Introduction
This paper explores the usefulness of describing South Africa's regional relations as a system of accumulation. Following from a chapter in New South African Review 4 (Van der Merwe 2014), which focuses on the role of regional parastatals within this system, this paper will attempt to describe the historical geographical development of such a system. The system, it is argued, is embodied in the concept of a ‘government–business–media' (GBM) complex. Although this system functions at the sub-national, national, regional and international levels (in fact, the mere act of disaggregating these levels is superfluous as such a system would not discriminate), the focus of this analysis will primarily be on the regional scale. The regional scale is important because it serves to foreground the functioning of such a system across space, and how such a system typically transcends national boundaries.
The analysis draws from Harvey's (2005) reworking of Marx's primitive accumulation in his concept of ‘accumulation by dispossession’ and Gramsci's (1971) notion of hegemony. The analysis is further guided by the critical geopolitical literature (Ó Tuathail 1996; Ó Tuathail and Dalby 1998). The usefulness of Marxian and Gramscian analyses to the materialist tradition needs no explanation; however, a brief explanation concerning the suitability of the geopolitical literature would be appropriate.
Despite various advances in the global geopolitical literature – such as the rise of critical geopolitics and its focus on discourse (Ó Tuathail 1996; Ó Tuathail and Dalby 1998), an instructive shift to geo-economics (Cowen and Smith 2009; Luttwak 1990), and the development of a critical literature on power networks (see next section) – geographers and scholars in general writing about southern Africa have surprisingly not made concerted efforts to explore the relationship between South Africa and the region using the analytical tools of power networks and discourse.
Carmody's (2012) analysis of South Africa's ‘geo-governance’ is probably the strongest example of the value of using these analytical tools to explain the relationship between South Africa and the region. Drawing from Paquet (1996, 3), Carmody (2012, 224) describes geo-governance as ‘the ways in which effective coordination is affected [between private, public and civic actors] in a world where resources, knowledge, and power are distributed through geographical space’. Recognising the need to look beyond a state-centric discourse, Carmody's analysis reminds us that a deeper exploration of the power relations of the state and its relationships with business, media and other civic actors is long overdue in the context of South Africa's relations in the region. Simon (1996) also contributes theoretically and methodologically to understanding South Africa's regional relations using the tools of power networks and discourse.
Geographer and political economist Patrick Bond (2004) manages to pull together the strands in government, business and media, and he does so in a manner which binds with global patterns of power relations. He does this by using the concept of sub-imperialism. Bond focuses on discourse to the extent that such sub-imperial powers use discourse to maintain their dominance and accumulate capital. Bond's analysis of South Africa in the region further underlines the importance of relating power networks such as a Pretoria–Johannesburg state–capitalist nexus to South Africa's sub-imperialism. However, in a useful critique of Bond's work, Samson (2009) suggests that Bond ‘does not explicitly elaborate a theorization of subimperialism’ and whilst there is some discussion concerning the continuities of South Africa's sub-imperialism over time, ‘Bond presents an ahistorical, unchanging conceptualisation of subimperialism’ (96). A similar argument has been made by Tandon (2014). To be fair, however, Bond's (2013, 2014) recent work suggests a deeper theorising of sub-imperialism.
As these brief examples highlight, a regionally aware system of accumulation would have to account for the multiple and varied actors and spheres involved in this process, giving rise to the description of a South African power network broadly spanning the areas of government, business and media. The focus on discourse (embodied in text, talk and practice) allows for a deeper understanding of how power operates between these spheres and across space more generally. Accordingly, the historical analysis presented in this paper will be prefaced with a discussion on some of the main sources of power in society which seek to manage and ‘control’ capital and resources, but also the flow and content of information systems. The argument is presented by means of a rethinking of the existing literature on ‘complexes’ as a defined sub-type of power networks.
Power networks: theorising the ‘complex’
Intrinsic to the concept of power applied in this paper is the understanding that power is multi-dimensional and originates from multiple sources and actors in society. Collectively, these spheres combine to form an enduring ‘logic’ and pattern of accumulation across time and space. This concurs with the work of Foucault (1980), who, as the leading theorist on discourse, stressed the prevalence of power networks through his power–knowledge nexus.
