From the CIAO Atlas Map of Africa 

email icon Email this citation


From ‘Total Strategy’ to ‘Human Security’: The Making of South Africa’s Defence Policy 1990-98

Gavin Cawthra

March 1999

Copenhagen Peace Research Institute



Since the onset of negotiations to end apartheid in 1990, and more particularly since the inauguration of the first democratic government in 1994, South African defence and security policy has been fundamentally reworked. The 1990s have been a decade of profound global change, and South Africa has not been alone in remaking its defence and security policy. The end of the Cold War, the information technology revolution which has accelerated globalisation, and limited and uneven international demilitarisation have led governments around the world to interrogate their assumptions about security, and scrutinise the roles, functions and postures of defence forces. This has been accompanied and to a minor extent influenced by a small revolution in the academic discourse around security and defence, characterised by the ‘widening’ of the concepts and referents of security (Buzan 1991, Buzan, Waever & De Wilde, 1998).

The extent of security review and revision has varied tremendously. In the United States, the basic assumptions of US foreign and defence policy appear to have remained largely unquestioned, despite the ‘end of history’ as some would have it (Fukuyama 1992). In many West European countries continuity of policy is more evident than change although efforts have been made to reorient the defence function and to adopt wider and more holistic approaches to security management. More profound reviews and policy revisions have taken place in countries such as Canada and Australia, and many former Soviet-bloc states and democratising developing countries have been obliged to fundamentally re-examine their security assumptions (although in some cases, particularly in Africa, limited policy capacity has resulted in desultory policy change, or simple diminution of the defence function).

South Africa has undertaken one of the most systematic defence and security reviews in the developing world. This has led to considerable demilitarisation (a 60 per cent reduction in defence spending for example), the abandoning of nuclear deterrence, a commitment to collective and common security, and the adoption of a new framework for the management of security. The South African process is remarkable in three ways: first in the way in which it consciously sought out a new concept or paradigm for security through engagement with academic discourse; second in the close attention it paid to questions of security governance and management and to civil-military relations; and third in the way in which the process was conducted—the South African defence review was an exemplar of a transparent and consultative policy process (although it was by no means open-ended).

This article concentrates on the process of making defence policy in the 1990s in South Africa, and touches on the conceptual debates and the way in which civil-military relations were recast. The process itself is periodised into three distinct phases. It is argued that a unique conjuncture of circumstances made it possible for an unusually open defence policy process to take place which resulted in a high degree of consensus around an approach to defence and security which owes much to critical security studies. Whether this will be a permanent characteristic of defence policy making in South Africa, or is merely a product of the peculiarities of a negotiated transition to democracy, remains to be seen.

Much of this article is based on the personal experience of the author, who was involved in all three phases of the policy process and assisted in drafting many of the policy documents which emerged. Where information is not referenced it can be assumed that the author is drawing upon this experience.


Overview of the Policy Process 1990-98

During the apartheid era, particularly in the period of the premiership of P.W. Botha (as Prime Minister 1979-84, and as State President 1984-89), the policy-making process in South Africa was significantly militarised. This occurred mainly through the mechanism of the National Management System, at the apex of which was the State Security Council, and which provided military, intelligence and defence officers and officials (‘securocrats’) with considerable leverage over the policy process. A militaristic ‘Total Strategy’ became the organising principle of policy—this essentially argued that all activities of state and society had to be orchestrated through a counter-revolutionary security agenda aimed to preventing a ‘communist’ takeover.

As might be expected, the military came to dominate national security and defence policy-making during the Botha era. Without a functioning Ministry of Defence, defence policy was largely the preserve of the defence general staff, under the overall purview of the Minister of Defence, General Magnus Malan, who himself had been the Chief of the South African Defence Force (SADF). Defence policy-making was carried out in a closed and secretive environment in which little public or political consultation took place (Cawthra 1986: 26-38; Grundy 34-57).

With the advent of the premiership of F.W. de Klerk and the onset of negotiations to end apartheid, a rapid reduction in the policy influence of the military occurred and the Total Strategy framework was largely discarded. The period 1990-94 saw multi-party negotiations which culminated in the establishment of the Transitional Executive Council (TEC) which oversaw the first universal democratic elections in South Africa in April 1994. During this period, defence policy was radically challenged and reviewed, both in the negotiating forums and also in civil society. A range of non-governmental organisations, many of them newly established, became involved for the first time in defence policy-making and a window of opportunity was opened for academics and other experts who had previously been marginalised from the defence process. This was a period of considerable policy fluidity and was characterised by a fairly open-ended debate, although at a formal level defence policy was to some extent protected from wider transitions, and ‘ring-fenced’ by the main actors in the interests of stability.

The Government of National Unity, installed in May 1994 and led by the African National Congress (ANC) instituted a major policy revision soon after its establishment in May 1994, in which a new White Paper on Defence was drawn up. This resulted in the adoption of a new security paradigm—‘human security’, which replaced the old Total Strategy framework. NGOs played a crucial role in drawing up the White Paper, but it will be argued that although public consultations took place, the process was in fact the preserve of a small elite of defence experts. The parliamentary defence committees—especially the Joint Standing Committee on Defence (JSCD) drove this process, while the role of civil servants in the Ministry of Defence was minimal.

