INTRODUCTION
Writing over 130 years ago, the eminent British historian and jurist F.W. Maitland (Reference Maitland1885: 105) remarked that, ‘The group of words, police, policy, polity, politics, politic, political, politician is a good example of delicate distinctions.’ Having quoted Maitland, Reiner (Reference Reiner2010: 32) goes on to argue that, since it is to do with the use of power, and notwithstanding the protestations of police officers to the contrary, policing is inherently and inescapably political. Again pace most police officers, the power exercised by the police on behalf of the state, and specific police powers to stop, search, arrest, detain and so on, tend to be exercised differentially reflecting social divisions of class, age, ethnicity and gender. Typically, those most likely to come into adversarial contact with the police are poor, young males from minority ethnic groups. Thus, as Reiner (Reference Reiner2010: 33) says, the police are not just political but partisan in the sense that, in an unequal society, they favour some groups at the expense of others. But it does not follow from this that policing should be the subject of political controversy, still less be partial as between competing political parties in a democratic society. Indeed, it is in this rather limited sense that Maitland's delicate distinction between politics and policing should be maintained.
This article considers the relationship between politics and policing, politicians and police chiefs in South Africa during the presidency of Jacob Zuma revealed in evidence given to a Commission of Inquiry appointed to look into the deaths of 44 people at the Marikana platinum mine in the country's North West Province in August 2012, and the Commission's findings based on that evidence. It does so by making use of the three-dimensional analysis of power provided by Steven Lukes in a study first published in 1974 but reproduced, together with two new chapters and an introduction, in a book that appeared some 30 years later (Lukes Reference Lukes2005). Following Lukes, the central argument of this article is that politicians may wield power over the police most effectively when the means by which they do so are least obvious. In other words, the politicisation of policing, its manipulation to serve the partisan interests of a politically dominant elite or party, is most profound under circumstances that require no orders, directives or instructions to be issued by politicians to police chiefs. It may even occur in the absence of any manifestation of pressure to do x, or refrain from doing y. Rather, as Lukes's analysis of political power suggests, the politicisation of policing is at its most complete when police chiefs think and behave like their political superiors, when they do not have to be told what to do, but do what is required of them because they have adopted the perspectives and priorities of the politicians.
The next section of the article begins by considering some recent literature on the relationship between politicians and the police drawing in particular on Bayley & Stenning's (Reference Bayley and Stenning2016) comparative study of police governance in six English-speaking democracies with common law traditions. It then goes on to outline the constitutional framework for governing the police in South Africa – another Anglophone, if still relatively new, democracy that draws on common law principles in a hybrid system alongside those from Roman-Dutch and customary law. A third subsection adopts a more critical, sociologically informed perspective and considers recent work on the practice of police governance in the context of the increasingly fragile political hegemony of the African National Congress (ANC). The last and longest section of the article uses some of the evidence to emerge from the Commission of Inquiry into events at Marikana that reported in 2015 to rethink the relationship between politicians and the police in South Africa. It begins with a brief chronology of events leading up to the massacre of 34 striking mineworkers on Thursday 16 August 2012. Two meetings are singled out for special attention in an effort to understand how politicians may seek to secure the compliance of the police by issuing explicit directives, by exerting more subtle forms of pressure; and/or by relying on police chiefs to share their concerns and priorities, and act according to their wishes. The lessons to be learned from Marikana, and the events leading up to the massacre, are summarised in the conclusion.
GOVERNING THE POLICE
Reflecting on recent changes in police governance in England and Wales, Wood (Reference Wood2016) draws attention to the tension between the liberal and the democratic impulses in policing in liberal democratic societies. While the former seeks to insulate the police institution from popular pressure and an overweening majoritarianism, the latter restrains inclinations towards elitism and encourages the alignment of police values and priorities with those of the public at large. In more practical terms, this tension finds expression in the competing demands of police (or constabulary) independence based on notions of professionalism and discretion, and of popular support encapsulated in the idea of policing by consent (Wood Reference Wood2016: 152). In modern representative democracies, elected governments at various levels from the local to the national are the principal, but by no means the only, vehicle for ensuring that policing is carried out in the public interest. Thus, for Bayley & Stenning (Reference Bayley and Stenning2016: 2), the fundamental problem of police governance in a democratic society is ‘to ensure that elected governments hold the police to public account while not at the same time abusing their authority’. Resolving this ‘democratic dilemma’ leads to attempts to stake out a division of responsibility between politicians who (prospectively) formulate policy and (retrospectively) hold the police to account for their actions, and police charged with implementing policy and providing ex post facto accounts of how they have used that measure of operational independence granted to them in doing so.
How this works in practice varies from place to place, time to time and issue to issue – as Bayley & Stenning (Reference Bayley and Stenning2016) go on to show. Their interviews with serving and retired police chiefs in Australia, Canada, England and Wales, India, New Zealand and the USA, reveal that attempts by politicians to direct police action are particularly likely to be resisted, and to lead to disputes, where they are concerned with personnel issues, criminal investigations and operational strategy and deployment (Bayley & Stenning Reference Bayley and Stenning2016: 96).Footnote 1 Political interest increases as the object of police attention extends from ‘threats to individuals to threats to society as a whole’; the chance of political interference is positively correlated with political sensitivity (Bayley & Stenning Reference Bayley and Stenning2016: 15). Bayley & Stenning (Reference Bayley and Stenning2016: 112) also found considerable variation in the ways in which politicians let their wishes be known. They suggest that the means of communication run along a continuum from ‘direct orders’ to ‘hints’ via ‘strong advice, suggestions, opinions [and] pointed questions’.