Mann (1993) discusses the multiple locations and sources of power in society – the political, ideological, economic and military forms of power – and how they converge and intersect with one other. There is also a body of critical literature drawn from various fields focusing on elite networks spanning the political and business spheres (Brayshay, Cleary, and Selwood 2006; Cox and Rogerson 1985). A strand within this literature focuses on the transnational nature of these networks through the promotion of a hegemonic capitalist class (Carroll and Carson 2003; Robinson 2005).
The most common power network discussed by academics is undoubtedly the military–industrial complex. The concept ‘military–industrial complex’ was coined by Mills (1956) in his book The Power Elite. It was also formally named by American president Dwight Eisenhower in his 1961 farewell speech. Although the concept did attract a great deal of attention, not much of it was from geographers (Farish and Vitale 2011, 778). However, a special edition of Antipode drew together a number of geographers to focus exclusively on the ‘complex’, and is worth further discussion.
Lutz (2011), who wrote the conclusion for the special edition, identifies ‘networks’ and ‘complexes’ as being largely synonymous, or, at the very least, ‘complexes’ can be viewed as a subsidiary of larger ‘networks’. Drawing from Der Derian (2009), Lutz states that the various aspects covered by the edition ‘were … shaped by cultural hegemonies that emerged within the military-industrial-media-entertainment network … or, more fully, the military-industrial-media-entertainment-academic network’ (903). Consistent with this approach, ‘complexes’ will be viewed as a means to identify related phenomena and how their interaction entrenches core productive relationships in a given society. These patterns give rise to what can be called a theory or system of accumulation.
The usefulness of using a ‘complex’ to understand the relationships between core productive sectors in society is particularly relevant in dominant or imperial societies, especially in explaining the historical relations embedded in state and capital. Vitale (2011), who also co-edited the special edition, focuses on the Westinghouse Electric and Manufacturing Company, a significant American arms manufacturer during World War II. He describes how throughout World War II, the state and industry developed the groundwork for a complex set of relationships that would go on to form the basis of the American military–industrial complex. Part of the strategy of the military–industrial complex was soliciting the help of citizens, often through their relationship to work.
In the South African context, geographers have also developed similar terms for studying dominant productive relations between the state and the private sector. Rogerson (1990) asserts that the rise of Armscor (the South Africa state-owned arms manufacturer) as a ‘military-industrial enterprise’ was at the centre of the post-1960 growth of the arms industry in South Africa. Although South Africa underwent escalating restrictions on the production of arms, owing to the advancement of the military–industrial enterprise the country had managed to reach a point of relative self-sufficiency in its military production. Rogerson concludes that the military–industrial enterprise was to play a defining role in the economic geography of the apartheid state, favouring the development of the ‘Johannesburg–Pretoria axis’ (241).
In addition to the interwoven relationship between the state and the private sector, the role of academia, information systems and media in processes of the ‘complex’ warrants attention. Also in the special edition of Antipode, Pinkerton, Young, and Dodds (2011) build on the themes of cold war social science, civic education and propaganda by examining Project Revere. Project Revere consisted of experiments in leafleting and the power of word-of-mouth message dispersal. The various experiments were carried out in targeted American communities. Building on what Senator William Fulbright termed the ‘military-industrial-academic complex’, they discuss ‘the manner in which universities, the military and intelligence agencies became bound up with one another at the start of the Cold War … ’ (822), and therefore, partners in the creation of propaganda to both domestic and foreign audiences.
Similarly, Peet's (2002) analysis of how globally constructed hegemonic discourses ‘colonise’ alternative ones, also focuses on academic and media processes associated with ‘complexes’. Peet develops what he calls the ‘academic-institutional-media complex’ through which he describes how a neoliberal discourse – strongly influenced by global financial institutions – ‘disciplined’ the African National Congress into adopting a neoliberal policy in the 1990s after the leftist Reconstruction and Development Programme was effectively abandoned. Another well-known ‘complex’ used to describe the South African political economy is the ‘minerals-energy complex’ developed by Fine and Rustomjee (1996). They note how historically the South African economy is centred on mining (see below). More recently, an influential special edition in the journal Transformation focusing on ‘varieties of capitalism' reasserted the importance of this concept in understanding South Africa's political economy (Padayachee 2013).
