Cooperation of South African Democratic Intelligence Oversight Structures for Good Governance By

o Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.

o NonCommercial — You may not use the material for commercial purposes.

o ShareAlike — If you remix, transform, or build upon the material, you must distribute your contributions under the same license as the original.

How to cite this thesis

Surname, Initial(s). (2012) Title of the thesis or dissertation. PhD. (Chemistry)/ M.Sc. (Physics)/ M.A. (Philosophy)/M.Com. (Finance) etc. [Unpublished]: University of Johannesburg. Retrieved from: https://ujcontent.uj.ac.za/vital/access/manager/Index?site_name=Research%20Output (Accessed: Date).

Cooperation of South African democratic intelligence oversight structures for good governance by

NAKAMPE MICHAEL MASIAPATO

A thesis submitted in fulfilment for the Degree of

Doctor Litterarum et Philosophiae

Public Management and Governance

at the

College of Business and Economics UNIVERSITY OF JOHANNESBURG

Supervisor: Prof F Cloete

Co-supervisor: Prof CJ Auriacombe

2017

DECLARATION

I certify that the thesis submitted by me for the degree at the University of Johannesburg is my independent work and has not been submitted by me for a degree at another university.

Nakampe Michael Masiapato

ACKNOWLEDGEMENTS

I would like to express my sincere thanks and appreciation to:   Almighty God who through His grace and mercy empowered me to successfully navigate and survive all kinds of weather. He gave me the ability, strength and resilience to complete this project.    Prof Fanie Cloete, my mentor and supervisor, for his patience, guidance and friendly advice during the entire project. 

 Prof Christelle Auriacombe, my co-supervisor for her resilience, patience and perseverance during trying times.

 My Apostle (Dr) Lucky Chawane and the entire KCA congregation for their prayers and understanding during difficult days.

 Ms Sarah Monyaki and Ms Karen de Lange for their inputs and contributions in IT and graphics work respectively.    All oversight, ombud, audit and other officials for the professionalism in answering the interview questions as posed.

 All oversight officers and external experts who availed themselves for an interview session and provided expert responses to the posed questions. 

 My parents, Mr Boy and Mrs Mmaselaelo Masiapato for their continued sacrifices, patience and encouragement during the entire project.

iii

 My friends and colleagues at the National Intelligence Coordinating Committee (NICOC) for their pieces of advice and inputs during the compilation of this thesis.

DEDICATION

I would like to dedicate this thesis to my lovely wife Ms Shirley Masiapato and our three boys, Quincy, Quinton and Mike Junior for their understanding during my absence in pursuit of this agenda.

ABSTRACT

Before 1994, members of the intelligence services did not operate within a particular set of statutory and regulatory frameworks. At the time, there was no clear separation of civilian, police and/or military intelligence services (Bruneau and Boraz: 2007). Their priorities were centred on the collection, analysis and dissemination of internal security threats posed by the liberation movements. Further, their preoccupation was to fiercely enforce the draconian statutory and regulatory frameworks to protect white supremacy and advance the apartheid agenda (Dombroski in Bruneau and Dombroski: 2001). Given the arrangement, the apartheid government did not make any effort to create any intelligence oversight structure to ensure the compliance of members of the intelligence services to the country’s statutory and regulatory prescripts.

However, after 1994 the ANC government undertook a number of measures to correct the situation. These measures gave rise to the creation of the clear civilian, police and military intelligence services with explicit sets of powers and legislated mandates. Through the Intelligence Services Oversight Act, 1994, the government established multifaceted intelligence oversight machinery. This was done to ensure the compliance of the intelligence services with the country’s statutory and regulatory framework. Unfortunately, the operational modalities of the intelligence oversight structures remain fragmented, uncoordinated and reflect a silo approach, thereby playing a minimal or no role in the promotion of good governance results (Dube: 2013); (Dlomo: 2004); (Netshithenzhe: 2005). This thesis seeks to find ways to improve the cooperation of these structures when overseeing members of the intelligence services.

To achieve the above-mentioned objective, the thesis firstly described the context - that is, the South African intelligence services and its oversight machinery. Secondly, a global benchmarking exercise on selected countries was conducted for the purpose of drawing lessons for South Africa. Thirdly, the thesis presented techniques, methods

vi and procedures for utilisation in the implementation of the research project. Fourthly, it assessed the rationale for, and state of cooperation among intelligence oversight structures. Fifthly, the thesis then assessed the nature of barriers and challenges facing oversight structures when attempting to cooperate with each other. Sixthly, the thesis presented proposals on what to do to improve the cooperation of these structures when conducting their oversight work. Lastly, a systematic cooperation model for guiding the future cooperation of intelligence oversight structures was presented to serve as an instrument for streamlining the work of oversight structures and guide their efforts towards the realisation of good governance results.

vii

TABLE OF CONTENTS Page Declarations ii Acknowledgements iii Dedication v Abstract vi List of key terms and abbreviations xv

CHAPTER 1: BACKGROUND AND SCIENTIFIC ORIENTATION TO THE STUDY

1.1 INTRODUCTION 1 1.2 OVERVIEW OF SOUTH AFRICAN INTELLIGENCE ARCHITECTURE 2 1.2.1 Pre-1994 period 2 1.2.2 Transitional period 3 1.2.3 Current intelligence architecture 5 1.3 Democratic intelligence oversight structures 11 1.4 Problem statement 15 1.4.1 Fragmentation 17 1.4.2 Lack of coordination 17 1.4.3 Lack of cooperation and collaboration 19 1.5 PURPOSE OF THE RESEARCH 21 1.6 RESEARCH QUESTIONS 23 1.7 RESEARCH OBJECTIVES 26 1.8 RESEARCH ETHICS 27 1.9 SIGNIFICANCE OF THE STUDY 28 1.10 CONTRIBUTION TO BODY OF KNOWLEDGE AND PRACTICE OF COORDINATION 30 1.11 DELIMITATION OF THE STUDY 31

viii

1.12 THESIS STRUCTURE 32 1.12.1 Chapter one: Background to the study 32 1.12.2 Chapter two: Review of good international intelligence oversight systems 33 1.12.3 Chapter three: Research methodology 33 1.12.4 Chapter four: Structural cooperation among oversight structures 34 1.12.5 Chapter five: Barriers to improved cooperation of structures 34 1.12.6 Chapter six: Strategies to improve cooperation of structures 34 1.12.7 Chapter seven: Proposed improved cooperation model 34 1.13 SUMMARY AND CONCLUSIONS 35

CHAPTER 2: CONTEXTUAL AND THEORETICAL VARIABLES INFLUENCING GOOD INTERNATIONAL INTELLIGENCE OVERSIGHT GOVERNANCE SYSTEMS

2.1 INTRODUCTION 36 2.2 GLOBAL BENCHMARKING OF DEMOCRATIC INTELLIGENCE OVERSIGHT 37 2.3 SELECTED INTERNATIONAL BEST PRACTICES ON INTELLIGENCE OVERSIGHT 38 2.3.1 Australia 40 2.3.2 Canada 42 2.3.3 New Zealand 44 2.3.4 United Kingdom (UK) 46 2.3.5 United States (US) 47 2.4 SUMMARY OF FINDINGS ON INTERNATIONAL OVERSIGHT STRUCTURES 49 2.4.1 Rationale for, and state of cooperation among intelligence oversight structures 49 2.4.2 Barriers to effective cooperation of intelligence oversight structures 51 2.4.3 Strategies for effective cooperation of intelligence oversight structures 52

2.5 Theoretical frameworks 54 2.5.1 Systems theory 55 2.5.2 Good democratic governance theory 59 2.6 CONCEPTUAL FRAMEWORK FOR SOUTH AFRICA 67 2.7 SUMMARY AND CONCLUSIONS 75

CHAPTER 3: METHODOLOGICAL APPROACH TO THE STUDY

3.1 INTRODUCTION 77 3.2 RESEARCH PARADIGM 78 3.2.1 Qualitative paradigm 80 3.3 RESEARCH DESIGN 83 3.3.1 Qualitative case study 83 3.4 RESEARCH PROCEDURE 87 3.4.1 Data collection instruments 88 3.4.2 Target population and description of respondents 93 3.4.3 Data presentation 94 3.4.4 Data analysis 95 3.5 RELIABILITY AND VALIDITY 99 3.5.1 Internal validity 99 3.5.2 Construct validity 100 3.5.3 External validity 101 3.5.4 Reliability 101 3.6 ETHICAL CONSIDERATIONS AND LIMITATIONS 102 3.7 SUMMARY AND CONCLUSIONS 103

CHAPTER 4: STRUCTURAL COOPERATION AMONG DEMOCRATIC INTELLIGENCE OVERSIGHT STRUCTURES IN SOUTH AFRICA

4.1 INTRODUCTION 105 4.2 SOUTH AFRICAN DEMOCRATIC INTELLIGENCE OVERSIGHT ARCHITECTURE 106 4.2.1 Intelligence oversight by the Legislature 107 4.2.2 Intelligence oversight by the Executive 110 4.2.3 Intelligence oversight by the Judiciary 112 4.2.4 Intelligence oversight by the Administrative Structure 116 4.2.5 Intelligence oversight by chapter 9 institutions 120 4.3 THE RATIONALE FOR EFFECTIVE INTELLIGENCE OVERSIGHT COOPERATION 127 4.3.1 Emergence of fourth generation warfare 132 4.3.2 Different intelligence oversight mandates 134 4.3.3 Varied intelligence oversight powers 136 4.3.4 Application of varied intelligence oversight methods 137 4.3.5 Different intelligence oversight intervals 138 4.3.6 Different reporting modalities 140 4.3.7 Potential for human rights violations 142 4.3.8 Need for transparency and public accountability 144 4.4 THE CURRENT STATE OF COOPERATION AMONG OVERSIGHT STRUCTURES 145 4.4.1 Cooperation gaps exposed by Project Avani scandal 145 4.4.2 Legislative framework on cooperation of oversight structures 150 4.4.3 Existing institutional mechanisms for intelligence oversight cooperation 153 4.4.4 Case referral mechanism 158 4.4.5 Intelligence oversight processes 161 4.4.6 Cooperation of chapter 9 institutions 163

4.5 SUMMARY AND CONCLUSIONS 165

CHAPTER 5: BARRIERS TO IMPROVED COOPERATION AMONG DEMOCRATIC INTELLIGENCE OVERSIGHT STRUCTURES IN SOUTH AFRICA

5.1 INTRODUCTION 168 5.2 BARRIERS TO IMPROVED COOPERATION OF INTELLIGENCE OVERSIGHT STRUCTURES 169 5.2.1 Structural barriers to the cooperation of oversight structures 175 5.2.2 Operational barriers to cooperation of oversight structures 194 5.3 SUMMARY AND CONCLUSIONS 206

CHAPTER 6: STRATEGIES TO IMPROVE THE COOPERATION OF INTELLIGENCE OVERSIGHT STRUCTURES IN SOUTH AFRICA

6.1 INTRODUCTION 208 6.2 RESHAPING THE KEY INTELLIGENCE OVERSIGHT STRUCTURES 210 6.2.1 Refocusing the Joint Standing Committee on Intelligence (JSCI) 212 6.2.2 Refocusing the Ministry of State Security 217 6.2.3 Refocusing the Inspector-General of Intelligence (IGI) and other structures 220 6.2.4 Strengthening the role of civil society and the media 221 6.3 IMPROVING THE OPERATIONAL COOPERATION OF INTELLIGENCE OVERSIGHT STRUCTURES. 225 6.3.1 Streamline legislative and regulatory frameworks 227 6.3.2 Empower intelligence oversight officers 231 6.3.3 Balance secrecy with the need for transparency 233 6.3.4 Improvement of trust among intelligence oversight structures 237 6.4 SUMMARY AND CONCLUSIONS 239

xii

CHAPTER 7: PROPOSED IMPROVED COOPERATION MODEL FOR INTELLIGENCE OVERSIGHT STRUCTURES IN SOUTH AFRICA

7.1 INTRODUCTION 241 7.2 SUMMARY OF RESEARCH FINDINGS AND RESULTS 243 7.2.1 Chapter one: Background to the study 244 7.2.2 Chapter two: Review of good international intelligence oversight systems 245 7.2.3 Chapter three: Research methodology 246 7.2.4 Chapter four: Structural cooperation among oversight structures 247 7.2.5 Chapter five: Barriers to improved cooperation of structures 248 7.2.6 Chapter six: Strategies to improve cooperation structures 249 7.2.7 Chapter seven: Proposed improved cooperation model 250 7.3 KEY RESEARCH CONCLUSIONS AND RECOMMENDATIONS 250 7.3.1 Theme 1: Absence of legislative and regulatory frameworks 250 7.3.2 Theme 2: Lack of the common doctrine 251 7.3.3 Theme 3: Undue political interference and ‘plausible deniability 252 7.3.4 Theme 4: Excessive secrecy versus transparency and public accountability 253 7.3.5 Theme 5: Proliferation of intelligence control and oversight structures 254 7.3.6 Theme 6: Unclear structural governance challenges 255 7.4 PROPOSED IMPROVED COOPERATION MODEL FOR OVERSIGHT STRUCTURES 257 7.5 IMPLICATIONS OF THE RESEARCH FOR THEORY AND PRACTICE 260 7.6 CONSTRAINTS ON THE RESEARCH AND DIRECTION FOR FUTURE RESEARCH 262 7.7 SUMMARY AND CONCLUSIONS 265

xiii

LIST OF FIGURES

Figure 1.1: The South African civilian intelligence architecture in 2017 7 Figure 2.1: Conceptual framework 74 Figure 3.1: Research methodology 78 Figure 4.1: South African intelligence oversight model 2016 107 Figure 4.2: Applications for judicial authorisation – 2008/2009- 2014/2015 114 Figure 4.3: Need for the cooperation of intelligence oversight structures 128 Figure 4.4: Underlying reasons justifying the need for the cooperation of various intelligence oversight structures 131 Figure 4.5: Current state of cooperation among intelligence oversight structures 149 Figure 4.6: Whether current legislative and or regulatory frameworks of the oversight structures require their cooperation with each other. 151 Figure 4.7: Institutional mechanism within which the current cooperation of the intelligence oversight structures is taking place 154 Figure 4.8: Frequency of the engagements between the various intelligence oversight structures 155 Figure 4.9: Extent to which the frequency of the engagement shaped the agenda of the interaction among oversight structures 156 Figure 4.10: Whether oversight structures have a formal or informal mechanism for referring cases to other structures 159 Figure 4.11: Whether the case referral mechanism functions optimally to enable speedy referral and timely responses 161 Figure 5.1: Whether there are certain impediments or barriers to the effective cooperation of the intelligence oversight structures 172 Figure 5.2: An outline of critical impediments or barriers to the effective Cooperation of intelligence oversight structures 174 Figure 5.3: Whether the oversight structures have been able to create Relationships of trust with other intelligence oversight structures 183

xiv

Figure 6.1: Whether there is a need to review the mandates, and roles of the various intelligence oversight structures to facilitate their cooperation with each other 209 Figure 6.2: Tabulation of particular structures whose mandates, roles and functions require reviewing to improve their cooperation with other structures 210 Figure 6.3: Tabulation of suggestions on what needs to be done to improve the cooperation of the various intelligence oversight structures 226 Figure 7.1: Whether there is a need for the creation of a new model to refine the South African architecture to include the systematic cooperation of their endeavour to cooperate with each other 258 Figure 7.2: Systematic cooperation model for intelligence oversight structures 259

LIST OF TABLES

Table 3.1 Research questions, data collection instruments and techniques 87 Table 4.1 Comparative analysis of the works of the Inspectors-General of Intelligence 118

LIST OF APPENDICES

Appendix A: Qualitative interview guide and results: oversight officers 267 Appendix B: Qualitative interview guide and results: external experts 276 Appendix C: Consent form for research participants 282

BIBLIOGRAPHY 283

LIST OF KEY TERMS AND ABBREVIATIONS AG – Auditor-General AIDS – Acquired Immune Deficiency Syndrome AIGI – Australian Inspector-General of Intelligence AIGIS – Australian Inspector-General of Intelligence and Security AISO – Australian Intelligence and Security Organisation ANC – African National Congress APLA – Azanian Peoples’ Liberation Army APP(s) – Annual Performance Plan(s) ARV – Antiretroviral drugs ASIO – Australian Security Intelligence Organisation ASIS – Australian Secret Intelligence Service AU – African Union AZANLA – Azanian National Liberation Army AZAPO – Azanian Peoples’ Congress BND – Bundesnachrichtendienst BOSS – Bureau for State Security BRICS – Brazil, Russia, India, China and South Africa CCO – Committee of Controlling Officials CDI – Chief of Defence Intelligence CI – Crime Intelligence CIA – Central Intelligence Agency CIGIE – Council of Inspector-Generals on Integrity and Efficiency CIO – Central Imagery Office CISSA – Committee of Intelligence and Security Services in Africa COMPUSEC – Computer Security COMSEC – Communications Security COSATU – Congress of South African Trade Unions CS – Corporate Services

xvi

CSIRC – Canadian Security Intelligence Review Committee CSIS – Canadian Security and Intelligence Service DA – Democratic Alliance DB – Domestic Branch DCAF – Democratic Control of the Armed Forces DG – Director-General DI – Defence Intelligence DIA – Defence Intelligence Agency DIGO – Defence Imagery and Geospatial Organisation DIO – Defence Intelligence Organisation DMI – Directorate of Military Intelligence DIS – Directorate of Intelligence and Security DNS – Department of National Security DOD – Department of Defence DSD – Defence Signals Directorate DSO – Directorate of Special Operations EFF – Economic Freedom Fighters EXCO – Executive Committee FB – Foreign Branch FBI – Federal Bureau of Investigations FIC – Financial Intelligence Centre FIS – Foreign Intelligence Service FISA – Foreign Intelligence Surveillance Act GCHQ – Government Communications Headquarters GCSB – Government Communications Security Bureau GILAB – General Intelligence Laws Amendment Bill GNU – Government of National Unity GNT – Governance Network Theory GST – General Systems Theory

xvii

HIV – Human Immunodeficiency Virus HIV/AIDS - Human Immunodeficiency Virus /Acquired Immune Deficiency Syndrome HRC – Human Rights Commission IA – Intelligence Academy ICCS – Intelligence Council on the Conditions of Service ICT – Information Communications Technology IG(s) – Inspector-General(s) IGI – Inspector General of Intelligence IGIS – Inspector-General of Intelligence and Security or IGI and Security IMSC – Inter Ministerial Security Committee IPID – Independent Police Investigative Directorate ISC – Intelligence and Security Committee ISS – Institute of Security Studies JSCD – Joint Standing Committee on Defence JSCI – Joint Standing Committee on Intelligence KGB – Komitet Gosudarstvennoy Bezopasnosti MISS – Minimum Information Security Standards MoU – Memorandum of Understanding MP(s) – Member(s) of Parliament MRC – Ministerial Review Commission NA – National Assembly NC – National Communications NCC – National Communications Centre NCCS – National Crime Combatting Strategy NCOP – National Council of Provinces NDF – National Defence Force NDU – National Declassification Unit NIA – National Intelligence Agency NICOC – National Intelligence Coordinating Committee

xviii

NIP – National Intelligence Priorities NIS – National Intelligence Service NPA – National Prosecuting Authority NRO – National Reconnaissance Office NSA – National Security Agency NSC – National Security Committee ODESC – Officials Committee for Domestic and External Security Coordination OECD – Organisation of Economic Cooperation and Development OIGI – Office of the IGI ONA – Office of National Evaluations PAC – Pan African Congress PAI – Promotion of Access to Information PAIA – Promotion of Access to Information Act 2 of 2000 PhD – Doctor of Philosophy PMG – Parliamentary Monitoring Group POSIB – Protection of State Information Bill PP – Public Protector RCMP – Royal Canadian Mounted Police RICA – Regulation of Interception of Communication and Provision of Communication- related Information Act 70 of 2002 RSA – Republic of South Africa SACP – South African Communist Party SADF – South African Defence Force SAHRC – South African Human Rights Commission SANAI – South African National Academy of Intelligence SANDF – South African National Defence Force SANEF – South African National Editors Forum SAPS – South African Police Service SASS – South African Secret Services

xix

SCOPA – Standing Committee on Public Accounts SDU – Self Defence Unit SIGINT – Foreign Intelligence SIRC – Security Intelligence Review Committee SOWETO – South Western Township SPU – Self Protection Unit SSA – State Security Agency SSA DB – State Security Agency Domestic Branch SSA FB – State Security Agency Foreign Branch TAC – Treatment Action Campaign TBVC – Transkei, Bophuthatswana, Venda and Ciskei TEC – Transitional Executive Council TECSEC – Technical Security TRC – Truth and Reconciliation Commission UK – United Kingdom UN – United Nations UNDP – United Nations Development Programme US – United States

CHAPTER 1 BACKGROUND AND SCIENTIFIC ORIENTATION TO THE STUDY

1.1 INTRODUCTION

In the post-1994 period, members of the intelligence services play a crucial role in the protection, promotion and maintenance of national security. It is their responsibility to assert the authority of the state and guarantee its territorial integrity. This is concretised by the expanded understanding of national security which embraces the interrelationship between securing the state and developing it. Countless scholars have confirmed this relationship by arguing that sustainable development happens in a secure environment. Accordingly, the efficacy of the intelligence services is measured on the basis of its contribution to the government’s policy agenda. This is mainly dependent on political hegemony and the integrity of intelligence practitioners during the performance of their legislative mandates. In this process, a key issue is the consistency of the intelligence services in observing and adhering to democratic practices and the rule of law. As a result, a government’s ability to establish and maintain disciplined and integrated intelligence machinery remains an important benchmark in sustaining any democratic agenda.

According to Bruneau and Dombroski (2001:325), “there is an intrinsic conflict between secrecy and accountability in the intelligence services in various societies”. In fact, excessive secrecy has the potential to undermine transparency and thereby threatening the tenets of democracy. When the intelligence services are not properly overseen, they have the tendency to undercut democratic principles, thereby advancing their own personal or group selfish interests. The contemporary intelligence governance model encourages the creation of a democratic intelligence oversight mechanism. This is meant to oversee members of the intelligence services in order to balance transparency, accountability and secrecy. This approach requires the correct

1 interpretation of national legislation as well as a clear understanding of organisational and bureaucratic behaviour.

This chapter starts by providing an overview of the South African intelligence architecture followed by a short assessment of the country’s democratic intelligence oversight structures. The chapter also provides the problem statement, its purpose and the research questions. The chapter also outlines the research objectives, significance and their epistemological contribution to the body of knowledge. As Agere (2000:95) puts it, “it is the behaviour and the operational modalities of the intelligence services which shape the public confidence in any government”. The chapter finally concludes with the thesis structure in the form of a breakdown of the contents of each chapter.

1.2 OVERVIEW OF THE SOUTH AFRICAN INTELLIGENCE ARCHITECTURE

The following sections introduce the South African intelligence architecture with a short description of the pre-1994 period, the transitional period up to its current form. The purpose of this discussion is to contextualise the subjects, ie those intelligence service structures that are overseen by the intelligence oversight structures. While the following discussion presents the ‘overseen’, a detailed discussion on the ‘overseers’ and their cooperation modalities will be made in Chapter 4.

1.2.1 Pre-1994 period

In the pre-1994 period, the South African intelligence services did not have a clear separation among the civilian, police and military intelligence services. Their primary mandate was to collect, analyse and disseminate intelligence on internal threats posed by the liberation movements. As described by Dombroski in Bruneau and Boraz (2007:245), “the civilian component had until the mid-1980 received intelligence officers from within the police and the military establishments”. The Special Branch of the South

African Police in particular was responsible for the recruitment, training and supply of intelligence officers to the other structures.

The primary objective of the intelligence services was to focus on the liberation movements (Dombroski in Bruneau and Boraz: 2007). These included, among others, the African National Congress (ANC), the Azanian Peoples’ Organisation (AZAPO) and the Pan African Congress (PAC). These movements were viewed as the primary threats to the state as they were seen to be controlled by external communist powers. During apartheid, the security branch was located within the civilian intelligence organisation called the Republican Intelligence. The organisation later became the Bureau for State Security (BOSS), the Department of National Security (DNS) and the National Intelligence Service (NIS) respectively (Dombroski in Bruneau et al.: 2007). Over time, state intelligence activities expanded to include elements from the Department of Information, the Department of Foreign Affairs, the Security Branch of the Railway Police and the Department of Prison Services (Bruneau and Boraz: 2007).

During this time, the intelligence services remained focused on a strict enforcement of legislation enacted to advance the apartheid agenda. These included, among others, the Suppression of Terrorism Act 83 of 1967; the Suppression of Communism Act 44 of 1950; and the Bantu Authorities Act 68 of 1951. In addressing the learners at Morris Isaacson High School in Soweto on 3 June 2015, the Deputy Minister of State Security, Ms Ellen Molekane indicated that, “the apartheid intelligence services were focused on the identification, arrest, torture and elimination of freedom fighters”. This reflection was quoted on the website of the State Security Agency (SSA) (www.ssa.gov.za).