The White Paper marked a fundamental break with past policies but it provided only a basic, and very normative, framework. A new process, the Defence Review, was thus instituted to add flesh to the bones of the White Paper, and in particular to recommend a force design specifying the size, configuration and equipment of the South African National Defence Force (SANDF). This process was largely driven by civil servants in the newly established Secretariat for Defence. However, the process was more consultative than that of the White Paper and a far wider spectrum of role-players were involved. The Secretariat consulted very widely with government and non-government stakeholders and interest groups and a series of well-attended meetings, workshops and conferences were held around the country.


Defence Policy-Making During the Transition, 1990-94

The onset of negotiations to end apartheid—made possible by the unbanning of the ANC and other liberation parties at the beginning of 1990, and the subsequent release of political prisoners and ‘normalisation’ of the political process—created a substantially different climate for defence management. A conjuncture of circumstances, internationally, regionally and domestically (the end of the Cold War, the independence of Namibia, moves towards settlement of civil wars in Angola and Mozambique, and the suspension of armed struggle in South Africa) made it possible if not necessary for civilians to reassert control over the political and policy process within the apartheid state. More importantly—since it would be an error to conflate militarism with military influence (civilians who were inured in Total Strategy were more than equal to the task of militarisation)—the militaristic political paradigm of the 1980s lost its utility.

The imperative to negotiate an end to the virtual civil war which accompanied the end of apartheid, which required the participation of organisations with popular legitimacy and anti-apartheid credentials, created a new space for policy work. The defence sector was not immune to this, even though the ANC appeared to tacitly accept some aspects of the National Party’s approach, which was to ‘ring-fence’ defence from the political process as a guarantor of stability, or, viewed another way, of entrenched interests.

The ANC had paid surprisingly little attention to the transformation of defence by 1990: it’s policy approach had been limited by its insistence (whether merely rhetorically or otherwise) that its revolutionary army, Umkhonto we Sizwe (MK) would seize power and form the basis of the new post-apartheid defence force. The onset of negotiations, and the realisation by the ANC that it would inherit the existing South African Defence Force (SADF) into which MK would at best be integrated, and possibly assimilated, led to a sudden requirement for new policy options. Capacity within the ANC in this regard was quite limited and was largely based within its Military Intelligence section, headquartered in Lusaka, Zambia. Some intelligence work and analysis of the SADF had also been carried out within the anti-conscription movement, especially in its London-based exile formation, the Committee on South African War Resistance (COSAWR), and the movement has inspired some liberal and radical academics based in South Africa to start examining security issues. The task of policy analysis for the ANC thus fell largely to a somewhat odd grouping of military intelligence operatives, anti-conscription activists and a small number of activist-academics, who operated within the broad political framework established by the ANC’s military and political leadership.

The approach which the ANC adopted, and which was eventually agreed to by most parties to the transition, had the following main features:

A small group of intellectuals, drawn mostly from MK Military Intelligence and the anti-conscription movement, framed much ANC policy in regard to these issues by articulating a new security paradigm, derived from the so-called ‘critical security studies’ which rose to prominence after the end of the Cold War. This conceptual framework was set out very clearly in the ANC’s 1992 policy framework (which essentially became its election manifesto), Ready to Govern (ANC, 1992: 71-73). In this document, the concept of security was broadened to include social, political and environmental issues and it was argued that the security of individuals and communities, and not just the state, should be considered. This turn away from military concepts of security is exemplified by the ‘human security’ approach which was being popularised at the time by the UN Development Programme (UNDP).

Some of the ANC’s policy groundwork was carried out by the Military Research Group (MRG), an ANC-aligned ‘think tank’ on defence policy, whose dozen or so members played an important role, individually and collectively, by workshopping policy options in the context of emerging international debates and then promoting these options within ANC structures. The ANC’s negotiators took many of these positions into the conference chambers. They were inevitably diluted during the process of negotiation, but is nevertheless possible to trace the passage of many policy formulations (often word-for-word) from the MRG, through ANC conferences, and then re-emerging as policy outputs from the multiparty forums and later from government. 1 A few other NGOs also played a role in policy formulation during this period—for example the Ceasefire Campaign (which emerged from the anti-conscription movement) promoted demilitarisation and some of its less radical proposals were taken up in multi-party forums. Another NGO, the Institute for Defence Policy (IDP, now the Institute for Security Studies) also soon began to influence defence and security debates through policy-oriented research.

During this period ‘forums’ became an important feature of the South African policy process. The old government structures clearly lacked legitimacy and in many cases could not catch up with current discourse, while policy capacity within the liberation movements was limited. Forums opened up the policy process and also facilitated the participation of NGOs and community-based organisations (CBOs): in some forums policy-oriented NGOs played a critical role and exploited this window of opportunity to make significant impacts on the shape of future government policy (Shubane & Shaw 1993: 12-13). The MRG—and to an increasing extent the IDP—also sought to do this. However, the defence policy process was sometimes of necessity less open in nature (especially in regard to intelligence and operational command) than other policy processes and, as explained below, there was tacit agreement during the multi-party negotiations that it was important to keep the defence force ‘on side’ by not threatening its institutional coherence. This inevitably constrained policy interventions.

During the debate over the nature of civil control over defence, these two NGOs, as well as a few academics who had remained abreast of international practice, put forward proposals not merely for the principles which should determine civil control but also the organisational structures and frameworks. This was a typical ‘forum’ process, consisting as it did of representation from government, the liberation movements, political parties, NGOs and experts. As a result of these deliberations, a decision was taken to establish a Defence Secretariat, which was to play a prominent role in later policy-making (JMCC 1994a).