The argument advanced in New South African Review 4 (Van der Merwe 2014, 361) was that ‘complexes’ should not be seen as mutually exclusive but rather complementary. Each concept highlights similar, yet slightly different, aspects of the functioning of a dominant society. However, it was further argued that, a broader, more integrated attempt at understanding South Africa's system of accumulation – as conceptualised through a ‘complex’ – could and should be made. Such a concept should attempt to ‘bring together’ the military, mining and information sectors as important yet incomplete manifestations of public–private interests operating across time and space. A more systematic treatment of the core sectors of government, business and media – as well as a focus on the temporal-spatio assumptions of the exchanges between these spheres – would shed light on a complex web of interactions between elite members spread across these spheres. The GBM complex proposes such a system.
The GBM complex is a network of state and corporate elite members who interact, collaborate or collude to facilitate capital accumulation located within a ‘fixed space’. The underlying premise of the GBM complex is that the state works in tandem with capital for purposes of accumulation over space. The state is not only an administrative and technical centre, but acts as a financial and institutional facilitator of capital accumulation. The elite members in government and business – along with their cohorts in media – are responsible for constructing and spreading discursive formations to sustain their dominance and accumulate capital. This is achieved largely through their influence over various media organisations and corporate-endowed research institutions. Class formations further reinforce these patterns of accumulation driving the interests of the dominant elite.
Geographies of the GBM complex operate through networks established between government, business and media elite members, both within the borders of South Africa, but also through the transnational flows of capital, goods and information between elite members. These networks play a crucial role in the decision-making and execution of the state; the character and location of business dealings and trade; and the role of the media in influencing citizens. Such geographies operate on trading floors, in boardrooms during high-level policy discussions and in newsrooms where editorial content is debated (Van der Merwe 2014, 362).
Early formations of the GBM complex 1652–1910: settler regimes, territorial conquest and commerce
The first successful attempt at European settlement in the Cape was made by the Dutch East India Company in 1652. The initial expansion into the African hinterland was characterised by the ‘cut and thrust’ of the relationship between the state and commerce. This territorial expansion was initially seen as expressing the Afrikaners’ wish to govern themselves, primarily through the Great Trek, which resulted in the establishment of three Boer republics, and then as an extension of British imperialism.
The discovery of diamonds in 1867 near Kimberley, and then gold in 1886 in the Witwatersrand or ‘Rand’ as it was commonly known, had a profound effect on the economic geography of southern Africa. The need to transport the raw minerals to the major ports and bring machinery from Europe to the mines sparked the rapid construction of railways. The infrastructure for a regional economy was therefore created through the burgeoning minerals trade. The city of Johannesburg soon became a regional focus of mining, finance and banking (Ally 1994; Mabin and Conradie 1987). The mining towns and cities became synonymous with the rise of an English mining elite commonly referred to as the ‘Randlords’. The more successful Randlords were Cecil Rhodes, Barney Barnato, J.B. Robinson and Alfred Beit. Despite a multiplicity of mining claims and counter-claims (some of them by Afrikaners), a deliberate strategy to monopolise ownership of the mines was devised by the Randlords.
The Randlords often had explicit political interests. Cecil Rhodes was elected governor of the Cape in 1890. Rhodes, with Beit, founded a new British colony to the north of the Transvaal named Rhodesia. Rhodes was obsessed with expansion and also founded the British South Africa Company (BSAC). The BSAC received a charter from the British government in 1899 which gave the Company effective sovereignty over Rhodesia, and was created with the extraction of minerals in mind. The BSAC was to give Rhodes control over large tracts of the sub-continent and became a way to open up new colonies in southern Africa.
Some Randlords openly supported the Transvaal government. However, those involved in deep-level mining were far less supportive owing to the government's inefficient and expensive practices relating to railways, dynamite and alcohol (Giliomee and Mbenga 2007, 198, 201). The British interests in the minerals of the Republic culminated in the Anglo-Boer War of 1899–1902, which the British won. In 1910 the former Boer republics merged with the Cape and Natal to form the Union of South Africa, thereby ensuring the continued extraction of resources by the British-based conglomerates.
Although at the time there was speculation that the British protectorates would later join the Union, this did not happen after the state created in 1910 effectively fixed the boundaries of modern South Africa. At that stage, the so-called British High Commission Territories (HCTs) were Bechuanaland (later Botswana), Basutoland (Lesotho) and Swaziland (BLS). Both Britain and South Africa envisaged that they would eventually join the Union, but with the development of apartheid after 1948 and South Africa's departure from the Commonwealth after becoming a republic in 1961, it became clear that this was politically impracticable. Southern Rhodesia voted not to join the Union in a referendum in 1922.
