Introduction

Developments towards the end of the 1980s were indicative of the unprecedented political changes ahead for both South Africa and Mozambique. The main content of these developments was the fact that ‘politics through war’ was gradually being pushed aside. It seemed that, once this process was completed, security would be guaranteed. However, war legacies, a surfeit of weapons, redundant soldiers and social dislocation — combined with poverty, environmental degradation and widespread epidemics — did not improve the security of individuals.

Of these legacies, the widespread availability of weapons has had the most far-reaching security effects. These weapons, previously used for political and ideological reasons, now fuelled criminal violence in both countries. As Cock15 explains, the social categories that sustain the demand for, and the use of light weapons have cultural and economic motivations. On the one hand, these weapons that were now used in the service of crime, acquired a high level of mobility that neither respected national territorial borders, nor political and ideological divides. On the other hand, the democratic transition period(s) was characterised by weak safety and security institutions, increasingly unable to protect the citizens and their property. The feeling of insecurity among the population clearly became a major problem. The lack of security threatened to jeopardise development and, with no development, political progress remained superficial and fragile.

These were the hard realities confronting the elected governments in Mozambique and South Africa during 1994. Enhanced co-operation became one of the most promising avenues to follow, but the road was not free of constraints. Historical factors and mutual suspicion stood in the way of a meaningful co-operation arrangement that only enlightened wisdom from both sides could remove.

On the positive end of the continuum was the fact that the general political and strategic environment was conducive to uncontroversial and quick political agreements. The end of apartheid provided South Africa with free entry into SADC and other regional arrangements. This new political environment that came into being in an uneasy region meant that the challenge ahead was to give practical and operational content to historically empty political agreements. In other words, while political agreements were extremely important — by virtue of defining strategic directions — the attitudes displayed during co-operation and implementation processes would ultimately determine their success or failure.

It is with these propositions in mind that this monograph examines the bilateral co-operation between the South African Police Service (SAPS) and the Police of the Republic of Mozambique (PRM) since 1996. The specific co-operation took the form of a joint weapons removal operation, code named ‘Rachel’. Firstly, this section describes the context in which the need for the joint operation emerged, based on a political environment that yielded easily, due to the convergent political processes in both countries and the perception of a common destiny. Secondly, the ad hoc implementation strategy is considered that was to become vital for the success of the initiative. Thirdly, the connections that were made during successive phases of this operation are considered. Finally, the potential of ‘Rachel’ is analysed in terms of its further development in the Southern African region.

Context: From Political to Criminal Violence

For the purpose of this section, it is assumed that Mozambique and South Africa were in states of civil war before 1994. The violence directly or indirectly related to these wars, is regarded as political violence.16 This proposition neither precludes the fact that criminal violence might have taken place during the conflicts, nor that political violence has come to an end in these countries. It is simply intended to emphasise that, prior to 1994, politically motivated violence overruled violence of other kinds.

South Africa

After peaking in the period between 1990 and 1994,17 political violence in South Africa, in general, and between the Inkatha Freedom Party (IFP) and ANC supporters in KwaZulu-Natal, in particular, declined considerably in the aftermath of the April 1994 elections. From an average of 244 deaths per month in the period prior to elections, politically motivated deaths declined to 144 in 1995.18 While political violence has eased since 1994, violent crime has increased.

According to the Nedcor/ISS Crime Index, five categories of the most violent crime seemed to have stabilised in the post-1994 period, but remain considerably high compared to the period prior to 1994.19 Weapons played an important role in these crime categories with, for instance, 39,8% of all reported murders during the first six months of 1995 committed with firearms. During the same period, 33 441 robberies were reported of which 26 563 (79,4%) were firearms-related.20 Furthermore, while the predominant weapons in KwaZulu-Natal were kwasha (home-made weapons) before 1990, political activists and criminals increasingly acquired AK47s, R4s and G3s since then, and "... gun running became a massive commercial operation."21 Most of these weapons entered the market as a result of ineffective disarmament and demobilisation processes both in Mozambique and South Africa.

These developments further increased the demand for illicit weapons for criminal activities in South Africa and, as a result, increasing numbers of licit weapons were required and acquired for self-defence. Statistics provided by the Central Firearms Register (CFR) show that, in 1996, South Africa had approximately 4,1 million licensed firearms in the hands of some 1,9 million owners. The statistics also show that the CFR received between 18 000 and 20 000 licence applications per month during this time.22 On the other end of the spectrum, rough estimates of illegal possession indicate the existence of between 400 000 and eight million small arms circulating in, through and out of South Africa.23

Mozambique

The end of the armed conflict in Mozambique left large numbers of redundant weapons in the hands of demobilised soldiers and civilians, as well as caches hidden in the bush.24 The failure of the UN Operation in Mozambique (ONUMOZ) to undertake a comprehensive disarmament process has been extensively reported.25 Irrespective of the reasons why disarmament has been problematic in one of the most successful UN peacekeeping missions, the hard fact, as Vines puts it, is that, "[a]lthough the conflict ended, the networks controlling light weapons supplies simply found new customers, using existing caches and networks to traffic weapons to neighbouring states, specially South Africa."26 The gap between the number of weapons thought to be in the country and weapons collected at the end of the peace process was so big that it clearly represented the potential for internal and external instability in Mozambique in the post-election period.

