Demand from different social categories means that there is a resilient market in both legal and illegal arms. Arms smugglers operate in secrecy but with relative ease. The market involves an expansive social network and incorporates diverse social groups. Money is not the only means of exchange since ivory, rhino horn, diamonds, drugs and even second-hand clothing may be tendered in exchange for weapons. This suggests that light weapons are a widespread source of currency. Much of the available supply dates back to the periods of armed conflict in the region.

Original suppliers of light weaponry during armed conflict

During armed conflict in the southern African region, there were three main sources of light weapons:

• Former Warsaw Pact countries, Cuba and China

These countries supplied arms to MPLA in Angola, PLAN in Namibia, MK in South Africa, ZIPRA in Zimbabwe, and Frelimo in Mozambique. This included rifles, carbines, AK-47s, land-mines, limpet mines, mortars, hand grenades, pistols and ammunition. ‘Control over this material was uneven’ (Williams, 1995:2).

‘From 1987 to 1991 the Angolan government imported $4,6 billion in arms, 90 per cent of them from the former Soviet Union’ (Morrison, 1995:712). The Soviet Union remains a major supplier since the conflict resumed in 1992.

A South African police source estimated that the former Soviet Union had dumped an estimated 300 000 AK-47s in Mozambique during the mid-1980s (Citizen, 1 September 1993).

• Western countries including the United States, West Germany, France, Britain and Israel

All these countries provided the apartheid state with military hardware. Various Western arms manufacturers sent clandestine military aid to South Africa in defiance of the United Nations arms embargo.

For example, during the 1970s two United States gun manufacturers shipped thousands of firearms and millions of rounds of ammunition to South Africa through front companies (Klare, 1988:23).

Apartheid South Africa and the United States were major sources of weaponry in Angola. According to Klare, in ‘1975/6 the CIA provided anti-Communist insurgents in Angola with 622 crew-served mortars, 42100 anti-tank rockets, 20900 rifles, and millions of rounds of ammunition . . . So abundant was this assistance that . . . Unita in Angola have been able to keep fighting for years after the cessation of United States aid with arms stockpiled during the Reagan period’ (Klare, 1995:15).

• The apartheid regime

From 1976 to 1990, the ideology of ‘Total Onslaught’ provided the underpinning for the militarisation of South African society. The apartheid state mobilised resources for war on political, ideological and economic levels. This process was spearheaded by the SADF. The SADF was at the centre of an undeclared war of destabilisation that was directed first against neighbouring states, creating what has been described as a ‘holocaust’ (Johnson & Martin, 1989:11), and later against the ANC inside South Africa.

As part of this process, the apartheid state supplied weapons to the SADF, SAP and various ‘homeland’ armies and surrogate forces inside the country, as well as to Unita in Angola, Renamo in Mozambique and other rebel movements in the southern African region. This was done via the former SADF Directorate of Special Tasks (operating under Military Intelligence). According to Roland Hunter’s court evidence, at least during 1982/3, four such projects existed – Operation Disa (support of Unita), Operation Drama (support of Zimbabwe dissidents), Operation Latsa (support of Lesotho Liberation Army) and Operation Mila (support of Renamo dissidents).

Weapons for these projects were procured either through the South African arms procurement agency, Armscor, or were captured by the SADF during direct military action in Angola and Namibia. Many of these arms were of Soviet and Eastern bloc origin and included AK- 47s, LMGs, RPG7s, hand grenades, mortar bombs and mines. Almost 40000 AK-47s were purchased from Poland, Romania, Bulgaria, Yugoslavia, Hungary and China between 1976 and 1986, specifically to be given to Unit a (Cameron Commission, 1995). AK-47s from Hungary and Bulgaria, obtained through Armscor, were also supplied to Renamo (personal communication with Hunter, 1995). Overall, the South African government amassed a large stockpile of captured light weapons.