1910–1994: entrenchment of capital accumulation
The early settler and imperial expansion had laid the platform for the GBM complex. Yet it was in the twentieth century that the character of the GBM complex really took root. Fine and Rustomjee's (1996) contention that South Africa's political economy is premised on a minerals–energy complex (MEC) demonstrates the concentration in ownership that was a significant feature of the GBM complex. This was to stem from the monopolisation of mining. Drawing from Davies (1979, 50 quoting Frankel 1938), Fine and Rustomjee (1996, 98) contend:
Out of the 576 gold mining companies floated on the Rand during the period 1887–1932 … only 57 remained in existence in 1932 … The 57 gold mining companies in existence in 1932 were, with some minor exceptions, controlled by six finance houses or groups.
These capital axes therefore shaped South Africa's experiences in the region and drove the development of its industries which enabled it to become the dominant trade power. Subject to the ebb and flow of global accumulation throughout the twentieth century, the processes of regional accumulation were further shaped by South Africa's foreign policy, multinational corporations (MNCs), class struggle within South Africa, the specific dynamics of capitalist accumulation and a state grounded in capitalism (see Samson 2009, 98–99 and generally). The emergent South African bourgeoisie consciously sought to integrate the regional economy through processes of uneven and combined development.
Control by these capital axes extended into media. The English press – largely owned by Anglo-American until the late 1980s – gave momentum to a new class associated with British imperialism. Rembrandt and Sanlam controlled the Afrikaans press and, with the state broadcaster after 1960, helped to consolidate National Party (NP) influence and supported Afrikaner capital accumulation (Tomaselli 1997, 23, 30). Discussing hegemony and the press in relation to the political economy of apartheid, Tomaselli, Tomaselli, and Muller (2000, 42) note that ‘mass media articulate[d] the ideology of apartheid which had the effect of naturalizing the class-assigned roles that define[d] the relationship between capital and labour.’ Mass media thereby assisted (with varying levels of success) in processes of accumulation by dispossession within a racialised, capitalist system.
Segregation and the first resistance to the GBM complex
A significant feature of South Africa during the previous century was the capital and class struggles, typically forming along racial, ethnic and gender lines. Race, ethnicity and gender were used instrumentally by the GBM complex in the mobilisation of capital and resources for the advancement of one group above another (Terreblanche 2002). Given the political and economic geography of a racially segregated society, Africans were to live in segregated areas close to large industries. This laid the foundations for a highly profitable cycle for the GBM complex. Several acts were passed to gain control of the flow of labour and ownership of land (Legassick 1974a, 1974b; Wolpe 1972).
The initial period of reconstruction was therefore about maximising the spatial and commercial gains made by the GBM complex in the late 1800s and early 1900s. Butler (2009, 14) explains: ‘During post-unification reconstruction, labour supply remained at the heart of government policy. Native reserves, “influx controls”, and compliant “traditional authorities”, became reliable policy instruments for a state supporting mining and agricultural capital.’ Drawing from Marx, Marais (1998, 10) talks about ‘reserve armies of labour’, further noting: ‘In essence the “native reserves” (and later the homelands) would subsidise capitalist growth in [SA]’ (see ‘homelands’ or ‘bantustans’ below). These ‘reserve armies of labour’ were not only for mining and agriculture, but also for manufacturing, which only really started to grow after 1933. Owing to the consolidation and centralisation of state capital as a result of unionisation, the GBM complex was increasingly able to broaden and diversify its activity in the economy. Marais (1998, 11) observes: ‘By the 1940s, manufacturing industry was growing in earnest, thanks to lavish support in the form of protective measures and tariffs, subsidies and major infrastructure projects that facilitated its growth. The shift from artisanal to mechanized production was rapid … .’ World War II increased demand in the North and aided in the rapid expansion of manufacturing in South Africa.