Secondly, given the weakness of the country’s economy, the social and economic reintegration of former combatants and persons returning from exile became a nightmare. A pilot study of the Refugee Studies Programme27 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 agricultural sector to guarantee the subsistence of rural families, forced former combatants to look for alternative income through the informal commerce of urban centres. The vulnerability to crime of all kinds in this setting is real, not only because the products on offer in the informal economy are stolen, but also because the temptation to use anything to guarantee survival, including weapons, is very high. Incidents of demobilised soldiers selling weapons to criminals and/or using weapons for criminal activities themselves were reported on many occasions.28 Besides, the simple fact that rural elements joined the unemployed ‘army’ in urban areas became a potentially destabilising factor in itself.

Thirdly, weak state institutions were unable to maintain law and order, in general, and the police were particularly unable to provide effective solutions to rising crime rates. Thus, weapons moved from war to crime with impunity, threatening to transform Mozambique into a society where only the logic of the powerful prevailed.29

Linkages between South Africa and Mozambique

The situation prevailing in Mozambique and South Africa in the aftermath of the general elections in both countries led to a synergy that impacted mostly on the illicit weapons market.

When political violence was eased as a result of the political engineering that culminated in the 1994 elections, the pipelines for illegal weapons transfers were already in place. Criminal motivations just replaced political ones in the use of these pipelines. The traditional political and ideological drives of regional arms deals were replaced by criminal ones. This apolitical, regional cross-border movement of weapons rapidly spiralled: weapons began to move from relatively stable areas to more unstable and violence-ridden ones in the region and further afield. Hence, the increased demand for weapons in South Africa was easily fulfilled through the already existing arms pipelines linking the two countries.30

It was thus that gun-running from Mozambique to South Africa became a major concern for both the South African and Mozambican police services. In response, both governments embarked on a variety of domestic and bilateral measures in attempts to curb arms proliferation and lessen its effects.

Domestic measures include a combination of prevention, confiscation, amnesty and buy-back initiatives, and heavy penalties for offenders. The South African Arms and Ammunition Act of 1969, was tightened in 1993, 1994 and 199831 and is currently being further debated in order to control and tighten the legal acquisition of arms. At the same time, South African National Defence Force (SANDF) and SAPS joint operations have been undertaken to combat the rising criminality in the country. SAPS figures of confiscated weapons support the importance of policing efforts. Indeed, as a result of the efforts of the firearms unit, 16 291 illegal arms were seized in 1995 alone.32

Comprehensive government policy was formulated, following these early attempts to curb crime and weapons. The National Crime Prevention Strategy (NCPS) was unveiled in 1996 and is managed by the Secretariat for Safety and Security. The NCPS is a multidepartmental initiative that centres on law enforcement and long-term prevention through improved policing, as well as the co-operation of the criminal justice system as a whole.

South Africa has taken steps to address the spread of illicit small arms in the country and the region from both an arms control and a crime prevention perspective. The South African Department of Foreign Affairs has recently released a position paper on small arms and light weapons proliferation. This document outlines the appropriate steps to stem the proliferation of these weapons in the viewpoint of the South African government. It suggests the need for:

an holistic approach with concurrent action at national, regional and international levels focusing on both licit and illicit small arms and light weapons; and



a regional initiative for the control of weapons proliferation in Africa.

The position paper emphasises that the proliferation of small arms and light weapons must be viewed from an inclusive perspective of arms control and disarmament, post-conflict peacebuilding, conflict prevention and socio-economic development. It proposes practical measures for co-ordination and co-operation at national, regional and international levels. At the national level, these include:

enhancement of legislation and regulation; and

reduction of the current number of existing weapons.

At the regional level, proposed actions are:

confidence-building and transparency measures;



steps to prevent the inflow of weapons to affected regions; and



co-operative partnerships between governments, international and regional organisations and non-governmental organisations.

The crime prevention approach as proposed in the NCPS, includes a strategy on firearms policy that aims at:

improving controls over the possession of legal firearms (both private ownership and those owned by state security structures);



preventing legal firearms from becoming illegal through criminal activity;



taking proactive steps to reduce the number of existing firearms in the country;



preventing the inflow of illegal weapons; and



mobilising public and political support for the above processes.

To date, the South African government has actively been improving controls over land borders and has reduced the number of international airports in the country. Efforts to improve the inspection and clearance of goods at sea ports are also under way. In addition, the current legislation regulating civilian ownership of firearms is under review and new legislation is expected to be tabled before Parliament in 1999.

In Mozambique, the authorities recognised small arms proliferation as a security challenge.33 The Mozambican Attorney-General stated that:

"... a large demand for illegal firearms in South Africa has prompted Mozambican gun-runners to extend their sources of such weapons and gun-running has formed part of several types of organised crime in [the] country, including drug trafficking, car theft, and money laundering."34

Manuel Antonio, then the Minister of the Interior, announced a national master plan in April 1995 to curb illegal weapons circulating in the country. The plan included the deployment of permanent police units to patrol the main roads, the re-establishment of district police commands and the creation of a special unit for the destruction of arms caches. In addition, an increasing awareness among civil society prompted the launch of anti-arms campaigns, such as the Arms into Hoes project of the Christian Council of Mozambique, and those of Gun-Free South Africa (GFSA).