Anti-personnel land-mines were among the most deadly light weapons produced and supplied by the SADF. As part of the apartheid state’s destabilisation strategy, the SADF delivered antipersonnel land-mines and other weapons, ammunition, ‘propaganda material’, maize seed, sugar and tobacco to Renamo. According to an informant who was involved in Operation Mila, the material supplied was fairly constant from month to month, and in August 1983 involved 500 AK-47 rifles, various other weapons and large quantities of ammunition (personal communication with Hunter, 1995). There are many reports on how South African assistance was paid for with ivory and rhino horn.

After 1990, the destabilisation strategy of the apartheid regime was turned inward to weaken the ANC and to block the democratisation process. A crucial element in this strategy was the training and arming of a surrogate force in the form of Inkatha vigilantes, who operated under the direction and control of what came to be known as a ‘third force’. The ‘third force’ was made up of elements of the army and police, and there is evidence that much of the township violence between 1990 and 1994 was organised by them. For example, there is clear evidence of SADF training of Inkatha hit squads which were deployed against the ANC.

Deaths from violent conflict (involving light weapons) between supporters of the ANC and the Inkatha Freedom Party (IFP) peaked between 1990 and 1993, when almost 10 000 people were killed. The apartheid regime was not the only source of weaponry. Several informants maintained that IFP members had bought arms directly from Renamo. In 1991, an extensive and sophisticated arms network was exposed, supplying IFP members in the Johannesburg area with AK- 47s, shotguns, Makarov pistols and scorpion sub-machine-guns (Weekly Mail, 7(15), 1991). The source of weapons was cited as ex-Mozambican soldiers and distribution was co-ordinated by an Inkatha supporter, with strong links to hostel residents.

Hostels housing mainly IFP-supporting migrants from Natal were crucial elements in the political violence on the Reef. The violence took various forms, including terrorist attacks on train and taxi commuters, and the ethnic cleansing of hostels as non-Zulu migrants were driven out. The hostels were used as launching pads to attack surrounding township residents and ANC supporters. Antagonism was rooted in the social construction of different political, ethnic and class identities.

In this ‘low-intensity conflict’ many migrant workers defined the conflict as a ‘war’ and several hostels became armouries. In a recent study of the relation between hostels and political violence on the Reef, all 31 residents of Meadowlands hostel interviewed mentioned one or more of the following weapons as frequently used by hostel residents against township inhabitants: AK-47s, R5, R4 and R1 rifles, pistols, shotguns, knives, axes, spears, knobkieries, sharpened iron poles, stocks, tomahawks, pangas and machetes (Xeketwane, 1995). In the same study, a third of the residents interviewed from Merafe hostel maintained that many of the weapons used in political violence were manufactured in the hostel by inmates. These handmade guns fired various objects, including nails and conventional ammunition (Xeketwane, 1995). Hostels became not only armouries where weapons were stored, but factories where they were manufactured.

While the apartheid regime, the United States, Eastern bloc countries, China and Cuba were the major sources of supply of light weapons throughout the region during armed conflict, the lack of effective post-conflict disarmament is behind much of the current proliferation of weaponry.

Ineffective disarmament in post-conflict peace building

The assembly, audit, control and disposal of weapons were an important feature of most comprehensive peace settlements in the region. While the United Nations played an important role, disarmament has not been totally effective. In Mozambique, Namibia, Zimbabwe and South Africa the cessation of armed conflict led to the creation of new national defence forces which integrated previously antagonistic guerrilla and conventional armed formations. The process of integration theoretically involved disarming rival armies and collecting stocks of weapons and ammunition.

In South Africa, this was particularly complicated as it involved the integration of seven different armed formations into a single, legitimate and representative defence force. The most significant of these armed formations were MK, the SADF and the four homeland armies. During this process there was meant to be full disclosure of arms caches established in South Africa by MK. However, a ‘leaking’ of weapons from MK and homeland military arsenals has occurred since 1990, and has undoubtedly contributed to criminal activity. MK arms caches were only cleared by the new defence force early in 1994. This was because, in South Africa, as in the course of peace negotiations in other countries, weapons and equipment have been held back for a variety of individual and political purposes, including the desire for an insurance policy if peace negotiations failed, or to maintain a material base for future political bargaining.