1.2.2 Transitional period

During the 1994 transitional period, the National Strategic Intelligence Act (Act 39 of 1994) was enacted by the government. It revised the mandate and the functions of the

3 intelligence services for a more democratic system. The legislation created the National Intelligence Agency (NIA) and SASS as civilian intelligence components. The NIA was given the domestic mandate while SASS focused on the international terrain. This Act also created NICOC to ensure the coordination of national strategic intelligence. Made up of heads of the security services, NICOC is supposed to avoid undue competition, duplication of effort, and the development of integrated national intelligence scenarios to advise cabinet.

All these activities were meant to legitimise and democratise the intelligence services, reorganise their work and redefine their functions. The initiative enabled the integration of the various intelligence services from the homelands and self-governing territories. Cawthra (2005:102) indicated that more than 1 000 intelligence operatives from the ANC, PAC and AZAPO were integrated into the newly created security services. In the new dispensation, the police and military components were renamed Crime Intelligence (CI) and Defence Intelligence (DI) respectively with clear legislative mandates. CI has the domestic mandate and operates within the South African borders with specific focus on crime and related matters. On the other hand, DI has the foreign mandate and works to support the National Defence Force (NDF) in preserving the country’s territorial integrity (Cawthra: 2005).

According to Dombroski in Bruneau and Boraz (2007:251), the establishment of the South African post-apartheid intelligence services was informed by the Constitution of the Republic of South Africa, 1996 (hereafter also referred to as the RSA Constitution of 1996, the Constitution of 1996 or the Constitution) and the White Paper on Intelligence of 1994. According to section 209(1) of the Constitution of 1996, “any intelligence service other than any intelligence division of the defence force or police service may be established by the President, as head of the national executive, and only in terms of national legislation”. The White Paper on Intelligence of 1994 provides “the philosophy, mission and roles of the Intelligence services in a democratic South

Africa”. It should be noted that the White Paper for Intelligence (1994) exclusively focused on redefining the responsibilities of the civilian intelligence structures within the domestic and foreign areas of responsibilities.

The reforms of the military and police intelligence services were addressed in the White Paper for Defence of 1996 and the South African Police Services Act 68 of 1995 respectively. To avoid any possible abuse of power, the Intelligence Services Oversight Act 13 of 1994 was enacted to establish the civilian intelligence oversight machinery.

1.2.3 Current intelligence architecture

Following the 1994 transitional period, the new intelligence dispensation is divided into three components, namely: civilian, police and defence intelligence. This was done in order to clearly delineate intelligence functions within the newly developed legislative framework. The following discussion provides a short discussion on the structures operating within each of these components.

1.2.3.1 Civilian intelligence

After his election as the president of South Africa in the 2009 general elections, President Jacob Zuma ordered another major restructuring of the civilian intelligence services since the 1994 transitional experience. As Africa (2012:98) puts it, “the 2009 restructuring saw the placement of the NIA, SASS, and South African National Academy for Intelligence (SANAI) and COMSEC under a single Director-General responsible for state security”. As a result, the NIA was replaced by the SSA DB and SASS got replaced by the SSA FB. Both branches maintain the same statutory mandates as their respective predecessors.

Further, increasing financial crimes forced the government to establish a Financial Intelligence Centre (FIC) through the Financial Intelligence Centre Act 38 of 2001. Its primary mandate is to counter financial crimes, such as money laundering, tax evasion and terror financing activities. While addressing learners at Morris Isaacson High School in Soweto on 3 June 2015, the Deputy Minister of State Security, Ms Ellen Molekane indicated that, “the intelligence services in the democratic dispensation looks for the crooks and thieves who want to destroy our country and bring them to book” (this was quoted on the website of the SSA). Further, SANAI was replaced by the Intelligence Academy (IA) and COMSEC by National Communications (NC), both reporting to the Director-General (DG) of the SSA (Cwele: 2012).

In justifying the restructuring of the civilian intelligence services, Dr Siyabonga Cwele, the Minister of State Security, indicated that “the restructuring was to correct the continued disproportionate allocation of huge resources to the corporate services (CS) of the intelligence services in the expense of the core business of intelligence tradecraft itself” (this quotation was sourced from the website of the SSA). In supporting this view, Dube (2013:16) indicates that “the 2009 restructuring was an attempt to position the intelligence services to effectively address the security challenges of the 21st century”. According to Africa (2012:126), the restructuring was authorised through a Presidential proclamation pending the finalisation of the General Intelligence Laws Amendment Bill, 2013 (GILAB) (www.ssa.gov.za). Figure 2.1 depicts the new South African civilian intelligence services after the 2009 restructuring:

FIGURE 1.1: The South African civilian intelligence architecture in 2016

Source: (www.ssa.gov.net).

This restructuring followed President Zuma’s renaming of the Ministry of Intelligence to the Ministry of State Security, and the subsequent reappointment of Dr Cwele as Minister. In the new architecture as shown in Figure 2.1, the Intelligence Staff Council (ISC) and intelligence coordination reside within the Ministry of State Security. All support structures are centralised within a single CS led by the Deputy DG. According to the Ministry, this was done in order to improve governance and reduce competition between the various structures of the intelligence services by optimising resource utilisation (State Security Agency: 2013). As Cwele (2012:16) puts it “the creation of the State Security Agency was necessary given the proliferation of statutory structures within the civilian intelligence services and this led to lack of coordination, overlapping mandates, unnecessary turf-battles, insufficient focus and wasteful duplication”.

According to Africa (2012:126), “the restructuring created a strong public perception that the intelligence services seek to depart from its founding principles as articulated in the White Paper for Intelligence of 1995”. This was followed by a series of scandals implicating members of the intelligence services (Africa 2012:126). This was also exacerbated by the re-introduction of the Protection of Information Bill of 2010 in Parliament during 2011.

1.2.3.2 Defence intelligence (DI)

According to Cilliers and Reihardt (1995:179), “the Government of National Unity (GNU) committed at the dawn of democracy to downsize and reform the Directorate of Military Intelligence (DMI) of the South African Defence Force (SADF)”. Ultimately, DMI became Defence Intelligence (DI) under the command of the South African National Defence Force (SANDF) and its internal oversight work is undertaken by the Defence Inspector- General for Intelligence. During the integration process, DI absorbed former intelligence personnel of the DMI, Directorate of Intelligence and Security (DIS), the Azanian People’s Liberation Army (APLA) and the Azanian National Liberation Army (AZANLA) as well as the intelligence officers from the militaries of the TBVC states and self- governing territories.

According to the National Strategic Intelligence Act 39 of 1994, the operational mandate of DI is to “gather, correlate, evaluate and use foreign military intelligence and supply foreign military intelligence relating to national strategic intelligence to NICOC”. Section 24(a) of the National Strategic Intelligence Act of 1994 “forbids DI from covertly collecting non-military foreign intelligence”. The Act further outlines the goals of DI as:

“the enhancement of political and military decision-making with regard to national security and military related developments which may have an impact on the area

of strategic interest in general, and on the Republic of South Africa (RSA) in particular”.

As Cilliers and Reihardt (1995:180) put it, “DI is responsible for the implementation of counter-intelligence measures within the SANDF”.

In doing its work, DI supports the provisions of section 200(2) of the Constitution which states that:

“the defence force is to defend and protect the Republic, its territorial integrity and its people in accordance with the principles of international law regulating the use of force” (RSA Constitution: 1996).

According to the Department of Defence’s (DOD) Annual Report (2006/7:85) the measurable objective of DI is to “defend and protect South Africa by providing both military and counter-intelligence products and services that meet the requirements of government”. According to Dombroski in Bruneau and Boraz (2007:260), “DI is forbidden to participate in any non-intelligence-related operations as the SANDF does not have the power of arrest, detention, or search-and-seizure”. As described in the DOD’s Annual Report:

“the Chief of Defence Intelligence (CDI) is responsible for the provision of defence intelligence and counter-intelligence by establishing, training and maintaining a prepared military intelligence service, auxiliary service and facilities for the SANDF” (Department of Defence (DOD)’s Annual Report (2006/7:85).

1.2.3.3 Police intelligence

According to Masiapata (2008:10):

“the advent of democracy in South Africa saw the demise of apartheid institutions and ushering in of the democratic structures built around the explicit values of good governance, transparency, and accountability”.

As a result, the police service went through a multiplicity of changes including the process of democratisation, transformation and integration. According to Masuku (2005:40):

“the integration process was characterised by the recruitment of individuals belonging to other subsidiary and informal policing structures such as the kits- konstabels, the railway and municipal police, as well as the militant youth of the township Self-Defence Units (SDU) and the Self-Protection Units (SPU)”.

Further, Masuku (2005:40) indicated that, “the integration resulted in over 28 791 officers being amalgamated coming from the police services of the 10 homelands”. In order to advance the accountability of the police there was a concerted effort to change the police force into a police service. As Rauch (1998:26) indicated, one of the reasons was to encourage the participation of community members in matters of safety and security. Even though this process included the demilitarisation of the police ranks, the decision was reversed in 2009 with re-introduction of the military ranks in the police services.

According to Dombroski in Bruneau and Boraz (2007:259), “the new dispensation saw the Intelligence Division of the South African Police Service (SAPS) being responsible for the management, collection and analysis of both crime and counter-intelligence

10 matters”. Further, section 16 of the South African Police Service Act 68 of 1995, outlines the responsibilities of CI as “the critical tool for fighting crime, which is the biggest threat to the country’s security and the democratic order”. As Dombroski in Bruneau and Boraz (2007:259) put it, “crime intelligence is also responsible for the provision of technical support towards crime prevention and investigations”. To effectively support the National Crime Combatting Strategy (NCCS), Dombroski in Bruneau et al (2007:259) argues that the “Crime Intelligence Division of SAPS had to shift its emphasis from reactive to proactive intelligence”. In fact, CI is expected to investigate suspected criminal activities before crime is committed.

1.3 DEMOCRATIC INTELLIGENCE OVERSIGHT STRUCTURES

At the dawn of democracy, the South African government enacted the Intelligence Services Oversight Act 40 of 1994 to authorise the establishment of an intelligence oversight model. This was necessary to ensure the full compliance of members of the intelligence services with the country’s legislative framework (Dube, 2013:12). However, the development of the oversight model was not an easy task as the difficulties were also felt by many democratic states across the world. For instance, Gill (2007:14) recalled the nature of questions asked by many countries across the world and they included the following:

“Should external overseers be located in the legislature or in some other body? What should be their mandate or over what precisely should they have oversight? How are they to be chosen and by whom? Must they be vetted? What powers will they have to obtain access to intelligence personnel and files? To whom do they report?”

Given the complexities of these questions and the need for democratic intelligence oversight, the South African government opted for a multiple structural approach guided

11 by its governance model of ‘trias politica’. Given this approach, the country placed intelligence oversight in a number of structures operating at various levels of governance. These include the legislature, executive, the judiciary, administrative, and institutions supporting constitutional democracy. As provided by the Act, the democratic intelligence oversight structures are expected to monitor, oversee and supervise the work of the intelligence services. This was emphasised by Dlomo (2004:55) when he argued that, “a coherent oversight model should provide comprehensive safeguards to protect the country’s constitutional dispensation”. Against this background, it is important to now summarise the South African intelligence oversight model as having the legislative leg, the executive, judiciary, administrative and the state institutions, supporting constitutional democracy.

At the legislative level, democratic intelligence oversight lies with the JSCI. According to section 2 of the Intelligence Services Oversight Act (Act 40 of 1994), “the JSCI is a multiparty parliamentary committee consisting of 15 members allocated through the principle of proportional representation” (Netshitenze: 2005). Since 1994, the ruling party has consistently chaired the JSCI. In Germany, the parliamentary oversight committee was chaired by the ruling party for the first year of its existence and the minority party took the chair after that (Dietrich: 2015). This is what is critical for avoiding the exploitation of situations for the advancement of party political agendas. Given the limited membership of the JSCI; the South African model forces parliament to always set up an ad hoc committee when it considers intelligence legislation. This is done to allow the participation of other political parties in the legislative processes of this important organ of state (Parliamentary Monitoring Group: 2015).

At the executive level, the Minister of State Security as appointed by the President is responsible for conducting intelligence oversight and control on members of the intelligence services. Section 209(2) of the Constitution empowers, “the President to appoint a member of Cabinet to assume the political responsibility to control and direct

12 the work of the intelligence services”. In turn, section 5(a) of the National Strategic Intelligence Act 39 of 1994, “empowers the Minister to do everything necessary for the efficient functioning of the intelligence services”. In consultation with the JSCI, the Minister of State Security is empowered by section 5(a) of the National Strategic Intelligence Act 39 of 1994 “to make regulations on matters of intelligence coordination, its production and dissemination”. Further, section 5(1)(a) of the same Act “authorises the Minister of State Security to advise the President and the national executive on the coordination of national strategic intelligence”.

At the judicial level, intelligence oversight work is conducted by an appointed judge whose primary function is to authorise the implementation of special intelligence operations of an intrusive nature. According to section 16(1) of the Regulation of Interception of Communications and Provision of Communication-related Information Act 70 of 2002 (RICA), “members of the intelligence services should seek judicial approval prior to the implementation of critical operations, particularly those that have the potential to infringe on people’s rights to privacy”. As such, the primary responsibility of judicial oversight is to grant warrants and authorise the use of intrusive methods of investigation. As Hannah, O’Brien and Rathmell (2005:5) put it, “judicial oversight ensures the legal compliance of the operations undertaken by members of the intelligence services”. This assertion implies the full observation of the country’s legislative and regulatory framework during the performance of one’s allocated responsibilities.

At the administrative level, the IGI is entrusted with the responsibility for ensuring the democratic oversight of members of the intelligence services. This is provided for in

13 section 210(b) of the RSA Constitution which indicates that:

“the Inspector-General is appointed by the President as head of the national executive following the approval of the resolution by the National Assembly with a supporting vote of at least two thirds majority”.

According to section 7(2) of the Intelligence Services Oversight Act of 1994, “the Inspector-General shall be a South African citizen who is fit and proper to hold such office with good knowledge of intelligence matters”. The IG is empowered by section 7(7)(c) of the Intelligence Services Oversight Act, 1994 “to receive and investigate complaints from members of the public, intelligence services and the JSCI”. More importantly, section 7(7) of the Intelligence Services Oversight Act, 1994 empowers the IG to:

“Oversee the compliance of intelligence services to the Constitution, applicable laws and all other relevant policies on intelligence and counter intelligence matters”.

Like judicial oversight, even administrative oversight monitors full compliance to the provisions of the country’s constitutional order.

In addition to legislative, executive and judicial oversight mechanisms, the democratic dispensation allows for the additional oversight of intelligence services by some of the chapter 9 institutions (also known as state institutions supporting constitutional democracy). In this case, special focus will be directed to the work of the PP, the AG and the HRC. They are by law responsible for overseeing the work of the organs of state of which intelligence service structures are no exception. Their role was emphasised by Van der Merwe and Du Plessis (2004:84) when they indicated that, “chapter 9 institutions act as watchdogs over government institutions by investigating

14 misconduct and monitoring the implementation of their respective legislative mandates”. As the law asserts, the work of chapter 9 institutions cuts across all organs of state and this includes the intelligence services themselves. This is not unusual, hence, Tsholofelo (2014:4) indicated that, “…any democratic dispensation should not exempt any organ of state from the oversight mechanism”. Despite national security imperatives, the legislative framework does not allow selective application, but does provide for certain operational exemptions or limitations based on certain peculiar circumstances.

Based on the above discussion, it is clear that democratic South Africa created robust intelligence oversight architecture with varied mandates and legislative powers. However, the model has major theoretical loopholes characterised by the absence of a clear cooperation framework in the fulfilment of its legislative mandates. The following discussion presents the problem statement which constitutes the focus and also played a fundamental role in underpinning both the relevance and significance of this study.

1.4 PROBLEM STATEMENT

In his interview with Marianne Thamm of Daily Maverick as reflected on the newspaper’s website (www.dailymaverick.co.za) on 8 June 2015, Laurie Nathan, the director of the Centre for Mediation at the University of Pretoria argued that:

“the South African intelligence services suffer from unreconstructed apartheid mentality and remain immersed in obsessive secrecy that precludes accountability and oversight”. He further indicated that, “party politics make them a threat to the country’s constitutional democracy”.

The former Minister of Intelligence Services, Mr Ronnie Kasrils had earlier shared similar sentiments as reflected on the website of the SSA (www.ssa.gov.za) by

15 indicating that members of SSA are caught “in the grip of political paranoia”. These sentiments came at a time when a dubious ‘blog’ published allegations that the PP, Advocate Thuli Mdonsela, the Commander in Chief of the Economic Freedom Fighters (EFF) Mr Julius Malema, and the former parliamentary leader of the Democratic Alliance (DA) Ms Lindiwe Mazibuko, were agents of the American Central Intelligence Agency (CIA) (Daily Maverick: 2014). Of serious concern was the sentiments expressed by the sitting Regulation of 2002 (RICA) appointed Judge Ms Yvonne Mokgoro who in 2013 indicated that, “intelligence services apply lax approach towards RICA application, as they cut too many corners and made her work more difficult” (Daily Maverick: 2014:6).

Born and Leigh (2005:17-21) state that “the work of the intelligence services is the fundamental and indispensable element of any democratic state”. However, any failure to effectively oversee their work has the potential to undermine the same democratic order it is meant to protect. According to Dietrich (2015), most scholars undertook studies on the international and domestic coordination of intelligence services when conducting their intelligence work. However, little focus has been placed on the coordination of intelligence oversight structures when conducting their oversight work on members of the intelligence services. Therefore, in this study, the problem statement emerged as follows: The South African democratic intelligence oversight structures are fragmented, uncoordinated, uncollaborated and operate in silos, thereby playing a minimal role in the promotion of good governance. This problem embraces three critical areas of concern, namely: fragmentation, lack of coordination, lack of cooperation and collaboration. The following discussion unpacks the three problematic themes in detail.

1.4.1 Fragmentation

Given the multiplicity of oversight structures and their work spread across various governance levels, literature indicates that they operate in silos and are heavily fragmented. This was demonstrated by certain occurrences in various cases. For instance, on 12 December 2009 Mr David Maynier, the Shadow Minister of Defence and Military Veterans of the DA and member of the JSCI wrote a letter to the IGI, Advocate Radebe. In the letter, he requested the IGI to investigate the granting of a security clearance certificate to Mr Paul Ngobeni, who at the time served Ms Lindiwe Sisulu, the Minister of Defence and Military Veterans in the capacity of special adviser (available on the website of the Parliamentary Monitoring Group). In her response, the IG indicated that such a complaint does not fall within her office’s mandate as provided for by section 7(7)(c) of the Intelligence Services Oversight Act of 1994.

In this case, the IGI’s response failed to refer the complainant to the relevant institution which is legislatively authorised to conduct such an investigation, to address the matter. As Alford and Fiedland (1985:110) put it, “the separation of powers and observation of equality before the law are critical instruments for building trust in a country’s regulatory quality and the proper legal framework”. This incident demonstrates the fragmentation and non-cooperation of the various structures tasked with overseeing members of the intelligence services. Based on this, it is clear that the various structures tend to concentrate on their narrow legislative mandates with little concern for the responsibilities of other oversight structures.

1.4.2 Lack of coordination

During his study of intelligence evolution in South Africa, Dube (2013:18) concluded that the “lack of institutionalised coordination mechanisms amongst intelligence oversight structures is a major limitation in the country’s oversight agenda”. This problem was

17 evidenced by the failure of these structures to identify the irregularities which emanated from the 2005/6 Project Avani scandal at the NIA (now the SSA DB) (Office of the Inspector-General of Intelligence (OIGI) 2006:5). As indicated by the OIGI (2006:5), this scandal revolved around the:

“unlawful interception of telephone calls, unauthorised surveillance of individuals and unwarranted circulation of e-mails damning factional political opponents by members of the intelligence services”.

These unlawful activities were conducted under the legally constituted Project Avani, whose primary objective was to evaluate the nature of the threat posed by the ensuing discussions on the presidential succession within the ruling ANC and the escalating service delivery protests (OIGI 2006:6). In the wake of the scandal, President Mbeki (2005:3) addressed the civilian Intelligence Services Day and indicated that:

“members of intelligence services should remain politically non-partisan and may not carry out operations which seek to undermine, promote or influence any political party or faction at the expense of the other” (www.SSA.gov.za).

As the OIGI (2006:6) puts it, “the scandal exposed the misuse of state resources, positions, power and subverted national intelligence apparatus for the gratification of personal and/or group political interests”. It also contradicted the provisions of the White Paper on Intelligence of 1995 which emphasised:

“the creation of effective, integrated, and responsive intelligence machinery that can serve the Constitution and the government of the day through the provision of relevant, credible and reliable intelligence”.

Dube (2013:29) adds that the “Project Avani scandal showed serious weaknesses in the design and the coordination modalities of the South African oversight model”. It also revealed gaps in the democratic intelligence oversight systems and the general governance of the intelligence services (Dube 2013:29). Although section 210 of the Constitution provides for strategic intelligence coordination, there is no legislative prescript that mandates the systematic coordination of intelligence oversight. Therefore, the need for integrated and systematic institutional oversight mechanisms for members of the intelligence services cannot be overemphasised.

1.4.3 Lack of cooperation and collaboration

In addition to problems of fragmentation and lack of coordination, the intelligence oversight structures lack effective cooperation and collaboration with each other. For instance, over the years the AG consistently failed to cooperate with the OIGI, until recently. During its audit work, the AG identified various limitations which hampered the effective auditing of intelligence sector accounts. As a result, the AG has consistently added a qualification to the intelligence services’ audit report as follows: “owing to the nature of certain transactions and the circumstances under which they are incurred and recorded as well as the circumstances under which assets and services are procured and utilised, the level of audit assurance will often be lower than is normally the case with ordinary audits” (Auditor-General Report 2010/2011).

This qualification demonstrates the frustrations of the AG in auditing the intelligence services’ accounts. This is based on the complexities associated with the utilisation of funds in the secret services accounts relating to the payment of sources, front structures and other covert operations. This situation reflects the failure of the AG to cooperate and collaborate its efforts with that of the IG and contradicts the provisions of the Public Finance Management Act 1 of 1999 which requires financial accountability and responsibility from all organs of state. The AG should have engaged the IG in order

19 to request its assistance in accessing certain financial information that was critical for the audit. As Africa (2006:124) indicates, “the IG has unfettered access to the information held by the intelligence services”. This problem could be easily solved by the effective collaboration between the offices of the IG and that of the AG. This was emphasised by Bryson, Crosby and Stone (2006) who indicated that the complexities of today’s challenges can no longer be solved by an individual organisation operating in silos. In addition, Thomson and Perry (2006) cited the increasing structural interdependence and dwindling resources as key global factors driving the increasing levels of collaborations among entities. Therefore, effective cooperation between the AG and the IG cannot be overemphasised (Parliamentary Monitoring Group: 2015). Since the 9/11 tragedy, Agranoff (2012:282) indicated that, “the US government emphasises collaborative arrangements amongst its structures to accomplish its public missions”. The incident taught the US government the significance of cooperation, coordination and collaboration among various structures in discharging their varied legislative mandates.

In its submission to the Ministerial Review Commission (2008: 214), the then SASS (now an FB of the SSA) complained of too many overlaps and duplications between the oversight work done by the AG and that of the IG. It complained of the lack of cooperation between the AG and IG when doing their oversight work on members of the intelligence services. It further submitted that the two structures put no effort into coordinating their functions, thus resulting in multiple reviews being undertaken by different structures while expecting the same reports from the same intelligence service structures throughout the year (Ministerial Review Commission, 2008: 214). This situation forces the intelligence services to put more effort into responding to the requirements of the oversight structures than focusing on their core business of intelligence production.

Further, the JSCI is expected to refer complaints of a human rights nature to the HRC responsible for the protection of human rights for further processing and reporting. However, the level of cooperation between the JSCI and the HRC has not yet been tested, as the JSCI has not yet received any complaint of human rights abuses by any member of the intelligence services (Records of JSCI deliberations retrieved from the Parliamentary Monitoring Group website) (Parliamentary Monitoring Group: 2015).

Despite these issues, it should be noted that these kinds of problems are prevalent in many democracies across the world. In 2012, Halchin and Kaiser (2012:2) “described the US intelligence oversight mechanism as dysfunctional and counter-productive”. This is a clear reflection of the nature of the challenges facing intelligence oversight work across the world. Dietrich (2015:11) argues to the contrary that “the German intelligence oversight structures complement each other and, in some cases, fully cooperate and collaborate their efforts”.

1.5 PURPOSE OF THE RESEARCH

The purpose of this research is to develop an improved systematic cooperation model to enhance the cooperation of democratic intelligence oversight structures when overseeing members of the intelligence services. This is done to facilitate their contribution to the promotion of good governance in the country. To achieve this, a number of things have to be done. The first is to evaluate the current state of cooperation among the intelligence oversight structures. As described earlier, this evaluation will cover the democratic intelligence oversight structures operating at parliamentary, executive, judicial, administrative and institutions supporting constitutional democracy. The second is to assess the nature of barriers hampering their cooperation. The third is to present recommendations on the best cooperation approach to be used by the various oversight structures when overseeing members of the intelligence services.