Following the launching of the master plan, Mozambican police were reported to have apprehended thirty armed gangs and uncovered 69 arms caches in three months. Between January and July 1995, the police seized more than 6 000 arms and 24 000 rounds of ammunition. Ambushes along the highway stopped shortly after the deployment of ‘Lightning Battalion’ (a police special unit created in response to ambushes and armed attacks along the highway between Mozambique and South Africa).35

However, none of these were enough to control and reduce the problem itself. Given the porousness of the countries’ borders, the existing supply networks and routes, and the interconnection between illicit arms and other cross-border crimes, such as vehicle theft and drug trafficking, any unilateral progress in either country clearly became insufficient. For example, SAPS estimates show that the South African police recover only ten per cent of illegal arms entering the country.36 Furthermore, in 1995, as Latham rightly described, "the government [of Mozambique] is destitute. It cannot afford to buy shoes for its policemen, most of whom walk the streets in an odd assortment of sandals, trainers and tennis shoes."37

In view of the above, it can be argued that the political transition processes in both Mozambique and South Africa were accompanied by:

increased small arms proliferation;

a shift in the use of weapons from war to crime;

an increase in violent crime;

an expansion of the illegal arms market within and between the countries;

a lack of state capacity to provide security for the public; and

ultimately, the potential for general social instability.

As a result, the democratically elected governments of Mozambique and South Africa were faced with increasing levels of violent crime exacerbated by the widespread proliferation of arms. Latham38 commented that

"... 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."

In the case of South Africa, the sources of weapons were clearly the domestic defence industry, the remains of the political conflict, and increasingly — but not exclusively — arms smuggled from Mozambique. As pointed out above, weapons could be found in Mozambique in the hands of demobilised soldiers and civilians, and cached in the bushes.39 It was recognised that the bulk of arms being smuggled out of Mozambique came mainly from caches. Hence, destroying caches before the weapons reached smugglers became the basis for the bilateral political willingness to co-operate. This led to the creation of Operation Rachel.

Political Contingencies and the need for Common Ground

Against the background of common problems described above, it has to be remembered that the recent history of the relationship between South Africa and Mozambique was characterised by hostilities. The two defence and security establishments fought an undeclared war for the previous two decades, and regarded each other as enemies. This made collaboration processes difficult, if not impossible in the past. Besides, co-operation between South Africa and its neighbours was long regarded with suspicion. Given its relative power, South Africa is perceived as having hegemonic aspirations. As a result of the sensitivity of security issues, co-operation in this area suffers the most from such a suspicious environment. Thus, even though the Mozambican and South African police services have attempted to co-operate before 1994, distrust and a lack of political clearance prevented the institutions to work together in a meaningful manner.40 Therefore, despite the fact that arms smuggling and proliferation between the two countries became a common security challenge, and co-operation became clearly not an option but an imperative, political awareness and willingness were prerequisites. Obviously, the new distended strategic environment, enhanced by the political transformation in both countries, laid the ground for settling these differences.

Although the political environment of the post-1994 period removed some of the obstacles for co-operation, it was the pressure brought about by the rise in violence and crime in both countries that sparked the final agreement. With the end of internal conflicts, both in South Africa and Mozambique, arms proliferation and its related violent crime were taken as factors which could jeopardise the political progress that was achieved. Shaw rightly argued that, in South Africa, "[c]rime is ... implicitly and explicitly seen as a central test of the capacity of the Government to rule and the new democracy to consolidate."41 The same became apparent early on in Mozambique with crime in Maputo spiralling out of control. How could democracies consolidate and develop when South Africa was losing an estimated R31,3 billion, or 5,6% of its gross domestic product in 1995 to crime,42 and Mozambique was not attracting a critical mass of investors due to its insecure environment? The toll of crime itself on business and potential investment was enhanced by its direct and indirect costs that society had to bear as a result of the prevailing insecurity. The use of scarce resources to tend to the treatment of firearms-related casualties in hospitals, for example, put a heavy burden on the Mozambican and South African health authorities alike.43 Furthermore, the lack of trust in the protection offered by the police generated a mushrooming of private security companies, which, in themselves, compounded the problem of the increased availability of firearms.

In recognition that neither country was achieving rapid progress in controlling its own internal security concerns, and because both were considered to be part of these problems, Mozambique and South Africa felt that they had no choice but to co-operate across borders. As a result, the presidents of the two countries, Joaquim Chissano of Mozambique and Nelson Mandela of South Africa, met in March 1995 to sign a co-operation agreement, In Respect of Co-operation and Mutual Assistance in the Field of Crime Combating.

The preamble to the Agreement states that the "... parties are desirous of concluding such an agreement in order to contribute towards peace, stability, security, prosperity and crime combating."44 For both countries, security, peace and stability are in part a function of the extent that crime can be prevented, controlled or fought. In Article 6, the Agreement provides that the parties, in "... recognising the incidence of organised crime and the need for close co-operation in addressing the problem ..." shall undertake:

the exchange of crime information on regular basis on arms;



the planning and co-ordination of operations, including covert operations; and



technical assistance and expertise where these are required for the purpose of criminal investigation.

Despite the common ground and improved political settings, the Agreement became an operational nightmare. A sound implementation strategy, therefore, became critical for the achievement of actual goals. The different security institutions were thus expected to work out this strategy. This was easier said than done in the context of the historical and structural differences between them.

