operating within Mozambican territory, but a special police unit was deployed instead.



C Shearing, Toward democratic policing: Rethinking strategies of transformation, 1995, <www.ncjrs.org/txtfiles/167024. txt>, 27 March 1999.



Ibid.



1990 Constitution, Articles 59-61, 106, 122a, and 135.



Just as an example, the PRM had only 144 vehicles, 114 motorbikes and 66 bicycles, fulfilling in just more than forty per cent of its requirements in 1998. Comando Geral da Policia da República, Balanço de Trabalho do Comando -Geral da Policia 1 — Semestre 1998, PRM, Maputo, Setembro 1998, p 28.



N Ball, Spreading good practices in security sector reform: Policy options for the British government, Saferworld, London, December 1998, p 24.



General Peace Agreement, Protocol IV I.i.2.a) and b), Rome, 1992.



E Young, The development of the FADM in Mozambique: Internal and external dynamics, African Security Review, 5(1), 1996, p 19.



Ibid.



For a further discussion of the unwillingness to join the FADM, see ibid.



Ibid, p 22.



Ibid.



Reuters News Service quoting BBC, 17 August 1995.



Comissao Nacional dos Assuntos Policiais (COMPOL), Relatorio Final de Actividades, COMPOL, Maputo, Dezembro 1994, p 3.



Ibid, p 6.



Reuters News Service quoting BBC, 7 March 1997.



Chimwenjes are a group of armed dissidents allegedly associated with opposition parties in Zimbabwe.



It is increasingly clear that the police in Mozambique accommodate military units as a strategy to avoid a perceived disloyal army. See Imparcial, 2 March 1999.



Several accounts have reported that the PRM is expelling many of its members as a result of misconduct. See, for example, 322 policemen expelled for misconduct, PANA, 6 November 1998.



D H Bayley, The contemporary practices of policing: A comparative view, in Civilian police and multinational peacekeeping: A role for democratic policing, Department of Justice, Washington DC, 1997, p 6.



What is regrettable, is the fact that CIVIPOL — with 1 144 members — spend about US $30 199 824 just to observe. This is about the same amount that has been budgeted for police reforms, but less than US $12 million has been guaranteed at this stage. If reforms had been started early enough and CIVPOL had been used for that propose, the process could be well ahead now. See Woods, op cit.



CIVPOL officers were denied access to sites where there was suspicion that military personnel and equipment were being converted for police use. In one case, an undeclared police training camp, where police recruits were being trained in machine-gun use, was inadvertently shown to a visiting Security Council delegation.



Projecto Moz/95/015 Document, UNDP, Maputo, 1998.



Initially, only Spain and the Netherlands funded the project, but Italy, German and Switzerland have been attracted to the idea by late 1998, as well as some regional partners who also seem interested in supporting the process of police transformation.