According to Born and Leigh (2005) and Caparini (2002) it is up to every democratic society to design and implement a vibrant democratic intelligence oversight mechanism. It should be made up of institutions which are committed to the key tenets of cooperation, coordination and collective effort. This suggestion is necessitated by the Intelligence Services Act, 1994 having authorised members of the intelligence services to use intrusive measures when implementing their operations. Ultimately, the primary objective of the oversight mechanism is just to ensure that members of the intelligence services discharge their responsibilities in full compliance with the country’s established legislative and regulatory prescripts. These prescripts include, among others; the RSA Constitution of 1996, the Intelligence Services Act of 1994, the National Strategic Intelligence Act of 1994, the White Paper on Intelligence of 1995 and the Intelligence Services Oversight Act of 1994.

With the enactment of these legislative platforms, the primary intention was to end up with the oversight structures seeking to contribute to the realisation of good governance results. Therefore, the theoretical consideration of this study will be based on both systems theory and good governance theory. The approach seeks to infuse the work of the democratic intelligence oversight structures and the preservation of the country’s constitutional order. In describing the importance of intelligence services, both Agere (2000) and Frederickson and Smith (2003) argue that, government requires intelligence analysis as it needs more than raw data and generic reports to do its work. The White Paper on Intelligence (1995:5) expects the intelligence services to “deliver the analytical base and required knowledge for assisting policy makers in their policy making processes”.

Van Niekerk, Van der Waldt and Jonker (2001b:56) add that, “the intelligence services play an important role towards the realisation of effective, efficient and productive policy-making processes”. In doing its work, the intelligence services guarantee human

22 security, human rights, freedom and the territorial integrity of the state. According to section 2(1)(2) of the National Strategic Intelligence Act 39 of 1994, the intelligence services are responsible for:

“the collection and analysis of information in order to detect and predict both internal and external security threats to national stability, and inform and advise the executive about the nature and the causes of such threats”.

This is important for their consideration by the policy-makers during policy-making processes. To guarantee the successful realisation of these functions, Born and Leigh (2005) had earlier indicated that members of the intelligence services require the application of a rigorous oversight mechanism by the vibrant cooperating structures.

1.6 RESEARCH QUESTIONS

As indicated by Schwella (1991), Punch (2005a) and De Vos (1998), any research inquiry requires the development of questions to address the problem and guide the research process. The following open questions attempt to provide answers to the above rationale, problem statement and guiding research question below:

 Is global benchmarking of democratic intelligence oversight systems helpful?  What does international best practices on intelligence oversight systems entail?  What is the rationale for and state of cooperation amongst intelligence oversight structures in general?  What does the role of the South African intelligence architecture and oversight structures entail?  What is the current state of cooperation among intelligence oversight structures in South Africa?

 What are the barriers to effective cooperation of intelligence oversight structures globally and in South Africa?  What are the strategies for effective cooperation of intelligence oversight structures?  Is there an existing approved theoretical framework or cooperation model to fulfil statutory and regulatory mandates and good governance criteria in South Africa, and what does it entail?  Is there a need for an integrated institutional oversight mechanism to oversee the members of the intelligence services in South Africa?  How can an improved conceptual and theoretical framework for the functioning of oversight structures be established?  How can a systematic cooperation model to guide cooperation of democratic intelligence oversight structures be developed?  What does systems theory and its application entail?  What does good democratic governance entail in terms of a public management tool and as a means to an end?  How is the South African democratic intelligence oversight architecture structured on legislative, executive (encompassing administrative) and judicial levels?  Which institutions and role players are responsible for intelligence oversight?  What is the role of these institutions and role players?  Who are the other constitutionally mandated institutions and stakeholders overseeing intelligence oversight in South Africa?  What is the rationale for effective intelligence oversight cooperation?  What does fourth generation warfare entail?  What does the context of intelligence oversight mandates entail?  Is there a need for transparency and public accountability in terms of intelligence oversight mechanisms?

 What is the current state of cooperation and oversight processes among oversight structures in South Africa?  What does the cooperation role of chapter 9 institutions entail?  What are the structural and operational barriers to improved cooperation of intelligence oversight structures?  How can the key intelligence oversight institutions and structures be reshaped/refocused in South Africa to improve the status quo?  What are the strategies to improve key intelligence cooperation structures?  How can an improved analytical model be developed to improve cooperation for intelligence oversight structures in South Africa?

Therefore, given the background, the problem and the purpose of this study, the main research question emerged as follows: How can the cooperation of the South African democratic intelligence oversight structures be improved for good governance results?

According to De Vos (1998 in Auriacombe, 2009:837) “coding represents the operations by which data is broken down, conceptualized and put back together in new ways. It is the central process by which theories are built from data”. For Charmaz (2007 in Auriacombe, 2009:837) “coding means categorizing segments of data [open questions] with a short name that simultaneously summarizes and accounts for each piece of data [or open question]. Your codes show how you select, separate, and sort data to begin an analytic accounting of them”.

In view of the above coding process, the open questions were sorted to respond to the main guiding research question, and the following three sub-questions emerged from

25 the coding process:

 What is the rationale for, and state of cooperation among the democratic intelligence oversight structures for good governance results?  What are the barriers for the effective cooperation of democratic intelligence oversight structures for good governance results?  What strategies can be developed to ensure the systematic cooperation of democratic intelligence oversight structures for good governance results?

Further, in order to properly respond to the above-mentioned sub-questions I developed various interview questions for consideration by the oversight, audit, and external experts. The questions are attached to this thesis as Appendices A and B.

1.7 RESEARCH OBJECTIVES

To adequately respond to the research questions outlined above, this research focuses on the realisation of the following objectives, namely:

 Establish the rationale and current state of cooperation among the South African democratic intelligence oversight structures.  Identify the nature of barriers for the systematic cooperation of democratic intelligence oversight structures.  Establish the nature of strategies to be adopted to improve the cooperation of democratic intelligence oversight structures.

Agere (2000:95) argues that, “when the laws of the country are disregarded and ultimately get flouted with impunity, especially by those in authority, it causes uncertainty and creates a state of near-anarchy”. Therefore a properly functioning and vibrant intelligence oversight mechanism would in essence deter any possible abuse of

26 power or anarchy as described by Agere (2000). In Germany, Dietrich (2015:65) confirms that “issues of administrative intelligence oversight and coordination are taken quite seriously”. This is critical to correct any illegal conduct by any member of the intelligence services.

1.8 RESEARCH ETHICS

In social sciences research, it is very important to ensure the full observation of ethical protocols and procedures in the research process. In this research, I have fully observed all relevant protocols and requirements for research ethics as provided for by the Ethics Committee of the University of Johannesburg. In terms of possible conflict of interest, I would like to confirm that I am currently working for NICOC as the Programme Manager responsible for strategic domestic intelligence coordination. Though working within the intelligence services, this research did not focus on members of the intelligence services, but on the intelligence oversight structures whose mandate is to oversee the work of members of the intelligence services.

Further, I should emphasise that the primary focus of this research is to assess the manner in which the intelligence oversight structures cooperate with each other during the implementation of their varied legislative mandates. Further, I should confirm that the research did not use any classified information from the intelligence services or oversight structures and this matter is fully discussed in Chapter 3 which outlines the manner in which the research unfolded. However, written approval was nonetheless granted allowing me to conduct the research in the intelligence environment.

During the research process, I tried to remain objective and impartial as required by the ethics protocols of the social sciences research. I have also tried to minimise or exclude any potential bias during my data collection, analysis, assessment and presentation of empirical data. In addition, I have taken steps to ensure that necessary

27 precautions are taken to ensure both the validity and reliability of the research and this is fully discussed in Chapter 3 under section 3.5.

1.9 SIGNIFICANCE OF THE STUDY

As Peters and Savoie (2000:56) put it, there is no question about the significance of intelligence oversight in the context of the governance discipline. This significance is predicated on the government’s continued possible use of intelligence structures in pursuit of their national interests in this competitive global world. Similarly, it will remain significant for so long as individuals can still use it to acquire business interests or other income generating measures for their own selfish ends. This is even more significant as members of the intelligence services themselves can use it to intimidate their opponents, instil fear and influence public perceptions. In addition, Frederickson and Smith (2003:1) argue that “public administration plays a critical role in the advancement of the activities of the human race”. The success of these activities can be attributed in no small part to the role and work of intelligence in supporting the government’s policy- making agenda.

More importantly, Dlomo (2004) indicates that the effective implementation of a vibrant democratic intelligence oversight mechanism has the potential to place the intelligence services under strict public scrutiny and ensures a better system of checks and balances. This mechanism has the potential to improve the intelligence services’ work and restore public confidence in the policy-making process. The global movement towards democratisation has contributed to the need for overseeing intelligence services. The significance of democratic intelligence oversight mechanisms gained prominence in many countries across the world as a result of various developments; most of which were negative.

Those developments are explained by Born and Leigh (2005: 1) as:

“scandals have in many states provided the impetus for change in the governance of intelligence services. This was the case in Australia, Canada, Norway and the US, where legislative and public investigatory committees exposed human rights abuses and pushed for strengthened intelligence oversight and control systems. However, constitutional reforms (e.g. South Africa), transition to democracy (e.g. Argentina, South Korea and Poland) and legal challenges brought by citizens (e.g. in The Netherlands, Romania and the United Kingdom (UK)) were other reasons why governments introduced intelligence oversight mechanisms”.

The issue of intelligence oversight has since 2006 become an international norm in the management of intelligence services across the world and in South Africa. As Born and Leigh (2005: 2) confirmed, “the model received the backing of international bodies such as the parliamentary assemblies of the Council of Europe and the Western European Union”.

According to Lowenthal (2003:4), the primary purpose of intelligence work is “to support the policy-maker, avoid strategic surprises, and provide long-term expertise on various issues of interest to the government”. However, in numerous states the intelligence services flouted these norms and became the instrument of individuals for the accomplishment of personal interests. The intelligence services of totalitarian and/or authoritarian regimes target and suppress the perceived political opponents of their regimes. Lowenthal (2003:5) indicated that “noteworthy examples include the former socialist regimes in East-Central Europe; the Komitet Gosudarstvennoy Bezopasnosti (KGB) in the former Soviet Union; the Stasi in the former German Democratic Republic; the Securitate in Romania under President Nicolae Ceausescu; and the Intelligence Services of Iran, Cuba, Chile, Nicaragua and Honduras, among others”. Given their repressive activities, these intelligence services gained a measure of infamy in their

29 respective societies as they penetrated “many spheres of social activity and systematically abused human rights” (Lowenthal 2003:5).

Based on these assertions, the significance and relevance of this study in the broader context of governance as a discipline cannot be overemphasised. In terms of South Africa, just reading the problem statement in section 1.4 above, the significance of this research remains unquestionable. The additional relevance of this research is reflected in its contribution to the body of knowledge discussed below.

1.10 CONTRIBUTION TO THE BODY OF KNOWLEDGE AND PROFESSIONAL PRACTICE OF INTELLIGENCE COORDINATION

In the past decade, governance gained prominence in disciplines such as political science, public administration and management (Agere: 2000). On the other hand, intelligence gathering is regarded as one of the oldest professions of human kind. However, intelligence oversight is a fairly new phenomenon. It gained prominence with the exposure of government’s repressive acts and the expansion of democratic practices across the world. The following literature review exposes minimal academic research in the field of intelligence oversight particularly on the African continent. There is however, sizable work that has been done in some developed countries of the world such as Australia, the US, the UK and Canada.

Good governance is according to Stoker (1998) and Nzongola-Ntalaja (2002), conceptualised as the reflection of effective rules, institutions and good democratic governments. To fulfil these requirements of good governance, intelligence tradecraft needs to implement a robust intelligence oversight mechanism. Unfortunately, reviewed literature reveals the existence of minimal interest in this subject hence few scholars conducted research on this discipline. While studying parliamentary oversight, Dlomo (2004) found a disjuncture between the work of the JSCI and that of the IGI on

30 intelligence oversight. Netshitenze (2005:101) acknowledged the significance of intelligence oversight, but concluded that the implementation of intelligence oversight is a very daunting task and asserts that oversight can be achieved when effective cooperation mechanisms have been developed for observation by the various democratic intelligence oversight structures (Netshitenze, 2005:101).

Therefore, the epistemological contribution of this study to the governance discipline and the professional practice of intelligence coordination is the development of a systematic cooperation model for guiding the cooperation of democratic intelligence oversight structures in doing their work. This is critical for the improvement of their contribution to the promotion of good governance. This approach emerged from Gladden’s (1961:12) assertion of the “need to interrogate the contribution of intelligence services in the policy-making and good governance agenda”.

This is also strengthened by Agere’s (2000:8) sentiments that, “the absence of an effective intelligence oversight mechanism could breed corruption, which could later breed poor governance”. The existence of poor governance serves as a major constraint for economic development and national security. Sustaining democratic governance is dependent on the ability to balance secrecy as the tenet for intelligence practice with the need for transparency and accountability as tenets of good governance (Sendall: 1997) and (Africa: 2009). This is an important benchmark as described by scholars like Fitzgerald, Mclennan and Munslow (1995); Fox and Meyer (1996); and Waldo (1994); for sustaining emerging democracies.

1.11 DELIMITATION OF THE STUDY

Without discounting the importance of intelligence services and the work they do, this study only focuses on the nature of cooperation among the democratic intelligence oversight structures when overseeing members of the intelligence services. As

31 described by Dietrich (2015), the coordination of intelligence services when conducting their strategic intelligence work has been fairly well studied. Furthermore, Born and Masevage (2012) indicate that the other studied area relates to the cooperation of intelligence oversight structures between countries. However, there has been little effort towards understanding the coordination of the various oversight structures during the performance of their domestic oversight work. This is also complicated by the fact that some countries do not have wide range intelligence oversight structures that South Africa has.

Democratic states should commit to examine not only the legislative mandates and the powers of oversight structures, but even the methods they use, the quality of their reporting, and the adequacy of their expertise and resources towards the performance of their oversight work. Looking at these various areas, this research focuses on issues of cooperation in general and, on cooperation of intelligence oversight structures when performing their oversight responsibilities on members of the intelligence services, in particular.

1.12 THESIS STRUCTURE

This thesis is divided into seven chapters and the following discussion provides a brief outline of the contents of each chapter.

1.12.1 Chapter 1: General background and scientific orientation to the study

This chapter contextualises the research by explaining the background and rationale to the research. It further provides the problem statement, the research purpose, research questions and research objectives of the study. The chapter contextualises the significance of the study, its contribution to the body of knowledge and professional

32 practice of intelligence coordination, and provides the chapter breakdown structure outlining the contents of each chapter.

1.12.2 Chapter 2: Contextual and theoretical variables influencing good international intelligence oversight systems

This chapter presents the findings of the global benchmarking exercise after a review of the international experiences of democratic intelligence oversight versus the South African oversight model. The chapter starts by providing the results of the global benchmarking exercise on the intelligence oversight work undertaken by Australia, Canada, New Zealand, the UK and the US, which comprise the existing best international practices. The chapter concludes with the presentation of a conceptual framework which guided the implementation of the research project.

1.12.3 Chapter 3: Methodological approach to the study

This chapter presents the selected research paradigm followed by the provision of clear reasons for choosing the paradigm appropriate for the purposes of this thesis. It continues to summarise the adopted research design followed by the justification of its relevance to the study. Further, this chapter presents a summary of the procedures which were followed in the research process. These include: data collection instruments, target population, sampling, data presentation and its envisaged analysis techniques. It then concludes by summarising the steps undertaken to ensure the research validity and its reliability, in order to substantiate the validity and credibility of the research findings and conclusions.

1.12.4 Chapter 4: Structural cooperation among democratic intelligence oversight structures in South Africa

Before assessing the cooperation of intelligence oversight structures, this chapter starts by summarising and assessing the current South African intelligence oversight model. It then assesses the research findings in response to the first research sub-question seeking to determine the rationale for, and current state of cooperation among the democratic intelligence oversight structures.

1.12.5 Chapter 5: Barriers to improved cooperation of the democratic intelligence oversight structures in South Africa

This chapter presents the research findings in response to the second research sub- question which seeks to establish the nature of cooperation barriers experienced by members of the democratic intelligence oversight structures in doing their work

1.12.6 Chapter 6: Strategies to improve the cooperation of democratic intelligence oversight structures in South Africa

This chapter presents the research findings in response to the third research sub- question which seeks to establish the nature of strategies to be adopted to improve the systematic cooperation among the democratic intelligence oversight structures.

1.12.7 Chapter 7: Proposed improved cooperation model for intelligence oversight structures in South Africa

Following the presentation of data on the three research sub-questions, this chapter presents its analysis and interpretation using qualitative, descriptive and explanatory analysis techniques. The analysis is expanded into four categories in accordance with

34 posed research sub-questions. After data analysis, this chapter provides an overview of the study propositions by drawing conclusions and recommendations from the assessment of the findings. The chapter concludes with the presentation and explanation of a proposed systematic institutional cooperation model for utilisation by democratic intelligence oversight structures. Towards the end, this chapter provides areas requiring further research and future interventions.

1.13 SUMMARY AND CONCLUSIONS

This chapter provided a summary of the study context by mapping the general background of the study, what the problem statement is, and how it will be addressed in terms of a specific research design and methodology. The discussion started with the introductory summary of the South African intelligence architecture and its democratic intelligence oversight mechanisms. This is important for the proper contextualisation of the study and a clear understanding of the difference between the overseen (intelligence services) and the overseers (oversight structures).

This was followed by clarifying the relevance and significance of the research for improving the current body of knowledge and professional practice of intelligence coordination through the different existing intelligence oversight structures. The chapter concludes with the presentation of a blueprint of the entire study by outlining the different contents of the various chapters of the thesis. Having understood the problem, the following chapter presents the findings of the literature review which focused on the good international intelligence oversight governance systems and practices.

CHAPTER 2 CONTEXTUAL AND THEORETICAL VARIABLES INFLUENCING GOOD INTERNATIONAL INTELLIGENCE OVERSIGHT GOVERNANCE SYSTEMS

2.1 INTRODUCTION

This chapter presents the findings of a literature review of different international intelligence oversight governance systems. The review was done as a global benchmarking exercise contextualising the state of international oversight experiences. During this process, I focused on various international best practices as applied in selected advanced democracies of the world.

The reviewed literature showed limited research and scholarly interest in the area of intelligence oversight, particularly on the African continent. Nonetheless, scholars and researchers such as Africa (2012 and 2009; Africa and Mlombile 2001), Netshithenzhe (2005), Dlomo (2004), Dube (2013) and Nathan (2010) have conducted some academic work on the functioning of intelligence services and their oversight platforms in South Africa. Globally, literature confirms the existence of a sizable scholarly body of work on intelligence tradecraft and oversight mechanisms especially in the developed countries of the world. These include countries such as Canada, the US and the UK.

The post-Cold War era saw the spread of democracies around the world and the emergence of asymmetric threats to the populace. These threats emphasised the relevance of intelligence tradecraft in assisting governments to develop mechanisms to mitigate the possible impact of these threats and support government’s efforts towards the achievement of good democratic governance. Most of these democracies are based on respect for the rule of law, people’s rights and fundamental values and civil liberties. To guarantee these values and ensure the maintenance of national security, efforts were made to establish democratic intelligence oversight structures to oversee

36 members of the intelligence services.

This chapter, starts by providing the results of a literature review that was conducted as part of the global benchmarking exercise on democratic intelligence oversight. The presentation includes data on the evolution of intelligence oversight in the selected countries.

2.2 GLOBAL BENCHMARKING OF DEMOCRATIC INTELLIGENCE OVERSIGHT

Following the end of the Cold War era after the collapse of the Berlin Wall, many countries observed major shifts and defining moments for intelligence tradecraft (Gerdes, 2004:157). This was accelerated by the implosion of the Soviet empire, the re-emergence of nationalist movements and religious fundamentalists, the problems of failed states, as well as the international networks of drug cartels and crime syndicates (Hedley, 1995:96). Assessing these shifts, scholars such as Lowenthal (2003), Filip (2006), Gill and Phythian (2006), Africa (2002), Dlomo (2004), Bruneau and Dombroski (2001) and Todd and Bloch (2003), began to emphasise the need for more sophisticated intelligence services with vibrant oversight mechanisms. These shifts were fast-tracked as a result of the emergence of suicide bombers whose activities became evident during the 9/11 attack on the World Trade Centre in the US (Africa, 2002:28). The need for intelligence oversight sought to ensure that intelligence work contributes to the realisation of good governance results. At this time, various countries adopted different intelligence oversight models, while others strengthened their long- existing ones. Therefore, the discussion focuses on the countries with long-existing intelligence oversight mechanisms which just had to be strengthened post-9/11.

According to Todd and Bloch (2003:209), the Australians adopted their civilian intelligence oversight and control mechanism in 1986 by setting up the OIGI. Subsequently, the Netherlands adopted their civilian intelligence oversight and control

37 and placed it under the parliamentary ombudsman (Todd and Bloch 2003:209). Further, Born and Caparini (2007:1-2) state that Germany adopted its intelligence oversight model and placed it under four parliamentary bodies loosely coordinated by the bipartisan Bundestag Control Commission. This new paradigm reached a number of countries such as Canada, Norway, the US, Argentina, South Korea, Poland, Romania and ultimately South Africa in the post-1994 period. On the other hand, Bruneau and Dombroski (2001:13) indicated that, “despite the new paradigm, some countries found it difficult to embrace the changes which seek to enforce intelligence accountability by continuing to embrace excessive secrecy during intelligence operations”. This is the case in many countries of the world including some African countries.

However, with the African Union’s (AU) establishment of the Committee of Intelligence and Security Services for Africa (CISSA) in 2002, there is now hope that African states will embrace the new paradigm and institutionalise oversight mechanisms for their intelligence services (African Union: 2010). During the literature review, it became clear that some countries have adopted intelligence oversight mechanisms which have been tried and tested over many years. Notably, such countries include Australia, Canada, New Zealand, the UK and the US. As a result, the following assessments present the results of the global benchmarking exercise which was centred on the oversight mechanisms of the selected five countries. This is an important exercise for the purposes of comparing the South African democratic intelligence oversight architecture with those of developed democracies of the world.

2.3 SELECTED INTERNATIONAL BEST PRACTICES ON INTELLIGENCE OVERSIGHT

From the global benchmarking of democratic intelligence oversight, five countries stood out as the key role models for consideration and those were Australia, Canada, New Zealand, the UK and the US. However, concerning these countries, the question may

38 arise why they were selected. Or, one may even ask why there is no developing country or even one African country among the selected countries. Firstly, it should be noted that even if intelligence gathering is regarded as one of the oldest professions of human kind, the issue of intelligence oversight is a fairly new phenomenon (Born and Caparini: 2007). As already discussed in Chapter 1 under section 1.9, when describing the significance of this study; the question of intelligence oversight was embraced by a few countries at their various transitional intervals. Unfortunately, most of the developing countries including African states, embraced this concept very late and in certain cases not at all. Therefore, in my quest to review good international intelligence oversight systems, I had to consider countries which have already embraced the systems and tested their applicability over time.

Therefore, the selected five countries have adopted and been applying their intelligence oversight systems for some years now. They therefore constitute good reference points for the strong democratic and good governance systems which South Africa attempted to emulate through its world renowned democratic Constitution of 1996. This commitment was further reflected in the country’s robust intelligence oversight machinery as adopted in the post-1994 transitional period. The selected five countries embraced the resilient oversight practices which are more comparable and tangible to assess than those of our Brazil, Russia, India, China and South Africa (BRICS) partners and/or our fellow African counterparts. In fact, most of these selected countries’ intelligence oversight systems evolved and improved many times since their adoption and this will be fully demonstrated by the summaries and assessments below.

The main focus is on the cooperation modalities of the intelligence oversight structures when overseeing members of the intelligence services. There are three main determinations to be made after the presentation of findings on the five selected countries. First, efforts are made to understand the state of cooperation among the various intelligence oversight structures. Second, it is important to understand the

39 nature of the barriers they face when attempting to cooperate with each other. Third, it is important to analyse the kind of strategies they adopted and used to improve their cooperation modalities over the years.

After the presentation and assessments of the comparative benchmarking findings, the chapter concludes with the formulation of the two theoretical approaches used in the interpretation of the research findings and the development of the conceptual framework for the rest of the study. Important policy and statutory provisions which established and regulate the functionalities of oversight structures will be directly quoted from the original sources.

2.3.1 Australia

According to the Australian legislative framework, there are six intelligence and security structures as mapped out by Flood (2004:53). They are:

“the Australian Security Intelligence Organisation (ASIO), the Australian Secret Intelligence Service (ASIS), the Defence Signals Directorate (DSD), the Defence Imagery and Geospatial Organisation (DIGO), the Defence: Intelligence Organisation (DIO) and the Office of National Evaluations (ONA) (Flood, 2014:53)”.

According to Flood (2004:53), “Australian intelligence oversight was conceptualised and established in 1986 by an Act of Parliament called the IGI and Security Act 101 of 1986”. According to section 6(2)(3) of the IGI and Security Act of 1986 “the Australian Inspector-General of Intelligence and Security (AIGIS) is appointed by the Governor- General on the recommendation of the Prime Minister after consultation with the leader of the opposition in the House of Representatives”. Together with the AG and the Ombudsman, the AIGIS is empowered by law to ensure that members of the

40 intelligence services fulfil their responsibilities in full compliance with the country’s statutory framework.