The debates over civil control, as well as most major debates on integration, force design, operational policy and threat analysis (the classic basis for planning military strategy), were carried out under the aegis of the Joint Military Co-Ordinating Committee (JMCC), a sub-committee of the Transitional Executive Council (TEC) which was established in October 1993 in the lead-up to the national elections.

While NGOs had influence over some aspects of the JMCC’s work, key issues—notably threat analysis and force design—were controlled by the former SADF. In part this was because these issues had been ‘ring-fenced’ and left until the last moment: it was only in January 1994, three months before the elections, that the JMCC began its strategic planning process. The JMCC relied heavily on SADF input and on the SADF’s planning approach: the same was true in the debate over integration, when the JMCC decided to rely on SADF facilities and processes (JMCC 1994b). As a result, the policies which emerged were strongly influenced by the status quo. It should also be noted that the participation of NGOs did not entail the ‘massification’ of the policy process: in general, the NGOs involved were themselves elite groups, small organisations which were expert-based. The ‘forums’ arguably represented a corporatist form of political decision-making, of ‘elite pacting’ in the transition to democracy, rather than a democratisation of the policy process.

The policy process in defence in 1990-94 was thus characterised by complex and sometimes contradictory processes. On the one hand, the ‘policy vacuum’ left by the collapse of Total Strategy and the discrediting of the existing security establishment allowed a small group of visionary activist academics based in NGOs or in the ANC to make a considerable impact on policy, particularly in the construction of macro-policy frameworks. These individuals could perhaps be described as ‘organic intellectuals’ in the Gramscian sense; it was certainly a particularly Gramscian ‘conjuncture’ that allowed them such influence (Gramsci 1971: 6, 12). However, with the exception of the process relating to the establishment of the Secretariat for Defence and the integration process, much of this policy making remained at the macro-level, and was very normative in nature: a declaration of intent rather than a framework for implementation. ‘Harder’ policy issues remained largely the preserve of the old bureaucracy and were more difficult for outsiders to penetrate: this pattern persisted in the next phases, as we will see. Furthermore, as noted above, democratisation of the process was limited as it remained the preserve of experts and elites.


The White Paper on Defence 1994-96

The inauguration of the Government of National (GNU) on 15 May 1994 radically changed the policy environment. Multi-party forums were no longer strictly necessary (although some continued, notably the national labour forum) as a legitimate government had been elected. Furthermore, the new parliament had high hopes of driving policy processes in a hands-on way.

A flurry of White Papers were initiated as the new ministers moved with varying degrees of urgency to establish policy frameworks and justify their departmental budgetary demands—the latter was an important factor in the decision by the Minister of Defence, Joe Modise, to initiate a White Paper on Defence. In theory, the newly established Secretariat for Defence, headed by the Defence Secretary, Pierre Steyn, should have been responsible for drafting the White Paper, as policy became one of the key functions of the new civil authority. In practice, however, there was little capacity in the Secretariat and responsibility for drafting the White Paper was devolved to consultants. The IDP wrote an initial draft which was almost immediately rejected by the Minister; subsequently the Centre for Conflict Resolution (CCR) at the University of Cape Town, led by Laurie Nathan, was appointed as the principal drafter.

Nathan, an activist academic and a member of the MRG, worked closely with the Joint Standing Committee on Defence (JSCD), a joint Senate/House of Assembly committee which had been set up through the 1994 Constitution as result of the ANC’s concerns about civil control over defence. The JSCD was very active during the White Paper process and insisted on ratifying the draft line-by-line. Nathan also worked closely with the Secretariat for Defence and the South African National Defence Force (SANDF).

The process became much more consultative after the first draft was published for public comment on 21 June 1995. The first draft was highly normative and its provenance in the MRG and CCR was evident. The draft had discussed through a working committee and with the Minister and Deputy Minister. Many written submissions on the first draft were received from the public, from NGOs, from academics and analysts, from companies involved in defence production and also from the defence arms of service and political. A second draft which incorporated some of these submissions then went to the JSCD which went through it paragraph by paragraph with the aim of achieving multi-party support for the document. The JSCD was unable to resolve a small number of issues which went to the Minister for decision, resulting in a final document which was presented to Cabinet on 8 May 1996, and subsequently approved by Parliament with multi-party support. This allowed the Minister to claim that the White Paper represented ‘a national consensus on defence policy’ (Minister of Defence 1996: 1).

While ‘consensus’ might be too strong a word, the White Paper did represent a remarkable compromise, essentially between those promoting demilitarisation and disarmament (and the institutional interests which naturally supported a strong defence force—the SANDF and the defence industry. This agreement was made possible in part by paying strict attention to the policy hierarchy: something approaching a national consensus had been achieved around the inter-departmental policy framework of the Reconstruction and Development Programme, and also around the Constitution, and the White Paper was careful to situate defence policy within this hierarchy.(Minister of Defence 1996: 5).

Historically, at least in the realist tradition, defence and foreign relations have often been regarded as ‘high politics’ and policies set in these fields have framed domestic and development policies: this was certainly the case during the Total Strategy period in South Africa. The new hierarchy—defence as a sub-set of socio-economic development policy—encapsulated the paradigm shift which had taken place. ‘The Reconstruction and Development Programme stands at the pinnacle of national policy and, consequently, defence policy,’ the White Paper proclaimed. It went on to elaborate the ‘widened’ agenda of critical security studies, and the ‘human security’ concept of the UNDP:

In the new South Africa national security is no longer viewed as a predominantly military and police problem. It has been broadened to incorporate political, economic, social and environmental matters. At the heart of this new approach is a paramount concern with the security of people.