Besides a large controlled labour force and, in order for the GBM complex to thrive, a significant proletarianisation of Africans had to occur. The concentration of ownership in the GBM complex steadily spread from mining and agriculture, to include manufacturing and finance. This meant that a greater skills base amongst Africans was required to sustain these industries as well as a larger African consumer base. This went hand-in-hand with the increased urbanisation of the African population and an almost total wiping-out of a self-sustained African peasantry (Bundy 1979; Morris 1977). However, the rise of a working class also led to union action, strikes and other forms of resistance towards the GBM complex, not only from Africans but also whites. Owing to the gradual spread of large capital into agriculture, further class stratifications along ethnic lines were to emerge amongst the whites. Thousands of Afrikaner farmers were forced to move to the cities in the face of economic depression. This tension between the GBM complex and the white working class was particularly prevalent in the mines. This culminated in the 1922 Rand Revolt, which was the result of dissatisfaction amongst the Afrikaners and English whites with capitalism, as well as the introduction of cheap African labour. It was suppressed by the military and cost 200 lives (Butler 2009, 14).
Integration of the regional economy
After union in 1910, the GBM complex began to focus on the region not only as a ‘fixed space’ to absorb the country's capital crises, but also as a growing source of minerals and labour. Thus another highly profitable cycle for the GBM complex was to be established, premised on asymmetrical regional development and expanded rounds of accumulation (Martin 1986, 1990). Industrial development in South Africa was to be subsidised not only by the exploitation of the local black labour force, but also at the sub-continent level through the migrant labour system. In 1905, 85.4% of the South Africa workforce was non-South African (Libby 1987, 38–39). This figure peaked again between 1970 and 1974 when it ranged from 75.8% to 79.0%. It was cheaper for large industries to attract workers from the region than to employ the locals. In addition, at that stage most regional economies were still in a pre-capitalist state of African peasantry and subsistence economies.
MNCs hoped that the other British territories and colonies would become ‘another Rand’. British-based MNCs moved to South Africa and, with the ones that were developed locally, started moving into the region. MNCs sought to make regional expansion part of their broader strategy, effectively viewing South Africa as a gateway to the region (Seidmann and Makgetla 1980). The largest corporation to operate in South Africa with interests in the region was the Anglo-American Corporation (AAC) of South Africa. AAC was formed in 1917. In the latter part of the twentieth century, AAC was to grow to control 56% of the South African economy (Cox and Rogerson 1985, 223). The AAC has significant interests in Botswana, Zambia and Zimbabwe. De Beers, which was founded by Rhodes and later acquired by AAC, is in Namibia and Botswana and also has strong interests in Angola and Tanzania (Hanlon 1987, 67–68).
AAC developed solid relations with governments throughout southern Africa and was a politically active corporation. AAC chairmen were generally regarded at the head-of-state level and they would frequently meet with regional presidents. AAC sometimes appeared to support (at least indirectly) the apartheid government (Hanlon 1987, 14–15). At other times AAC played a strong role in opposition politics. They were closely linked to the Progressive Party and its successor, the Progressive Federal Party, in the 1980s, when the economic absurdities of apartheid were increasingly apparent (Hackland 1980, 11–12).
The other English monopoly group which had significant interests in the region was South Africa Mutual (or Old Mutual), the largest life insurance company in South Africa and in Botswana, Lesotho, Swaziland, Malawi and Zimbabwe. Towards the end of the previous century, after several amalgamations, SA Mutual gained control of the region's international shipping (Hanlon 1987, 223). The amalgamations made it a force in regional tourism through Sun International, which controls casinos and hotels in Swaziland, Lesotho, Botswana and Mauritius as well as South Africa's bantustans (Hanlon 1987).
The GBM complex created various ways to keep the region, and especially the HCTs, as a ‘fixed space’ in which to invest its surpluses and from which to extract minerals. Although the HCTs were not formally part of a South African federation, by the early 1900s they were locked into a customs union without proper consultation. The Southern African Customs Union (SACU) came into being after the signing of the SACU Agreement in 1910. SACU played a large role in perpetuating the dependencies of these states on South Africa. The major problem in SACU was the Revenue Sharing Formula, which was to only be reviewed three times in a century. The 1910 Agreement resulted in South Africa receiving 98.7% of the joint revenue, whilst the HCTs received collectively 1.3% (Gibb 1997, 75, referring to Cattaneo 1990). This was to remain fixed until 1969 despite fluctuations in trade.
South West Africa attained independence in 1990, at which stage it became Namibia. Although it had been effectively a member of SACU since 1915, it formally joined SACU upon independence and united with the BLS states in arguing that SACU countered their developmental efforts (Gibb 1997, 77). Before that, its trade and foreign exchange statistics were generally incorporated into South Africa's as it was directly controlled by South Africa (Ibid., referring to Simon 1991, 385). During World War I the British government asked South Africa to take South West Africa from Germany. After the war, under the terms of Article 22 of the Covenant of the League of Nations, South Africa was given the authority to administer the country.