Article 2 provides that:

"This agreement shall in no way be construed as derogating from any provision of: (a) laws of respective parties regarding extradition; (b) any extradition agreement which is in force or may be entered into between the parties; and, (c) the co-operation agreement entered into between the government of the Republic of South Africa and the government of the Republic of Mozambique on 20 July 1994."45

Article 3, however, gives broad discretional powers to the safety and security portfolio ministers. As a matter of fact, paragraph 3.2 states that:

"The Ministers shall consult each other and advise their governments as to how the legislative or administrative steps that may be necessary for the implementation of the provisions of this agreement and remove any legal obstacles or impediments that may be found to exist in the execution of the provisions of this agreement."

Thus, the Agreement actively attempted to create an environment conducive to effective working relations between the safety and security institutions of the countries. A peculiar characteristic of the Agreement is that it not only outlined what needed to be done, but also carefully set out the conditions that would allow co-operation between two very different and distrustful agencies. The Agreement could not rely on existing ad hoc arrangements, but became a confidence-building measure in itself aimed at improved working relationships between the two parties to the Agreement. Whereas other regional arrangements had started off as ad hoc measures that were eventually institutionalised (for example, SARPCCO), Operation Rachel became the opposite — an institutionalised arrangement providing an umbrella for ad hoc co-operation.

The Agreement and the political discourse surrounding it were tailored in such a way that it would foster a sense of ownership at all levels. According to Director Naude46 of the SAPS firearms special unit, the opposition in Mozambique — RENAMO in particular — was consulted prior to the operation. Even though the agreement between Eastern Transvaal (now Mpumalanga) and the Governor of Gaza province has nothing to do with the joint operation, the need for local authorities to support the operation led Mike Bester, then acting police commissioner, to consider that "... weapons raids across border had been authorised by the 12 June [1995] agreement between Eastern Transvaal Premier Mathews Phosa and Mozambican authorities [Gaza Province Governor, Eugenio Nhumaio]."47

Operation Rachel: structure, functioning, cost and Results

As is clear from the previous section, the governments of South Africa and Mozambique made sure that a common ground was defined, before the police agencies started to work together. It was explicitly recognised that the security challenges emerging from the illegal flow of firearms into South Africa and the potential of existing arms caches to disrupt rural safety in Mozambique should be put above any political agenda. For South Africa, it was important that caches were identified and the weapons destroyed to prevent them from being smuggled into its territory were they fuelled violent crime.48 For Mozambique, rural safety, the eradication of violence and the general disarmament of people in rural areas were the main aims.

At the outset, it was vital to identify common concerns. Problems had to be defined in such a way that neither country would be seen as helping the other, but that each was in the process of serving its own country’s security needs. In other words, the destruction of arms caches in Mozambique by SAPS had to be interpreted as part of its mandate in maintaining law and order within South Africa. SAPS officials working in Mozambican territory were fulfilling their duties and they had to commit themselves fully to their task, as they would have done if they were policing Johannesburg streets. For the Mozambican police service, this operation was just an additional effort in the country’s process of demilitarising society.

Having conceived the problem in this way, the question was how to go about this task in the light of the characteristics of each country. Mozambique had neither the financial resources nor the expertise to destroy arms caches. SAPS lacked the knowledge of the Mozambican terrain and had no legal right to operate inside Mozambique. The combination of problems became a strength: SAPS would not only supply the bulk of the financial resources, but also landmine-resistant vehicles and other specialised equipment, as well as highly trained senior police officers. Mozambique would gather intelligence and with its knowledge of the local conditions, facilitate contacts with local communities. Given the circumstances around arms caches, it was decided, in addition:

The arms caches destruction operation would be intelligence-driven. It was agreed that both Mozambican and South African police forces would gather intelligence about these caches. The two teams jointly plotted the caches on a global position system (GPS) map. A team of Mozambican and South African police experts would subsequently be deployed to the field to destroy the weapons in situ.



The operations should have an unorthodox policing approach. While one would expect individuals holding arms caches to be prosecuted in a traditional proactive policing operation, individuals were co-opted, worked with and generally rewarded for disclosing arms caches in this case. There were several reasons for this approach, for instance, the belief that most of the cache informers knew of more than one cache. Hence, "if you prosecute at the outset, you lose his/her co-operation in disclosing other caches,"49 argued a South African police officer. In the Mozambican post-conflict situation, a proactive police operation also ran the risk of being easily politicised, for most of the caches had been kept for political objectives.50 For the sake of reconciliation, an undeclared amnesty was therefore introduced as a component of the operation. Finally, given the poverty prevailing in the rural areas of the country, it was decided to reward people disclosing arms caches. A modest buy-back component was also introduced.



An equal partnership should be fostered between the two police forces.



Hence, a joint command structure should be put in place.

With this structure in mind, the first operation, code named ‘Rachel’, was launched on 11 August 1995.

Rachel I

The operation started with an exchange of information related to arms caches. The Mozambican side had to verify this information by contacting informers and confirming the existence of these arms caches. Between 11 and 26 October 1995, 45 police officers from the SAPS Task Force (Air Wing, Crime Intelligence Service and Bomb Disposal Unit) were deployed. The team was divided into two groups — one operated in the Ponta-d’Ouro area and the other moved northwards to the Massingir area. The South African government provided R 301 223 for the operation. A joint command comprising one senior police officer from each country was established in Maputo. While the joint command was responsible for co-ordinating the actual work, including taking decisions regarding current operational contingencies, each team representative referred to its respective police headquarters for clearance.