Security is an all-encompassing condition in which the individual citizens live in freedom, peace and safety; participate fully in the process of governance; enjoy the protection of fundamental rights; have access to resources and the basis necessities of life; and inhabit an environment which is not detrimental to their health and well-being (Minister of Defence 1996: 5-6).

However, while the White Paper adopted a broad definition of security, it abandoned the approach of critical security studies when it came to spelling out the roles and functions of the defence force, instead reverting to the classical realist formula of international relations and traditional security studies:

The SANDF [South African National Defence Force] may be employed in a range of secondary roles as prescribed by law, but its primary and essential function is service in defence of South Africa, for the protection of its sovereignty and territorial integrity (Minister of Defence 1996: 6).

This formulation was motivated by a desire to delimit as narrowly as possible the role of the military. During the Botha era the defence force had become involved in a range of ‘civic action’ tasks in support of the civil authority and this was widely as seen as contributing to the politicisation of the military and the militarisation of politics. There was a great fear on the part of those seeking to widen the security agenda that what Buzan and Waever (1998) now call ‘securatization’ would take place again if the logic of the argument was followed through in terms of defining the roles of security institutions. South Africa’s intelligence agencies, for example, were initially very excited by critical security studies, as they saw it as a way of broadening their purview and legitimising an expanded role for themselves (after all, if security concepts are applied to economic, social and environmental issues, and to individuals and communities, then surely the intelligence agencies, the police and even the military have a legitimate claim to activity on these terrains?).

This ‘solution’ adopted by the Defence White Paper to this conceptual conundrum was unsatisfactory and led to inherently contradictory policies. On the one hand, the White Paper talked of ‘a compelling need to reallocate state resources to the RDP.... [and to] rationalise the SANDF and contain military spending’, while on the other it called for a defence force which was balanced, modern and technologically advanced in order to carry out its primary task of conventional defence. This contradiction became evident when it came to designing the new defence force and establishing what equipment it needed (see below), and was eventually to lead to a decision by the South African cabinet in late 1998 to authorise a R30 billion rearmament programme to equip the SANDF with new front-line fighter aircraft, submarines, corvettes, jet trainers and helicopters. Apart from the resource implications of this decision (rationalised on the basis that rearmament would be linked to inward investment deals with supplier countries), a case could be made that a ‘baroque arsenal’ (Kaldor 1983) was in the making, as the defence force was being equipped for tasks it was never likely to carry out and not being equipped for what it was actually doing—support to the police and civil authority, border patrol and peacekeeping.

The White Paper dealt more successfully with principles for civil-military relations and for the organisational architecture of defence management, which laid a firm basis for the next phase of the policy process, the Defence Review. The White Paper provided for the establishment of a Defence Secretariat, staffed ‘predominantly with civilians’ although without precluding the involvement of military officers, which would be largely responsible for the policy function. It also affirmed that parliament ‘has a range of significant powers regarding military affairs in order to assert democratic control over the armed forces and defence policy’. It further specified that:

Parliamentary and public scrutiny and debate will only be meaningful if there is sufficient transparency on military matters. A measure of secrecy will undoubtedly be necessary... However, the governing principle is ‘freedom of information’... The DoD recognises that has a positive duty to provide sufficient information to ensure adequate parliamentary and public scrutiny and debate on defence matters (Minister of Defence 1996: 10).

These principles strongly informed the Defence Review, which was driven by the newly established Secretariat for Defence.

The White Paper also set out important principles for defence in a democratic South Africa, including:

The White Paper was incomplete as a policy document. It was strongly normative in content and in some places open to various interpretations, and it avoided dealing with key ‘hard’ issues such as force design. Other issues, such as policy on the defence industry were relegated to other policy processes (see below). No targets or time frames were set and there was no indication as to how policy would be implemented, monitored and evaluated. The price of policy ‘consensus’ is often vagueness and the White Paper was no exception in this regard. It was therefore necessary for the DoD to institute another phase in the policy process.


The Defence Review 1996-98

The concept of a defence review had been widely discussed in the lead-up to democracy. Some of those involved in the policy process proposed that a review should precede the White Paper, as a national consultative process (the Canadian defence review which immediately followed the end of the Cold War, necessitating a reassessment of Canada’s entire strategic geography, was often cited as an example). However, as we have seen, the White Paper was drawn up without a national consultative process (although drafts were published for comment) and in a relatively closed policy framework in which it sought to ‘pull down’ defence policy from national policy (the RDP).

The Defence Review took place within an even more constrained ‘policy envelope’ since it was essentially aimed at generating second-order policy to implement the framework already established in the White Paper, as well as considering force design, posture and budget (Defence Review Drafting Team 1997:2). Proposals which were at variance with the White Paper (such as some advocated by the IDP, now Institute for Security Studies, which argued for a different concept of military roles and tasks) were simply not considered.