In several countries there was weak control over the former guerrilla armies and their weaponry in the lead-up to the integration process. ‘Many armouries and caches established prior to independence were not claimed during the post-election period and either lay dormant (and were incrementally reclaimed for various uses) or “leaked” into civil society for use in a range of criminal and political activities’ (Williams, 1995:2). Williams cites the following examples:

• The weapons depots in central and southern Mozambique were established by the Zimbabwe African National Liberation Army (ZANLA) during the liberation struggle in Zimbabwe. In 1980, with the advent of independence in Zimbabwe, ‘many of these armouries and caches were left in the rear bases as ZANLA combatants returned home to be integrated into the new Zimbabwe National Defence Force. These caches were to be used, variously, by Frelimo, Renamo and commercial arms smugglers, depending on access and identification during the 1980s and the early 1990s’ (Williams, 1995:3).

• The arms caches established within South Africa by MK included AK-47s, pistols, land-mines, hand grenades and limpet mines. Williams gives particular weight to the weaponry smuggled in during Operation Vula. This was an ANC initiative carried out between 1988 and 1990 to establish MK operatives with access to weaponry inside the country to promote the armed struggle against the apartheid regime. Williams estimates this weaponry to be in the region of 20 tons, and this was confirmed by another source from the state intelligence service. Williams maintains that ‘the mass repatriation of guerrillas during the 1990/5 period, the inadequate accounting of MK inventories within the country after 1990, and the use of weapons by besieged ANC communities during the political violence of the 1991-4 period, saw the gradual and uncontrolled decimation of these caches inside South Africa’ (Williams, 1995:3).

Since 1994, there have been frequent press reports of arms caches discovered near the Swaziland, Lesotho and Mozambican borders. During 1992,25 arms caches were uncovered by the police (SAIRR, 1994:301). In 1993, a cache of about 10000 rounds of AK-47 ammunition was discovered near the Lesotho border. A police source said they suspected the ammunition was hidden years ago by the Lesotho Liberation Army (Citizen, 17 June 1993). Arms continued to be smuggled into South Africa after the cessation of hostilities by MK. In January 1993, police discovered arms, ammunition and explosives hidden in the false bottom of a car allegedly belonging to a member of the ANC on the Transvaal-Swaziland border. Five alleged members of MK were arrested in the same month in connection with arms smuggling (SAIRR, 1994:300).

• An SAP source estimated that there were some 1,5 million AK-47s in existence in Mozambique in 1993, remnants of the war (Citizen, 16 July 1993). According to a number of informants, the integration and demobilisation process in Mozambique involved substantial weaponry leakage. Hidden arms caches were of major concern to the Cease Fire Commission as Mozambique prepared for national elections. According to one source in Mozambique, many weapons were stored in the assembly areas and not secured. (This is in contrast to Namibia where, according to Batchelor, weapons were moved from the assembly areas to a central state armoury and subject to very strict verification and control procedures.) A total of 186000 weapons were collected by the United Nations, registered and handed over to the Mozambican government. The serial numbers were supplied to the South African government. However, ‘many weapons were kept in unguarded buildings’, and there was no independent verification of the storage. According to one source, the United Nations later located some 200 undeclared arms caches. One source pointed out that 186 000 weapons are considerably more than the number required by the new national defence force, which has a force level of 12 000.

The lack of effective disarmament in the negotiated conclusion of armed conflicts has been a major problem in the region.

Inadequate control over new armed formations

The problem of light weapon proliferation is sometimes due to inadequate control of the armed forces. Only South Africa, Namibia, Botswana and Zimbabwe possess well-organised, cohesive armies.