Instructive though this analysis is, two glosses need to be put on it before events at Marikana, and what they can tell us about the relationship between politicians and the police, are examined in detail. The first comes from the work of Otwin Marenin (Reference Marenin1982) and his notion of the relative autonomy of the police. In a classic analysis, he argues that the police may act as guardians of both general order (‘the interests of all in regularity’) and specific domination (the interests of a specific class or other dominant group within a given social formation in maintaining their position) (Marenin Reference Marenin1982: 258–259).Footnote 2 They may also act in their own interests as a distinct organisational and occupational group. For Marenin (Reference Marenin1982: 259) then: ‘[T]he fundamental question is this: what is being enforced in specific situations and for whom are the police acting as agents – general order, specific domination or their own interests?’ The argument advanced here is that, in the ‘specific situation’ that arose at Marikana – a situation involving what the police, their leaders and members of local and national elites saw as insurrectionary action (Dixon Reference Dixon2015) – the South African Police Service (SAPS) were acting, at least in part, as agents of specific domination on behalf of a dominant group extending across government, business and trade unions.Footnote 3 This is not to say that the police were not also acting to preserve general order and/or in their own interests in what was, as the following account will seek to demonstrate, an extremely complex and volatile situation. Nor is it to suggest that the scale of the violence unleashed on the strikers, and the deaths of 34 of their number, served elite interests (it did not). The emphasis here is on the decision to act against the strikers rather than the fatal consequences of that action.
If Marenin (Reference Marenin1982) directs our attention to the role of the police in a particular social formation and why politicians may try to interfere in policing, the second gloss on Bayley & Stenning's (Reference Bayley and Stenning2016) analysis of police governance adds an important new dimension to the ways in which police chiefs become aware of, or attuned to, the wishes of politicians. Steven Lukes's (Reference Lukes2005) broad, three-dimensional view of power is rarely referred to in the literature on policing, yet it provides an important antidote to the behavioural focus of more conventional analyses.Footnote 4 Lukes (Reference Lukes2005: 28, emphasis in original) is interested in ‘the many ways in which potential issues are kept out of politics, whether through the operation of social forces and institutional practices or through individuals’ decisions’.
To put the matter sharply, A may exercise power over B by getting him to do what he does not want to do, but he also exercises power over him by influencing, shaping or determining his very wants. Indeed, is it not the supreme exercise of power to get another or others to have the desires you want them to have – that is, to secure their compliance by controlling their thoughts and desires? (Lukes Reference Lukes2005: 27)
Thus, for Lukes (Reference Lukes2005: 1), power is exercised most effectively when it is least observable.Footnote 5 In the context of police governance, Lukes's insight suggests that politicians may not need to communicate their wishes explicitly in the form of orders, hints or any of the other means on the continuum identified by Bayley & Stenning (Reference Bayley and Stenning2016), if they can influence, shape or determine the actions of the police by ensuring that chiefs think like them, and will act in accordance with their wishes. Evidence for believing Lukes's analysis is a useful way of understanding the background to the shootings at Marikana will be presented in a moment. But before that task is tackled, the legal framework for governing the police in South Africa must be outlined.
GOVERNING THE POLICE IN SOUTH AFRICA
South Africa's post-apartheid constitution (known formally as the Constitution of the Republic of South Africa and dating from 1996) gives ‘political responsibility’ for policing to a member of the Cabinet, who must ‘determine policing policy’ (section 206(1)). ‘Control’ over the police service is vested in a National Commissioner appointed by the President of the Republic ‘as head of the national executive’ (section 207(1)). The relationship between the two, politician and police chief, is provided for in section 207(2): ‘The National Commissioner must exercise control over and manage the police service in accordance with the national policing policy and the directions of the Cabinet member responsible for policing.’ In addition to these specific provisions in relation to the police, the Constitution also states (in section 199(7)) that neither the country's security services, nor any of their members, may, ‘in the performance of their functions’: ‘(a) prejudice a political party interest that is legitimate in terms of the Constitution; or (b) further, in a partisan manner, any interest of a political party’.
The Chief Justice of South Africa has expressed the view (albeit in a minority judgement in the case of Glenister v President of the Republic of South Africa and Others [2011] ZACC 6 at para. 129) that ‘far from requiring insulation from the political sphere, it is a fundamental requirement of our legal system that there is political oversight over the police’. What is not so clear from Chief Justice Ngcobo's statement, or the Constitution itself, are the limits of that oversight, or precisely where the line between policy and political direction on the one hand, and management and control on the other, lies. The South African Police Service Act 68 of 1995 confirms the substance of section 207(2) of the Constitution (section 11(1)) but does little to clarify the position.