Section 21(1)(a) of the IGI and Security Act of 1986 in Australia empowers the IGI to “prepare reports indicating his/her conclusions and recommendations based on the findings of a particular undertaken inquiry”. These reports are meant to provide an independent assurance of the legal compliance of the intelligence services to the Prime Minister, senior ministers, and parliament. This is done by inspecting, inquiring and reporting on the activities of the intelligence services. Regarding cooperation among oversight structures, section 16 of the IGI and Security Act of 1986 provides that “before conducting any inquiry on any intelligence service, the AIGIS commits due regard for the work of the Auditor-General and the Ombudsman in relation to the service under consideration”.

This is done in order to avoid any duplication of roles and functions by any of the oversight structures. By default, the AIGIS spends most of its time overseeing the work of Australian Intelligence and Security Organisation (AISO) given its domestic focus. In fact, the AISO has more than a 70% chance of infringing the rights of ordinary Australians than any other service with external focus (Flood: 2004). Ironically, this is also the case in South Africa wherein the IGI spends most of its time overseeing members of SSA DB and SAPS CI given their domestic focus rather than on members of DI and SSA FB given their foreign mandates (Ministerial Review Commission: 2008).

The Australians also established a National Security Committee (NSC) which is responsible for conducting the overall oversight work on the intelligence services and is chaired by the Prime Minister (Flood 2004:52). In this model, it is clear that institutional cooperation is legislated for, and this has the potential to prevent undue fragmentation, promotes coordination and cooperation. Despite doubts in other countries, Weller (1999:500) indicates that “greater oversight of the Australian intelligence services

41 increased their staff morale, client satisfaction and general efficiency”. He further argued that, greater oversight increased their general image, effectiveness and relevance within the government and the country at large.

2.3.2 Canada

In Canada, the Canadian Security and Intelligence Service (CSIS) was established in 1984 by section 3(1) of the Canadian Security and Intelligence Service Act of 1984. This Act was “passed as a consequence of the McDonald Commission which advocated the separation of security intelligence work from policing” (Den Boer, 2012: 21). According to the CSIS Act of 1984, “the CSIS is responsible for the collection and analysis of information and security intelligence from across the country and abroad; it reports to and advises the Government of Canada on national security issues and activities that may threaten the security of Canada”. Further, the CSIS is responsible for the provision of security evaluation “to all federal departments and agencies with the exception of the Royal Canadian Mounted Police (RCMP)” (Canadian Security and Intelligence Services, 2016: 18). The CSIS website also provides that, “the CSIS is accountable to Parliament through the Ministry of Public Safety”. In terms of intelligence oversight, the Canadians established the Parliamentary Security Intelligence Review Committee (SIRC) and the IG, which serve alongside the federal court system as the country’s intelligence oversight mechanism.

The SIRC is responsible for the oversight of the CSIS as an external independent review body and prepares reports on the operations of the service for the attention of Parliament (Canadian Security and Intelligence Services: 1984). The SIRC has five members appointed by the Federal Government and meets monthly, with its day-to-day operations managed by an Executive Director. Whereas the CSIS has, “extraordinary powers to intrude on the privacy of individuals, the SIRC must ensure that these powers

42 are used legally and appropriately in order to protect people’s rights and freedoms” (Canadian Security and Intelligence Services, 1984: 19).

According to the legislative framework, “the SIRC has absolute authority to examine all information concerning CSIS activities, no matter how sensitive and highly classified that information may be”. Section 48(1) of the CSIS Act of 1984 states that the “SIRC privately investigates complaints through quasi-judicial hearings presided over by a committee member, assisted by members of staff”. Further, Section 41(2) of the Act provides that “any person may make a complaint to the Review Committee regarding any action by any member of the Intelligence Service”.

Regarding civilian intelligence oversight, section C-23 of the CSIS Act of 1984 “provides for the establishment of the IGI who is responsible for overseeing members of the intelligence services”. According to the CSIS Act of 1984, “the IGI is responsible to the Deputy Solicitor-General, not to the Prime Minister”. As an intelligence overseer, “the IG is entitled to have access to any information under the control of any intelligence service that relates to the performance of his/her duties and functions with the exception of the confidences of the Queen’s Privy Council”. In terms of cooperation, both the SIRC and the IG have full access to the files of the intelligence services (Canadian Security and Intelligence Services: 1984). As is the case in other states, section 30 of the CSIS Act of 1984 “mandates the IG to monitor the compliance of the intelligence service to its operational polices”.

In terms of the cooperation of intelligence oversight structures, Den Boer (2012) argues that the Canadian oversight mechanism provides for the four overlapping oversight mechanisms. For instance, the IG monitors the compliance of CSIS to operational policies, the SIRC reviews the “CSIS activities and investigates complaints against the service”, the Federal Court of Canada authorises the use of special investigative measures and the Minister of Public Safety reports CSIS activities through its Annual

Statements on National Security (Den Boer, 2012: 11). Like in South Africa, it is also clear that Canada has no legislative or systematic mechanism which provides for the cooperation of intelligence oversight structures in doing their work.

2.3.3 New Zealand

The New Zealand Government Communications Security Bureau (GCSB) has existed since 1977 and was legislated for through the GCSB Act of 1993 (New Zealand Government Communications Security Bureau: 2015). According to the Act, the GCSB is responsible for Communications Security (COMSEC), Technical Security (TECSEC), Foreign Intelligence (SIGINT) and Computer Security (COMPUSEC). Since 1977, the GCSB was located within the Department of Defence as a covert structure until its separation as an independent agency in 1989. According to section 8(a)(b) of the GCSB Act of 1993, “the GCSB is responsible for information assurance, cyber security, foreign intelligence and cooperation with and assistance to other entities”.

In 2013, Rebecca Kitteridge undertook a review to determine the compliance of the GCSB with the country’s legislative framework. This resulted in the “enactment of the Government Communications Security Bureau and Related Legislation Amendment Bill of 2013” (New Zealand Government Communications Security Bureau: 2015:17). The Bill amended various provisions of the “Government Communications Security Bureau Act of 2003, the IGI and Security Act of 1996 and the Intelligence and Security Committee Act of 1994” due to persistent challenges around the interpretation of certain sections (New Zealand Government Communications Security Bureau, 2015:14).

Regarding intelligence oversight, the GCSB website states that “throughout the history of foreign intelligence capability in New Zealand, there has been a comprehensive system of intelligence oversight and control by the government”. Since 1977, the GCSB has been overseen and controlled by the Committee of Controlling Officials (CCO)

44 chaired by the Head of Department in the Prime Minister’s Office. The CCO was “responsible for the operational supervision, policy guidelines and the general overview of the GCSB management” (New Zealand Government Communications Security Bureau, 2015: 16). However, in 1989, the CCO was disestablished and its responsibilities were assumed by the newly established Officials’ Committee for Domestic and External Security Coordination (ODESC) (New Zealand Government Communications Security Bureau: 2015).

At the parliamentary level, intelligence oversight takes place through the “Parliamentary Intelligence and Security Committee (ISC) as established by the Intelligence and Security Committee” (Section 5 of the Intelligence and Security Committee Act 46 of 1996) (hereafter the ISC Act of 1996). Section 5 of this Act empowers parliamentarians to “oversee the functions of the Intelligence services and ensure their compliance to the country’s legislative framework”. Further, section 7(1)(2)(3) of the ISC Act of 1996 states that the “ISC consists of (a) the Prime Minister, (b) the leader of the opposition and (c)(d) three other members of the House of Representatives”. Section 6(1) of the same Act authorises the ISC to examine the “policy, administration, and expenditure of the intelligence services”.

Like in South Africa, the New Zealand civilian intelligence oversight is provided by the IGI and Inspector-General of Intelligence and Security (IGIS) as established by the IGI and Security Act 47 of 1996. Section 5(1)(2) of this Act provides that “the IGIS is appointed by the Governor-General on the recommendation of the Prime Minister following consultation with the leader of the opposition”. Section 5(3) of this Act provides that “no person shall be appointed as the Inspector-General unless that person has previously held office as a Judge of the High Court of New Zealand”.

According to section 11(1)(a) of the IGIS Act 47 of 1996, “the IGIS is responsible for, amongst others, to inquire, of the Inspector-General’s own motion or at the request of

45 the Minister, into any matter that relates to the compliance by an intelligence and security agency within the law of New Zealand” (New Zealand Government Communications Security Bureau: 2015). In accordance with section 11(6) of the IGIS Act 47 of 1996, the IGIS’s reports are presented to the Prime Minister who later can publish them in the national library. In this model, the responsibility of intelligence oversight is concentrated within the office of the IGIS; hence no provision is made for the cooperation of the intelligence oversight structures.

2.3.4 United Kingdom (UK)

In the UK, the Intelligence services are established by an Act of Parliament called the Intelligence Services Act c.13 of 1994. Sections 1 and 3 of the Act provide for the establishment of the Secret Intelligence services (MI-5 and MI-6) and the Government Communications Headquarters (GCHQ) under the authority of the Secretary of State. In terms of Britain’s oversight mechanism, Born and Caparini (2007:1) indicate that “their Parliament always considered it as the competence of the executive and therefore remained hardly concerned with intelligence matters whether operations and/or oversight”.

However, Bruneau and Dombroski (2001:14) indicate that other ombudsmen-related institutions guarantee a high level of intelligence accountability. Although the UK does not have an IGI, Weller (1999:387) indicates that “the Prime Minister appoints Commissioners who are responsible for the work of the Intelligence services”. These commissioners are responsible for authorising intelligence activities outside the British Isles. Weller (2007) indicates that the investigations of complaints about members of the Secret Intelligence services and/or GCHQ are conducted by a tribunal. According to the Intelligence Services Act of 1994, “the decisions of both the Commissioners and/or the tribunal are not subject to appeal or liable to be questioned by any court of law” (Weller, 1999: 387).

Further, the Intelligence Services Act of 1994 provides for the establishment of the “Intelligence and Security Committee (ISC) which is appointed by the Prime Minister after consultation with leaders of the opposition” (Gill and Phythion, 2006:161). Section 10(1)(a)(b)(c) of the same Act indicates that “the ISC is responsible for the examination of the expenditure, administration and policy of the Secret Services and GCHQ”. According to Gill and Phythion (2006:161), “the ISC shall prepare an annual report on the work of the intelligence services for the attention of the Prime Minister who shall bring it to the attention of each house of Parliament”.

In the post-9/11 period, Gill and Phythion (2006:165) indicate that the “ISC committed to intensify its work in overseeing and controlling members of the intelligence services and GCHQ in order to improve their levels of accountability”. Even though the UK does not have an office of the IGIS, it is clear that the appointed commissioners and the tribunal perform similar functions to those performed by the IG in other countries. In this model, the ISC plays a central role in overseeing members of the intelligence services. In the UK model, institutional cooperation is not as pressing as in South Africa, as the ISC plays a central role in overseeing the intelligence services.

2.3.5 United States (US)

According to Weller (1997:384), “the Americans introduced intelligence oversight in the mid-70s after the emergence of intelligence scandals involving members of their intelligence services”. The scandals revolved around “illegal domestic spying on anti- Vietnam war protesters and revelations of illegal covert operations against the protesters by the CIA” (Weller, 1997:384). This move was fast-tracked by the findings of the Church Committee on the illegal operations of the Federal Bureau of Investigations (FBI) and the CIA. The committee revealed that “FBI documents used Orwellian language to label Martin Luther King Jr. the leader of a ‘Black Nationalist Hate Group’ and secretly set up mechanisms to destroy him” (Gill, 2012:69). Further, the Committee

47 found that the CIA was hiring ‘Mafias’ to permanently eliminate the leader of Cuba, Mr Fidel Castro (Gill: 2012). However, their attempts remained permanently unsuccessful.

Based on the findings of the Church Committee, the government enacted the IG Act 1101 of 1978 as the primary intelligence oversight legislation for guiding the work of the oversight structures. Section 3(a) of this Act provides for the establishment of the office of the IGI who “is appointed by the President with the advice and consent of the Senate”. As provided for by the Act, “the Inspector-General has the mandate to oversee various agencies such as the CIA, the FBI, and the Defence Intelligence Agency (DIA)”. Further, section 3 of the Act provides for the establishment of the IG within the DIA who is responsible for overseeing the Central Imagery Office (CIO), the National Reconnaissance Office (NRO), and the National Security Agency (NSA) (United States IGI: 2016).

As in South Africa, Born and Caparini (2007:15) assert that “the American oversight framework provides for executive, parliamentary and judicial oversight of the Intelligence services”. To assist the parliamentary oversight mechanism, the IG issues a variety of semi-annual reports for consideration by Congress. In terms of judicial oversight, “Congress enacted the Foreign Intelligence Surveillance Act of 1978 (FISA) which establishes the special courts” which grant the intelligence services judicial approval for their special operations such as the interception of communication media and electronic surveillance (Born and Caparini, 2007:15). This is meant to prevent the misuse of the system for selfish personal or group interests.

Further, the IG Act of 1978 provides for the establishment of the Inspectors-Generals’ Council on Integrity and Efficiency (CIGIE) to deal with allegations of misconduct against any IG. This initiative is meant to address the question of ‘who shall watch the watchers’ or ‘who shall supervise the supervisors’. In the US, the literature review reveals that there is no institutional mechanism to facilitate cooperation among various

48 oversight structures. This anomaly resembles similar challenges faced by the South African intelligence machinery and its oversight platforms.

2.4 SUMMARY OF FINDINGS ON THE INTERNATIONAL INTELLIGENCE OVERSIGHT SYSTEM

As already indicated earlier, the summary of these findings is grouped into three main categories in accordance with the three research sub-questions and the study objectives. Globally, there is sufficient literature emphasising the significance and relevance of intelligence tradecraft and the need for its effective oversight. The leading scholars in these writings include: Born and Caparini (2007); Born and Leigh (2005); Born and Wills (2012); Flood (2004); Gill (2012); Den Boer (2012); Weller (1999) as well as the Democratic Control of the Armed Forces (DCAF) Intelligence Working Group (2006), among others.

However, there is a clear shortage of literature emphasising the need for the cooperation of intelligence oversight structures when overseeing members of the intelligence services. Despite this shortage, the selected five countries put formal and/or informal measures in place to facilitate the cooperation of these structures when doing their work. The following section presents a comparative assessment of these international mechanisms with the intention to draw lessons for South Africa in its quest to improve the cooperation of its intelligence oversight structures for good governance results.

2.4.1 Rationale for and state of cooperation among intelligence oversight structures in the selected case studies

Like many other countries, Australia created multiple intelligence oversight structures with varied legislative mandates. However, to advance their proper coordination the

49 country established the NSC which is chaired by the Prime Minister. The NSC serves as the main coordination structure on the work of the IGI, the AG and the Ombudsman’s office. As indicated by Weller (1999), this coordination model avoids the duplication of efforts and creates synergy in the work of the intelligence oversight structures. On the other hand, Canada’s oversight mechanism is housed in the federal court system (Canadian Security and Intelligence Services: 1984). This is to ensure the speedy prosecution of members who flout the country’s legislative framework for personal or group factional interests. Like Australia, even Canada has multiple intelligence oversight structures with varied mandates (Den Boer: 2012). In terms of the cooperation of those structures, Canada does not have a clear legislative or regulatory framework for guiding the cooperation of intelligence oversight structures in doing their work.

As for New Zealand, civilian intelligence oversight is concentrated in the office of the IGI while parliamentary oversight is handled by the ISC. There is no legislative framework to ensure the systematic cooperation of the IGI, the ISC and other structures such as the AG in the fulfilment of their legislated mandates. However, the IGI is expected to submit reports to the ISC for consideration. In the UK, while intelligence oversight is the preoccupation of commissioners and the tribunal, they established the ISC to ensure their coordination. To ensure the clear separation of powers, the US spread their intelligence oversight across the executive, parliament and the judiciary. This resembles the approach which is used by South Africa. However, the US does not have a clear statutory or regulatory framework to facilitate the systematic cooperation of intelligence oversight structures. Instead, they created the CIGIE to supervise and monitor the integrity and efficiency of the multiple IGIs when overseeing various intelligence or security service structures.

2.4.2 Barriers to effective cooperation of intelligence oversight structures

Before looking at the country specific barriers based on the literature review, it is important to consider what Agranoff (2012:282) summarised as key global barriers to the cooperation/collaboration of government structures. He cited the following:

 “the structure’s protection of its turf,  risk aversion,  lost time and opportunity cost,  human relations cost,  the exertion of power by agencies and publics,  resource hoarding, and  public policy barriers”.

Operationally, the Australian IG spends most of its resources conducting proactive intelligence oversight (Australian IGI and Security Act 101 of 1986) (hereafter also referred to as the AIGIS). It does regular inspections and monitoring activities in order to proactively identify areas of concern and develop mitigation measures before the manifestation of the problems. However, it does not put much effort into cooperation or collaboration requirements with other structures of similar functions. Instead, it sees the ISC as the relevant structure responsible for such coordination work.

Unlike Australia, Canada’s approach is fully post-facto as the SIRC conducts investigations into the past operations of the intelligence community with the view to uncover irregularities. This is similar to the approach used in the work of the AGs across the globe. Therefore, being legally empowered to access to all sorts of information without any restrictions the SIRC does not see the value of cooperating with any structure in the performance of its legislative mandate. This view is necessitated by the allocation of an internal auditing function of the intelligence services’ books to the

IG’s office. It is for this reason that the IGIs do not see the value of fully cooperating with one another during the implementation of their respective mandates.

Unlike other countries, the New Zealand IGI does not have the legislative powers to enquire about sensitive intelligence operational matters such as collection and production methods including the sources of information. It is for this reason that its reports get published in the public national libraries for public consumption. As a result, attempts to cooperate with other structures remain futile. The UK opted to use commissioners for the authorisation of operations and the tribunals for the investigation of complaints. The operational interface of authorising (commissioners) and investigating (tribunals) reflects seamless synergy. Therefore, any conventional cooperation and/or coordination of functions remain futile.

In the US, intelligence oversight has been spread across the various overseen structures. It has also spread into the two houses of congress. Like other countries, the US does not have a clear legislative or regulatory framework to guide the cooperation of these structures. The respective founding legislation of the various structures does not allow room for inter-structure coordination. Instead, each structure has its specific oversight structure. Therefore, to fully address the underlying barriers to structural cooperation, collaboration and coordination, there is a need to consider the various issues raised by Agranoff (2012) above.

2.4.3 Strategies to ensure effective cooperation of intelligence oversight structures

Over the years, the above selected five countries managed to test their strategies for improving their intelligence oversight efforts. The main details of these efforts are covered in the discussions above. Globally, there is general agreement that intelligence oversight plays a crucial role to advance good governance. Therefore, the

52 continued search for a good intelligence oversight governance system is very relevant. This global benchmarking exercise revealed the absence of a clear systematic approach in the cooperation of intelligence oversight structures. However, the creation of the NSC chaired by the Prime Minister in Australia is something interesting to look at.

The exercise revealed a global absence of concerted efforts for ensuring the building and/or financing of systematic cooperation approaches for intelligence oversight structures. As a result, this situation leads to little or non-contribution of intelligence oversight structures to global good governance agendas. On the other hand, the comparative benchmarking exercise summarised above revealed the existence of concerted efforts for ensuring the cooperation of operational inter-agency work particularly among law enforcement structures. This exercise on international experiences of intelligence oversight practices presents critical lessons for consideration. First, in cases of multiple intelligence oversight structures, there is a clear need for the creation of a mechanism to facilitate the cooperation of such structures. Second, as in the UK, there is an option to separate the overseer from the one to investigate non-compliance, thus separating oversight from remedial intervention. Third, the other option is to choose one of the key oversight structures to stand as the convenor or coordinator of the various intelligence oversight structures as in the case of Australia (with the ISC) and the US with the CIGIE.

Further, the findings of the benchmarking exercise together with the reviewed literature confirm the significance of systems and democratic governance theories as prerequisites for a well-functioning intelligence oversight model. Therefore, to correctly address the problem statement, I have chosen to use the two complementary theories of systems theory and good governance theory. In this research, the systematic nature of the discussion remains the most critical part of the entire discourse. This is an important variation to understand as various structures can easily claim cooperation only to find that their cooperation processes are clumsy, unstructured and mostly work

53 according to personalities. Therefore, to ensure coherence, resilience and robustness, I will be using the two theories to interpret the empirical data findings.

After all these discussions, it is clear that governments require an effective good democratic governance system where various intelligence oversight structures can work together for the realisation of common objectives. It is clear that the realisation of effective and efficient cooperation of various intelligence oversight structures has the potential to achieve the key tenets of good governance. These include tenets such as accountability, transparency, openness and the observation of the rule of law, among others.

2.5 THEORETICAL FRAMEWORKS

Globally, there is general consensus among democratic countries on the role, mandate and functionalities of members of the intelligence services. They mainly agree that intelligence machinery should serve to preserve national security in accordance with the country’s constitutional and founding prescripts. Similarly, South Africa sees its intelligence services as the guardians of its constitutional order as provided for in the Constitution of 1996 and the White Paper on Intelligence of 1994.

In addition, this dispensation provided for the establishment of the intelligence oversight machinery through the promulgation of the Intelligence Services Oversight Act, 1994. It empowers members of these structures to oversee members of the intelligence services to ensure their compliance to the country’s legislative framework. Considering this framework, I intend through this research to assess the cooperation modalities of these structures when overseeing members of the intelligence services through the lenses of both systems theory and good governance theory. On the one hand, systems theory is used to understand the existing systems used by these structures when cooperating with each other. On the other hand, governance theory is used to

54 understand the significance of their cooperative activities in the betterment of intelligence oversight outcomes and the promotion of democratic agendas.

2.5.1 Systems theory

Before summarising the main elements of systems theory as applicable in this research context, I found it important to understand the meaning of the word ‘system’. According to Anderson, Carter and Lowe (1999:4), a ‘system’ refers to “an organised whole that is made up of components which interact in a way distinct from their interaction with other entities and this endures over some period of time”. Laszlo and Krippner in Jordan (1998: 67) regard a ‘system’ as “the complexity of interacting components together with the relationship among them that permit the identification of a boundary-maintaining entity or process”.

According to Robbins, Chatterjee and Canda (2006), the word ‘system’ emerged from the work of Emile Durkheim in his early study of social systems. Fook, Ryan and Hawkins (1997) and Wakefield (1996) argued that ‘system’ originated from the work of Ludwig von Bertalanffy in social work discipline. This was also supported by other writers like Bruce; Friedman and Allen in Brandell (2011) who argued that Von Bertalanffy developed systems thinking in social work, but from his background as a biologist. The social psychologist Uri Bronfenbrenner later adapted this thinking when examining human biological systems within an ecological environment (Brandell: 2011).

Other scholars such as Talcott Parsons were also credited for playing a very significant role in the emergence and adoption of the current system’s approach. In his work, Parsons introduced structural functionalism and argued that “social structures involve interaction and relationship among actors who are characterised by functional imperatives” (Brandell, 2011: 47). Anderson et al (1999: 34) described systems theory as a mechanism that enables people to “understand the components and dynamics of

55 client systems in order to interpret problems and develop balanced intervention strategies”. Meyer (1976) indicated that systems theory serves as an organising conceptual framework or meta-theory that is used to understand the interactions of various systems. It is regarded as meta-theory given its applicability in different species, ages, domains and grades of analysis (Thelen and Smith: 1994).

Despite sometimes conflicting arguments about its origins, various scholars do concur that “systems” has to with the working together of various components, organisms or structures for the realisation of common objectives. It is for this reason that the following discussion clarifies the actual origins of “systems” and its theoretical imperatives by mapping its evolution.

2.5.1.1 Evolution of systems theory

The evolution of systems theory can be traced to the contributions of seminal thinkers such as Alfred North Whitehead, Ludwig von Bertalanffy, Anatol Rapoport, Kenneth Boulding, Paul A. Weiss, Ralph Gerard, Kurt Lewin, Roy R. Grinker, William Gray, Nicolas Rizzo, Karl Menniger, Silvano Arieti. According to Laszlo and Krippner in Jordan (1998), the recent advances of systems theory includes the evolutionary systems design, the use of cognitive maps and evolutionary learning systems.

In its earlier evaluation, von Bertalanffy’s systems’ thinking was influenced by the earlier pioneers of sociology such as Max Weber and Durkheim. Ironically, the latter is credited by some scholars as the actual originator of systems theory through his scholarly work on social systems and their interactions. Both Durkheim and Weber are regarded as the main scholars who borrowed systems theory from its original application in biological organisms for utilisation in human social systems. As argued by Bruce, Friedman and Allen in Brandell (2011: 36), Durkheim wanted to understand

“how societies were organised and how they maintained cohesion or group identity over time”.