Nevertheless, the Defence Review was a remarkably transparent and consultative process and the Secretariat went to considerable lengths to ensure public participation, even flying around in military aircraft large numbers of civil society representatives (directors of NGOs, church and community leaders etc) to make sure that they could attend conferences and workshops. One reason for this was undoubtedly the fact that some of the key activist intellectuals who had taken part in the first two policy processes remained actively involved in the third phase, and some of them had by this stage moved into key positions in the ministry of defence. Many of the new members of the Defence Secretariat had been involved in anti-apartheid NGOs and maintained close links with civil society. One of the key players in the Defence Review, for example, was Rocky Williams, a former Umkhonto we Sizwe commander and an academic and NGO employee, who had moved from the MRG into the SANDF and then into the Defence Secretariat.

What the Ministry called ‘stakeholder consultation’ took a considerable amount of time and effort. As the process evolved, the Secretariat made a distinction between ‘stakeholders’ and ‘interest groups’. The former were defined as those who had an immediate material interest in the process and its outcomes (government departments, trade unions, the military, the defence industry, parliament) and the latter as those who had an interest but not necessarily a significant material stake (academics, NGOs, research institutes and civil society organisations). This allowed the Working Group to concentrate on achieving consensus with stakeholders but not necessarily with interest groups, although their views were taken into account.

The Working Group was initially restricted mainly to government departments and the JSCD, but a decision was taken at an early stage to invite a few academic consultants to sit on the group and to assist in the drafting process, and to invite ‘interest groups’ to sit on the various sub-committees (in general, each chapter of the Review was drafted by a different sub-committee) (Defence Review Drafting Team 1997a: 3). The Working Group maintained close links with JSCD but by this stage some key individuals on the committee had left parliament and it took a less detailed interest in the process than it had in the White Paper. In part this was due to increasing pressure of work, but also by this stage the initial enthusiasm of MPs for policy making had waned, and the Executive was becoming a more important locus of policy making than the parliamentary committees.

On some working groups (notably human resources), NGOs were strongly represented and argued their case vigorously, sometimes with better information or analytic skills at their disposal than the government officials or military officers present. There is certainly a case to be made that this had a direct influence on the policy outcome. For example, feminist and gay rights organisations were strongly represented on the Human Resources Working Group and played an important role in drafting policy in this chapter. The Defence Review contains some of the most progressive policy in the world on the rights of women and homosexuals within the military (Department of Defence 1998: Chapter 10).

On other working groups, NGOs were largely absent and in the more technical groups, SANDF officers and defence officials clearly had the advantage of information. This is particularly true of the working group dealing with force design. The SANDF had developed a sophisticated computer modelling programme, Project Optimum, to assist in force design, and this formed the basis for the generation of four options which were presented to the JSCD and then to parliament as a whole and the Minister for a decision. A cynic might argue that it was hardly an accident that Project Optimum identified the SANDF’s preferred force design (which included the corvettes and military aircraft it had long been agitating for) as the most cost-effective and which was therefore endorsed by parliament, as the complex calculations which costed the force design were not understandable to many outsiders The result of this process was that the recommended design for the SANDF looked very similar (if scaled-down) to what had existed during the apartheid period, and this became the rationale for the rearmament package which was approved in 1998 (Department of Defence 1998: Chapter 7).

The most notable feature of the Review process was the extent of civil society participation. The Working Group deliberately went beyond the narrow concept of NGOs as those organisations which are geared to political and policy interventions, to involve organisations such as religious and sporting groups, and to get out to the provinces—on the (probably correct) assumption that the previous policy processes around defence had catered mainly to the urban elite in Gauteng and the Western Cape. A series of provincial workshops were held in all nine provinces, which varied considerably in character. In KwaZulu-Natal, for example, the workshop was organised by the SANDF’s Part-Time Forces Council, and held at Natal Command, while in the Western Cape it was organised by CCR with a strong anti-militarist tradition. In addition, the Ministry organised four national consultative conferences to consider various drafts and reports of the Defence Review.

Ironically, the success of the Working Group in involving civil society in this policy process almost proved its undoing. Some members of the legislature became concerned that the Ministry was bypassing parliament in its dealing with civil society, and that NGOs were gaining an inordinate influence over the policy process, since they were actually drafting documents. The Ministry thus attempted to back-pedal on NGO involvement but by this time the NGOs had become accustomed to being in the policy loop and firmly resisted. It took some adroit political footwork by the deputy minister to retrieve the situation. This incident, however, pointed to an important issue in the policy process, one which had become obscured by the ‘abnormal’ conditions of the transition: the relationship between government, political society and civil society (see for example Dye 1995: 44-73).

The process in each working group differed, with a result that the product was somewhat uneven, especially in its first draft. The Human Resources and Part-Time Forces chapters, for example, contained a series of recommendations which were highlighted for the JSCD to discuss and rule on, while in other chapters little distinction was made between sections describing existing policy and those in which new policy was recommended, with a result that there was sometimes little discussion on these issues in the JSCD. With the exception of the Force Design chapter, few policy options were presented and their implications weighed up (Department of Defence 1998). Policy-makers, in this case mainly the JSCD, were simply given a ‘yea or nay’ choice or none at all.

Despite, or perhaps because of, the consultative nature of the process, some of the NGOs involved became frustrated when they were unable to achieve their policy objectives—either because these had already been ‘set in stone’ by the White Paper, because they were outnumbered and outmanoeuvred by government, or because the Department of Defence was determined to restrict the parameters of debate and reach a policy resolution within a limited time-frame. The IDP, for example, argued that the Department had ‘tried to cram into some six months [it eventually went on for two years] a process which requires several years in established democracies’ and that ‘the Review occurs without the benefit of a national security policy framework’, and it was unable to have its proposals for force design considered because these were at variance with the White Paper (Institute for Defence Policy 1996:1).