THE AFRICAN NATIONAL CONGRESS AND THE POLITICS OF POLICING
Important though it is to keep the constitutional and legislative scheme of police governance in mind – not least because it was relied upon so heavily in defence of their conduct by both the then Minister of Police and the police chiefs with whom he interacted in giving evidence to the Commission of Inquiry – it is impossible to make sense of what happened at Marikana without pausing to consider the impact on the police generally, and on relationships between politicians and police chiefs in particular, of political developments in South Africa in the 12 years leading up to the massacre. This is not the place for a detailed account of the changes that have taken place within the ruling African National Congress (ANC) since it first came to power in 1994.Footnote 6 But one seasoned observer has argued that the emergence of ‘neo-patrimonial politics’ in the ANC has led to a hybrid system of personal rule and bureaucratic order, of modern democratic procedures and legal rationalities intermingled with highly personalised relationships between rulers and ruled built on more traditional loyalties and reciprocities (Lodge Reference Lodge2014). According to Lodge (Reference Lodge2014: 1–2), evidence of neo-patrimonialism can be seen, for example, in the use of public powers for private purposes, in contests between individuals and rival groups and factions for positions of power and influence within the ANC and in national, provincial and local government, and in the acquisition of business interests by politicians and their families through the manipulation of tendering procedures for government contracts. Beresford (Reference Beresford2015: 228) makes similar points when he talks about ‘gatekeeper politics’ with reference to ‘how political leaders in positions of authority within the ruling party or in public office control access to resources and opportunities in order to forward their own political and economic ends’.
However they are characterised, these developments have had a profound effect not just on the SAPS but on the security service and parts of the criminal justice system such as the National Prosecuting Authority (NPA). The ANC's distrust of the police has a long history stretching back to the days of the old South African Police (SAP), which played a vital, violent, often murderous, part in suppressing opposition to apartheid (Brogden & Shearing Reference Brogden and Shearing1993; Brewer Reference Brewer1994). Its determination to exercise control over the police, and other state agencies and institutions, is evident in this passage from an ANC policy document, State, Property Relations and Social Transformation, published in 1998:
Transformation of the state entails, first and foremost, extending the power of the NLM (national liberation movement) over all levers of power: the army, the police, the bureaucracy, intelligence structures, the judiciary, parastatals, and agencies such as regulatory bodies, the public broadcaster, the central bank and so on. (Quoted in Johnston Reference Johnston2014: 178)
This aspiration, the document asserts, does not contradict constitutional provisions (section 199(7) prominent among them) to the effect that most of these bodies should be independent and non-partisan. It seeks, rather, to ensure that they operate in accordance with constitutional precepts, are ‘guided by new doctrines’, reflect South Africa's demographics and, crucially, ‘owe allegiance to the new order’ (ANC 1998 quoted in Johnston Reference Johnston2014: 178). In the view of the ANC then, independence and non-partisanship are taken to apply only to the extent that they are compatible with the values espoused in section 1 of the Constitution. These include respect for human dignity, equality, non-racialism, non-sexism, constitutional supremacy, the rule of law and accountability, responsiveness and openness in government. The tensions built into such a formulation are evident, and become all the more pronounced when account is taken of the ANC's tendency to see itself as the sole legitimate representative, and defender, of the ‘new order’. In the increasingly Manichean world view of large sections of the liberation-movement-turned-ruling-party, ‘independence and non-partisanship’ come a poor second to loyalty to the ANC as principles for effective governance. As Butler (Reference Butler2005: 720) observes, ANC leaders, in common with those in many other new democracies, have found that electoral success and financial gain, both for themselves and their party, lie in colonising and controlling the state rather than entrenching its objective authority.
More recent work by Steinberg (Reference Steinberg2014) and von Holdt (Reference Von Holdt2014) adds important detail to this picture of the ANC's growing dominance of the state and its agencies. Steinberg (Reference Steinberg2014: 177) argues that, with the appointment of Jackie Selebi as National Commissioner of the SAPS in 2000, ‘the ruling party began to use the organs of high policing primarily to police itself’.Footnote 7 Selebi was appointed to succeed George Fivaz, a career policeman, from outside the SAPS. He had no understanding, or prior experience of policing and, as a veteran cadre of the ANC, owed his appointment to the patronage of the then-President, Thabo Mbeki (Steinberg Reference Steinberg2014: 177). This pattern of external appointments of ANC loyalists was repeated when Bheki Cele succeeded Selebi in 2009, and again on 12 June 2012, some two months before the shootings at Marikana, with the appointment of Mangwashi Victoria (‘Riah’) Phiyega. Both Selebi and Cele left their posts facing allegations of corruption. The former was eventually convicted and sentenced to 15 years imprisonment. At the time of writing, the latter has been reincarnated as Minister of Police, neatly illustrating the overriding importance of political reliability in appointments to both roles.Footnote 8 The nub of Steinberg's argument is that, as the ANC began to fragment and corruption reach new heights within the organisation's hierarchy:
controlling the agencies that investigated corruption was fast becoming a crucial tool of control inside the ANC. For the question of whom criminal justice agencies went after and whom they left alone became critical to determining who would control the ANC in the near future. The discretion of the leaders of investigative agencies became explosively political. (Steinberg Reference Steinberg2014: 185)
After Selebi set about dismantling the formal structure of the SAPS he had inherited from Fivaz he took to assembling ad hoc bodies to ‘bring down’ political heavy-hitters, as an informant of Steinberg's (Reference Steinberg2014: 186) put it to him shortly after Selebi's appointment. From facing down the opposition to apartheid in its former guise, the new SAPS had transmogrified into a powerful mechanism for managing a diverse and disputatious ruling party.
Drawing on his and colleagues’ research on the policing of popular protest (von Holdt et al. Reference Von Holdt, Langa, Molapo, Mogapi, Ngubeni, Dlamini and Kirsten2011), von Holdt (Reference Von Holdt2014) comes to a similar conclusion. He argues that conflict within a nascent black bourgeoisie over access to tenders, and other means of accumulating wealth dependent on winning and holding on to political power, has led to the intensification of struggles for control over the coercive instruments of the state. The erosion of the ANC's authority and its inability to manage its internal and external environment has led to the redoubling of efforts to control key centres of power, amongst other things by deploying trusted cadres to leading positions in policing and criminal justice (von Holdt Reference Von Holdt2014: 600–601). It is against this background that the relationship between the SAPS (in the shape of National Commissioner Riah Phiyega and the Provincial Commissioner for the North West, Zukiswa Mirriam Mbombo) and the Minister of Police, Nathi Mthethwa, revealed by the Marikana Commission of Inquiry unfolds.