The rise of Afrikaner capital
After the founding of apartheid in 1948, the NP government promoted the advancement of the Afrikaner people and Afrikaners were given preference in the economy. In the 1940s a small number of Afrikaners occupied the top white-collar positions and the majority were working class. This was typically due to lower levels of education and fewer opportunities. However, as a result of job reservation, Afrikaners started to move into white-collar employment in the cities, with a slightly diminished focus on agriculture. This was to lead to wholesale changes in the class structure as they obtained far greater social mobility (see O'Meara 1983).
There was also state support for aspirant Afrikaner capitalists, farmers and small-business owners. The main beneficiaries of state-backed empowerment of Afrikaner capital were represented by four important groups, three of which (Volkskas, Sanlam, Rembrandt) were centralised and involved themselves in a number of diverse and overlapping activities (Fine and Rustomjee 1996, 149). In the 1950s the Afrikaners were almost entirely absent from mining, but a significant presence had been established in commerce. With the exception of financial institutions, they could not compete with the strength of English capital. However, the apartheid government was to aid in strengthening Afrikaner financial institutions (148–150). Through the use of government contracts and bank accounts, the apartheid government was able to boost Afrikaner business (Hanlon 1987, 11).
A disjuncture developed within the GBM complex between English and Afrikaner capital. Historically the mining companies such as AAC were English. But by the 1960s Afrikaners played an important part in mining. This was initially focused on the state-backed extraction of coal through the parastatal Eskom, but then spread to include other types of mining (Fine and Rustomjee 1996, 160). The tension between English and Afrikaner capital started to diminish in the 1960s, leading to even stronger concentration in ownership. The ‘interpenetration’ of capital was largely due to the withdrawal of foreign capital after the 1960 Sharpeville massacre (Fine and Rustomjee 1996).
A powerful ‘clique’ of Afrikaner state and English and Afrikaner corporate elite members became dominant within the GBM for many decades. This rather narrow, ethnically and racially defined strand within the GBM complex, led by a conservative NP elite, was to forge a particularly close relationship between the government and business. Increasingly, the focus was on embedding their relationship in the region. Hanlon (1987, 68) notes that of the three Afrikaner monopoly groups, Sanlam was the most important regionally. Sanlam had close links with the enlightened wing of the NP and its regional interests included mining, retail and construction, amongst others.
Entrenchment of capital accumulation: state-owned enterprises
The relationship between the state and private capital was strengthened through the state's investments in sectors related to mining, such as energy and transport. This allowed the GBM complex to sustain the conditions for monopolistic control of the region. Eskom was therefore founded in 1928 and mandated to sell low-cost energy to South Africa's enterprises. An efficient and low-cost transport system was also vital to the competitiveness of South Africa's MNCs. Accordingly, the South African Transport Services, the predecessor to Transnet, was to develop a monopoly of regional transport (and therefore trade) covering railways, harbours, airlines and some road transport (Van der Merwe 2014, 365).
The monopolistic control exercised by the GBM complex was further apparent in respect of basic services and goods, such as oil, coal, iron, steel, water and later telecommunications. Agencies such as Sasol (oil and coal), ISCOR (iron and steel) and vast water schemes, such as the Lesotho Highlands Water Scheme, were created to extract such resources from the neighbouring countries. The state industrial corporations were instrumental in lowering the costs of production for South Africa's MNCs, thereby helping them to establish and retain regional markets (Van der Merwe 2014, 365–366).
Apartheid's early regional policies
For many decades, the GBM complex had been protected by what was often termed a cordon sanitaire of European-controlled states immediately to the north. These states provided a ‘protective’ layer from the threat of African nationalism starting in the 1960s. The GBM complex had also sought to use MNCs in maintaining this buffer zone, by trying to increase their leverage over these states and thereby hoping to restrain liberation movements in these territories (Libby 1987, 63–64). Of particular interest to the GBM complex was securing its immediate power base, the so-called British High Commission territories. These territories remained under British control until their independence in the 1960s. The apartheid government sought to reduce them to the status of its bantustans. However, when it became clear that in a time of decolonisation Britain would not accede to these demands, the Verwoerd government suggested in 1963 the establishment of a ‘common market/commonwealth’ in southern Africa (Davies and O'Meara 1990, 183). This initiative failed. Yet increasingly close links were forged, primarily as an attempt to install governments that would not be critical of Pretoria (184). In the 1960s, the newly independent BLS states chose to remain part of SACU. The 1969 SACU Agreement resulted in a gradual increase in the amount of revenue going to the BLS states, particularly after minor renegotiations in the 1970s (Gibb 1997, 76). However, South Africa's complete domination remained uncontested.