Many anti-militarist groupings also became seriously disillusioned: despite their minor victories, they felt that the Review had been presented in a triumphalist way by the DoD as a national consensus on the value of defence, and that they were in danger of being co-opted. While all the groups remained within the process until the end, many became increasingly critical, especially the Cape-based Coalition for Defence Alternatives.

Whether this was a result of the wider consultation (which brought in what might be called the ‘silent majority’ of civil society groupings who do not necessarily share the more progressive vision of politically-oriented NGOs), or whether it was a result of the increasingly technocratic nature of the process which favoured the ‘Project Optimum’ spin-doctors, or both, is a moot point. What is clear, however, is that the DoD found a successful formula for incorporating civil society into the defence policy-making process without compromising its own interests and professionalism.

The Review sets out frameworks for implementing the policies of the White Paper, as well as identifying budgetary, human resource and equipment targets and requirements. Its key features include:

Defensive posture: Doctrine is based on ‘a strategic defensive posture’ which nevertheless includes ‘capabilities to reverse the effects of offensive actions’ which means that ‘appropriate offensive capabilities will be required at operational level’. In other words, the Review makes a distinction between strategic posture, which is defensive, and operational posture, which will include an offensive capability. The latter is framed in the context of an adherence to international law, including an explicit recognition of the prohibition on acts of aggression, as well as a belief that ‘the defensive orientation of the SANDF is reinforced by Executive and Parliamentary control’ (Department of Defence 1998: Chapter 2). Core force: The concept of the SANDF’s primary function of self-defence against aggression, as set out in the White Paper, is articulated to mean the maintenance of a conventional core capability which is ‘balanced’ and ‘sustainable’ (i.e. within realistic defence budget allocations). This ‘peacetime’ force can be expanded ‘should the threat situation deteriorate significantly’. In other words, the SANDF will seek to retain most of the capacities it would need to fight a war against a similar power, but only in nucleus form. Any secondary functions as set out in the White Paper (and including internal security tasks) will be performed using these core conventional capabilities. Good intelligence coupled with the maintenance of all-round technical and training capacities will allow for fairly rapid upgrade and expansion in the event of a conventional threat (Department of Defence 1998: Chapter 3).

Force design: In the light of the doctrinal assumptions and the core force approach, a force design is arrived at, which is broken down by arms of service. The recommended option, which the SANDF insists is ‘acceptable’ in the light of budgetary restrictions but not ‘prudent’, includes a radically scaled-down full time personnel component of 22 000, bolstered by 70 000 part-timers. Most of the strength will remain in the army, which will be whittled down to a mobile division, two rapid deployment brigades (one mechanised and the other parachute), a special forces brigade and a larger part-time volunteer territorial force including 14 light infantry battalions and 12 territorial or motorised infantry battalions, as well as 183 locally based area protection units. The air force will require 16 light fighters and 32 medium fighters, as well as a reconnaissance, helicopter, support, transport and airspace control capacity, while the navy will need four submarines, four corvettes and six strike craft as well as support and smaller craft (Department of Defence 1998: Chapter 8). It was on the basis of this force design that Cabinet was persuaded by the DoDefence during the course of 1998 to authorise the estimated R30 billion rand needed for capital equipment, although payment for this will be spread over 15 years and is contingent on ‘counter-trade’ or related inward investment.

Regional and international security co-operation: The Review reiterates South Africa’s commitment to building a common security regime in Southern Africa, based on confidence- and security-building measures including multilateral treaties on disarmament (Department of Defence 1998: Chapter 4). It also commits the SANDF to regional and international peace support operations of various types, including peacemaking, peace building, peace-keeping, humanitarian relief operations and peace enforcement operations. However, the Review is very careful to spell out detailed procedures and requirements before such operations can be authorised by the Cabinet. In particular it argues that operations should be carried out under UN auspices and that peace enforcement operations in particular should have UN Security Council authorisation (Department of Defence 1998: Chapter 5). These principles were expanded upon in a White Paper on Peace Missions, drawn up largely by the Department of Foreign Affairs and approved by parliament at the end of 1998. Nevertheless, the policy framework did not prevent cabinet ministers from ignoring virtually all the principles set out in the Review when, on 22 September 1998, they authorised a SANDF military intervention into tiny Lesotho to put down a military coup. Although this was carried out at the request of the Lesotho government and nominally endorsed by SADC (and included the participation of troops from Botswana), there was no UN mandate or involvement and the military fiasco which resulted indicated that the careful planning called for in the Review before such operations were mounted had either not been done or had been done very badly (‘Countdown to Lesotho invasion by SA troops’, Business Day, Johannesburg, 23.9.98).

Secondary functions: The Review spells out the conditions under which the SANDF will carry out secondary tasks, including support to the police and essential services and various maritime tasks. It argues that some such operations may be ‘economically inefficient and politically unwise’ but are nevertheless necessary in ‘extreme situations’ (Department of Defence 1998: Chapter 7).