GOVERNING THE POLICE: THE CASE OF MARIKANA
The official, if not definitive, account of events leading up to the shooting of 34 striking miners at Marikana on 16 August 2012 is contained in the report of the Commission of Inquiry established by then-President Jacob Zuma and chaired by Ian Farlam, a retired judge of South Africa's Supreme Court of Appeal.Footnote 9 Other in-depth accounts provided by Alexander (Reference Alexander2016), Bruce (Reference Bruce2015) and Marinovich (Reference Marinovich2016) offer critical commentaries on the Commission and its findings. None of them pay very close attention to the relationship between the SAPS, its commanders and their political superiors. But they all agree that the then Minister of Police, Nathi Mthethwa, played a decisive role in the fatal events of 16 August without being able to specify exactly what it was (Bruce Reference Bruce2015: 45; Alexander Reference Alexander2016: 832; Marinovich Reference Marinovich2016: 146, 149).Footnote 10
Some of the details of what happened at Marikana, and why, are still contested, so the following brief chronology attempts to avoid controversy and is offered only by way of context for the argument that follows. Thursday 9 August was a public holiday in South Africa. Rock-drill operators (the title of the job is self-explanatory) working at the Lonmin-owned platinum mine at Marikana met at a local sports stadium that day and decided to take strike action in support of a wage claim. The stay-away began the next day and a series of confrontations took place between strikers and officials of the National Union of Mineworkers (NUM), Lonmin security staff and employees attempting to get to work that Friday and over the weekend. The first fatalities occurred on Sunday 12 August when two of the company's security personnel were killed in the course of one of these confrontations. Two other employees died overnight as they attempted to report for work. Five more people (three strikers and two police officials) died the next day, Monday 13 August, in chaotic circumstances when the police intercepted, and attempted to disarm, a group of strikers on their way back from one of the mine's shafts to the ‘mountain’ (in reality a small hill or koppie) they were using as a base. On 14 and 15 August unsuccessful attempts to get the strikers to leave the mountain were made involving a police negotiator and the leaders of the NUM and another trade union, the Association of Mineworkers and Construction Union (AMCU). Lonmin itself steadfastly refused to negotiate with the strikers while they insisted that they would not end their protest unless and until the company agreed to talk to them. Another death came to light on the afternoon of Tuesday 14 August when the body of a man accused of being a spy by the strikers was found. Thus, 10 people had already died before 16 August and the bloody denouement to the strike that saw 34 Lonmin workers shot dead by the police – 17 at each of two ‘scenes’.Footnote 11
Key meetings
The key to understanding how and why a decision was taken to disarm and disperse the strikers camped out on the koppie on Thursday 16 August, if necessary by force (what became known as the ‘tactical option’), lies in meetings that took place on the previous two days. The first was held on the afternoon of Tuesday 14 August. Its main protagonists were the SAPS Provincial Commissioner for the North West, Lieutenant General Mbombo, and Lonmin's Executive Vice President for Human Capital and External Affairs, Barnard Mokwena.Footnote 12 Two other Lonmin executives were also present: Graeme Sinclair, the Group Mining Emergency and Security Manager; and Jomo Kwadi, Senior Manager of Employee Relations. Unbeknown to the Provincial Commissioner, the meeting was recorded by a member of the company's staff. Neither Mbombo nor Mokwena mentioned the meeting in their initial statements and the Commission (Report: 160, para. 1) described the former's explanation for this omission (that it was an ‘informal discussion’ on which nothing turned) as ‘unsatisfactory’. A transcript was eventually made available to the Commission by the SAPS in compliance with a subpoena.Footnote 13 Commenting on the significance of the transcript in their heads of argument, the evidence leaders for the Commission (546, para. 997) observed that statements made by Lieutenant General Mbombo at the meeting were of particular importance because ‘she was unaware that her remarks were being recorded and would one day be closely examined, with the result that she was probably speaking freely’ – something she proved either unable or reluctant to do in giving evidence before the Commission.
The second meeting took place just over 24 hours after the first in the early evening of Wednesday 15 August. Convened as an ‘Extraordinary’ Session of the SAPS's National Management Forum (NMF), which had met earlier that day, it was attended by the National Commissioner, General Phiyega, all nine Provincial Commissioners, including the North West's Lieutenant General Mbombo, and three of the most senior officers responsible for police operations and crime investigation nationally. The minute of the session approved by Phiyega is less than six lines long.Footnote 14 It records that, at the request of the National Commissioner, Mbombo briefed those in attendance on ‘issues of labour unrest’ at Marikana. It then goes on to say that, ‘After deliberations the meeting endorsed the proposal to disarm the protesting masses and further indicated that additional resources must be made available upon need identification by [Mbombo]’.