Leakages from official armouries not only occurred during the demobilisation process, but continued after the formation of new national armies. The new Mozambican army is not strong or cohesive. ‘Low military salaries, inadequate discipline, low levels of morale, and a ready market for weapons in South Africa, has led to senior members of the Mozambican armed forces being implicated in illegal arms deals . . . the Mozambican armed forces have in the past, according to one source, “leaked like sieves.” South African members of the joint task force . . . estimate that the rate of recovery of weapons is fractional and isolate Mozambique as the major and unaudited ongoing source of arms transactions in the region’ (Williams, 1995:5).

The failure of demobilisation policies

The lack of control over new armed formations is related to the failure of demobilisation policies in Zimbabwe, Mozambique and Namibia particularly. Demobilisation rarely has involved effective social integration in the sense of restoring ex-combatants to their communities, with access to employment, supportive social networks and a culture of peace and respect for human rights. Instead, many ex-combatants throughout the region have reported a sense of marginalisation and social dislocation (Cock, 1993; Tapscott & Mulongeni, 1990; Alao, 1991).

In Mozambique, the United Nations set up 49 assembly points where Renamo and Frelimo soldiers were demobilised and disarmed. Some 90 000 soldiers were demobilised, supplied with transport to the district of their choice and 18 months’ salary ‘as a personal incentive to actively reintegrate into economic and social life’ (Republic of Mozambique, 1992:4). However, integration has been problematic given the general lack of economic opportunity in Mozambique.

Several informants reported that demobilised soldiers had sold their weapons to support their families. ‘The sale of weapons spells cash to buy transportation, food, shelter and medical equipment for those who have left war behind’ (Rana, 1995:14).

The SANDF has begun the process of reducing its present inflated force level of 100 000 to 75 000. This is planned to occur over four years and will involve cash payments based on years of military service, and low-level skills training in the Service Corps. Angola has begun the integration of the 150 000-strong government army and Unita’s 90 000-strong rebel army under United Nations supervision. It is planned that the new Angolan army will have force levels of 70 000, leaving 170 000 soldiers to be demobilised; each is scheduled to receive $2 000 per year for two years.

There is a clear need throughout the region not only for effective disarmament, but for demobilisation policies which provide for the effective social integration of ex-combatants. Such social integration involves re-socialisation and recasting of social relations – a process which is far more complex than one-off cash payments.

In summary, the tragic case of Angola illustrates the social processes outlined above. Armed conflict in the region, ineffective disarmament in the negotiations ending these conflicts, inadequate control over the new armed forces and ineffective demobilisation policies, all provide important sources of supply for the diverse social categories which represent markets for light weapons.

In South Africa, as elsewhere, the arms market is ‘deep black’ and shades of grey: dead-of-night smuggling of undocumented weapons and covert shipment of arms licensed under false pretences. While it seems unlikely that the illegal arms market in southern Africa is controlled by some co-ordinated cabal of ‘deep black’ operators, what is striking is the scale on which it occurs and the links between illegal arms and a range of economic activities connected to trade in ivory, diamonds, teak, drugs, endangered species, rhino horn and second-hand clothing.

The outcome of this legacy is that the supply of guns is deeply embedded in the South African social and economic order. Furthermore, the indigenous arms industry was built on links with most of South Africa’s major manufacturing companies.

Legal arms marketing and manufacture

Many light weapons are locally produced. A range of domestically manufactured small arms is available commercially and includes the production of semi-automatic rifles modelled on the R4, in standard use in the South African army, and small handguns. They are an important part of South Africa’s indigenous arms industry, which is a crucial and often overlooked source of supply.