On the other hand, von Bertalanffy’s idea was to focus on the entire system and seek to understand the manner in which it relates to other parts of the system (von Bertlanffy: 1968). This was new thinking at the time as most theorists were then reductionists by nature and believed that the whole can only be understood by breaking it into smaller parts. At the time, many scientists and scholars believed that to properly understand a phenomenon or system one should isolate or break it into smaller parts or elements. However, von Bertalanffy felt that understanding would come, not from the separated parts or elements, but from the relationship among them. In fact, Durkheim believed that when individual members of society agree to meet ordinary societal expectations people get sustained. On the other hand, Weber believed that “bureaucracies are coercive in nature and get maintained through the use of legitimate violence or force” (Brandell, 2011: 6). During its evolution, systems theory was viewed as part of General Systems Theory (GST) as advanced by von Bertalanffy in 1968.

This was mainly used in scientific explanations and classical physics wherein organisms were viewed as part of the whole or as a system (von Bertalanffy: 1968). In Weber’s systems approach, he cautions people to remain mindful that the interest of “individual or client they serve might be in conflict with the interest of those in power” (von Bertalanffy, 1968: 47). As a result, there is a need to find ways to harmonise one’s interests with those of the authorities. Further, Nobel chemist IIya Prigogine also contributed to the evolution of systems theory. In his work, Prigogine was mainly interested in the physics of complex systems that were far from the thermodynamic equilibrium (Thelen and Smith: 1994).

2.5.1.2 Application of systems theory

Based on literature review, systems theory can be seen as a broader phenomenon than just a theory; hence some people see it as meta-theory. It is seen as a world view which seeks to adopt systems’ thinking which looks at the world and see the objects as interrelated with one another (von Bertalanffy: 1968). This represents the frame of reference which will be used to evaluate the operational modalities of the various democratic intelligence oversight structures and their relationship to one another. Despite its meta-theoretical posture, Thelen and Smith (1994) argue that systems theory can be applied as a specific theory seeking to understand how human beings can gain knowledge from their everyday actions. In its application, Wolff (1987) indicates that systems theory observes the emergence of patterns and orders from the interaction of the components of a complex system without explicit instructions, either in the organism itself or from the environment.

In this study, the application of systems theory is for guiding the development of a resilient, coherent and systematic model for marshalling the cooperation of intelligence oversight structures. As already discussed, the current intelligence oversight architecture consists of various structures with varied mandates and powers. However, it does not provide for a systematic cooperation of these varied structures. Therefore, this study seeks to use systems theory in the development of its envisaged model in which the IGI plays a central role. Once developed, the model would provide the systematic approach in which the various oversight structures engage each other towards bettering the coordination of intelligence oversight work. Considering the definition of “systems” as articulated by Anderson et al (1999:4) above, the envisaged model would be the organised whole that is made up of components which will be the various intelligence oversight structures which interact with each other in a clearly distinct way towards the realisation of common objectives. At the end, a systematic

58 approach would not singularly complete the puzzle but good governance theory would have to assist.

2.5.2 Good governance theory

According to Schwella (1996), governance is an all-encompassing phenomenon that is flexible enough to be able to embrace a multitude of influences and inputs. He adds that good governance can be realised when public participation contributes to the improvement of people’s quality of life. Schwella (2014) argued that persuasive communication can be used as a learning tool in the advancement of governance practices. From what Schwella (1996) argued, it can be concluded that governance embraces a number of key concepts for the advancement of humanity. These include ethical leadership, accountability, responsiveness, transparency and respect for the rule of law. Similar arguments were made by various scholars such as Agere (2000), and Cloete (2004). The emergence of governance as a field of study did not lack controversies. Louw’s (2002) main criticism of governance has been centred on its applicability in almost every discipline. Similar sentiments were uttered by Klijn and Koppenjan (2012) when describing the emergence of Governance Network Theory (GNT). They argued that if GNT is “to evolve as a theory of everything, it will lose its explanatory power, ending up being a theory of nothing” (Klijn et al., 2012:1). However, in this research I am more interested in its ability to reconstruct the interaction process of various interdependent structures with divergent interests towards the realisation of common goals (Klijn and Koppenjan: 2012).

Further, Van Niekerk, Van der Waldt and Jonker (2001:64) state that governance has to do with the maintenance of law and order, defending society against external enemies and advancing the broader welfare of community members. Olowu, Dele, Sako and Soumana (2002:37) view governance as “…a system of values, policies and institutions through which society manages its economic, political and social affairs”. Cloete,

Moller, Dzengwa, Davids and Ismael (2003:7) also concur and simplify governance to mean “the coordination of all the available resources in the public, private and voluntary sectors”. In addition, Agronoff (2012) says such cooperation can be achieved through the network model wherein organisational functions are being synchronised to facilitate the overall realisation of intended objectives. As Cloete and De Coning (2011) emphasised, even public policy-making is no longer the purview of a single policy- maker, but a product of discussion among networks of policy stakeholders. Therefore, even members of the intelligence services are supposed to make unique contributions to the work of government in its policy agenda towards the realisation of good governance. Both intelligence services and their oversight structures coordinate, cooperate, and collaborate their efforts in their environment.

As Agere (2000:2) argues, governance has recently come into regular use in particular disciplines such as political science, public administration, and development management. According to Adejemboi (1998:26), governance consists of “…public accountability, transparency in government procedures, abiding by the rule of law and efficient and effective public service leadership and management”. The United Nations Development Programme (UNDP) (1997) describes governance as “the exercise of political, economic, and administrative authority in the management of a country’s affairs at all levels”. Further, Van Niekerk, Van der Waldt and Jonker (2001:64) consider governance to include the ordering or management of a group, community, or society by a public authority, including the “…maintenance of law and order, the defence of society against external enemies and the advancement of the broader welfare of community members”.

Olowu et al (2002:37) see governance as “the system of values, policies, and institutions by which society manages its economic, political, and social affairs through interaction within and among the state, civil society and the private sector”. According to the World Bank (2010:102), good governance is epitomised by “predictable, open, and

60 enlightened policy-making; a bureaucracy imbued with professional ethos acting in furtherance of the public good; the rule of law; a transparent process and a strong civil society participating in public affairs”. According to Stoker (1998) and Nzongola-Ntalaja (2002), good governance is regarded as the reflection of effective rules, institutions and good democratic governments.

Agere (2000:1) takes it further by indicating that the governance paradigm advocates for an increased managerial autonomy, reduced central agency controls, rewards both organisational and individual performance and recognises the importance of providing human and technological resources. Hyden and Braton (1993:7) add that the style of governance can be evaluated by the society’s “degree of trust in their government; the level of responsiveness pertaining to the relationship between government and civil society; the degree of accountability to its voters and the nature of the authority it exercises over its people”. In distinguishing public and private sector governance, Agere (2000:2) argues that the paradigm of governance is open-minded and very clear regarding the public functions that should be performed by public servants as opposed to those performed by the private sector.

To emphasise this point, Olowu et al (2002:37] argue that governance relates to the system of values, policies and institutions “by which society manages its economic, political and social affairs through interaction within and among the state, civil society and private sector”, with the assistance of the intelligence officers. Governance provides for a clear definition of the relationships between government and the markets, and government and the private sector. Cloete et al (2003:7) then argue that the full realisation of the governance paradigm requires domestic and international democratic mobilisation and the application and coordination of all available resources in the public, private and voluntary sectors. From what has been argued thus far, it can be deduced that governance can either be good or bad. Therefore, the following discussion describes good governance as a theoretical consideration to be used by this study.

2.5.2.1 Understanding good governance

According to Cloete (2000:62), good governance relates to the “achievement of the most appropriate developmental policy objectives by a democratic government in order to ensure the sustainable development of society”. Stoker (1998:17) describes it as a “governing style in which the boundaries between and within public and private sectors have become blurred”. Peters and Pierre (1998:232) see it as “a process that is concerned with the creation of conditions for ordered rule and collective action”.

According to Nzongola-Ntalaja (2002:28), “good governance is meant to operate at every level including household, village, municipality, nation, region, or globe”. In contrast, Born and Leigh (2005:22) describe bad governance as “the reflection of arbitrary policy-making, unaccountable bureaucracies, unenforced or unjust legal systems, the abuse of executive powers and widespread corruption”.

To contextualise good governance Naidoo (2004:180) describes effective governance as “the system which ensures the popular welfare and promotion of the greatest happiness of the greatest number of people in society”. In addition, Cloete (2007:2) argues that “building trust is the foundation of good governance”. Given its continued emphasis on the creation of efficient and accountable institutions, Cloete (2007:2) sees good governance as a “sine qua non” for development. He considers it to be the effective, transparent and accountable discharge of responsibilities within the established legislative framework. From what has been argued thus far, it is clear that good governance can be realised through a major reform of the institutions of state, the private sector and civil society. Those reforms should unanimously adopt the formal networks approach instead of informal networks. In this case, formal networks are consciously created with a built-in mechanism for creating platforms with binding agreements for participation while informal networks are mainly created by chance and do not have any binding agreement for participation (Andrade and Joia: 2012).

Naidoo (2004:181) argues that, “good governance can be achieved by ensuring the full observation of the rule of law, improving the efficiency and accountability of the public sector as well as tackling corruption”. As argued by scholars like Cloete (2007), Naidoo (2004), Stoker (1998), Born and Leigh (2005) and Nzongola-Ntalaja (2002), governance is about systems, structures and processes by which organisations or governments make decisions. But good governance only occurs when those decisions are made in a transparent, accountable and effective manner in conformance with the country’s legislative framework.

2.5.2.2 Good governance as a public management tool

According to the World Bank (1994), “public governance is the style of interaction between a government and the society it governs”. This is supported by Agranoff and McGuire (2004) who emphasise that the complexities of today’s communities force governments to facilitate the effective collaboration of various structures across sectorial boundaries for the benefit of humanity. According to Fox and Meyer (1996:55), “public governance also includes a process by which governments are chosen, monitored, and changed”. This includes the government’s ability to enforce its will and policy direction as mandated by the electorate. Fox and Meyer (1996:55) define government as a “body of persons, systems, and institutions that make and apply all enforceable decisions for a society”. Fitzgerald et al (1995:195) describe “public governance as a process through which power and authority are exercised between and within institutions in the state and civil society with regard to the allocation of resources”.

Therefore, to ensure the successful implementation of public governance a concerted effort is required to maximise the involvement of members of the intelligence services in a manner that is compliant with the country’s legislative framework. In his address at

Intelligence Services Day on 28 November 2008, Minister of State Security, Dr Cwele, maintained that “members of the intelligence services should closely cooperate with other government departments in order to contribute positively to the achievement of the government’s national priorities and objectives as set out in the programme of action”. The consistent call for inter-structure cooperation, coordination and collaboration has graduated to embrace policy network, collaborative network and ultimately governance network, in the Public Administration literature (Isett, Mergel, LeRoux, Mischen and Rethemeyer: 2011). Among others, collaborative network focuses on bringing together a multiplicity of structures when one could not single- handedly achieve the intended results (Agranoff and McGuire: 2004).

According to Gildenhuys (2004:153), “the public service is a fixed establishment with defined posts that has been created for the normal and regular requirements of the department”. Section 197(1) of the South African Constitution of 1996 provides that “in public administration there is a public service for the Republic which must function and be structured in terms of national legislation and which must loyally execute the lawful policies of the government of the day”. Further, section 7(2) of the Public Service Act 35 of 1994, provides “for the structures and organisational components of the public service in both national and provincial departments”. Further, the “theory of public service in modern democracies requires public officials to adhere to democratic principles and a set of ethical standards and be further guided by community values and common law, which have developed into a separate set of rules, known as administrative law” (Gildenhuys, 2004:154). The White Paper on the Transformation of Public Service Delivery of 1997, also known as the Batho Pele Paper, provides that “public officials should ensure that services are provided in a transparent and responsible manner in order to enhance the accomplishment of a better life for all”. To

64 enhance the achievement of good governance, section 195 of the Constitution provides that:

“the public service must be non-partisan, career-oriented and function according to fair and equitable principles; promote an effective and efficient public administration that is broadly representative of all population groups; serve all members of the public in an unbiased manner; functioning in terms of the conditions of service and be organised in departments”.

In his address at Intelligence Services Day on 28 November 2008, Minister of State Security Dr S Cwele concurred with section 195 of the Constitution and argued that “members of the intelligence services should display the highest level of discipline and professionalism as they are the crucial component and an important role-player in the strategic goal of creating a non-racial, non-sexist, democratic, and prosperous South Africa”. As described by scholars like Waldo (1994), Stillman (1980), Frederickson (1999) and Gladden (1961), public governance complies with both the principle of an enquiry and also an activity. To unpack this assertion, Frederickson (1999:702) indicated that “for a theory to exist, it must describe and explain the boundaries for enquiry as determined by the activity itself”.

2.5.2.3 Good governance as a means and/or end

According to Agere (2000:4), “good governance can be used as both the means and the end”. It is the ‘means’ when considering its contribution to economic growth, human development, social justice, institutional development, capacity-building, coordination, decentralisation and the relationship between politicians and appointed officials. It is the ‘end’ when it successfully addresses all of its major elements in a way that society embraces the route that was undertaken for arriving at the solutions, even if society may not necessarily agree with the methods used and conclusions reached. As Agere

(2000:5) puts it, good governance can be attained when governments guard against the following characteristics of poor governance:

 “Failure to make a clear distinction between what is public and private, hence a tendency to divert public resources for private gain.  Failure to establish a predictable framework of law, government behaviour and the rule of law.  Excessive rules and regulations which impede the functioning of markets.  Priorities inconsistent with development, resulting in a misallocation of resources.  Excessively narrow-based or non-transparent decision-making processes.  Lack of a code of conduct in managing the affairs of the state.  Lack of clearly defined policy assumptions”.

According to Agere (2000:5) good governance is “a precondition of sustainable development operating in an ethical, accountable, and appropriately regulated environment which facilitates competition in the market place”. He further argues that “lack of accountability and corruption could lead to the dissolution of poverty alleviation programmes and environmental protection, resulting in the suffering of ordinary members of the public”. From the discussion thus far, it is clear that failure to recognise poor governance could inhibit or delay the adoption and implementation of good governance practices. However, it should be noted that there is no monolithic international best practice to achieve good governance, as each environment requires special attention. As the means or the end, the International Centre for Integrated Mountain Development (2010) describes governance as “the power relationship among various stakeholders such as communities, transitional institutions and the local formal institutions over the ownership of and access to natural resources”.

Among its plethora of attributes, good governance advocates for the full respect of the rule of law and that all persons are presumed innocent until proven guilty by a

66 competent court of law (Agere: 2000). This assertion resembles one of the core values of the South African democratic dispensation. For instance, section 198 of the RSA Constitution of 1996 emphasises the creation of the governance model underpinned by the country’s legal prescripts. On the intelligence front, section 7 of the Intelligence Services Oversight Act of 1994 emphasises the creation of an intelligence dispensation that operates within the boundaries of the country’s legislative framework.

Another key attribute of good governance is the question of accountability. In doing their work, members of the intelligence services are expected to be accountable for their actions. This was explicitly cited in the White Paper for Intelligence of 1995 which emphasised the need to create an intelligence community that is accountable to the country’s constitutional imperatives. In being accountable, organs of state will also be conforming to the principle of transparency or openness in terms of the work they do. Therefore, this study could not have chosen a better theoretical framework than good governance. But of course for complementarity, systems theory was also chosen to complement good governance particularly when considering issues of cooperation among intelligence oversight structures. As the central theme of this study, concepts such as oversight and control, among others, are fully dissected below. As fully outlined in Chapter 1, under section 1.5, the problem with failure to effectively oversee and ensure the transparency and accountability of members of the intelligence services, is that their activities can easily compromise the same democratic dispensation they are meant to protect.

2.6 CONCEPTUAL CLARIFICATIONS

A number of scholars identified various concepts which underpinned both the systems and good governance theories (eg Agere 2000, Born and Leigh 2005 and Meyer 1976). The concepts oversight, control, transparency and accountability, can be operationally

67 defined as follows for use in this study:

Oversight has two basic meanings according to Oleszek (2010: 67):

“First, it implies the failure to notice something or to inadvertently omit something. Second, it denotes some form of legislative supervision or watchfulness of the delegated authority to the executives within government”.

It is the latter definition which orientates the focus of this study. According to Gill and Phythian (2006:151) oversight is “a process of overseeing members of the intelligence services in order to ensure their compliance to the overall policies and the country’s legislative framework”. This process does not include the daily management of members of the intelligence services. Further, Gill et al (2006: 151) add that “oversight includes the reviewing, monitoring, guiding, and controlling of intelligence practitioners in order to ensure that intelligence services do not become the sinister organ of state that is alien to democratic ethics, values, and ethos”. However, Gill (2007:46) distinguished oversight from review by indicating that ‘review is an ex post facto process while oversight encapsulates supervision which takes place in ongoing activities”.

Control is regarded as the “creation of conditions that would lead to the achievement of agreed standards of performance, desired results, and proper compliance with the requisite legislative framework” (www.gov.za). Born (2004) describes intelligence control as the act of being in charge of the daily functioning of the intelligence services. For instance, parliament exercises intelligence oversight through the JSCI while government exercises intelligence control through the Minister of State Security. Greg Hannah, Kevin O’Brien, and Andrew Rathmell (2005:5) in Baldino (2010:9), described the formal instruments for an effective oversight mechanism as “executive accountability, parliamentary oversight and legal compliance mechanisms”. According to Dlomo (2004), oversight mechanisms are instituted as “checks and balances to

68 ensure that members of the intelligence services do not exceed their constitutional and legal mandates”.

Similar views were earlier expressed by Lowenthal (2003) and Caparini (2002) who argued that establishing effective and vibrant oversight structures serves as a deterrent for any possible abuse of state power for personal or group interests. Smist in Schwarz Jr (2007: 87) divided oversight into two categories; namely, institutional and investigative. In his view, the institutional model “applies oversight as a cooperative relationship between the overseer and the overseen while investigative model views oversight as involving the adversarial relationship between the overseer and the overseen” (Schwarz Jr, 2007:87).

In South Africa the work of oversight is constitutionally allocated to the various organs of state whose mandates are to scrutinise the executive action and ensure compliance to the rule of law. By Smist’s understanding, the South African oversight model carries both the institutional and investigative posture in the application of its legislative mandates. Even though some scholars might be using oversight and control interchangeably, these are two different concepts. For instance, the oversight of the intelligence services is done by all oversight structures and the control function is the responsibility of the Minister of State Security (Intelligence Services Oversight Act: 1994).

Cloete (2007:12) conceptualises transparency as “the availability of information to the public on the transactions of the government, including its decision-making processes”. This is expanded by Combs (2009:42) who describes it as “the government’s opening of its books for members of the public to see exactly where their money is going”. However, Combs (2009) argues that opening books is not only limited to publishing expenditure data in great quantities, but also examining the reporting practices to ensure quality financial management. In line with Combs’s (2009) arguments,

69 transparency involves “ready access to reliable, comprehensive, timely, understandable, and internationally comparable information on government activities which is necessary for sound government and good governance”. In the contemporary environment, Cloete (2007:12) argues that “the most promising new strategy in promoting transparency is information technology in the form of an integrated electronic and mobile government”.

According to Kin (2009), transparency is a “multifaceted concept which includes being open; truthful, accurate, rational and clear in articulation”. At the government level, Schwella et al (1996:16) consider it as “the extent to which the functioning of government institutions is open to public scrutiny”. Premchand (1993:16) expands the latter view by indicating that government transparency is “when the availability of information to the public on the transactions of government and the openness of the decision-making processes is unquestionable”. In South Africa, the need for transparency is highlighted by section 195(1)(g) of the Constitution of 1996 which emphasises the need for transparency in public service delivery.

This emphasis is important as transparency in government decision-making and public policy implementation has the potential to reduce uncertainty and inhibit corruption among public officials. The Asian Development Bank (2009) considers transparency to include “the availability of information to the general public and clarity about government rules, regulations, and decisions”. With its proper application, transparency complements and reinforces predictability.

Accountability is conceptualised by Agere (2000:7) to include “the holding of elected or appointed individuals and organisations charged with a public mandate responsible for specific actions, activities, or decisions to the public from whom they derive their authority”. In this case, accountability can be viewed in either its narrow sense or broader sense. In the narrow sense, Agere (2000:7) lauds it for its “ability to explain the

70 allocations, their use and the manner in which resources are dispatched within the context of budgeting, accounting and auditing”. It also focuses on good corporate governance (Agere, 2000:7).

In contrast, Ekpo (1979:46) describes “corruption as not just a deviation of behaviour from accepted norms, but the deviation from norms and established patterns of ethical conduct and accountability”. In view of the Intelligence Services Oversight Act of 1994, the IG is given the legislative mandate to ensure the accountability of members of the intelligence services in the performance of their functions. As Kin (2009) puts it, “accountability starts with full compliance to the provided legislative framework and results in a transformed individual and institution”. In its parliamentary oversight model document, the South African parliament defines accountability as follows:

“the social relationship where an actor (an individual or an agency) feels an obligation to explain and justify his or her conduct to some significant other (the accountability forum, accountee, specific person or agency)”.

In clarifying answerability and accountability, the Asian Development Bank (2009) argues that “public officials should be answerable for their behaviour and remain accountable to the entity from which their authority comes from” (African Development Bank Group: 2003). Although accountability can apply to the administrative, moral and managerial levels, the political accountability of broader government, its civil servants, and politicians to members of the public remains crucial (Van Niekerk et al., 2001:125). This is further expanded by Van Niekerk et al (2001:125) who argue that “politicians and public servants should be accountable for their actions to their constituencies and their superiors respectively and, more broadly, to the public in general”.

This notion is emphasised by the White Paper on Transforming Public Service Delivery of 1997 (1997:9) which states that “citizens should be given full and accurate

71 information about the public service they are entitled to receive”. Accountability has two main dimensions distinguished in the literature, namely political accountability and administrative accountability. According to Agere (2000), “political accountability is generally associated with the electoral mandate which citizens give politicians to constitute the executive and legislative arms of government”. On the other hand, Agere (2000:26) describes administrative accountability as the obligation to fulfil “assigned and accepted duties within the framework of the authority and resources provided”.

According to Solomon (1992:24), cooperation and collaboration includes a voluntary arrangement in which two or more entities (individual, business) engage in a mutual beneficial exchange instead of competing. It mostly happens when a group of institutions or structures cooperate for their mutual benefit instead of competing for selfish benefit (Solomon, 1992:26). From a business point of view, effective cooperation can build into cooperatives which can be owned, controlled and/or operated by a group of people for their own benefit. This view is supported by Smith, Carrol and Ashford (1995:22). In practical terms, cooperation plays a critical role in the broader management success and simplifies the complexities associated with today’s world (Smith et al., 1995:22). Angranoff and McGuire (2003: 4) describe collaboration as the “facilitation of the multi-structure arrangements to solve problems which cannot be easily solved by a single structure”.

According to Smith et al (1995), cooperation and collaboration can be sub-divided into five principal types: first, direct cooperation, this refers to a situation where people do things together because they bring social satisfaction; second, indirect cooperation, this refers to a situation where people bring together different competencies for the realisation of common good; third, primary cooperation, this kind of cooperation is mostly found among family and friends, it includes the sharing of a particular identity and means and goals; fourth, secondary cooperation which is mostly found among social groups and is highly formalised and specialised, it is mostly seen when people

72 assist each other for the ultimate benefit of the entire group; and lastly tertiary cooperation which is mainly found among political parties or even tribes in which two or more groups can cooperate and work together for antagonistic goals.

After this extensive discussion on the various crucial concepts of this study, it should be noted that the central concept is cooperation and collaboration which refers to the working together of various components or entities for the realisation of common good. Comparatively, cooperation is the process of openly working and acting together for the realisation of common good while competition is the process of fighting between people, nations, and tribes and even social groups given their disagreements about something. In sum, Thomson and Perry (2006) differentiate between cooperation, coordination and collaboration. They indicated that cooperation mainly relates to the sharing of information in a short-term and less formal manner. Coordination has to do with the working together of organisations towards the realisation of a common goal in the long-term (Thomson and Perry: 2006). Collaboration involves the willingness of parties and stakeholders to enhance each other’s capacities for the realisation of common goals (Himmelman: 2001). In fact, Thomson and Perry (2006:23) regard collaboration as the “cooperation of mutual goals while achieving individual goals and additional outcome which are shared differently by individuals”. However, this study strives for the realisation of cooperation, coordination and collaboration of intelligence oversight structures when overseeing members of the intelligence services. Fleishman (2009) cautions that, members of any network tend to willingly participate in a network when the provided resources, such as financials and/or technical expertise, are satisfactory.

Based on the contextual and conceptual clarity as provided above, this study seeks inter alia to determine the state of cooperation among the democratic intelligence oversight structures. The ultimate end is to design an improved systematic cooperation model to facilitate the cooperation of intelligence oversight structures. Therefore, the

73 following discussion operationalises the focus of the study through a conceptual framework built on governance, systems and network frameworks, as understood within oversight structures in South Africa. In addition, the above concepts are variables influencing the theoretical framework as discussed. Figure 2.1 presents the conceptual framework detailing the manner in which this study unfolded.