Based on information that was gathered, appointments with informers were arranged. Before proceeding to the caches, local authorities and the population were made aware of the nature of the operation. The briefing sessions were also an opportunity for public awareness raising. Informers guided the police teams to the caches, where most of the weapons had been buried under the ground. Depending on the assessment made by the experts of the quantity of arms found, the accessibility of the location and the security of the people in the vicinity, the weapons were destroyed in the original location or moved to a more appropriate place. Sometimes, a public destruction session was arranged where the media were invited as witnesses. The public destruction sessions were also used as a means for raising public awareness of the operation.

As far as informers were concerned, a cash reward was given. No rigid criteria were used to define the value of the reward. It depended mostly on the value of the cache which, in turn, was determined at the discretion of the team through negotiations with informers. Broadly speaking, the quantity and quality of the weapons that were found, played an important role in determining the amount of the reward. Since rewarding was used as an incentive for disclosing caches, the value was also greatly influenced by indications that the informer might know of other caches. Informers were further mobilised to disclose more caches that they were aware of, and were also encouraged to pass on their experience to other people.

According to Monguela,51 while the involvement of local police officers has been a feature of all the operations, the general co-ordination had to be left in the hands of senior officers from the Mozambique Police General Command in Maputo to ensure that the visiting police team was properly treated. The need to involve local police officers was mainly based on their knowledge of the area and, particularly, the imperative to establish sound relationships with local communities.

The support of local communities is deemed a critical factor, because they are valuable sources of information. Director Naude, in appreciation of the role of local populations, stated that, if operation Rachel has been successful, this success should be attributed to the people of Mozambique who, tired of war and violence, did not want more weapons freely available in their country.52

At the outset, difficulties of all kinds were experienced. Some were of a structural and others of a contingency nature. The former were mainly those related with resources and the general context in which the operation took place. For example, South African police officers deployed in Mozambique found the working conditions somewhat inhuman (lack of basic infrastructure and general living conditions). In fact, when Lieutenant-General Wouter Grove, head of the SAPS Crime Intelligence Service, flew to Maputo shortly after the first deployment, he went not only to confer with senior Mozambican officials, but also

"... to visit SAPS task team on the ground in Mozambique [in order to ascertain] their welfare first-hand to ensure that everything possible is being done to support them in the field [where they are] ... working in extremely difficult and primitive conditions."53

Another structural problem was related to language. Communication between the members of the two police teams was not always easy, due to language differences. This made the selection process much more difficult for the Mozambican police service because, besides expertise, language had to be one of the criteria. It was important to make sure that the task force teams communicated between themselves, as a lack of dialogue could cause minor problems to deteriorate into major conflicts that would jeopardise the whole operation.

Furthermore, the differences in operational skills between police officers were potential sources of misunderstanding. Indeed, as Director Naude pointed out, some specialised SAPS officers found it difficult to work with their Mozambican counterparts on specific issues at the beginning, due to their weak technical skills, for instance, in handling explosives. A final structural difficulty worth mentioning was the lack of resources among Mozambican officers. Indeed, the illicit firearms unit at the general command of the Police of the Republic of Mozambique, which was responsible for actions around arms caches since the end of the armed conflict in 1994, has a budget of only about R15 00054 a month, with no available vehicle. This makes their work — required to find arms caches as part of the intelligence gathering operation — mainly dependent on South African resources.

Other day-to-day contingencies had the potential of hampering the relative success of the operation. Examples were personality-related conflicts, cultural differences, perceptions evolving from the prevailing prejudices, to name but a few.

As pointed out above, the two police forces had regarded each other as enemies before and neither was sure to what extent the other had changed. This certainly resulted in speculation whether there were hidden agendas or not. Distrust thus characterised the first encounter between the two police forces. For example, during the first operation, Mozambican police officers implicitly questioned why SAPS officials insisted on identifying the origin of the weapons that were found, only when they were sure that these weapons were not of South African make. This suggests that they perceived the South African task force team to have had a parallel agenda of dismissing old allegations that one of the major suppliers of weapons for the conflict in Mozambique was South Africa. If this view had prevailed, the operation could have been politicised.

Another example related to the way in which Mozambican police officers regarded their counterparts, due to the fact that they hardly ever saw black SAPS officers as members of the teams. There were also claims that some of the police officers deployed for Rachel I displayed racist behaviour.

From the South African side, even though interviewed SAPS officers have dismissed the point, it seems that frequent reports about rampant corruption within the security forces in Mozambique had created an image of a corrupt police force that deserved no trust.

A final example is one related to cultural differences. According to SAPS police officers, there were a few small problems regarding food preferences at the beginning. While some of these complaints may arguably be true, there is no doubt that they were fuelled by old perceptions. Superficial as this may be, it had the potential to endanger the working relationship between the parties.

Notwithstanding these difficulties, the operation recovered 1 120 weapons of various calibre and other assorted war materiel, including mortar tubes and bombs, AK-47 rifles, RPG-7 rocket launchers, hand grenades, landmines, limpet mines, PPSH sub-machine guns, explosives and 23 182 rounds of ammunition (see Table 1).