The majority of informants with licensed guns obtained them – at prices ranging up to R8 000 – from one of the over 600 licensed commercial gun dealers. Sometimes they did so on hire purchase terms of 10 per cent deposit and 12 months to pay. Most guns were imported but small arms are also locally produced. For example, the Denel subsidiary Lyttleton Engineering in Verwoerdburg produces R4 and R5 automatic assault rifles, which are developments of the AK-47. 1t also produces the Vektor small arms range which sells about 15 000 pistols a year to the civilian market. This includes the Vektor CPl 9 mm parabellum compact pistol, which was available for R2 500 in early 1996. One of the main weapons of violent crime is the 9 mm pistol.

The illegal arms market

No one has any real idea how many illegal firearms are in circulation. According to a 1993 report, ‘whites are at the head of many of the illegal arms smuggling rackets in southern Africa’ (Citizen, 21 April 1993). The state is using a combination of amnesty, reward and heavy penalties to deal with the proliferation of illegal firearms.

There are three main sources of supply of illegal arms:

• Cross-border smuggling

Smuggling of light weapons across the porous borders of Swaziland, Namibia and Mozambique into South Africa is common. According to the SAPS Centre for the Analysis and Interpretation of Crime Information, and numerous informants, the main source of illegal weapons is Mozambique. There are also reports of young women exchanging AK-47s for second-hand clothes on the Namibian border with Angola (information supplied by Karen Hansen). The exchange of guns for food by Angolans and Mozambicans is also said to be contributing to the dramatic increase in armed criminal violence in Zambia (Sunday Mail (Zambia), 20 August 1995).

Weapons are smuggled into the country by air, rail, road and foot. According to one source, smugglers use many ingenious methods, including hiding weapons in specially adapted fuel tanks. According to another source, smuggling into KwaZulu-Natal is organised as a large-scale commercial operation and involves ski-boats, sugar-cane trucks and private aircraft. This source also reported off-loading weapons on the coast.

In 1992, the police established a special task force to deal with cross-border arms trade, and in January 1995 President Mandela and President Chissano of Mozambique signed an agreement for cross-border police co-operation to find illegal weaponry. Operation Rachel was launched in June 1995, and within the first three months traced 1 164 weapons, 685 of them AK-47s (The Argus, 22 August 1995).

• Illegal imports

Illegal exports from various countries, including the United States, to South Africa in defiance of United States law and the United Nations arms embargo are another important line of supply. There have been media reports of illegal trade in weapons from the United States to South Africa, including shotguns made by United States companies, which have been used in township political violence. In May 1992, a sergeant in the KwaZulu police was arrested in connection with an arms cache which included shotguns made by a United States gun manufacturer, Mossberg & Sons of Connecticut (Weekly Mail, 18 September 1992). According to the Africa Fund, hundreds of semi-automatic pistols, revolvers, rifles, magazines and hundreds of thousands of rounds of ammunition, worth millions of dollars, left the United States but never arrived at their stated destination of Harare, Zimbabwe.

In June 1995, the former United States gun dealer Robert Mahler was sentenced to 18 months’ imprisonment for illegally shipping more than 200 guns from Oregon to South Africa. SAP reports stated that Mahler belonged to the right-wing Afrikaner extremist group, the AWB. In October 1993, South African authorities seized a container on a plot near Pretoria belonging to Mahler that held more than 220 rifles, pistols and shotguns and 46 983 rounds of ammunition (The Star, 15 June 1995).

• Leaks from state armouries and security force personnel

There is weak control over weapons issued to the SANDF and inadequate control over state armouries. The pistol which killed Chris Hani was obtained in a raid from a state armoury. According to one informant, attacks on members of the security forces are more common than is publicly admitted. The Annual Report of the Commissioner of the South African Police in 1992 registered a 10-14 per cent increase in the number of firearms stolen from the national security armoury.

Several informants maintained that security force personnel were involved in the illegal arms trade, largely for profit. There are also reports of sales of R4s and R5s by ‘high ranking former SADF officers’, who had obtained them when the SADF left Namibia and Angola. Several sources maintained that poorly paid black policemen frequently sold their semi-automatic rifles (R5s), shotguns and handguns on the black market to the highest bidder. There are also reports of Mozambican policemen selling their weapons (Sunday Times, 27 August 1995).