Figure 2.1: Conceptual Framework

Source: Adapted from (Bruce, Friedman and Allen in Brandell: 2011)

Figure 2.1 above depicts the inflow of oversight and control in the form of systemic inputs towards members of the South African intelligence community. This is done by a multiplicity of democratic intelligence oversight structures based on their respective legislated mandates to be fully discussed in Chapter 4. Further, this figure depicts the 74 results of each structure’s feeding into the work of the JSCI as the ultimate custodian of intelligence oversight and control residing at the parliamentary level. As already indicated, this study seeks to improve the cooperation of intelligence oversight structures for good governance. The whole exercise will be based on the provisions of systems theory and its contribution towards the realisation of the tenets of good governance. Systems theory provides an operating model while good governance portrays the desirable social environment that characterises democratic South Africa. Therefore, this study is located within the context of good governance and consequently applied a systems theory approach to the development of a systematic cooperation model to enhance the cooperation of various intelligence oversight structures when overseeing members of the intelligence services.

2.7 SUMMARY AND CONCLUSIONS

This chapter provided a detailed comparative analysis and assessment of the intelligence oversight work in advanced democracies such as Australia, Canada, New Zealand, the UK and the US. This is an important intervention as the assessment will be used in the upcoming Chapters 4 to 6 to juxtapose the South African oversight architecture vis-à-vis the global experience. A literature assessment confirmed the global need for intelligence oversight but recognised the absence of concerted efforts to enhance the cooperation of intelligence oversight structures in doing their work.

The above assessment also summarised and applied the role of good governance and systems theory in the development of the conceptual framework portrayed in Figure 2.3. In summary, for the purposes of this study, systems theory requires the application of a systems approach in the cooperation of intelligence oversight structures. A revised, applied systematic model to facilitate such cooperation is proposed here, customised for the South African situation. As for good governance, it should be a focus of such a model to ensure the contribution of such structures to promote a

75 democratic agenda. Therefore, these multiple oversight structures should systematically cooperate when overseeing members of the intelligence services and thus contribute to the promotion of good governance results.

Having completed the global review of international intelligence oversight systems; the next chapter proceeds with the methodological approach that was utilised during the research design and the collection of empirical data.

CHAPTER 3 METHODOLOGICAL APPROACH TO THE STUDY

3.1 INTRODUCTION

Following the previous chapter’s presentation of the literature review findings and conclusions, this chapter provides the adopted methodological approach for driving the successful completion of this project. The adoption of the methodology was mainly guided by the problem under consideration, the research purpose and its accompanying objectives. As already discussed, the main purpose of this study is to improve the cooperation of democratic intelligence oversight structures for good governance results. The methodology serves as the blueprint of how the research unfolded. It outlines the procedure which was followed during the implementation of the research process.

The discussion starts by presenting the selected research paradigm and design which were used in this research process. It includes the description of the data collection instruments, and the target population, including the manner in which sampling was done. Further, the discussion provides reasons for the adopted paradigm and the accompanying design. This is followed by a clear articulation of the undertaken interventions to ensure that the research results are reliable and valid. This chapter concludes with the discussion of possible ethical considerations and the limitations of the research. The discussion starts by presenting Figure 3.1 which depicts the entire research methodology for this study.

FIGURE 3.1: Research Methodology

3.2 RESEARCH PARADIGM

According to Welman and Kruger (1999:32), “scientific research involves the application of various methods, strategies, approaches and procedures to create scientific knowledge”. The validity of this scientific knowledge largely depends on the adopted research paradigm which resembles the manner in which data was collected. Welman and Kruger (1999:33) are of the opinion that “social sciences researches are mainly 78 based on qualitative, quantitative or participatory research paradigms”. However, participatory (or action) research is also based on qualitative, quantitative or mixed approaches. According to Kuhn (1996), a paradigm resembles a worldview which reflects the framework of beliefs, values and methods within which the research takes place. In differentiating qualitative from quantitative paradigms, Badenhorst (2008:45) argues that “qualitative data is about reasoning and quantitative data is about statistics”. Badenhorst (2008:45) clarifies that the distinction between the two paradigms does not mean that a qualitative researcher cannot use statistics to argue a point, or that a quantitative researcher cannot interpret data for qualitative meaning. In other words, quantitative methods can be used in a qualitative research design and vice versa (mixed-methods). Most research designs, in fact consist of either a predominantly qualitative design using both qualitative and quantitative data collection, analysis and presentation tools, or a predominantly quantitative design using both qualitative and quantitative data collection, analysis and presentation tools. The design itself can also be mixed, comprising both a qualitative and quantitative dimension.

Schwandt (2001) and Krauss (2005) identify three philosophical differences between the two paradigms:

 “Firstly, they argue that a qualitative researcher believes in multiple realities which depend on meaning and interpretations, while quantitative researcher tends to believe in a single reality which is based on truth.  Secondly, they argue that a qualitative researcher believes in researching what is out there in its context and complexities, while a quantitative researcher believes in researching the world in smaller bits, thereby contributing to the body of knowledge.  Thirdly, they argue that a qualitative researcher believes that immersing oneself in the research context is a good way to understand meaning, while a

quantitative researcher believes that distance between the researchers and the researched is essential to achieve truth”.

After close consideration of these philosophical differences, coupled with the nature of the research problem, the purpose, the questions and the objectives at hand, I opted to use a qualitative research paradigm. The use of a qualitative paradigm was complemented by some quantitative data collection and analysis methods. For instance, during data presentation I counted the number of emerging interview responses and themes and grouped them accordingly. The choosing of a qualitative paradigm is in line with Cavaye’s (1996) assertion that a chosen methodology depends on the problem at hand rather than a commitment or just the liking of a particular paradigm. This is also supported by Falconer and Mackay (1999) who argue that a good researcher decides on the research paradigm based on the nature of the phenomenon under investigation rather than the methodology itself. The following discussion describes the main attributes of the qualitative paradigm followed by the reasons to justify its appropriateness for this study.

3.2.1 Qualitative paradigm

According to Creswell (1994:18), “a qualitative paradigm is a process of enquiry whose understanding is based on the distinct methodological traditions of enquiry that explores a social or human problem”. This paradigm allows for the building of a comprehensive picture based on the information sourced from the respondents in their natural settings. In 1998, Creswell (1998:15) re-emphasised that “qualitative research seeks to understand and explore people’s social realities as they experience them in their daily lives”. In supporting these views, Leedy and Ormrod (2005:95) assert that “the results of qualitative research tend to focus on the interpretation of social meaning through representing the social world of the research participants”.

Denzin and Lincoln (1994:6) earlier argued that “a qualitative research paradigm is a multi-dimensional methodology involving an interpretative naturalistic approach to its subject matter”. Further, Leedy and Ormrod (2005:95) argue that “qualitative research helps to understand complex situations and is exploratory in nature, as the findings can be used to build theory”. These points imply that a qualitative researcher is able to engage respondents in their natural environment and also interpret their inputs without any further interference.

In an attempt to simplify this paradigm, Strauss and Corbin (1990:17) assert that qualitative research is “any kind of research that produces findings not arrived at by means of statistical procedures or other means of quantification”. They argue that whereas a quantitative researcher seeks to predict and generalise the results of their findings, a qualitative researcher seeks to illuminate and explain the situation at hand. Despite the differences, Strauss and Corbin (1990:17) clarified that “both paradigms can be used together in a complementary manner towards each other”. That is, a qualitative paradigm can be used to identify variables which might later be tested quantitatively. As for this research, I used a qualitative paradigm for the reasons advanced below.

3.2.1.1 Why a qualitative paradigm?

Drawing from the problem at hand, the posed research questions and research objectives, I have decided to use a qualitative paradigm as the appropriate method to address the issues. A qualitative paradigm was not chosen as an easy substitute for a statistical or quantitative paradigm. It was mainly chosen for the following reasons. These reasons evolved from the various articulations made by scholars such as

Strauss and Corbin (1990), Denzin and Lincoln (1994), and Leedy and Ormrod (2005).

 First, in order to determine the rationale for, and the state of cooperation among democratic intelligence oversight structures, a qualitative paradigm allowed me to spend many hours in the field collecting extensive data in order to gain an insider perspective on the need for the cooperation of the oversight structures. It also allowed me to determine their current state of cooperation.  Second, in order to establish the nature of cooperation barriers experienced by the democratic intelligence oversight structures, a qualitative paradigm allowed me to conduct semi-structured interviews with oversight officers and external experts to obtain the required empirical data.  Third, in the quest to gather proposals among various role players on what strategies can be adopted to improve cooperation among democratic intelligence oversight structures, a qualitative approach allowed me to engage a multiplicity of stakeholders and elicit their suggestions on how to improve the cooperation of oversight structures.

On the basis of the above-mentioned reasons, I found it appropriate to use a qualitative research paradigm for this study. This has assisted me to address the problem under investigation and achieve set objectives. This paradigm allowed me to engage the oversight officers in their natural settings through direct gathering of data from the respondents themselves. Most importantly, the qualitative paradigm allowed me to become an active learner able to narrate the respondents’ messages from their own view points, rather than an outsider who is just passing judgement on them. I do concur with scholars such as Eisner (1991) and Cronback (1975), who once said that a quantitative approach cannot by itself cover the developments taking place in the social environment.

Similar views were also expressed by scholars like Creswell (1994) and Patton (1990), who in most of their writings affirmed the qualities of a qualitative paradigm and its ability to effectively address the social problems. The following summary explains the adopted research design for this study.

3.3 RESEARCH DESIGN

As Welman and Kruger (1999: 78) put it, “research design describes the plan of action indicating the specific steps that are necessary to provide answers to the research questions, test hypotheses, and thereby achieve the ultimate research purpose”. Creswell (1994: 67) divides qualitative research paradigms into five main types of designs, namely: “biography, phenomenology, grounded theory, ethnography and case study”. From among these five, I opted to use the case study design for the successful investigation of the problem at hand. Based on Creswell’s (1994) assertion, I identified the case and ensured that the selected case is worth researching. The following discussion unpacks the nature of qualitative case study followed by its relevance in addressing the problem at hand.

3.3.1 Qualitative case study

By definition, Thomas (2011:35) regards a case study as a “research strategy that empirically enquires and investigates a phenomenon within its real life context”. It can be used to analyse a person, events, decisions, period and/or even a project. Even though scholars like Burgelman (1983), Pettigrew (1973), Penrose (1960) and Chandler (1962) have praised a case study approach as an important tool for generating and testing theories in social sciences, scholars like Campbell (1978), March, Sproull and Tamuz (1991), Miles (1979), Draft and Lewin (1990), raise concerns around its methodological rigour concerning how valid and reliable its results are. Based on Yin (1994) and Eisenhardt’s (1989) work, Bergh, Perry and Hanke (2006) also argue that a

83 lack of rigour gets exposed when the interface of the various variables are expanded and tested. In 2000, Scandura and Williams (2000:1263) cautioned that “without rigour, relevance in management research cannot be claimed”.

However, after 20 years of inaction, scholars like Eisenhardt and Graebner (2007), Weick (2007) and Siggelkow (2007) contributed to the revival of the case study method for generating and testing theories. With this development, scholars like Pettigrew (1973) felt vindicated, as he had argued as far back as 1973 that the power of the case study is that it seeks to study a phenomenon in its context rather than independent from its context. Similar views were expressed by Leedy and Ormrod (2005:95) who argued that “the power of a case study approach is that its analysis is done on things in their natural settings and this enables me to make sense of the researched phenomenon”.

According to Yin (2009) and Stake (1995) the notion of a constructivist paradigm does underpin the qualitative case study which then recognises the element of subjectivity in people’s creation of meaning. Interestingly, it does not reject outright the idea of fairness and objectivity. The constructivist paradigm is built upon the premise of a social construction of reality (Searle: 1995). As Crabtree and Miller (1999) put it, a constructivist paradigm allows the working together of the researcher and the participants and/or respondents while allowing them to tell their story. It is on the basis of these arguments that I was able to understand the participants’ decisions and actions.

Further, Feagin, Orum and Sjoberg (1991) describe the case study as a triangulated approach which can be done through multiple data sources whether at theoretical level and/or methodological level. In this research, I have triangulated data through acquiring it from multiple sources and these include the oversight officers, external experts and from the management of various oversight institutions. This approach is advanced by scholars like Yin (1994) and Stake (1995), who argued that collecting data from various

84 respondents improves the validity and reliability of results. As described by Yin (2003:23), a case study can be used:

 “to answer the how question,  to avoid the manipulation of respondents,  to cover the contextual conditions given their relevance to the phenomenon under investigation, and  to determine the boundaries between the phenomenon and the study context”.

These four situations were found to be relevant for this research, hence the decision to adopt a case study approach. The following section provides the main reasons why a case study approach was chosen for addressing the problem at hand.

3.3.1.1 Qualitative case study approach

This approach was chosen as it allowed me to study and understand the problem in detail to enable me to present comprehensive detail on the nature of the findings at hand as articulated by Van Wynsberghe and Khan (2007), and Yin (2009). A similar understanding is expressed by scholars like Hammersly (1989) and Ragin (1992), who see a case study as an approach that is capable of facilitating information gathering from multiple respondents to enable the better understanding of a particular phenomenon. Therefore, based on the need to determine the rationale for and state of cooperation among the democratic intelligence oversight structures, I used the qualitative case study approach. The description of a case study fits with what this study seeks to achieve, including the need to engage a range of respondents. A case study approach assisted me to establish the nature of the barriers experienced by the various intelligence oversight structures when cooperating with each other. It also helped in the extraction of suggestions as to what strategies could help to improve the cooperation of the various intelligence oversight structures in doing their work.

With this approach, I studied the democratic intelligence oversight structures at various levels of governance. In order to understand their state of cooperation, I used empirical data collected from the respective structures and derived certain conclusions about the problem at hand. The option of studying a variety of cases complies with Creswell’s (1994) assertion that the researcher could choose a sizable number of cases to investigate. He later cautioned that researching more than one case could dilute the overall analysis and depth in a given case. However, this assertion does not apply to this research, as the intention was not to exhaust each case, but to understand how various cases cooperate with each other. As Yin (2003) puts it, multiple cases enable the researcher to explore the differences within and between cases and the findings can be replicated across the cases. As Creswell (1998), Creswell and Miller (2000) and Yin (2009) put it, a qualitative case study enabled me to investigate and gain an in- depth understanding of the problem from key informants involved in the activities being studied.

Yin (1994:4-6) sub-categorises case studies into three main categories and those are: exploratory, descriptive and explanatory case studies. This research attempted an exploratory case study in answering the ‘how’ question as articulated by Yin (1994:67). He asserts that; “an exploratory case study explores the situation in which the interventions being evaluated have no clear or single set of outcomes”. I analysed the state of cooperation among the democratic intelligence oversight structures with the intention to improve their cooperation. The research adopted an un-sequenced approach in which the various structures were studied to understand their efforts to ensure effective cooperation among them. This approach is supported by scholars like Yin (1994), who advises that a sequenced approach requires that all extracted data from the individual cases be pulled together to ensure the integrated understanding of the problem under investigation. The following discussion outlines the research procedure which was followed in this study.

3.4 Research procedure

A research procedure provides the manner in which the research was conducted from beginning to end. It clearly spells out what I did at what point in the research process. In social sciences, the research procedure remains central in guaranteeing how reliable and valid the research results are. This matter was fully articulated by scholars like Newman (2011), Yin (1994) and Cook and Campbell (1979). The following Table contains a summary of the research procedure adopted and the interface between the data collection instruments, research questions and the analytical techniques. As already outlined in Chapter 1, the main research question is posed as follows: How can the cooperation of the South African democratic intelligence oversight structures be improved for good governance results?

Table 3.1: Research questions, interview questions, data collection instruments and techniques Three research sub- Interview questions Data collection Objective Analysis questions instrument technique 1. What is the rationale 1.1 Refer to  Interv  To establish the  Qualitative for, and state of Appendix A and B as iews rationale for, and state of  Descriptive cooperation among attached  Docu cooperation among the  Explanatory democratic ments review democratic intelligence intelligence oversight oversight structures for good structures for good governance results. governance results? 2. What are the 2.1 Refer to  Interv  To identify the nature  Qualitative barriers for the Appendix A and B as iews of cooperation barriers  Descriptive systematic attached  Docu experienced by democratic  Explanatory cooperation of ments review intelligence oversight democratic structures for good

intelligence oversight governance results. structures for good governance results? 3. What 3.1 Refer to   To elicit suggestions  Qualitative strategies could be Appendix A and B as Interviews to develop strategies to  Descriptive developed to improve attached  Docu improve the cooperation  Explanatory the systematic ments review among democratic cooperation of intelligence oversight democratic structures for good intelligence oversight governance results. structures for good governance results?

3.4.1 Data collection instruments

According to Bless and Higson-Smith (1995:99), “the determination of data collection instruments is based on the nature of data to be collected”. In fact, data can either be primary or secondary. In this research, both primary and secondary data was collected through the use of various data collection instruments. Yin (2009) provides examples of data collection instruments as interviews, documents review, participants’ observations, focus group discussions and archival records. In an attempt to respond to the above- mentioned questions, this study used both interviews and documents review for the collection of primary and secondary data respectively.

3.4.1.1 Primary data

Bless and Higson-Smith (1995:99), describe primary data as “the kind of information that is directly collected by a particular researcher to directly answer his/her questions to address his/her own problem under investigation”. In this research, the primary data was collected through structured and semi-structured interviews. The following discussion describes the manner in which the interviews were used to collect primary data.

3.4.1.1.1 Sample and interviews

According to Neuman (2010:219) a sample is “a smaller set of cases which a researcher selects from a pool and generalises to the population”. Neuman (2010) further asserts that a sample is just a group of a selected few from the broader population and should be as representative as possible. Crouch, Mira, Mackenzie and Heather (2006) argue that in qualitative research the determination of a sample is concerned with creating meaning rather than making generalised hypothesis statements. They further argue that a qualitative approach takes time and involves analysing huge quantities of data which at times takes too much time and often is difficult to implement. Compared to a quantitative sample, qualitative research uses a generally small sample size and in this case only 35 respondents were sampled.

Further, Lewis and Ritchie (2003) point out that a qualitative sample has a point of diminishing returns. This implies that as the study continues, the consistent collection of more information after saturation does not bring new narratives. It is on this basis that Mason (2010) argues that, saturation should be used as a guiding principle in the determination of a sample size within qualitative research. As far back as 1967, Glaser, Barney, Strauss and Anselm (1967) argued that any faithful qualitative researcher should follow the concept of saturation in determining a sample size. On the other

89 hand, Charmaz (2006) argues that a sample size is determined by nothing else but the aim and the objectives of the study. In addition, Morse (1995:4) argues that “the sample size is determined by the research scope, nature of topic, quality of data and the study design”. Morse (1995) wrote that qualitative researchers often claim to have reached saturation, but could not provide an overt description of what it meant or how it was achieved. Based on these assertions, it is clear that scholars do not agree on what constitutes a sufficient sample size in qualitative research.

This conundrum is inextricably linked with the debate on the sufficient number of interviewees within a qualitative research approach. Therefore, scholars such as Guest, Bunce and Johnson (2006: 59) indicate that, saturation is found understandable when at the conceptual stages but not during data collection. Based on the discussions, it is clear that saturation can be limited to the conceptual stages of the research and not included in the actual research process.

Mason (2010) also added his voice to the conundrum of sampling in a conventional piece of social science research. While studying the sample sizes and the saturation of Doctor of Philosophy (PhD) studies using qualitative interviews, Mason (2010) started by expressing concern about the absence of scholarly consensus regarding the number of purposive interviews to be conducted during qualitative research. However, he acknowledged that various scholars do provide guidelines on this matter. Morse (1994:225) suggests that “in ethnography and ethno science based researches, the researcher can interview between 30 and 50 interviewees”. Further, Bernard (2000:178) asserts that “most ethno science studies conducted between 30 and 60 interviews”. On grounded theory, Creswell (1998:64) suggested between 20 and 30 interviews, while Morse (1994:225) suggested between 30 and 50 interviews on the same method. In phenomenology studies, Creswell (1998: 64) suggested between five and 25 interviews, while Morse (1994:225) suggested only six interviews. Guest, Bunce

90 and Johnson (2006:10) suggested that 15 interviews should be the acceptable minimum sample size in qualitative research.

Although the above discussion presents guidelines, the scholars did not provide many coherent reasons for their respective suggestions and guidelines. Further, even for the scholars who argue that certain methodological approaches will need more respondents, no coherent arguments were advanced to support the suggested numbers of interviewees. Therefore, in this study I purposively selected 35 respondents in total from both structured and semi-structured interviews.

Given the nature of the problem at hand, most of the useful responses were solicited through the semi-structured interventions. As Bogdan and Biklen (1982), and Patton (1990) put it, semi-structured questions have no predetermined responses and therefore allow the researcher to probe and further explore within the predetermined scope of the inquiry. Of the 35 respondents, 28 were drawn from the seven intelligence oversight structures as described in Chapter 4, under section 4.2. In each structure, four people were interviewed of which two were from management while the other two were oversight officers/auditors/investigators or overseers themselves. This kind of categorisation and approach used in this research is supported by various scholars such as Ndimande (2012), Muthuma (2011), and (Sangster-Gormley: 2013). They affirm the importance of soliciting responses from a multiplicity of respondents as this has the potential to complete the picture and allow for a good triangulation process. The remaining seven interviewees were drawn from among the academics, commentators, scholars and researchers specialising on intelligence tradecraft and/or oversight.

3.4.1.2 Secondary data

According to Bless and Hinson-Smith (1995:99), secondary data is described as “information collected by other researchers concerning their own research problems and gets used by other researchers to address their problem under consideration”. As for this research, secondary data was collected from relevant publications on the democratic intelligence oversight structures, the Constitution, annual reports and other relevant legislative and regulatory prescripts. A further batch of secondary data was collected from books, journals and/or periodicals. This was mainly done to determine the rationale for, and state of cooperation among intelligence oversight structures in overseeing members of the intelligence services. The following discussion describes the manner in which document review was used for the collection of secondary data.

3.4.1.2.1 Documentary review

As Yin (1994) argues, there are various advantages of using document review as a data collection instrument. It represents a neutral piece of data which existed even before the conceptualisation of the research. Therefore, such data can be analysed over and over again until a particular picture emerges. Procedurally, I started by collecting secondary data through different document reviews in order to be adequately prepared to advance probing questions during structured and semi-structured interviews. This was an important approach mainly for triangulation purposes, as documentary evidence was juxtaposed with the answers received from respective interviews (Yin: 1994). One of the most important advantages of using document review for data collection was that it allowed me to adequately answer the question of the rationale for the cooperation of intelligence oversight structures, in addition to the answers of the respondents.

During the research, I studied relevant documents ranging from academic books, journals, periodicals, annual reports, strategic plans, pieces of legislation and regulatory frameworks on intelligence and its oversight in particular. With the uniquely adopted oversight model, a clear understanding was sought in terms of the nature of cooperation among the various intelligence oversight structures operating at various levels of governance. This was an important venture in order to understand issues around the cooperation barriers and develop strategies to improve it. This approach is also advanced by scholars like Patton (2002) who argues that it is important to study the external environment in order to have a clear understanding of the internal realities from the holistic picture. The findings of the documentary reviews of intelligence practices in the selected international case studies have been summarised and assessed in Chapter 2.

3.4.2 Target population and description of respondents

According to Charles (1988:73), a target population refers to all cases or people who may be included in an investigation. Further, McMillan and Schumacher (1989:161) describe the target population as a larger group of cases from which a sample can be selected or drawn. To expand this assertion, Sekeran (1992:225) argues that, “a target population can be the entire group of people, events or things of interest which I may wish to investigate”. In this study, the targeted population included the oversight officers, members of management within the intelligence oversight structures. It included experts, academics, researchers and commentators in the field of intelligence tradecraft. As described by Neuman (2010), Patton (1990), Lincoln and Guba (1985) and Merriam (2002) this large group was sampled in order to draw participants from the entire research population.

In terms of this study, the respondents could be described in the following manner. Oversight officials referred to individuals employed within the various democratic

93 intelligence oversight structures whose job description revolves around overseeing members of the intelligence services. They are local officials at the coal face and have first-hand experience in overseeing the operations of members of the intelligence services. However, in this category the respondents ranged across various levels of responsibilities including those with functional and line managerial responsibilities. Further, the members of management refer to individuals serving in the Executive Committees (EXCO) or top management in these intelligence oversight structures. These represent the individuals whom the oversight officials report to. External experts are individuals with specialised and expert knowledge on intelligence tradecraftor oversight and are located outside the oversight/overseen structures themselves.

3.4.3 Data presentation

After the completion of data collection, the collected data was presented in order to facilitate its analysis. In this research, data presentation was arranged based on the applicable data collection instruments and the research sub-question as posed. In this research, data was presented in a narrative format supported by evidence from raw data in the form of direct quotes or paraphrases (Bryman: 2008). In terms of data acquired through interviews, I searched for patterns of evidence, identifying the similarities and differences and then packaged them accordingly. During the presentation of data, it was codified for ease of organisation and the identification of themes which served as the basis for data categorisation. During the packaging process, I looked for emerging categories and presented them to enhance easy reading. Further, I looked for the emergence of new concepts, constructs, propositions and possibly theoretical imperatives. These were juxtaposed with existing literature, and the conceptual and theoretical framework.