Having secured its ‘fixed space’ and having thwarted the threat of decolonisation in the immediate region, the GBM complex sought to forge cooperative links with governments further north. This was also in response to increasing criticism from the West. The attempts at a more cooperative foreign policy were manifested in Prime Minister B.J. Vorster's ‘outward-looking’ policy, founded in 1967. The policy was an attempt to build relations with African states that were prepared to ignore South Africa's domestic affairs (Nolutshungu 1975, 114). It was from 1976 onwards that some of the so-called homelands or bantustans were created by the apartheid government as nominally independent territories. The apartheid government claimed that interaction with them was equivalent to interacting with independent African states (116). The ‘outward-looking’ policy's greatest success was forging diplomatic ties with Malawi. The dialogue initiative was eventually rejected by an Organisation of African Unity resolution in 1971, but did gain some support (Davies and O'Meara 1990, 185).
Pretoria's efforts to secure support were becoming increasingly strained, especially with the Portuguese withdrawal from southern Africa in the mid 1970s leading to a crumbling of the cordon sanitaire. The GBM complex thus needed a strategy to defend its interests. This was to take the form of détente which was a diplomatic and political initiative constructed to further a ‘cooperative’ stance towards the region, but also included increasingly clandestine and military means (Davies and O'Meara 1990, 185). The period of détente signalled a broadening of the GBM complex to include, in a more deliberate and expansive fashion, the information and intelligence agencies, and the media. The GBM complex started placing greater emphasis on the importance of information and intelligence agencies in ‘manipulating’ public opinion (Pfister 2005, 68). In the 1970s the South African Broadcasting Corporation produced a 24-hour external radio service, Channel Africa, which was funded by the Department of Foreign Affairs. This service broadcast in the three historically colonial languages as well as in three widely spread African languages. The channel covered most of sub-Saharan Africa (Teer-Tomaseli 2008, 93).
The concerns sparked after the Portuguese revolution, and the rapid decolonisation which followed, had resulted in significantly increased military spending after 1974. This led to the ill-conceived intervention of South Africa in Angola in an attempt to prevent the Marxist People's Movement for the Liberation of Angola (MPLA) from consolidating its rule. South Africa faced international condemnation and was forced to retreat in early 1976 as the MPLA received help from Cuban forces. South Africa's military intervention in Angola in 1975–76 effectively ended the policy of détente (Jasper 1988, 68). After the brutal suppression of the 1976 Soweto uprising, South Africa was almost totally isolated on the continent (Davies and O'Meara 1990, 186).
The ‘Total National Strategy’
The rising faction of securocrats in the GBM complex in conjunction with monopoly capital created the so-called ‘Total National Strategy’ in the mid 1970s in defence of ‘Western capitalist values’. This allowed the GBM complex to equate attacks on apartheid as Marxist in nature. Besides the economy, the other three branches of the total strategy involved diplomacy, social and psychological aspects, and security institutions (Hanlon 1987, 9). Given the strong collusion in these spheres, a closer and more strategic alignment between capital and the security institutions led to a militarisation of the GBM complex. When the Minister of Defence P.W. Botha became premier of South Africa in 1978, state reorganisation encompassed new decision-making structures in which the military played a key role (Davies and O'Meara 1990, 181–182). The total strategy significantly strengthened relationships within the GBM complex. The policy of the parastatal Armscor was to maximise the use of the private sector. Armscor increased its use of sub-contractors from 800 in 1978 to 3000 in 1982 (Hanlon 1987, 14).