In 1993, two policemen were arrested in connection with a large illegal arms network (The Star, 30 September 1993). In 1995, five policemen were arrested in connection with the theft of weapons from the police training college at Koeberg on the Cape’s west coast. In all, 38 firearms, 203 magazines and more than 2 000 rounds of ammunition were taken in the theft. A police spokesman said he believed the theft had been ‘for financial gain and not for any political reasons’ (The Star, 20 July 1995).

In 1993, it was reported that a Ciskei state armoury had been broken into and R4 rifles stolen (Business Day, 13 October 1993). Several informants maintained that large numbers of G3s and ammunition issued to the KwaZulu police had entered the black market.

In 1994, an official investigation found that only 3514 of the 5634 firearms issued to police stations in the former homeland of Transkei could be accounted for. The 2 000 missing firearms included R4 rifles and various handguns (Sunday Times, 10 September 1995).

The current problem of proliferation is thus not only a legacy of the apartheid regime. In several ways, the post-apartheid state is encouraging proliferation and, paradoxically, contributing to the erosion of its own authority. It is doing so through its support for the import and manufacture of small arms, encouragement of private gun ownership in the form of liberal licensing, extensive arming of the security forces and weak control over their weaponry. Several informants maintained that security force personnel – especially poorly paid policemen – were involved in the illegal arms trade.

The consequent proliferation of the means of violence is one of the most distinctive features of contemporary South African society. The ‘diffusion’ of arms suggests their dispersion and recirculation through multiple channels to all levels of society. It follows that solutions to the problem of gun violence have to be far reaching.

Solutions to the problem of gun violence

It is increasingly clear that the proliferation of light weapons is a destabilising force throughout the world. Establishing controls will be difficult since both the legal and illegal arms trade are embedded in intricate social relations, institutions and material interests. On the supply side, governments, manufacturers and individual dealers will want to continue making enormous profits. On the demand side, there are strong economic interests, cultural meanings, identities and social practices attached to the possession of light weapons. Meaningful arms control must be part of a broad process that emphasises demilitarisation, an unravelling of social identities that legitimise violence, and a shift in social values towards peace and human rights, as well as economic development and political legitimacy. In short, arms control has to be understood as part of the transformation of social relations in post-apartheid South Africa. The culture of violence and secrecy created by apartheid has to be replaced by a respect for human rights and the transformation of key institutions – parliament and Armscor – into accountable, transparent structures.

It is also crucial to establish the legitimacy of these structures. ‘Politically, the critical issue in dealing with the use of small arms in intra-state conflicts is to bring small arms back under the authority of

the state, functioning through a democratic government which enjoys broad public support’ (Rana, 1995:18). This has particular pertinence given that one of the strategies of resistance to apartheid was a level of lawlessness to make the country ‘ungovernable’. Consequently, one of the legacies of the apartheid regime is a distrust of authority and minimal public respect for the law.

South Africa needs ‘an integrated policy of transparency, oversight and control’ (Goldring, 1994:34). But the country also needs sustainable development: probably the single most important factor stoking the demand for arms is the prevailing gross disparity in income, wealth and natural resource capital. With an unemployment rate of 34 per cent and one of the most unequal distributions of income in the world, South Africa illustrates this factor, but there are high levels of poverty and unemployment throughout the region.

None of these measures will be achieved without an indigenous demilitarisation movement that involves the kind of mass mobilisation that marked the anti-apartheid struggle. The most important solution to the current problem of light weapons proliferation in the region is social: it means creating values and social organisations to work for national consensus on the need for a comprehensive programme of demilitarisation.

As the supply of guns is socially organised, the demand for guns is socially constructed and embedded in various social practices and cultural forms. Guns are connected to various overlapping social identities, particularly those defined by gender, race, class, age, political affiliation, nationality and ethnicity. All of these are strong representations of common interests and carry powerful mobilising sentiments. They are all relational identities; they involve boundaries demarcating ‘us’ from ‘them’; they mark lines of exclusion and difference which perceptions of external threat and access to weaponry make potentially lethal.