As this research seeks to improve the cooperation of democratic intelligence oversight structures for good governance, the collected data was packaged in accordance with

94 three broad categorical imperatives. The first category included all data which described the rationale for, and state of cooperation among the democratic intelligence oversight structures. The second category reflected the nature of barriers faced by democratic intelligence oversight structures. The third category reflected the suggestions from the respondents on strategies to improve the cooperation of democratic intelligence oversight structures. Within each broad category, I developed sub-categories based on the themes from the collected empirical data.

After the coding of each data set, I applied a quantitative approach by counting the relative frequency of the type of responses. This was important as the frequency assisted me to decide what weight should be given to a particular category of responses. I also related certain metaphorical phrases from the raw data to the specific realities. Furthermore, I also explored the presentation of data through charts, images, photographs, diagrams, maps and drawings. This approach was only used with specific collected data sets.

In terms of writing, I adopted a major to minor approach in data presentation as described by Chenail (1995). In this approach, Chenail (1995) advises that the most important findings should be presented first with minor discoveries presented last. All this was done through the application of a plausibility test in which I used my own experience and knowledge to determine the plausibility of certain data sets in order to ensure accurate data presentation. The following discussion outlines the manner in which collected data was analysed.

3.4.4 Data analysis

Data analysis is needed to make sense of the data in relation to the problem under investigation as part of managing the data. This process is explained by scholars like Bogdan and Biklen (1982), Patton (1990), Brown (1996), and Strauss and Corbin

(1990), who argue that qualitative data analysis is all about data management. In this process, I organised collected data into “sizable manageable units, synthesized it, searched for patterns, discovered what is important” and decided on what to communicate to the reader (Patton, 2002: 34). However, I started the process of data management during my data collection phase in the field. This was emphasised by scholars like Merriam (2002), Patton (2002), Stake (1995), and Miles and Huberman (1994), who argue that preliminary data analysis begins at the field level where the researcher prepares field summaries during the data collection process.

In drafting field summaries, I elevated key issues relevant to the problem under investigation. This was critical for recording key findings as they emerge during interviews and documents review. Further, I recorded all emerging concepts and themes which were interpreted on the basis of the literature and theories as presented in the theoretical framework (Rubin and Rubin: 1995). For data analysis, I used qualitative, descriptive and explanatory data analysis techniques starting with qualitative data analysis. In both techniques, I followed an explanatory approach.

3.4.4.1 Qualitative data analysis

During data analysis, I observed the emergence of three epistemological and phenomenological ‘voices’ which had to be properly managed. Those were the ‘voice’ of the collected empirical data, the ‘voice’ of the conceptual framework, and my own internal ‘voice’. At that point, I dissected the emerging themes and codes by juxtaposing the meaning emanating from data and the conceptual framework to construct scientific knowledge. During this process, I avoided imposing my “views by setting aside preconceived knowledge and remained open, sensitive and empathetic to the participants’ responses” (Creswell 1994: 78). This kind of posture is advanced by Creswell (1994) who advises that, a qualitative researcher should record his subjectivities and acknowledge them in the research report. However, in analysing

96 qualitative data, Creswell (1994) concedes that the general understanding of the researcher and his experiences and judgements remain a critical tool in making sense of the collected empirical data. This is necessitated by the absence of a clearly defined manner in which to analyse opinions, choices, descriptions and feelings (Merriam: 1998).

During this process, I generated new forms of meaning in order to transform perspectives and actions (Becker: 1996). This conforms to Kohlbacher’s (2006) assertion that qualitative data analysis allows categories to emerge out of data and also emphasises the understanding of meaning. I did all this in order to understand the rationale for, and the state of cooperation among intelligence oversight structures, their barriers, and required strategies to improve their cooperation modalities (Merriam: 1998). This approach is supported by scholars like Sidani and Sechrest (1996), who assert that the primary objective of a qualitative data analysis is to unpack and simplify the complicated social environment by the researched themselves.

Hence, I developed various qualitative measurement scales to facilitate my interpretation of the collected data. I developed these scales in line with Stanley Smith Stevens’ distinction among nominal, ordinal, interval and ration measurement techniques (Stevens: 1975). I used qualitative measurement scales that classify responses on different questions posed. I then used the elementary quantification of responses in real numbers and percentages to populate those scales in order to make more accurate sense of different meanings that my respondents attended to. In the presentation of responses, I used various types of charts whether pie or otherwise as presented in Chapters 4 to 6. The summaries of the findings are attached as Appendices A and B. The following section explains in more detail the manner in which I used descriptive approaches in analysing the collected data.

3.4.4.2 Descriptive data analysis

As already emphasised by Creswell (1994), qualitative data analysis seeks to derive meaning from a complex social reality. Erikson (1963) had earlier divided meaning into two subcategories: common meaning and unique meaning. However, it should be noted that what might be common to one person may be unique to somebody else. Therefore, with descriptive data analysis, I identified the contributions of individuals or a group’s unique meaning based on their area of operation. For instance, Intelligence Oversight Officers from the OIGI remained worried about discussing security modalities while officials from the office of the PP and those from HRC remained frank and open. Therefore, in line with Merriam’s (1998: 67) assertion, I used descriptive data analysis “to generate new forms of meaning in order to transform perspectives and actions”. By using descriptive data analysis, I wanted to understand, among others, the barriers for the effective cooperation of the various intelligence oversight structures. During this process, I also used quantitative descriptive data analysis whereby I reflect my findings in the form of percentages for easy understanding. Though qualitative in posture, my presentation of findings embraces the lowest level of numerical considerations through the utilisation of numbers and percentages to interpret findings.

As Becker (1996: 47) puts it, “the ability of descriptive data analysis to generate meaning makes it a unique and powerful epistemological tool for understanding even seemingly mundane experiences”. Simply put, descriptive data analysis allowed the movement from raw data into forming explanations, understandings or interpretations of the people and the situation under investigation. With descriptive data analysis, I ensured that tabulation of enough evidence to fully support drawn conclusions was a critical tool for ensuring the validity and reliability of results.

3.5 RELIABILITY AND VALIDITY

To ensure the reliability and validity of the research results, I used a qualitative research approach within the context of the positivist tradition as championed by scholars like Cook and Campbell (1979), and Behling (1980). They argue that reliability and validity can be achieved when general reliability, construct validity, and both internal and external validity are fully observed. This approach has been adapted and used by scholars like Yin (1994) in his various qualitative case studies to enhance his results’ reliability and validity. Neuman (2011:208) describes reliability as “dependability, consistency and truthfulness which is concerned with how well an idea fits with the actual reality”. Similarly, I undertook the various activities described below to ensure the results, reliability and validity of this research.

3.5.1 Internal validity

To ensure internal validity, I provided compelling logical argumentative reasoning in this thesis which explains and clarifies how the research conclusions were reached. This report has attempted to present and maximise the logical consistency of the causal linkages between the research problem, the research question and sub-questions, the research objectives, findings and conclusions that are contained in the descriptions, comparisons, analyses and assessments of the issues addressed in the thesis. As noted before, I tried to remain neutral and removed from the research context; in order to arrive at objective conclusions. As Cook and Campbell (1979) and Yin (1994) put it, internal validity or logical validity investigates the causal relationship between the various variables and research results. Further, Yin (1994:105) clarifies that the question of internal validity mostly “applies at the analysis phase of the research”. Therefore, I allowed the ‘voice’ of data to take precedence during the data categorisation, theme development and its subsequent analysis. However, Lincoln and Guba (1985) argued that, it is sometimes difficult to determine internal validity given that

99 one will have to precisely know the nature of that reality. Ironically, if one already knows a particular reality there would be no need to undertake scientific study to test such reality. The following section discusses ensuring construct validity.

3.5.2 Construct validity

In terms of this validity type, I had to ensure the high quality of conceptualisation or operationalisation of the concepts as described by Gibbert, Ruigrok and Wicki (2004:1466). The linkages among critical concepts like oversight, evaluation, accountability, transparency and cooperation were clearly spelled out in Chapter 2, under section 2.5 titled conceptual framework. These linkages were used to illustrate the conceptual interfaces between these concepts and the theoretical framework. Like internal validity, Denzin and Lincoln (1994) argue that construct validity also applies at the data collection phase of the research. Therefore, I tried to ensure synergy between the research questions and the actual investigation or data collection. This was important in order to ensure that the research addresses issues it intended to address. All of the above were based on procedural correctness, which led to the accurate interpretation of reality as suggested by Denzin and Lincoln (1994). However, Yin (1994:41) argued that most case study researchers have a “tendency of not developing a well-considered set of measures, thereby allowing their subjective judgements to take precedence”.

To address the problem of subjectivity, I attempted to adduce adequate clear and objective incidences of evidence which can be checked and deconstructed to understand and validate the final conclusions. Further, I triangulated the findings by looking into the different angles of the same phenomenon on the basis of responses received from different documentary and empirical data sources (Denzil and Lincoln: 1994) and (Yin: 1994). The collection of data through interviews and documentary

100 reviews served as the first basis for coherent triangulation to enhance construct validity. The following section discusses ensuring the external validity of the research findings.

3.5.3 External validity

To ensure external validity, I provided clear rationale for the selection of case studies as articulated by Cook and Campbell (1979:83). This was important for ensuring an appreciation of the sampling choice, which is summarised in more detail in section 3.4.1.1.1 above. In this study, I applied cross-case analysis which provided a good basis for analytical generalisation. As described by Yin (1994), cross-case analysis is a tested approach which allowed me to analyse various sub-cases among the multiplicity of intelligence oversight structures as described in Chapter 4, under section 4.2.

As Calder, Phillip and Tybout (1982), and McGrath and Bringberg (1983) argue, external validity or generalisability seeks to derive theories to account for phenomena not only where they were studied, but also in other settings. In fact, the three validity types are closely linked to each other as mooted by Cook and Campell (1979:83) and Mook (1983:379). This was captured by Scandura and Williams (2000:1261) who indicated that, “without clear theoretical and causal logic (internal validity) and without a link between theoretical conjecture and empirical observations (construct validity), there can be no external validity”. The following section summarises the activities I undertook to ensure the reliability of my research findings.

3.5.4 Reliability

According to Denzin and Lincoln (1994), reliability is considered achieved if a subsequent researcher can conduct the same study and follow the same steps and arrive at the same conclusions. To ensure results reliability for this study, I produced a case study protocol through “careful documentation and clarification of research

101 procedures” as suggested by Yin (1994:89). Denzin and Lincoln (1994) further argued that, reliability remains concerned with issues of transparency and replication. Therefore, to ensure replication in this study I kept a case study database which included notes, documents and narratives collected during fieldwork and arranged them in such a way that they are easy to retrieve for future use. This was done to eliminate the possibility of any random errors as cautioned by Leonard-Barton (1990).

3.6 ETHICAL CONSIDERATIONS AND LIMITATIONS

On ethical considerations, Sangster-Gormley (2013) cautions that before the collection of any empirical data, the researcher should obtain an ethical clearance certificate from the university’s ethics committee. He further indicated that the researcher should clearly indicate how they will gain access to the research setting wherein the empirical data will be collected. This research project has fully complied with the requirements of the University of Johannesburg’s ethics committee. During the research, I worked closely with the identified informants or research participants in the collection and presentation of the accumulated data. Yin (2009) describes this process as gaining entry and creating a relationship with the identified research participants.

Therefore, in addressing possible ethical concerns I observed the four ethical considerations as articulated by Beauchamp and Childress (1983) and those are: respecting the rights of respondents, doing good at all times, deciding not to do any harm and remaining ready to pursue equity at all times. Brikci and Green (2007) regard consent and confidentiality as the two key ethical issues to be considered in any project. In this study I ensured that all participants freely consented to participate in the research without any coercion or undue pressure. All participants were fully informed of their participation protocols which included the fact that they had the right to withdraw at any point of the process without compromising any service that had been received at any point in time.

102

Before any written consent form was signed, participants were given the opportunity to decide whether they wished to sign the form or not. This option served to avoid any possible suspicions or any element of discomfort among participants. As for confidentiality, I protected the identities of all respondents as advocated by Rocha (2004). To achieve this, Rocha (2004) recommends the full password protection of respondents’ information stored in notebooks or computer files. As a result, all notebooks and computers where I stored respondents’ information were password protected and stored in a safe place.

In terms of limitations, the very nature of the intelligence topic made it difficult for me to source all relevant data to adequately answer the research questions. However, I did not seek to acquire classified operational data on the democratic intelligence oversight structures, but to explore mechanisms to improve the cooperation of intelligence oversight structures for good governance results. In case of the need for acquiring classified information; I am familiar with the applications processes to acquire such information in terms of the Promotion of Access to Information Act 2 of 2000, the Protected Disclosures Act and the Minimum Information Security Standards (MISS). However, during this research process this need did not arise. As an employee of the NICOC, I remained well versed on all protocols regarding information classification, declassification and distribution. This matter has already been addressed in Chapter 1 under section 1.8.

3.7 SUMMARY AND CONCLUSIONS

This chapter provided a succinct summary and explanation of the adopted research paradigm together with the research design used for this study. Based on this summary, the study used a qualitative research paradigm which served as the point of departure during data collection, analysis, representation and assessment. In terms of research design, the study used a qualitative case study approach which covered a

103 multiplicity of intelligence oversight structures.

The chapter clarified the main data collection instruments used in the study, which were interviews and documentary reviews. Descriptions of the target population, sample size and the nature of the respondents were also provided. In terms of data analysis, this study used qualitative, descriptive and explanatory data analysis techniques and assessment tools. Further, the chapter clarified the process to ensure result validity and reliability.

While this chapter focused on the applicable research design and methodology, the following chapter focuses on the rationale for, and the cooperation of intelligence oversight structures. Its starts by presenting the South African democratic intelligence oversight architecture before providing details about why South Africa needs a vibrant intelligence oversight mechanism characterised by the systematic cooperation of these structures. It also makes a determination on the current state of cooperation among the prevailing intelligence oversight structures.

104

CHAPTER 4 STRUCTURAL COOPERATION AMONG INTELLIGENCE OVERSIGHT STRUCTURES IN SOUTH AFRICA

4.1 INTRODUCTION

This chapter starts by providing a concise summary of the South African governance system within which the current intelligence oversight model operates. This is followed by a summary and assessment of the prevailing South African intelligence oversight model. This is important for the proper assessment of the cooperation of the oversight structures in overseeing members of the intelligence services. The assessment is sub- divided according to the locations within which the various structures operate ranging from the legislature, the executive, judiciary, administrative and the chapter 9 institutions. This chapter also provides the rationale for the cooperation of intelligence oversight structures, followed by the determination of their cooperation modalities.

While the previous chapter summarises the research methodology outlining the procedures which were followed in the implementation of this research project, this chapter presents and assesses the research findings related to the first research sub- question which was posed as follows: RQ 1: What is the rationale for, and the state of cooperation among democratic intelligence oversight structures for good governance results?

In South Africa, the issues relating to the cooperation of intelligence oversight structures gained prominence after the ruling elites got accused of misusing intelligence structures for their own factional political agendas (OIGI: 2006). Despite the concerns and public debates in South Africa about the abuse of intelligence structures, very few scholarly studies have so far been undertaken to analyse and assess the existing cooperation

105 mechanisms of the various intelligence oversight structures (Nathan: 2010 and Dube: 2013).

The concept of cooperation is explained and assessed in Chapter 2, under section 2.5. As indicated there, cooperation is conceptualised by Rogers (2012) as the working together of two or more entities in a mutually beneficial exchange for the realisation of common goals. Cooperation also implies an effort to avoid competition and advance the spirit of cooperative governance in line with the country’s constitutional imperatives (RSA Constitution: 1996).

4.2 THE SOUTH AFRICAN DEMOCRATIC INTELLIGENCE OVERSIGHT ARCHITECTURE

After 1994, the ANC government adopted an explicit intelligence oversight mechanism to oversee the work of the intelligence services. The mechanism is made up of various structures operating at various levels of governance. As indicated in Chapter 1, under section 1.3 these structures operate within the legislature, executive, judiciary, administrative and chapter 9 institutions. Figure 4.1 visually illustrates the location of these structures and their legislative and regulatory frameworks. This is important to properly contextualise the rationale and the state of cooperation among these intelligence oversight structures.

106

FIGURE 4.1: South African Intelligence oversight model 2016

Source: (www.ssa.gov.net)

4.2.1 Intelligence oversight by the legislature

At the legislative level, intelligence oversight is undertaken by the JSCI which consists of Members of Parliament. The JSCI together with the Joint Standing Committee on Defence (JSCD) are the only two standing committees established by parliament in terms of the Constitution, legislation and rules, given the sensitive nature of issues discussed in these committees. According to parliament’s website (www.parliament.gov.za), parliamentary oversight is meant to monitor government’s work towards the achievement of goals set out by the relevant pieces of legislation. As Born (2004:58) argues, legislative oversight does not include the day-to-day management of the intelligence services, but the monitoring of the executive and the

107 intelligence services in terms of their compliance with the country’s legislative framework. In the US, Gill (2007) indicates that the adoption of a legislative oversight mechanism was subject to serious scepticism and suspicion as some politicians believed that parliamentarians might leak intelligence information for selfish political gains.

However, in Germany, Dietrich (2015) indicates that the application of a parliamentary oversight function for intelligence services can also be used as part of the political mechanism to control members of the services. Therefore, members of parliament should not only be concerned about illegalities but also focus on the role of intelligence services in preserving the country’s constitutional democracy. During the researcher’s global benchmarking exercise which was summarised in Chapter 2 (section 2.2 and 2.3), a similar parliamentary mechanism was found in New Zealand called the ISC. The following discussion seeks to outline the work of JSCI as a parliamentary oversight structure in the South African intelligence oversight model.

4.2.1.1 Joint Standing Committee on Intelligence (JSCI)

Section 3 of the Intelligence Services Oversight Act of 1994 established the JSCI as a parliamentary oversight mechanism and outlines its legislative mandate as follows:

 “To obtain audit and other reports from the Auditor-General and to consider financial statements of the intelligence services.  To obtain a report from the Evaluation Committee on the evaluation conducted on the intelligence services with accompanying comments or recommendations.  To obtain from the designated Judge a report regarding the functions performed in terms of the Regulation of Interception of Communications and Provisions of Communication-Related Information Act of 2002, including statistics of interception requests made by the intelligence services.

108

 To consider and make recommendations on the report and certificates presented to it by the IGI.  To obtain reports from the Ministers responsible for the Intelligence services.  To consider and make recommendations on all proposed legislation relating to the Services and Agencies and to initiate legislation in connection with such Services and Agencies.  To order investigation by and to receive a report from the Head of a Service or the Inspector-General regarding any complaint received by the Committee from anyone”.

In accordance with the NA Rule 137(2) and the National Council of Provinces (NCOP) Rule 102(2), “each committee conducts its business on behalf of the House and should therefore report to the house on the matter referred to it for consideration”. According to Hannah et al (2005:23), “The South African JSCI is similar in function to the Canadian Security Intelligence Review Committee (CSIRC)”. Further, section 4(1) of the Intelligence Services Oversight Act of 1994 “empowers the JSCI to have access to intelligence, information, and documents in the possession or under the control of the Intelligence Service”. These processes are used by parliament as important oversight instruments for monitoring the functioning of the intelligence services. Section 5(1) of the Oversight Act further states that “the JSCI shall conduct its functions in a manner consistent with the protection of national security, hence the need to duly ensure the vetting of committee members”.

However, Dlomo (2004) indicated that due to resource constraints and the high demand for vetting, it is not always possible to vet all JSCI members. Further, section 3(a)(ii) of the Intelligence Services Oversight Act of 1994 “empowers the JSCI to receive the Evaluation Committee’s report on the evaluation of the secret services accounts and expenditure of the SSA”. As indicated by the JSCI’s Annual Report (2009/10), “the committee has a Memorandum of Understanding with the Auditor-General of which its

109 members are also trained by them”. Of concern was Dlomo’s (2004:75) finding that “there is high level of absenteeism in the attendance of JSCI meetings by members of the committee”.

This is mainly due to the allocation of senior party members to the committee who often have other party responsibilities to fulfil. Gusy is of the opinion that “not only are the parliamentary oversight institutions blind guardians, but they are guardians without a sword” (Gusy in Dietrich 2015:135). This is mainly caused by a lack of expertise among members of parliament on intelligence related matters.

4.2.2 Intelligence oversight by the executive

In the South African oversight model, the Minister of State Security is responsible for the executive oversight and control of the intelligence services following their appointment by the president. This is done in accordance with section 209(2) of the 1996 Constitution. According to this section, the president shall appoint a member of cabinet to assume the political responsibility to oversee, control and provide direction to members of the intelligence services. According to Born (2004) executive control of the intelligence services focuses on giving them direction, tasks, priorities and resources. This also happens in Canada where the CSIS is accountable to parliament through the Ministry of Public Safety (as indicated above in Chapter 1 in section 2.3.2, which summarised and assessed the Canadian oversight experience).

However, New Zealand has a hybrid arrangement in which the executive oversight occurs through the office of the Prime Minister. Similarly, the Australians established the NSC which is chaired by the Prime Minister as a coordinating mechanism on the work of various intelligence oversight structures (Flood 2004).

110

4.2.2.1 The Minister of State Security

Section 5(a) of the National Strategic Intelligence Act of 1994 “empowers the Minister of State Security to do everything necessary for the efficient functioning of the intelligence services”. Further, section 6 of the same Act empowers the Minister “to make regulations in consultation with the JSCI on matters such as intelligence coordination, the production and dissemination of intelligence as well as the coordination of counter- intelligence”. Section 5(A.5) of the National Strategic Intelligence Act authorises the Minister of State Security to, after consultation with the JSCI, make regulations regarding the following matters:

 “Protection of information and intelligence;  Carrying out of security screening investigations;  Coordination of intelligence as an activity;  Production and dissemination of intelligence;  Coordination of counter-intelligence by the SSA DB (former NIA);  Coordination of crime intelligence; and  Execution of any other matter necessary for the effective administration of this Act”.

This was emphasised in Table 1 of the Institute of Security Studies’ (ISS) submission to the Ministerial Review Commission (MRC) (2007:18) which argued that “the Minister may, in consultation with the JSCI, make regulations regarding the protection of information, security screening investigations, and the general coordination of intelligence”. These regulations are meant to ensure the effective compliance of the intelligence services with the country’s statutory framework. Section 5(a)(1) of the National Strategic Intelligence Act of 1994 further states that “the Minister of State Security shall advise the President and the national executive on national strategic intelligence and coordination of intelligence”. In accordance with section 7(7)(c) of the

111

Intelligence Services Oversight Act of 1994 “the Minister of State Security may oblige the IGI to perform all functions designated to him/her regarding intelligence oversight”.

However, the MRC (2008:84) found that “there is no intelligence legislation which compels ministerial approval for implementation of intrusive operations to be undertaken by members of intelligence services”. Although this could be viewed as an omission in the executive oversight mechanism, the minister is not supposed to be involved in intelligence production. For example, there is no legislation which requires the intelligence services to send strategic intelligence products and evaluations directly to the minister, but to the NICOC, which in turn provides strategic intelligence advice and priorities to cabinet. At the same time, section 5(a)(1) of the National Strategic Intelligence Act of 1994 states that the Minister of State Security is to “advise the President and the National Executive on national strategic intelligence matters” (MRC 2008:87).

4.2.3 Intelligence oversight by the judiciary

At the judicial level, the South African intelligence oversight is conducted by an appointed judge. This is in line with Agere (2000:9) who argues that a judicial system in which the laws are clear and uniformly applied through an objective and independent judiciary is regarded as pro-governance and pro-development in nature. Therefore, the exceptional powers granted to members of the intelligence services should be based on the legal framework and a system of effective legal oversight (Agere 2000:9).

Dietrich (2015) argues that in Germany, should a member of the Bundesnachrichtendienst (BND) violate the provisions of the penal code during the fulfilment of their functions, such member can be brought before the German court. The question of judicial oversight is therefore not unique. In the UK, intrusive operations are authorised by appointed commissioners while the investigation of complaints is

112 done by a tribunal, and the decision of the commissioners and the tribunal are final and cannot be appealed or questioned by any court of law (Weller: 1999). As already discussed in section 2.3.2 in Chapter 2, the Canadian model empowers the federal court system to conduct oversight on the work of the intelligence services. The following section summarises the functions of the South African appointed judge in overseeing members of the intelligence services.

4.2.3.1 Appointed Judge

As already discussed, members of the intelligence services are granted special powers to acquire information through intrusive means which have the potential to infringe on people’s rights to privacy and dignity. Therefore, the Regulation for the Interception of Communication and Provision of Communication-related Information Act of 2002 compels members of the intelligence services to acquire judicial authorisation from the appointed judge prior to conducting interception of communication media. Under the RICA, 2002 the designated judge may authorise;

 “the interception of direct or indirect communications by way of an interception direction in terms of section 16 of RICA, including the interception of real time communication in terms of section 17 of RICA.  combined interception of direct or indirect communication, including real time communication related to archived communication as directed by section 18 of RICA.  the decryption of intercepted information by means of a decryption direction in terms of section 21 of RICA  entry warrants for the purposes of entering premises for the placing of interception devices in terms of section 22 of RICA”.