Organisation: The organisation and structure of the Department of Defence is set out in some detail. The organisation of the Defence Secretariat is spelt out in somewhat different terms to the ‘civilian’ structure originally envisaged (and specified in South Africa’s Constitution). Instead, an integrated civilian-military head office, along the lines of the British Ministry of Defence, is described. The ‘civilian’ Secretary for Defence (the incumbent is actually a former general) and the Chief of the SANDF divide the supervision of 19 functions between them. Functions such as policy, finance, personnel, acquisition, communication, legal services, intelligence and logistics fall under the Secretary and are usually carried out mostly by civilians, while issues related to the provision, preparation and deployment of forces reside under the Chief of the SANDF (Department of Defence 1998: Chapter 9).

Personnel: The Review, as might be expected, pays detailed attention to the issue of transforming the DoD to make it more representative (at present around 30 per cent of the Department is white, and whites predominate overwhelmingly at senior officer level). The Review also puts forward a plan to reduce the size of the SANDF from its current strength of over 90 000 personnel to about 70 000. This is clearly an issue which will be fraught with political tensions, as the relative weight of the various ‘former forces’ (the old South African Defence Force, the liberation movements, KwaZulu Self-Protection Forces associated with the Inkatha movement, and former ‘independent homeland’ forces) remains a closely contested issue. However, through careful management of the downsizing process, the SANDF hopes to achieve both a more representative force and one in which former SADF personnel are less prominently represented (they current constitute over 60 per cent of the force). The Review does not shirk from addressing the commitments to non-discrimination on the grounds of race, gender or sexual orientation which are entrenched in the South African Constitution. It sets out a range of measures to prevent discrimination on these grounds and to promote affirmative action for previously disadvantaged groups (Department of Defence 1998: Chapter 10). These policies, including the policies related to sexual orientation, were introduced with remarkably little resistance from the SANDF, but it remains to be seen what their effect will be. The Review also pays attention to the challenges involved in building up a representative part-time volunteer force to supplement the full-time component (Department of Defence: Chapter 11).

Land and environment: The Review has a chapter dealing with land and environment issues. As one of the largest users of land in the country, some of which was acquired for military use by displacing black communities, the DoD has been under considerable pressure to rationalise its land use and restore some land to previous owners. A strategy linked to down-sizing, and involving local communities and NGOs, is set out (Department of Defence 1998: Chapter 12).

Acquisition management: The apartheid-era policy of ‘self-sufficiency’ in armaments production (in reality never achieved) is finally put to rest in the Review, which argues that only ‘strategically essential capabilities’ need to be retained in-country. However, the Review defines these in rather broad and vague terms. It also argues that the SANDF should support technological innovation in the defence industry rather than production, and describes an acquisition management system in which the role of the previously all-powerful state armaments acquisition agency, Armscor, is substantially reduced and procedures for more open tendering are set out (Department of Defence 1998: Chapter 13). The Review does not deal with defence industrial issues in any depth—this is left to another policy process, the White Paper on Defence-Related Industries, which was drafted under the auspices of the Cabinet’s National Conventional Arms Control Committee during 1997 and 1998. This process is not yet completed, but preliminary indications are that it will take forward the non-autarkic approach set out in the Review. The use of the term ‘defence-related industries’ rather than the ‘defence industry’ of the Review indicates a further shift in government thinking away from regarding defence production as a special sector which requires support.

The Defence Review, approved by parliament in April 1998, is by any standards a comprehensive and well-argued policy document. There are some basic flaws: insufficient attention to implementation strategies and in particular to monitoring, evaluation, time frames and (especially) financing. The latter is critical, for this carefully constructed defence policy will come to little if more important policy priorities—the need to maintain fiscal discipline in terms of the Growth, Employment and Redistribution Strategy—take precedence. While a plan is in place to finance the capital expenditure needed for the agreed force design, this is contingent on a number of questionable assumptions, including the ability of the SANDF to slough off at least 20 000 jobs and the relative stability of the rand on international exchange markets.

It is unlikely that any future defence policy process will be quite as inclusive, consultative and transparent as the Defence Review. The need to legitimise policy through public consultation is not likely to be as pressing in future, the links between civil society and government—which depend to a large extent on personalities who share histories in the the struggle against apartheid—are gradually weakening, and as capacity in government grows it will be less reliant on academics and expert-based NGOs. Nevertheless, government—and in particular the DoD—seems still to be committed to the involvement of NGOs and academics in policy processes. For example, the drawing up of the White Paper on Defence-Related Industries was delegated to a steering committee which involved not only relevant government departments and relevant industries, but also some university departments and NGOs. Civil society organisations and trade unions were also consulted for input. This White Paper is likely to released for public discussion during 1999, in a similar way to the other policy papers discussed in this article.



In many countries, even today when the information revolution makes such considerations somewhat absurd, defence policy is conducted within a cocoon of secrecy and within the confines of a limited number of state organisations. This is particularly the case in many weak states, where regime weakness and perceptions of vulnerability lead to an over-emphasis on secrecy, even though these are probably the countries which would benefit most from an opening up of the policy process.

South Africa’s DoD has demonstrated that, in the context of a democratising state, it is possible to open defence policy to much wider scrutiny and to assimilate wider social and intellectual viewpoints, in fact, to deal with defence policy in the same way as policy on education or culture or transport. Indeed, in some ways the DoD has been even more open and consultative than other government departments. This may not have led to consensus, but it is certainly a remarkable achievement that virtually all South African political parties, and the vast majority of civil society organisations, have endorsed the White Paper and the Defence Review. The DoD, once seen as a bastion of apartheid and militarism, has undoubtedly managed to legitimise itself, although on the basis of playing a much smaller role in the state ensemble.