In political terms, gun violence is ultimately about the contestation of power invested in these identities. The relation between power and violence is an ambiguous one. The power that ‘grows out of the barrel of a gun’ in Mao’s phrase is often the obverse of moral authority and political legitimacy. Paradoxically, the perpetrators of criminal violence, such as robbery and car hijacking, belong to marginalised and powerless social groups to whom guns represent the power to enforce compliance.

Any solution to the proliferation of guns has to deal with these social relations and contested identities. At present, the state has established a number of structures to come up with policy proposals to reduce the number of firearms available to the general public. But a control policy that ignores the historically and socially constructed meanings attached to firearms will be ineffective: the allegiances and identities which underlie acts of gun violence need to be altered. Thus the concern of traditional politics, Lenin’s question of ‘What is to be done?’, is inextricably bound up with the question of ‘Who am I?’

Contemporary social relations and identities have been shaped by the legacy of war in the region. The proliferation of small arms is one material legacy of war; antagonistic social identities and an ideology of militarism are part of the ideological legacy. Ex-combatants are often perceived as the direct bearers of these material and ideological legacies – they are marked by their experience of war, their training in the means of violence, lack of marketable skills and access to weaponry. Frelimo ex-combatants in impoverished Mozambique admit that ‘guns can mean food’ and ‘a way to survive’. However, ex-combatants are only part of the problem; the failure to provide for their effective social integration, in the sense of restoring them to their communities with demilitarised social identities and access to employment and supportive social networks, is only one symptom of the broader failure to create a common society and a new collective identity for South Africans. To do so requires confronting the legacy of war through an indigenous demilitarisation movement.

This exists in embryonic form in organisations such as Gun-free South Africa and Cease Fire. However, the movement is marked by a ‘social shallowness’, being extremely small, fragmented, and mainly white and middle class. This movement is demanding a shift of power and resources away from the military and is challenging militarist values and social practices – especially the notion that guns are socially acceptable.

A limited process of state demilitarisation has been under way in South Africa for several years. However, it is a complex and uneven process and has had some contradictory consequences. For example, the increasing emphasis on arms exports is partly a response to the reduction in domestic defence procurement. Paradoxically, state demilitarisation is contributing to global militarisation.

Policy solutions that are overtly statist ignore the plurality of institutions and social relations with which the state must engage. The process of demilitarisation needs to go beyond the restructuring of state institutions and encompass a much broader project of social transformation. This involves de-linking guns and masculinity, a challenge posed by Virginia Woolf when she asked, ‘How can we alter the crest and spur of the fighting cock?’ Today, many feminists are insisting on the need to create new gender identities. This cannot be achieved through equal rights feminism – a stunted feminism that is focused on specific issues such as women’s access to combat roles. Nor can it be achieved through radical feminism which focuses narrowly on domestic violence against women. It requires a transformative feminism which confronts the connections between private and public spheres, questions the understanding of ‘difference’ and challenges the relation of gender identities to violence and power.

However, feminism is somewhat ‘contaminated’ as a ‘Western’ notion, and the social mobilisation necessary to create a demilitarisation movement is particularly difficult in South Africa. During the process of ‘elite-pacting’ which marked South Africa’s transition from authoritarian rule, an alliance of militarists from the various armed formations, but particularly the SADF and MK, was firmly established. No strong grass roots anti-militarist movement emerged during the 1990-4 period to challenge this alliance. A militaristic nationalism has been inherited in South Africa which links prestige in international relations to military power. Subverting this requires the erosion of current antagonistic ethnic identities and the creation of a common society bound by shared values of peace and justice, democratic participation, equality of economic opportunity and respect for cultural diversity. Recognising light weapons as an issue of human security is one step on a long road.