The Commission of Inquiry, and the evidence leaders on its behalf, made strenuous efforts to obtain a fuller account of proceedings, but an audio recording stored on a memory stick went astray while in the possession of an SAPS Brigadier (see Report: 449–552, paras. 68–73 for details).Footnote 15 When they came to give evidence to the Commission, Phiyega and Mbombo were closely questioned about the discussion at the extraordinary session of the NMF but were unable or, as the Commission was forced to conclude, unwilling to help. An attempt to get more information from the others present at the meeting by means of a questionnaire elicited a response described by the Senior Evidence Leader as ‘a disgrace’ – a view the Commission itself was happy to endorse (Report: 184, para. 4). In the absence of an account of what transpired at the extraordinary session beyond the terse official minute, the Commission concluded that the decision to remove the strikers from the koppie by force if they did not voluntarily lay down their arms was not taken by operational commanders on the scene at Marikana on 16 August as the SAPS had claimed but by Lieutenant General Mbombo at some point the previous day.Footnote 16 It was then endorsed by the national leadership of the SAPS that evening (Report: 183, para. 2, cf. p. 144, para. 61). More damning still was its view that ‘the only reasonable inference’ to be drawn from the failure of those who attended the extraordinary session of the NMF to explain themselves was that ‘they are hiding something’ (Report: 452, para. 73). The disappearance of the memory stick only fortified that inference.
Exercising political control
What do these two meetings tell us about the ways in which politicians exercised control over the police in relation to matters that, on the face of it, appear to be of an operational nature, that is when and how to disarm and disperse a group of striking workers? Two possible modes of control emerge from the work of the Marikana Commission; but it will be argued here that a third, based on Lukes's (Reference Lukes2005) three-dimensional analysis of power, provides a more persuasive interpretation of the available evidence.
Guidance of the executive
The Marikana Commission received no direct evidence that Minister of Police Mthethwa, or any other politician, ordered or directed the police (in effect the National and/or Provincial Commissioner) to disarm and disperse the strikers on 16 August, if necessary by force. On the contrary, all three of the principal players, Mthethwa, Phiyega and Mbombo, categorically denied that the former had played any part in the decision to adopt the ‘tactical option’. Asked by his counsel whether he had sought to ‘prescribe how SAPS perhaps should manage what was unfolding in Marikana’, Mthethwa replied: ‘Well, that's the how part. You as the minister, that's not your province how operationally you have to carry your tasks. As police officers that's your job. I don't enter into that terrain. It's not my terrain (Mthethwa Evidence, Day 255: 32086, line 23–32087, line 2).
Under cross-examination by senior counsel for two non-governmental organisations and the family of one of the miners killed on 16 August, General Phiyega was asked whether the decision to confront the miners that day had any connection with ‘any communication or directions you received from the Minister’ (Phiyega Evidence, Day 70: 7503, lines 17–20). The National Commissioner's characteristically enigmatic response was, ‘Not to any of my knowledge and information’ (Phiyega Evidence, Day 70: 7503, lines 19–20). And finally, when the senior evidence leader asked the Provincial Commissioner about a telephone conversation she had with the Minister on 12 August, Lieutenant General Mbombo replied that Mthethwa had informed her about a call he had received from Cyril Ramaphosa, then a member of the ANC's National Executive Committee and the chairman (sic) of an investment company, Shanduka Group (Pty) Ltd with a 9% stake in Lonmin (Mbombo Evidence, Day 181: 21741, lines 1–10).Footnote 17 He had asked her whether she was attending to matters at the Marikana mine. But he did not encourage her to take action against the strikers. A second conversation the next day, 13 August – by which point five more deaths had occurred – also ended without Mthethwa expressing a view on how she should respond to the situation (Mbombo Evidence, Day 181: 21741, line 11–21742, line 15).
In the final written submissions made on his behalf, Minister Mthethwa's counsel referred to the evidence given by Phiyega and Mbombo before stating that he ‘did not in any way direct the National or Provincial Commissioner on how to discharge their duties and responsibilities in so far as the Marikana incident is concerned’ (Mthethwa Written Submission: 14, para. 29; 21, para. 42). However, the Commission was unable either to find in the Minister's favour on this point, or to exclude the possibility that ‘the guidance of the executive’ had been at least one of the factors behind the decision taken at the extraordinary session of the NMF to endorse Lieutenant General Mbombo's proposal to take decisive action against the strikers the following day (Report: 452, para. 74). It also found that, if such ‘guidance’ had been given, it was ‘probable’ that it had been conveyed to the police by Mthethwa. The Commission's findings here are hedged about with qualifications and double negatives and its report emphasises that ‘it is not finding that such ‘guidance’ was given’ (Report: 452, para. 75). The phrase ‘guidance of the executive’ itself was borrowed from evidence given to the Inquiry by a Dutch policing expert called by the SAPS, Cees de Rover. Based on his experience of working in over 70 countries over 22 years, he had told the Commission that ‘in no democratic country’ he knew of would the police have intervened in a situation with such serious national security and economic ramifications without first having sought, or been given, some ‘guidance’ from the ‘executive’ (De Rover Evidence, Day 286: 37082, lines 7–10; 37071, lines 1–5; 37075, lines 15–20). By ‘the executive’, he meant the Minister of Police, and ‘maybe even the President’, to whom the police, as ‘agents of the state’, were accountable (De Rover Evidence, Day 286: 37072, lines 9–13). And it seems clear from the context that the kind of ‘guidance’ he had in mind was tantamount to a directive.