Whether defence policy itself is the ‘better’ for the consultative process is debatable, although many analysts would support the view that policies which are widely canvassed are likely to withstand the buffeting of political and social processes better than those which are not. One result of the active role played by academics and NGOs has certainly been that the incorporation into defence policy of the discourse of critical security studies, even if the DoD has shied off some of the more radical policy implications of such analyses.



ANC, 1992, Ready to Govern: ANC Policy Guidelines for a Democratic South Africa, Johannesburg: African National Congress.

Buzan, B., 1991, People, States and Fear: An Agenda for International Security Studies in the Post-Cold War Era, 2nd ed, Boulder: Lynne Reinner.

Buzan, B., Waever, O., & De Wilde, J., 1998, Security: A New Framework for Analysis, Boulder: Lynne Reinner.

Cawthra, G., 1986, Brutal Force: The Apartheid War Machine, London: International Defence and Aid Fund.

Cawthra, G., 1996, Securing South Africa’s Democracy: Defence, Development and Security in Transition, Basingstoke, Hampshire: Macmillan.

Defence Review Drafting Team, 1997a, ‘Defence Review: First Report on Defence Posture, Functions and Force Design’, Pretoria: Department of Defence.

Defence Review Drafting Team, 1997b, ‘Defence Review: Second Report’, Pretoria: Department of Defence.

Department of Defence, 1998, South African Defence Review, Pretoria: Department of Defence.

Dye, T.R., 1995, Understanding Public Policy, 8th ed., Englewood Cliffs: Prentice Hall.

Fukuyama, F., 1992, The End of History and the Last Man, Harmondsworth: Penguin.

Gramsci, A., 1971, Selections from the Prison Notebooks (Transl. Hoare, Q. & Smith, G.N.), London: Lawrence and Wishart.

Grundy, K.W., 1988, The Militarization of South African Politics , Oxford: Oxford University Press.

Institute for Defence Policy, 1996, ‘IDP Submission to the National Consultative Conference: The Defence Review and the Design of the Future SANDF’, Midrand: IDP.

JMCC, 1994a, ‘Post Election Control and Ministry of Defence Structure’, Pretoria: Joint Military Co-ordinating Committee, Transitional Executive Council.

JMCC, 1994b, ‘Strategic Planning: Situation Analysis’, Pretoria: Joint Military Co-ordinating Committee, Transitional Executive Council.

Kaldor, M., 1983, The Baroque Arsenal, London: Sphere.

Minister of Defence, 1996, Defence in a Democracy: White Paper on National Defence for the Republic of South Africa, Pretoria: Ministry of Defence.

Shubane, K. & Shaw, M., 1993, ‘Tomorrow’s Foundations? Forums as the Second Level of a Negotiated Transition in South Africa’, Johannesburg: Centre for Policy Studies.


List of Acronyms

ANC African National Congress
CBO Community-based organisation
CCR Centre for Conflict Resolution
COSAWR Committee on South African War Resistance
CSBM Confidence- and security-building measure
DoD Department of Defence (South Africa)
GNU Government of National Unity
IDP Institute for Defence Policy
ISS Institute for Security Studies
JMCC Joint Military Co-ordinating Committee
JSCD Joint Standing Committee on Defence (Parliament)
MK Umkhonto we Sizwe
MRG Military Research Group
NGO Non-governmental organisation
PAC Pan-Africanist Congress
RDP Reconstruction and Development Programme
SADF South African Defence Force
SANDF South African National Defence Force
TEC Transitional Executive Council
UNDP United Nations Development Programme


Contact Details for Defence-Related NGOs in South Africa

(Conflict resolution, peacekeeping training)
c/o University of Durban-Westville
Private Bag X54001
South Africa
Tel: 204 4816 or 262 9340
Fax: 204 4815 or 262 9346
E-mail: (Cedric De Koning)

(Anti-militarist and demilitarisation campaigns and research)
P O Box 31740
South Africa
Tel: 403 5315
Fax: 339 7863

Centre for Conflict Resolution
(Conflict resolution, policy-oriented defence and security research, peacebuilding, monitoring of military expenditures)
c/o University of Cape Town
Private Bag Rondebosch
Cape Town
South Africa
Tel: 22 2512
Fax: 22 2622
E-mail: (Peter Batchelor)

Defence Management Programme
(Education, training and research on civil-military relations, defence management and peacekeeping)
Graduate School of Public and Development Management
Faculty of Management
University of the Witwatersrand
2 St David’s Place
Parktown, Johannesburg
P O Box 601
Wits, 2050
South Africa
Tel: +27 11 488 5519
Fax: +27 11 484 2729

Group for Environmental Monitoring – Defence and Development Project
(Policy-oriented research and projects on defence conversion and environmental security)
P O Box 30684
South Africa
Tel: 403 7666
Fax: 403 7563

Institute for Security Studies
(Wide range of security-related policy-oriented research and programmes; extensive publications, conferences and seminars)
P O Box 4167
Halfway House
South Africa
Tel: 315 7096
Fax: 315 7099

Note: The Military Research Group is defunct, but some of its work has been taken on by the Defence Management Programme.



Note 1: For example, in 1992 the MRG devised a set of ‘Principles of Defence in a Democracy’which became ANC policy and then re-emerged virtually word-for-word in the White Paper on Defence as the main principles underpinning the DoD’s policy framework.  Back.