The first part of the total strategy was largely modelled on increased ‘cooperation’ with Pretoria. It ran from the end of 1978 until mid 1980. It saw the launching of the Constellation of Southern African States (CONSAS) (Davies and O'Meara 1990, 195). CONSAS was Pretoria's attempt at forming a regional economic community supposedly encouraging ‘regional integration’. President Botha asked business to support the initiative by forming a Southern African Development Bank to fund regional infrastructure projects. The notion of ‘mutual reciprocity’ underlying CONSAS was, however, just rhetoric. By referring to ‘cooperation’, Pretoria really meant that inclusion in CONSAS would further secure the involvement of South Africa's MNCs in the countries concerned (Libby 1987, 66). It was initially envisioned that CONSAS would consist of the entire southern African region, but this was later reworked to include just the ‘moderate’ states (Jasper 1988, 82).
The CONSAS concept, however, failed before it even started. This was primarily due to the formation of the Southern African Development Coordination Conference (SADCC) instead, and the fact that ZANU-PF won the Zimbabwean election in 1980 installing a newly liberated, ‘radical’ government (Davies and O’ Meara 1990, 196). Despite its failure, the ideas behind CONSAS were crucial to strengthening the relationships within the GBM complex.
Destabilisation
The softer approach of trying to ‘co-opt’ the region failed and a harder approach known as ‘destabilisation’ was adopted. Already in the 1980s, the ‘policy think tank’ community or the ‘state intelligentsia’ provided some of the rationale and motivation for the actions of the GBM complex in the region (a possible example of this is the work commissioned or produced by the Institute for Strategic Studies of the University of Pretoria in the early 1980s). The destabilisation policy included a range of strategic economic and military interventions (either directly or through proxies) designed to debilitate or stunt the development of the region. The policy went through phases of intensified and selective destabilisation throughout the 1980s (Davies and O'Meara 1990, 198–199). This culminated in the 1984 Nkomati Accord, which was a ceasefire agreement signed between South Africa and Mozambique. South African capital benefited from the investment opportunities presented by the peace accord. Military sabotage would effectively occur parallel to the more ‘magnanimous’ acts by the GBM complex leading to a ‘two-faced’ foreign policy. This is a defining feature of the GBM complex, effectively saying one thing but doing another. South Africa was simultaneously presenting itself as regional ‘peace-maker’ and ‘stabiliser’ whilst pursuing its interventionist policies (Johnson and Martin 1989, 159). The military would destroy and debilitate societies, whilst capitalists would take the lead in recreating them. The information branches of the state and state-linked media would legitimise these actions and further create an environment conducive to capital accumulation. This is how power within the GBM complex would ‘congeal’, in a mutually beneficial and mutually reinforcing manner. The ‘technology’ between the three spheres was ‘instinctively’ driven to accumulate capital.
A major development objective behind the formation of SADCC was to lessen the region's economic dependence on South Africa. The year 1992 saw SADCC transform itself into a ‘Development Community’ and in 1994 South Africa became a member. By that stage, divisions within the GBM complex were beyond repair. English and Afrikaner capital started realising the inevitability of the political changes. This happened under steady international pressure and internal opposition, and owing to the heightened need to create an environment truly favourable for an expanded round of capital accumulation in a post-cold war era.
Conclusion
The historical development of the GBM complex was premised on a spatially and commercially expanding state controlled by a ruling English and Afrikaner oligarchy. This oligarchy's power was centred on conglomeration in the sectors of mining, manufacturing and finance. It was secured as a result of strong links with government. The conglomerates were able to exercise monopolistic control of the region on account of their command over capital, labour and production.
Although this paper has sought to build on the literature related to ‘complexes’, it has also attempted to break out of this mould. It was demonstrated how the most common power network discussed by geographers and scholars in general is the military–industrial complex. However, in this paper and elsewhere, it has been argued that such an analysis need not be fixed to the military and industrial relations when it is clear that the relationships and agents in regionally dominant or sub-imperial societies cover numerous other sectors and operate through more subtle means of persuasion and influence. A concept which encompasses other types of industries, media and branches of the state would service greater explanatory power for such societies.
The GBM complex therefore builds on concepts such as the minerals–energy complex. Whilst the MEC describes the primary site of accumulation in South Africa, the GBM complex attempts to frame South Africa's political economy as a broader system operating across time and space. By treating the sectors of government, business and media in a more wholesale manner – and by drawing attention to the temporal-spatio elements, as well as the agents acting within this system – it is hoped that the concept of the ‘GBM complex’ will open up further debates and new analyses on South Africa's political economy.