If such a directive or ‘guidance’ was given and prompted Mbombo's proposal to move against the strikers, the evidence leaders suspected that it was conveyed to her in the course of a 175 second telephone conversation she had with Minister Mthethwa at 6.50 am on the morning of 15 August, the day of the NMF meeting (Heads of Argument: 539, para. 983.4; 542, para. 990).Footnote 18 Notwithstanding his protestations to the contrary, it seems that the Minister was not averse to giving what Bayley & Stenning (Reference Bayley and Stenning2016: 147) characterise as ad hoc directions on contentious issues in areas where the potential for disagreement between politicians and police chiefs is greatest. So, for example, it has been alleged that he ordered Phiyega's immediate predecessor, Acting National Commissioner Nhlanhla Mkwanazi, to reinstate Richard Mdluli, a senior officer accused of a raft of offences including murder and corruption, as the SAPS's Head of Crime Intelligence (Institute for Security Studies 2012).
Political pressure
The evidence that the police acted on ‘the guidance of the executive’ communicated to them, probably by the Minister of Police by telephone on the morning of 15 August, is no more than circumstantial. In the absence of a smoking gun that could be placed in the hands of Mthethwa or one of his ministerial colleagues, the case for believing that the police moved against the strikers at the time and in the way that they did was, at least in part, a result of political interference comes to rest on evidence of pressure exerted on the National and Provincial Commissioners by senior figures in government, the ANC and its trade union allies, namely: Cyril Ramaphosa; the Minister of Mineral Resources, Susan Shabangu; the President of the NUM, Senzeni Zokwana; and, of course, the Minister of Police, Nathi Mthethwa. Two accounts of the nature and effects of such pressure emerge from the work of the Commission. The first, strong version was put forward by representatives of the approximately 270 mineworkers injured and arrested at Marikana during and after the protests of 9–16 August, led by senior counsel, Dali Mpofu. Their description of a ‘chain of political pressure’ is summarised in a section of their heads of argument that asks, ‘Who gave the orders?’ (Injured and Arrested Persons Heads of Argument: 223–224, para. 598). It consists of two links connecting politicians to the fatal events of 16 August. First Ramaphosa abused his political position by contacting the Ministers of Police and Mineral Resources in order to have events at Marikana characterised as criminal thus justifying draconian action by the police against the strikers.Footnote 19 And, second, Mthethwa then transmitted this pressure to Phiyega and Mbombo leading the Provincial Commissioner to order, and the National Commissioner to approve, the measures taken on 16 August.
The strong implication of this account is that Ramaphosa and Mthethwa not only did exert pressure on the two commissioners to take urgent and, if need be, forceful action, they intended to do so. In effect, they, through members of the SAPS chain of command from Phiyega and Mbombo downwards, ‘gave the order(s) for the killings’.
The evidence leaders took a more lenient view. They accepted that a telephone conversation between Ramaphosa and Mthethwa in the early evening of 12 August in which the former expressed concerns about the situation at Marikana, and passed on a request from a Lonmin executive that more police be deployed to prevent further loss of life (two Lonmin security guards had been killed earlier that day), ‘[might] be seen as the first in [a] chain of calls’ (Evidence Leaders Heads of Argument: 549, para. 1005). They acknowledged that this conversation, taken together with those that Mthethwa went on to have with the President of the NUM and the National and Provincial Commissioners were ‘likely’ to have been a factor in the decision to move to the ‘tactical phase’ on 16 August if the strikers did not lay down their arms and leave the koppie of their own accord: ‘senior police officials [felt] the need to act and be seen to act’, as they put it. But the evidence leaders also argued that there was nothing ‘improper or inappropriate’ either in Ramaphosa trying to persuade the Ministers of Mineral Resources (with whom he had a face-to-face meeting on 15 August) and Police that there was a criminal aspect to the events unfolding at Marikana that merited police attention, or in Mthethwa discussing the need to deal with the violence with the National Commissioner (Evidence Leaders Heads of Argument: 556, paras 1019.1 and 1019.2). They concluded that there was no evidence to suggest that Ramaphosa could have anticipated, let alone intended, that the SAPS ‘act precipitately’ as a result of his call to Minister Mthethwa, or that 34 people would die as a result of that action (Evidence Leaders Heads of Argument: 549, para. 1005). Nor, they believed, was there ‘any evidence before the Commission’ (words that may well have been carefully chosen) to show that Mthethwa had urged the police to remove the strikers from the koppie without delay (Evidence Leaders Heads of Argument: 545, para. 995). The Commission agreed with the evidence leaders on these issues, adding only that there was no reason to believe that the SAPS would have reacted any differently to the events of 11–16 August if Ramaphosa had not intervened (Report: 436, para. 37). As for Mthethwa, it noted his insistence that he did not feel that Ramaphosa had put any pressure on him, or that he, in turn, had pressurised Phiyega or Mbombo to take any particular course of action (Report: 442, para. 55). Incomplete as it may well be, the evidence available to the Commission supports the weak rather than the strong version of the ‘pressure’ argument, and the conclusions drawn from it by the evidence leaders rather than counsel for the injured and arrested mineworkers.