Summary

As conceived, the operation simultaneously had to tackle the security concerns of South Africa, as well as those of Mozambique. Each of the task force teams had to be able to see the results in terms of the solution of security problems in its own country. The operation therefore had to be undertaken in such a way that it fulfilled this objective. The only way that this could possibly be achieved, was to concentrate on the immediate borders of both countries. This would allow both the South African and Mozambican authorities to link the retrieval of any weapon to criminal images or the probable illicit trade in their respective countries. South Africa, for instance, could easily convey the message that arms were found and destroyed literally on the border, thus showing that, if not destroyed, the potential of easily crossing the border and being used in criminal activities or for other purposes inside the country is clearly high. The head of the National Crime Investigation Service, Wouter Grove, explicitly conveyed this idea by commenting that "[t]he operation was a major success for President Nelson Mandela’s community safety plan announced in May."55 This seems to explain that, although most of the caches were clearly in the central part of the country (Sofala and Zambezia),56 the first Operation Rachel took place around Maputo and Gaza provinces.

Both countries presented the results as a breakthrough in their fight against crime. Both at the political and operational levels, the enthusiasm to repeat the experience was clear.

Rachel II

Although the second Rachel operation followed in the steps of the first one, there were some subtle differences. The prevailing philosophy was that the first operation should not be judged in terms of the quantity of arms recovered, but in terms of the lives and property which could be lost if these weapons were put to use and the operation’s contribution to stability in these countries. In this way, any result could be taken as adequate to justify a second Rachel operation. Though the results may not have been as successful in absolute terms as it has been claimed, Rachel was meant to continue. SAPS stated that "the operation had shown the success that could be achieved through co-operation ... in the fight against crime,"57 thus proving this point.

Public reports emphasising successes suggested that the operation was above question. Negative factors were being worked out diplomatically in the period between Rachel I and Rachel II. These problems have never been publicly acknowledged. However, a reading between the lines of the statement that "[o]ur breakthrough in establishing a good working relationship with our Mozambican counterparts and citizens of Mozambique will definitely contribute to stemming the flow of criminal activities into South Africa,"58 suggests that, after a closer look at Rachel I, changes had to be made.

From the statement it is quite clear that most of these changes were related to improving relationships between the two teams and the population, as well as among police officers themselves. A one-week training course was undertaken. The content of the training course included, among others, explosives and booby-trap handling techniques, techniques for safely digging under ground caches, and communications. Above all, the training was aimed at teambuilding.

When Rachel II was launched a year later, some important changes had been made. A global positioning system (GPS) was introduced for general use. The leadership had to make sure that unconfirmed allegations that "... a series of scandals involving tipping off traffickers prior to raids and involvement of both South African officials and their Mozambican counterparts with these traffickers,"59 were dealt with in a diplomatic manner.

Based on each party’s complaints, personality clashes, for example, were seemingly sorted out by removing some of the officers on the police task force team from the operation. Although this may seem a weak argument, Mozambican officials have indicated that, for the success of the operation, some of their colleagues from both SAPS and the PRM had to be removed.60 The indication by Director Naude61 that the operation demanded a specific kind of person who is friendly and can persevere, supports this argument.

Another example of change was the fact that, in response to the potential problems resulting from the insistence on tracking the origins of weapons that were found, this was no longer pursued in Rachel II. Spontaneously, or by rational decision, the emphasis was placed on destroying weapons as opposed to finding their origins. The causes of the problem were to be set aside, as they were a legacy of the past. The concern was how to get rid of these weapons. To this end, contacts between the two teams were eased by way of simplifying the bureaucratic process. This demonstrated the need for concentrating on the common interests and overlooking the factors which could divide the parties. A common uniform and insignia, in Portuguese and English, indicating South Africa/Mozambique police joint operation, were also introduced to foster a sense of unity. Above all, by making small changes, the leadership demonstrated their responsiveness towards both teams that, from the point of view of confidence-building, had a great impact.

With all these issues in mind, the exercise of intelligence gathering was repeated. Arms caches were jointly plotted on a GPS map. The police teams were deployed between 30 September and 5 October 1996. Maputo, Gaza, and Inhambane provinces had been earmarked for Rachel II and a joint command was established in Xai-Xai, Gaza. A two-way radio communication station was established at the bordering area of Pafuri aimed at providing communication links between the teams on the ground and the police headquarters in Pretoria and Maputo. South Africa made available some R565 033 and other resources, including helicopters for the operation. The teams consisted of 36 SAPS and 22 PRM officers. Areas covered by Rachel II included Funhalouro and Maputo Province in the south. At the end of the operation, a fair number of weapons were recovered and destroyed (see Table 2).

Despite the improved environment, the quantitative success was less than expected (See Table 5). Rachel II recovered and destroyed less than half the number of firearms destroyed in Rachel I. The reasons for the lack of success seem to lie beyond the operation’s effectiveness. This may be due to the fact that, while people may be predisposed to collaborate with the police for crime prevention and combat, the willingness to disclose the position of hidden caches is mediated by general confidence in the twin processes of pacification and democratisation, as well as the prevailing economic hardship in the country.

Indeed, because the general elections in October 1994 were successful in Mozambique, the political environment of 1995 was marked by an overall consensus on the need for joint efforts to reconstruct the country. The most important manifestation of this mood was demonstrated, among others, by the meeting between President Chissano and Dhlakama62 to discuss the arms proliferation problem and the parliamentary consensus over the government’s five-year programme.