The meeting between Provincial Commissioner Mbombo, Lonmin's Barnard Mokwena and other company executives on 14 August gives the clearest indication of what was going through the Lieutenant General's mind immediately before she made her proposal to the NMF, and the transcript of their conversation unearthed by the Marikana Commission deserves careful attention. What it reveals is that she seems to have been considering a range of often related factors, some practical, many more or less explicitly political. The first, and most prosaic, of these considerations was the cost of maintaining a police presence at Marikana that, as the Commission's report records, had grown to 532 personnel by 14 August (Exhibit JJJ-192: 16, lines 9–13, Report: 162, para. 4). With her background in financial management, it is hard not to sympathise with Mbombo on this score. The Marikana operation was costing the SAPS a fortune and diverting scarce resources from other policing priorities. Given the responsibility of the police for maintaining public order, her fears that the contagion of industrial unrest would spread not just to other Lonmin mines, but to those owned by rival companies such as Anglo-American and Samancor, was equally understandable, even if it showed a degree of sympathy with the mining houses that she did not seem to extend to their workers (Exhibit JJJ-192: 17, lines 7–22). Consistent with her view that the SAPS and Lonmin had a shared interest in bringing the protest on the koppie to a speedy conclusion, the Provincial Commissioner agreed with the company's representatives that ‘we [the SAPS and Lonmin] need to show our employees … that we are still in control’ (Exhibit JJJ-192: 14, lines 19–21). It was important, she said, to avoid ‘creating a situation where these people feel that we [the strikers] are in control’ – a situation in which their employer did not ask them to come to work, and the police did not arrest them when they committed acts of criminal violence (Exhibit JJJ-192: 5, line 27–6, line 1). The strikers had to comply with the law as employees of Lonmin (Exhibit JJJ-192: 8, lines 2–4).
In Mbombo's defence, it must be recalled that two of her police colleagues had lost their lives the day before and, as she remarked at one point in the meeting, ‘The emotions are very high’ (Exhibit JJJ-192, p. 7, line 5). Her identification with Lonmin and its interest in resolving an increasingly violent situation may be forgivable in the circumstances. However, and to take only the most obvious example, her failure to raise the possibility of the company opening negotiations with its employees – their core demand and the surest and safest way of ending the stand-off – is evidence of a marked lack of even-handedness in her approach towards workers and employers. A similar degree of partiality is apparent in Mbombo's attitudes towards the two unions active at Marikana, and her warning to Mokwena (based, apparently, on an earlier discussion involving the National Commissioner and another executive) that the company should give AMCU no ‘leeway’ lest it be thought that Lonmin was part of a wider initiative across the mining sector to replace the NUM (Exhibit JJJ-192: 9, line 27 – p. 10, line 8). There was, she said, a ‘political view’ that such a move was afoot following a recent dispute at a mine owned by another company, Impala or Implats (Exhibit JJJ-192, p. 9, lines 24–27). It should be noted at this point that, unlike AMCU, its rival for the support of the workforce at Lonmin, Impala/Implats and elsewhere, the NUM was (and still is) affiliated to South Africa's largest trade union federation, the Congress of South African Trade Unions (COSATU), itself part of a strategic tripartite political alliance with the South African Communist Party (SACP) and the ANC.
Dangerous though this territory was for a senior officer in a supposedly non-partisan police organisation, Mbombo did not stop there. In another exchange with Lonmin's Mokwena she gave the impression, firstly, that a ‘politically high’ individual, Cyril Ramaphosa, was ‘pressuring’ the Minister of Police and, secondly, that something needed to be done in response to that ‘pressure’ since Ramaphosa was both a shareholder in Lonmin and a member of an ANC committee that had expelled a young political firebrand by the name of Julius Malema (Exhibit JJJ-192: 10, lines 19–30).Footnote 20 As she then went on to remind Mokwena and his colleagues, the significance of Ramaphosa's prominent role in that process was that Malema had been instrumental in bringing an end to the dispute at the Impala mine referred to earlier. He had also called for the mines to be nationalised (Exhibit JJJ-192, p. 11, lines 5–17). ‘So’, Mbombo concluded, in words that came back to haunt her in giving evidence to the Commission, ‘[I]t has got a serious political connotation that we need to take into account, but which we need to find a way of de-fusing. Hence I just told these guys that we need to act such that we kill this thing’ (Exhibit JJJ-192: 11, lines 17–20).
Her eagerness to prevent opposition politicians from taking advantage of the situation at Marikana did not end with Malema. When they discovered that they had both been telephoned by a member of parliament for the African People's Convention, Themba Godi, Mbombo and Mokwena soon agreed that arrests needed to be made, and the strikers’ protest ended, because, as Mokwena put it, ‘the longer it goes it is giving all the other opportunists to comment and seize the opportunity and then it gets out of control’ (Exhibit JJJ-192: 13, lines 8–12).
The Commission followed its evidence leaders in finding that the transcript ‘clearly showed’ that Lieutenant General Mbombo ‘took into account irrelevant political considerations in approaching the situation at Marikana’ (Report: 167, para. 8, quoting Evidence Leaders Heads of Argument: 268, para. 543). She did so by saying that she did not want Lonmin and other mining companies to be seen to support AMCU (thus undermining the NUM), that a response was needed to the ‘pressure’ being exerted by the politically well-connected Ramaphosa, and that the stand-off with the strikers at Marikana should be brought to an end before it could be turned to the advantage of Julius Malema, an opposition member of parliament and other people she seemed content to hear described as ‘opportunists’. Her attempts to explain herself in giving evidence to the Commission were dismissed as ‘unconvincing’ and ‘at odds’ with what she had said at the meeting (Report, p. 166, para. 8, quoting Evidence Leaders Heads of Argument: 267–268, para. 542.1 and 542.2). The essence of her explanation was revealed at the end of a lengthy and irritable series of exchanges with Dali Mpofu, senior counsel for the injured and arrested persons, and the Chair of the Commission (Mbombo Evidence, Day 186: 22411, line 12–22412, line 2). Although she conceded that the record of her conversation with the Lonmin executives quoted above was accurate, it was not correct to suggest that she saw the need to ‘kill this thing’ as in any way linked to the ‘serious political connotation’ of Malema's potential involvement. Finally – and agreeing with the evidence leaders once again – the Commission found that her superior, General Phiyega, also ‘participated in inappropriate discussions about political considerations’ with Mbombo (Report: 168, para. 9). To do so was, the Commission stated, ‘inconsistent with our constitutional and statutory regime which requires that policing be conducted in an impartial and unbiased manner’ (Report: 169, para. 10, quoting Evidence Leaders Heads of Argument: 270, para. 546).
Third-dimension power
Lieutenant General Mbombo's reference to Cyril Ramaphosa's contact with the Minister of Police at her meeting with Barnard Mokwena and his colleagues is certainly evidence of weak pressure of the kind identified by the evidence leaders. But, taking the conversation as a whole, it is clear that there is more to the subjection of Mbombo and Phiyega to the will of a dominant ANC-aligned elite than that. It is here that Lukes's conception of power as three-dimensional comes into its own. An initial point to note is that neither of the police chiefs in overall command at Marikana had any significant operational experience. By mid-August 2012, the National Commissioner had been in post for only just over two months. Her training and professional background were in social work and human resources management not policing. It is safe to assume from this that, like her predecessors, Jackie Selebi and Bheki Cele, Riah Phiyega's principal qualification was her loyalty to the ANC and, in her case, to the faction within it aligned to President Jacob Zuma (Marinovich Reference Marinovich2016: 149). Unlike her superior, the Provincial Commissioner was a career police official having started out as a constable under apartheid in what was then Umtata, the ‘capital’ of the nominally independent ‘homeland’ of the Transkei. But she had risen up through the ranks in financial and administrative roles leading eventually to her appointment as Provincial Commissioner in the Northern Cape Province in 2005. It seems likely, then, that Mbombo's elevation to the rank of Lieutenant General and Provincial Commissioner owed more to her talents as a politically reliable bureaucrat than her reputation as an operational commander. Despite their different backgrounds, Phiyega and Mbombo were both products of, and vehicles for, the developments within the ANC, and the changes in its relationship with the SAPS and other state institutions, identified by Butler (Reference Butler2005), Lodge (Reference Lodge2014), Steinberg (Reference Steinberg2014), von Holdt (Reference Von Holdt2014) and Beresford (Reference Beresford2015) – and what Bruce (Reference Bruce2016: 28) describes as the politicisation of the police.
Consider too the range and scope of the political considerations playing on Mbombo's mind at the time of her meeting with the Lonmin executives. Her grasp of the political exigencies of the situation – the need to face down the militancy of workers threatening the profitability of a foreign-owned company in which members of local elites had significant financial interests, to defend the ANC's allies in the NUM from what appeared to be an insurgency led by an upstart rival, and to forestall attempts by political foes old and new to gain support among organised labour, a core constituency for the ruling party – is indicative not of someone succumbing to political pressure (still less reluctantly obeying orders) but of a conscious and willing political actor. Mbombo was well aware of the risks entailed in asking police members incensed by the loss of two of their colleagues to disarm a group of armed strikers. She did not ‘want a situation where 20 people will be dead’ if instructions about the police use of firearms were ignored in the emotionally charged circumstances prevailing at Marikana (Exhibit JJJ-192: 7, lines 10–12). Yet, within 24 hours of saying this to Mokwena, she was prepared to order an intervention that led to 34 deaths, most plausibly because she was also aware of the need to ‘kill it’ immediately and before ‘any Jack and Jay from a political angle’ could become involved (Exhibit JJJ-192: 11, lines 22–25).Footnote 21 Here again, it is hard to see Mbombo's contributions to the meeting with Lonmin, and her subsequent decision to order that the strikers should be disarmed and dispersed, as the behaviour of a police officer whose professional judgement had been overridden by a higher political authority. On the contrary, as Lukes's (Reference Lukes2005) analysis of the third dimension of power or compliance with domination would suggest, her conduct was consistent with that of a police chief whose thoughts were at one with those of her political superiors and in close alignment with the interests of the ANC and its allies, her actions intuitive rather than coerced.
CONCLUSION
The relationship between police chiefs and politicians lies at the heart of policing in democratic societies. If politicians have too much power, the rule of law may be compromised under a tyranny of the majority. Allow the police too free a hand and democracy itself may be threatened. Bayley & Stenning (Reference Bayley and Stenning2016) provide invaluable insights into how this dilemma is resolved in practice from the police chiefs’ perspective and document some of the ways in which political requests are communicated. What they do not allow for is the possibility that, as Lukes's three-dimensional analysis would suggest, political control of policing may be at its most effective when it is least visible – when no communication is necessary because police chief and politician are of one mind. Careful reading of the evidence presented to the Commission of Inquiry into events at the Marikana platinum mine in August 2012, and the conclusions it reached on that evidence, suggests that the decision to disarm and disperse a group of striking mineworkers that led to 34 deaths may well have been taken largely if not entirely because senior police officers were so well attuned to the political exigencies of the situation that they did not need to be told or induced to act; they knew what had to be done, and that it had to be done quickly. To make the same point more broadly in the terms used by Marenin (Reference Marenin1982), the police may act as agents of specific domination in the interests of a powerful group without that group or its representatives directing them to do so in any observable way. In the case of Marikana, in the face of insurrectionary action by the Lonmin strikers (Dixon Reference Dixon2015), the SAPS responded on behalf of an ANC-aligned political elite with close ties to international and domestic capital. The result was 34 deaths for which, at the time of writing, no-one, neither police official nor politician, has been held to account before a criminal court.