Abrahamsson and Nilsson named this positive political mood among the politicians the "spirit of reconciliation policy."63 No matter what the reasons might have been, the political environment seems to have determined a fairly high level of people’s confidence in the peace and democratisation process at the time. This is especially true for those who might have kept arms for political reasons. The reintegration support scheme also still provided some sustenance for demobilised soldiers. In such circumstances, the political value of weapons decreased and hence, whoever might have kept a weapon for political reasons was tempted to hand it in at the first opportunity.64 This may explain the relative success of Rachel I.

However, the popular confidence in the peace and democratisation process in Mozambique that coincided with Rachel I, decreased from the end of 1996 as the political environment clouded. The remaining expectations of Renamo to play a considerable political role in the country, were jeopardised by the discussion around local elections. The government was tough in its attempts to force Renamo to vote against the local elections bill package. Renamo, in turn, threatened to block the elections. The tension increased up to the point that the language used by both parties suggested a potential resort to armed actions.65 At the same time, the reintegration subsidies were no longer paid. These issues had increased the political value of arms caches. Hence, political control over such caches seemed to have tightened and cache caretakers were less willing to disclose their existence. This suggests that the politically harsh environment since the second half of 1996 was less conducive for politically held caches to be handed in, hence the difference in quantities of weapons recovered during Rachel II.

Despite the less spectacular results of Rachel II, the commitment to proceed remained unchanged in both countries. This was partly due to the fact that the parties were convinced that more weapons were still lying in the Mozambican bush, and partly because the operation had gained a momentum of its own.

Summary

During Rachel II, the responsiveness of the leadership enhanced the confidence between police teams and among police officers. South African police helicopters, which initially had to refer to Maputo for clearance every time they crossed the borders, no longer needed to go through all these bureaucratic red tape. Contacts between police officials were reportedly taking place on a daily basis. A clear shift from functional partnership to amicable friendship was taking root among police officers. Finally, an unconfirmed decline in arms smuggling into South Africa was interpreted as a result of the Rachel operations. Raul Freia, spokesperson for the General Command of the PRM believed that, if the operation continues, Mozambique will be free of weapons in a couple of years.66

The image of attaching arms caches in Mozambique to crime in South Africa was maintained. South African Minister for Safety and Security, Sydney Mufamadi reiterated this view: "One shudders to think what the consequences might have been if these weapons had made their way across the border and into the sea of illegal weapons that is contributing to South Africa’s crime problem, as well as afflicting Mozambique’s own cities and towns."67 In a meeting between Mufamadi and his Mozambican counterpart, Almerino Manhenje, joint efforts in fighting crime were further emphasised. As far as the Rachel operations were concerned, it seems that a decision to expand northwards to search for weapons caches in the central Mozambican area was taken at this time. Certainly, there was a belief that once there were no more weapons in the southern provinces of Mozambique, chances were that arms smugglers could still operate as far afield as Sofala and Zambezia, bringing weapons into Maputo and then into South Africa.

The rationale became clear: there would be no rest before the Mozambican territory was free of arms.

The operations had also managed to attract civil society’s attention. Companies in South Africa expressed willingness to provide incentives such as sweets for children and women handing weapons or ammunition to the team. Moreover, Director Naude noted an increasing voluntary and unpaid collaboration by the local population, particularly women and children, with the operation. The willingness of the local population helped to sustain the morale of the task team and provided the foundation for Rachel III.

Rachel III

At this junction, the information gathering had become a routine, ongoing activity that fed information to the joint GPS maps. The third operation was launched on 21 July and lasted until 9 August 1997. It was aimed at covering Maputo, Gaza and Inhambane in the south, and Sofala and Manica in the centre of the country. A group of 24 SAPS officers, joined by twelve PRM officers, were deployed. For the sake of team cohesion, there was a need to use the same officers as before and avoid bringing in new officers unless circumstances demanded it.

Rachel III was launched against a background of:

a reiterated political commitment;

a sound working relationship between members of the joint team;

improved skills and methods of arms collection among police officers;

an improved understanding of and support by the community;

an ever sensible civil society; and

conversely, increased cost.

The prevailing political commitment and improved relationships among the police officers have been described above. Although it may seem that the removal of Interior Minister Manuel Antonio late in 1996 had nothing to do with the operation, it has been seen as a further signal by the Mozambican government that it will fight corruption within its security apparatus. With the appointment of the new minister, relations improved even further.

As far as skills development was concerned — explosive disposal, booby-trap handling and other security measures — the Mozambican team was reported to have acquired the necessary skills to carry on with the operation on its own at this stage.68 The need for intelligence and enhanced community participation was further recognised. Given the increasing role played by women and children, new incentives were used. Various companies from the South African private sector started supporting Rachel by providing foodstuffs to be given to the local population as incentives. The importance of communication for the success of the operation led to the introduction of satellite phones that made contacts much easier.

However, while these factors suggested a conducive and co-operative environment for the operation, costs started to escalate. Indeed, Rachel III cost R600 452 — just R266 000 less than the cost of Rachel I and II combined. Two factors seem to have contributed to the increased costs:

The expansion of the operation northwards meant that the further afield it moved, the more expensive it became. Almost all of the costs incurred, are directly proportional to distance. In addition, the problem of accessibility also influenced costs. In Mozambique, the further north one ventures, the worse the roads become and the more difficult to gain access to remote areas. This increased the demand to use helicopters. As result of these factors, South African police officers have voiced their concerns over costs, which threaten the continuation of the operation.



The longer the operation takes, the more expensive intelligence gathering and rewarding informers become. According to Monguela, information about arms caches is becoming more costly as time passes. Informers are aware that they can get material and financial benefits from the fact that they know where weapons are. As they sense a demand for their services, it is logical that the price will escalate. This is especially true in a context of rampant poverty. There is no doubt in Naude’s mind that many of these people sell weapons in the first place (or try to get as much out of the Rachel as they can), to meet their basic needs. He argues that one needs to build schools, hospitals and roads, and provide clean water to reward these communities that suffered severely from the war and are now willing to get rid of weapons.69 This added another factor to the operation. While security calculations had determined the political will, the commitment of police officers, as well as people’s willingness to support the initiative at the beginning, these were no longer sufficient to keep the momentum going. Human needs and socio-economic development imperatives have to be considered. This adds the debate around the security/development dichotomy, showing that security problems cannot be effectively dealt with, without addressing general socio-economic development issues.

With this mixture of enthusiasm and concern about costs, Rachel III took place. The results were much more encouraging than those of Rachel II (see Table 3). These results were immediately published in police statements, both in South Africa and Mozambique.

According to Director Reg Crewe of the SAPS Detective Service, this operation was by far the most successful.70 Among the joint operation team members everything seemed more on track than before. Furthermore, the team increasingly believed that further operations were needed, since the number of reported caches had increased significantly as additional provinces were targeted and people clearly showed their willingness to co-operate. Monguela indicated that more than sixty caches had been identified and plotted on maps, but resource constraints loomed large on both sides.

Summary

The relatively poor results of Rachel II and the increasing costs did not prevent a further operation from taking place, thanks to the political commitment, the sound working relationship among police officers, and an increasing awareness and willingness by the community to support the operation. The operation’s structure and procedures have taken root, and its philosophy has been internalised among police officers. In all, the operation has gained a momentum of its own. This environment was further enhanced by the success of Rachel III, providing an uncontroversial basis for the following operation.

Rachel IV

At the end of Rachel III, the enthusiasm among both officers and the leadership became more pronounced. Everything seemed to be going according to plan. The results were regarded as particularly successful, feeding the hope that, if the operation continued, the country would soon be free of arms. In turn, South Africa would free itself from one of the sources of illegal firearms in the near future. Given the field reports of reported caches, there was no doubt that the next operation would be even more successful. These factors allowed South Africa to commit more resources to the next operation. Indeed, 31 well-trained bomb disposal, explosives and firearm experts and task force operatives, assisted by twelve Mozambican colleagues were deployed for a three-week long operation in Rachel IV. The operation cost R1 120 144 or about twice as much as Rachel III.

More than sixty caches plotted on the GPS map and spread through southern and central Mozambique filled the team’s work schedule. Rachel IV was classified as the biggest cross-border operation SAPS had ever undertaken. The team was divided into two groups. One started from the central province of Sofala and Zambezia moving southwards, and the other moved from Maputo towards the north. The two groups would meet halfway.

An initial clearance problem was experienced. According to Mozambican police officers co-ordinating the operation from the Mozambican side, every operation of this kind has to be cleared at the highest political level possible. The delay this time was due to the fact that the starting dates coincided with the absence of both Minister Almerino Manhenje and the President. Thus, "... at our level, our South African colleagues were on standby until the clearance was given."71

This delay meant that the team could not fulfil appointments arranged with informers. As a result, the team had to look for informers once they arrived on site. Further delays were experienced because, according to Assistant Commissioner Suiker Britz of SAPS, inhabitants of remote rural areas were reluctant to co-operate with the team.72 This was also the result of the team being split into two groups, with some of the contact people on the police’s side thus not present in the specific site where their informers were. Informers had problems dealing with strange police officers. In one instance in Mapai, caches were not excavated until Monguela,73 who had made the initial contact, was present.

Another new problem was that of communication and co-ordination between the operation teams and other security agencies within Mozambique, particularly the army. Indeed, according to a SAPS officer, some of the caches that were already plotted on the GPS map, were destroyed by the Mozambican defence force. While the fact that the caches have been destroyed is good, the lack of communication led to a duplication of efforts and a waste of resources. Notwithstanding these organisational problems, the operation has been considered successful (see Table 4).

Summary

At this stage, there are suspicions that weapons remain in the country, mainly north of the Save River. Within police circles, the willingness to proceed is maintained. The Mozambican Minister of Home Affairs, Almerino Manhenje, has indicated that the top priority for his ministry in 1999, remains the retrieval of weapons from caches.74 The South African NCPS also remains committed to fight illicit weapons. There are indications that SAPS have earmarked R1,5 million for Rachel operations in 1999.

However, doubts exist as the operation is becoming extremely expensive and common ground is being lost. As a matter of fact, arms smuggling from Mozambique into South Africa are reported to have decreased substantially. The further north the cache is found, the weaker the argument that one can attach weapons recovery to South African security concerns. With this in mind, the prospects that South Africa will remain committed to pay for the operation in the north of Mozambique are slim within the present parameters.

The Mozambican police are ready to continue this work, provided that financial resources and the technical means are made available. Whether the Mozambican government is in a position to cover the costs of the operation in the near future is doubtful, unless a generous aiding hand is extended. No matter what happens from now on, Operation Rachel has been an unprecedented success, not so much because of the number of weapons that were retrieved, but because it has worked effectively.