113

According to the ISS submission to the MRC (2008:18), “judicial oversight of the intelligence services is to ensure that intelligence operations comply with the rule of law and legal standards”. This was facilitated by the promulgation of RICA, 2002 whose purpose is to protect the subjects of the Republic against unlawful interception of their private communication. On the other hand, the Act compels the electronic communication service providers to provide a service that allows for interception and able to store communication related information. According to the annual report of the JSCI (2008/2009), the appointed judge received a total of 221 cases for judicial authorisation, of which 96 were from SAPS, 109 from NIA, six from SASS and 10 from DSO (Parliamentary Monitoring Group: 2016). Further, in the 2008/09 financial year, the appointed judge received 98 applications from SAPS CI, 70 from NIA, 14 from the disbanded Directorate of Special Operations (DSO) and seven from SASS. Figure 2.3 summarises these applications graphically and shows the number of approved and declined requests by the appointed judge (JSCI Annual Report: 2010).

Figure 4.2: Applications for judicial authorisations: 2008/2009 to 2014/2015

Requesting Approved Declined Total Agency Requests Requests SAPS CI 84 14 98 NIA 63 7 70 Disbanded 13 1 14

DSO Source: (JSCI Annual Report 2008/2009: AvailableSASS at www.pmg.org.za 6 1) 7

Financial Years 2013/2014 Requesting Agency Approved Requests Declined Requests Total SSA 226 5 231 SAPS CI 385 0 385 FIC 3 0 3 SANDF DI 4 0 4

114

Financial Years 2014/2015 Requesting Agency Approved Requests Declined Requests Total SSA 338 10 348 SAPS CI 386 0 386 FIC 17 1 18 SANDF DI 6 0 6

Comparatively from the 2008/2009 financial years to 2014/2016, there is a clear increase of the requests for approval of interception by the designated judge. In her annual report attached to the JSCI Annual Report (2015/2016), Justice Mokgoro complained of the increases of the applications. She indicated that the interception of communication media should not be considered as the investigating method of last resort by law enforcement agencies, but should be used as the last option after the failure of conventional methods (Judge Annual Report: 2014/2015 in JSCI Annual Report: 2015/2016). Within the South African legal framework, RICA does not provide for user notification, however, countries such as the US, Japan, Austria and Chile do provide for user notification. Therefore, most of the applications for the interception of communication media are taken at face value, hence the lowest rate of declined requests (Duncan: 2017).

During his interview with Brendan Boyle of the Sunday Times (5 April 2009), the former IGI, Mr Ngcakani emphasised the significance of intercepting people’s communication media in accordance with legislative mandates. He further indicated that failure to acquire such authorisation constitutes a serious offence and is punishable by law. However, this emphasis was not observed during the interceptions of telephone conversations during Project Avani. This project was legally constituted to investigate the security implications of the presidential succession debate within the ANC and the escalating service delivery protests (OIGI: 2006).

115

After the investigation into the Avani matter, the IG concluded that “the interception of voice communications by the National Communications Centre (NCC) was not only a gross abuse of facilities, but also illegal, as the requisite authority of the judge was not obtained” (OIGI 2006:16). This is a clear demonstration of the value and contribution of the appointed judge in the intelligence oversight architecture. However, as Dlomo (2004) observed as early as 2004, the JSCI has not been consistent in its publication of statistics on judicial authorisations by the appointed judge. The persistent non- publication of such statistics resembled the prevalence of non-compliance to the country’s legislative protocols on parliamentary oversight. However, following pressures on this matter, the JSCI released their 2015/2016 statistics in their annual report to the National Assembly. As part of improving the work of the designated judge, there is a need to appoint a regulatory body to ensure the compliance of the electronic communication service provider with the RICA (JSCI Annual Report: 2015/2016).

4.2.4 Intelligence oversight by administrative structures

In South Africa, the IGI also has a statutory mandate to undertake administrative oversight on members of the intelligence services. According to Hannah et al (2005:23), an IG should have “access to documents, budgets, reports and other classified information”, including information on sources. This is the kind of powers given to the IGI in countries such as New Zealand and Australia (see Table 4.1 below).

The following discussion describes the work of the South African IGI in overseeing members of the intelligence services.

116

4.2.4.1 Inspector-General of Intelligence (IGI) ] The IGI is appointed by parliament through the Joint Standing Committee of Intelligence (section 7 of the Intelligence Services Oversight Act 1994). The IGI is operationally accountable to the JSCI and administratively accountable to the Minister of State Security. According to section 7 of the Intelligence Services Oversight Act of 1994, “the Inspector-General reviews the activities of the intelligence services and monitors their compliance within policy guidelines and other mandates and principles”. In terms of section 7(7) of the Intelligence Services Oversight Act of 1994, the legislative mandate of the IG includes the following:

 “Monitoring compliance by any Service with the Constitution, applicable laws and relevant policies on intelligence and counter-intelligence;  Reviewing and investigating complaints from members of the public and members of the Services;  Performing all functions allocated by the President or any Minister responsible for a designated intelligence Service; and  Submitting certificates to the Minister responsible for the intelligence services concerning the activities of the Service for a prescribed period”.

In order to successfully fulfil these functions, section 8(a) of the Intelligence Services Oversight Act of 1994 “empowers the IG to have access to any intelligence, information or premises under the control of any Service”. However, to gain such access, the IG must first obtain a warrant issued in terms of the Criminal Procedure Act 51 of 1977. The MRC (2008:46) summarised the legislative mandate of the IG into three key components:

“first, a compliance or ombudsman role which entails the monitoring of the intelligence services for their compliance with the legislative framework; second, the

117

monitoring of significant intelligence failures in terms of which the heads of services are expected to report failures and any corrective actions taken to the IG; and third reviewing the intelligence and counter-intelligence activities of any service”.

According to the Intelligence Services Oversight Act of 1994, “the IG can receive complaints from members of the public, the JSCI, Human Rights Commission (HRC), Independent Police Investigative Directorate and any other structure deemed necessary to do so”. However, the MRC (2008:125) “found the IG’s complaint-handling mechanism distorted”. As part of its work, the office of the IG monitors and reviews the use of intrusive techniques by members of the intelligence services as this may infringe upon human rights. Section 7(7)(d) of the Intelligence Oversight Act 1994 empowers “the IG to submit certificates of performance to the responsible Minister concerning the activities of the intelligence services for a prescribed period”.

The following Table provides a comparative analysis of the role of the South African IGI with the IGs in the US, Canada, Australia and New Zealand. The details of the following areas of work have been fully discussed in Chapter 1 under section 2.3 focusing on the best practices of intelligence oversight as practiced by developed democracies of the world.

Table: 4.1: Comparative analysis of the work of the Inspectors-General of Intelligence in five countries including South Africa Area of work US Canada Australia New South Zealand Africa Ensure compliance with current legislation Yes Yes Yes Yes Yes Ensure compliance with standards of propriety Yes Yes Yes Yes Yes Carries out audits, investigations and inspections Yes Yes Yes Yes Yes Prevents and detects waste, fraud and abuse Yes No No No Unclear Promotes economy, effectiveness and efficiency Yes No No No No

118

Reviews compliance with executive directive and Yes Yes Yes No Yes operational policies Ensure compliance with warrants authorizations No N/A Yes Yes No Reviews operational activities Yes Yes Yes Yes No Reviews pending legislation and regulations Yes No Yes No No Reports regularly to the agency head(s) Yes No No No No Reports regularly to executives, legislature or oversight Yes Yes Yes Yes Yes commission Reports in response to requests by legislature or Yes Yes Yes Yes Yes oversight commission Investigates complaints about the agency Yes No Yes Yes Yes Ensures proper regard to human rights Yes No Yes No Unclear Ensure compliance with regulations on release of records Yes No Yes No Unclear and information Has access to all records and information of the agency Yes Yes Yes Yes Yes Initiates investigations on own initiative Yes No Yes No Yes Administers oath for taking testimony Yes No Yes No Yes

Source: (Intelligence and Security Committee (UK) Annual Report for 2001-2002 - CM 5542 Appendix 3)

It is clear from this analysis that IGIs play a central role in the provision of administrative oversight to members of the intelligence services in many countries of the world. It is also clear from Table 4.1 above that the statutory framework governing the work of the South African IGI is fairly congruent with the IG’s work in the leading democracies of the world. It is only in three areas where the Intelligence Services Oversight Act is not clear about the responsibilities of the OIGI compared to the countries reflected above. Those areas embrace the following questions as refined from the Annual Report (2001/2002: 18) of the UK Intelligence and Security Committee.

“Firstly, does the South African IG have the mandate to prevent and detect waste, 119

fraud and abuse? Secondly, does the IG have the mandate to ensure proper regards to human rights? And thirdly, does the IG have the mandate to ensure compliance with regulations on release of records and information?”

When closely reading these questions, it is clear that their answers are not explicitly provided for in the South African Intelligence Services Oversight Act of 1994 when analysing the work of the IGI. More work therefore needs to be done to improve the role and responsibilities assigned to the South African IGI to bring it in line with other offices that operate fairly similarly across the world. It is only in the US where a unique phenomenon was observed during the global benchmarking exercise. This relates to the US IG Act’s (1978) provision that a CIGIE be established to monitor the functioning of the IGs across the country.

In addition to the conventional work of intelligence oversight structures, it is also important to look into the work of some chapter 9 institutions in South Africa in overseeing members of the intelligence services.

4.2.5 Intelligence oversight by chapter 9 institutions

In the post 1994-dispensation, democratic South Africa established chapter 9 institutions in the Constitution to support and safeguard its newly-established constitutional democracy. These institutions are well positioned to investigate sensitive political matters as opposed to Members of Parliament, who are subject to the discipline of their political parties which deploy them to parliament. The following section summarises and assesses the work of the main chapter 9 institutions that are supposed to also oversee members of the intelligence services. They are the PP, the AG and the HRC.

120

4.2.5.1 The Public Protector (PP)

The PP is established by section 181(1)(a) of the South African Constitution as a state institution supporting constitutional democracy. According to section 193(4) of the Constitution “the Public Protector (PP) is appointed by the President on the recommendation of the National Assembly for a non-renewable period of seven years, in accordance with section 183 of the RSA Constitution”. According to section 182(1)(a) of the Constitution of 1996, “the PP has the power, as regulated by national legislation, to investigate any conduct in state affairs or in public administration of any sphere of government that is alleged or suspected to be improper or to result in any impropriety or prejudice”. Sub-section 5 of the same Act provides that “any report issued by the PP must be open to the public unless in exceptional circumstances determined by national legislation, a report must be kept confidential”.

According to the Annual Report of the PP (2012/2013), “the primary mandate of the PP is to strengthen democracy by ensuring that all state organs are accountable, fair and responsive in the way they treat all persons and handle service delivery issues”. In Section 2(a) of the PP Act 23 of 1994, it is stated that “the PP shall report in writing on the activities of his/her office to the National Assembly at least once a year”. Further, section 2 of the PP Act 23 of 1994 empowers her “to submit reports at any time to the National Assembly on the findings of a particular investigation”.

The role and functions of the PP are provided for in section 181(1) of the Constitution of 1996 and section 6(4) of the PP Act of 1994 and emphasised in the PP’s Vision 2020 and its three-year strategic plan. Those functions are as follows:

 “Investigation (complaint or own initiative) and dispute resolution regarding alleged maladministration or improper conduct by the state, through mediation,

121

conciliation or negotiation under the Public Protector Act, with a view to provide remedial action or redress;  Investigation of alleged violations of the Executive Ethics Code in terms of the Executive Ethics Act;  Resolution of disputes through mediation, conciliation or negotiation under the Promotion of Access to Information Act 2 of 2000;  Receiving protected disclosures from whistle-blowers under the Protected Disclosures Act; and  Investigating alleged violation of the Prevention and Combating of Corrupt Activities Act”;

As outlined on the website of the PP (2010), “the office is regulated by section 182 of the Constitution, and five other key pieces of legislation, namely: the Public Protector Act of 1994, the Executive Members Ethics Act of 1998, the Promotion of Access to Information Act of 2000, the Protected Disclosures Act of 2000 and the Prevention and Combating of Corrupt Activities Act of 2004”. Section 15 of the PP Act of 1994 “empowers the PP to ensure a fair, unbiased and proper investigation of matters brought to his/her attention”. This empowerment applies to all public authorities except court decisions. Further, section 3(1) of the Executive Members’ Act 82 of 1998 “empowers the PP to investigate any breach of the executive code of ethics by any executive member of state following the receipt of such a complaint”.

Based on the above summary, it is clear that the PP is given powers cutting across various organs of state. It is clear that the PP’s statutory mandate authorises it to also protect the public from possible harassment by the intelligence services. However, there is no clarity on the operational interface of the work of the PP versus that of the IGI. In terms of oversight, the law requires the PP to refer intelligence-related complaints to the IG for investigations. Therefore, to facilitate this interface both structures signed a Memorandum of Agreement (MoU) (Public Protector (PP) 2010).

122

Despite its statutory powers, the PP therefore refrains in terms of the MoU from launching full investigations into the functional modalities of the intelligence services, and just refers complains to the IGI for investigation. This arrangement contradicts the provision of the PP Act which empowers it to deal with complaints across all organs of state. It is this kind of uncertainty which emphasises the significance of this study to develop a better model to facilitate the cooperation of such structures for their realisation of the common good. Compared with other countries, the South African ‘Ombudsman’ (the PP) in practice does not look into intelligence related matters, but leaves it to the IGI.

4.2.5.2 The Auditor-General (AG)

Section 181(1)(e) of the Constitution provides for the establishment of the AG as one of the state institutions supporting constitutional democracy. Section 6(1) of the Public Audit Act 25 of 2004, states that: “the AG is appointed by the President following recommendations by the National Assembly, and he/she remains accountable to Parliament”. Like all state institutions supporting constitutional democracy, section 188(2) of the Constitution describes the AG as “an independent body which is subject only to the Constitution and the law and should exercise his/her powers and functions without fear, favour, or prejudice”. Section 188(1) of the Constitution states that “the AG must audit and report on the accounts, financial statements and financial management of all national, provincial and local entities; including any other institution required by law to be audited by the AG”. Section 189 of the Constitution further stipulates that “the AG is appointed for a fixed, non-renewable term of between five and 10 years”. As a result, the intelligence services are not immune from this constitutional provision.

123

The website of the SSA (2016) explicitly states that “the Auditor-General and the Standing Committee on Public Accounts (SCOPA) monitor relevant intelligence budgets in accordance with the provisions of the Public Financial Management Act, 1999 (Act 1 of 1999)”. Section 3(a)(i) of the Intelligence Services Oversight Act of 1994 further stipulates that “the JSCI obtain from the AG an audit report compiled in accordance with section 4(6) of the Auditor-General Act, 1995 (Act 12 of 1995)”. This report is critical to enable the JSCI to oversee the expenditure patterns of the intelligence services. According to section 10(1)(c) of the Public Audit Act of 2004, “the AG must annually submit a report to the National Assembly on his/her activities and the performance of his/her functions, including on the institutions and accounting entities wherein such services have been conducted”.

According to section 4 of the Public Audit Act of 2004, “the AG is regarded as the supreme audit institution of South Africa whose work aims to strengthen democracy by enabling oversight, accountability and governance in the public sector, thereby building public confidence”. In the AG’s Annual Report (2008/9:7), “the AG reaffirms its commitment to identify the impact of the findings of their reports on those they audit and ensure that their audit reports are relevant, simple, clear and more usable”. In practice, the AG had only started scrutinising the books of the intelligence services in 2010. In its inaugural audit, the AG raised concerns about the security of information contained in the NIA databases (Auditor-General’s Annual Report: 2008/9).

The MRC (2008:225) concluded that “the AG undertakes two kinds of audit, namely regularity audits and performance audits”. As for the intelligence services, the report indicates that “the AG has only undertaken regularity audits which focuses on testing system controls, and the performance of value-for-money procedures”. While performance audits have not been done yet, it however focuses on “information regarding the performance of the audited body against specified objectives” (MRC 2008: 225).

124

The AG’s regularity audit reports are not published even though their audit of the intelligence services is to detect any possible abuse of financial resources allocated to the services. In fact, the kind of audit work undertaken by the IG on the books of the intelligence services is superfluous. It is not allowed to access certain types of information during its audit process. For instance, the AG is not able to access information regarding source databases despite having significant amounts of money being spent through that mechanism (MRC: 2008). There is a clear need for the creation of a better cooperation mechanism to facilitate the collaboration between the AG and the IGI during the audit processes. However, an interim mechanism in the form of another MoU between the AG and the IGI is currently in operation (Respondent 10 of the Oversight Officers).

4.2.5.3 The Human Rights Commission (HRC)

Section 181(1)(b) of the Constitution establishes an HRC “ as one of the state institutions supporting constitutional democracy”. It is required “to report to Parliament on its legislative obligations in terms of the Promotion of Access to Information Act 2 of 2000, and the Promotion of Equality and Prevention of Unfair Discrimination Act 4 of 2000”. Section 184(1) of the Constitution outlines the primary mandate of the HRC as follows:

 “To promote respect for human rights and the culture of human rights;  Promote the protection, development and attainment of human rights; and  Monitor and evaluate the observance of human rights in the Republic”.

125

To ensure the effective fulfilment of this mandate, section 184(2) of the Constitution empowers the HRC to perform the following functions:

 “To investigate and report on the observance of human rights through its International and Parliamentary Affairs Programme; Research, Policy Analysis and Documentation Programme; and Information and Communications Programme;  To take steps through its Legal Services Programme to secure appropriate redress where human rights have been violated;  To carry out research through its Research, Policy Analysis and Documentation Programme; and  To provide human rights education through its Education and Training Programme”.

According to section 4(1) of the Human Rights Commission Act 54 of 1994, members of the HRC are expected to “serve impartially and independently and exercise or perform their powers, duties and functions in good faith and without fear, favour, bias or prejudice and subject only to the Constitution and the law”. Further, section 83 of the Promotion of Access to Information Act of 2000 “empowers the HRC to monitor compliance with this Act, make recommendations and facilitate the realisation of the right to access of information”. In its submission to the Ginwala Commission (2007:1) which was established to assess the viability and resilience of the then NIA and the SASS towards the accomplishment of their statutory mandate, the HRC expressed concern about the inadequate compliance of the intelligence services with section 14(1) of the PAI Act of 2000. It requires “the information officer of the public body to compile, in at least three official languages, a manual containing among others, the description of its structure and functions, its postal and street address, as well as a description on how members of the public can access such a structure”. However, the intelligence

126 services are exempted by the Ministry of Justice from doing as required in section 14(5) of the PAI Act of 2000, to which the HRC objects.

Regarding the HRC’s work on intelligence oversight, section 3(g) of the Intelligence Services Oversight Act of 1994 states that: “the JSCI may refer any matter related to human rights to the HRC and receive reports from the Commission concerning the matter”. However, the JSCI has not yet referred any matter to the HRC for consideration; hence, no such report has been generated thus far. Given its legislative mandate, the HRC should be able to protect members of the public when their fundamental rights are being violated, but the absence of a statutory or regulatory framework for guiding the cooperation of intelligence oversight structures remains a problematic constraint.

4.3 THE RATIONALE FOR THE EFFECTIVE INTELLIGENCE OVERSIGHT COOPERATION

In order to determine the views of intelligence specialists about better cooperation of intelligence oversight structures, I asked all 28 respondents whether it is a necessity for the various intelligence oversight structures to cooperate with each other when overseeing members of the intelligence services. A total of 25 officials (89%) agreed with the need for better cooperation of intelligence oversight structures, while only two officials (7%) found it unnecessary for the oversight structures to cooperate better with each other, and only one official (4%) was not sure whether oversight structures should cooperate with each other or not. A similar question was posed to the seven external experts, who all (100%) agreed with the need for better cooperation of intelligence oversight structures. Figure 4.3 provides the schematic presentation of the responses from interviewed officials and external experts.

127

Figure 4.3: The need for better cooperation of intelligence oversight structures

As mentioned earlier in Chapter 2, Born and Caparini (2007) concluded that the creation of an intelligence oversight mechanism underpinned by a clear cooperation framework could correct public misgivings about the work of intelligence services. This was also supported by Dietrich (2015:46) who argued that intelligence services operating within a strict oversight framework increase their operational legitimacy towards the community. In an address to members of the civilian intelligence services, the then President of South Africa, Mr Thabo Mbeki (2005:2) emphasised that “the activities of the intelligence services should be conducted in accordance with the Constitution, the Bill of Rights and the democratic ethos of society”. This can be fully realised when members of intelligence services operate under the watchful eye of cooperating intelligence oversight structures.

According to Caparini (2002), the primary purpose of intelligence oversight is to provide assurance of legality, proportionality and propriety for intelligence activities that are mainly conducted in secret. This view was also expressed by Webster (1988), who concluded that democratic societies need systematic checks and balances to ensure

128 that members of the intelligence services do their work in compliance with the country’s legislative framework. In addition, Born and Wills (2012) indicated that the efficacy, effectiveness, strength and resilience of an intelligence oversight structure is mainly dependent on its operational powers, independence, full access to information, ability to maintain secrets and the power of its support staff. However, in this assertion Born and Wills (2012) omitted to mention the significance of its networking, engagement and consultative trajectories, including its skills set.

Rogers (2012) explicitly indicated that the efficacy of any oversight structure is primarily dependent on its ability to cooperate, coordinate, and integrate its efforts with other structures with similar objectives. Born and Caparini (2007) did conclude, however, that even though intelligence is regarded as the second oldest profession in human history, attempts to oversee its activities is a fairly new phenomenon. This is probably one of the main reasons that many intelligence oversight structures have not yet come to terms with the idea of fully cooperating with one another in their professional practice, despite statutory mandates to do so. The envisaged oversight model that is the result of this research project will suggest how members of the intelligence services could comply better with applicable laws in the performance of their legislative mandates.

In a later publication, however, Born and Wills (2012:53) cited the following arguments as key justification for the need for better cooperation among intelligence oversight structures:

 “Ensuring a healthy balance between the secrecy of the intelligence services and transparency around their mandates and the parameters of their powers  Minimising misadministration and abuse of power  Ensuring effective control and public accountability in the funding of the intelligence services  Ensuring effective control over the use of intrusive methods of collection

129

 Developing a new identity and consciousness on the part of the intelligence officer”.

To test these considerations of Born and Wills (2012), I asked the 28 oversight officials to identify what they regard as the underlying motivations justifying the need for better cooperation of intelligence oversight structures during the implementation of their varied legislative mandates. In their responses, five (18%) of them cited oversight mandates, another five (18%) cited transparency and accountability, four (14%) cited human rights violations, three (11%) cited 4th generation warfare, another three (11%) cited oversight methods, and another three (11%) cited reporting modalities. Although two respondents (6%) could not cite any specific reason, one official (4%) cited varied oversight intervals as one of the key reasons justifying the need for the cooperation of the intelligence oversight structures. In addition, I also asked the seven external experts to cite their reasons for the need for better cooperation of oversight structures. Their responses agreed with those outlined by the interviewed officials. Of the seven external experts, two (29%) cited human rights violations, another two (29%) cited transparency and accountability while the remaining three scored 14% each citing varied oversight mandates and powers, varied oversight methods and modalities and reporting modalities respectively. Figure 4.4 provides a schematic representation of the major thematic areas that have emerged in the interview responses with both oversight officers and external experts:

130

Figure 4.4: Underlying reasons justifying the need for the cooperation of various intelligence oversight structures

131

These themes can be assessed as follows:

4.3.1 Emergence of fourth (4th) generation warfare

As reflected in Figure 4.4 above, the emergence of the 4th generation warfare was cited by three officials (11%) as one of the key reasons justifying the need for the cooperation of intelligence oversight structures. In citing this reason, Respondent 2 indicated that this reason does not only reflect the need for intelligence oversight or their cooperation but also the need for intelligence tradecraft itself. He further argued that the need for intelligence tradecraft cannot be separated from the need for intelligence oversight. To illustrate the point, 23 (82%) respondents of the oversight officers referenced the interwoven nature of intelligence tradecraft and the need for its oversight. Further, Respondent 12 of the officials indicated that as much as there is a need for intelligence tradecraft and its oversight machinery, there is also a need for the cooperation of intelligence oversight structures especially in democratic countries like South Africa. To reinforce this point, Lustgarten and Leight (1994) argue that the 21st century is fraught with more perils, uncertainties and unpredictability than any other time in world history. As a result, policy-makers need accurate intelligence on emerging threats, dangers, risks, and broader national security challenges to make sound judgements in policy discussions (Lustgarten and Leight, 1994: 12). He further argues that “since the end of the cold war, governments have confronted a multiplicity of new security challenges manifesting in less calculable conflicts” (Lustgarten and Leight, 1994: 12).

Further, Born and Leigh (2005:16) argue that “democratic states are no longer threatened by foreign military invasions but by organised crime, terrorism and the spill- over of regional conflicts or failed states”. Similar views were earlier expressed by Gerdes (2004:154), who argues that “the global intelligence services face the fourth- generation warfare which manifests in highly compartmentalised, cellular and predatory

132 adversaries manifesting in networks outside the framework of traditional nation states”. To illustrate the point, Thornton (2007) mentions Hezbollah in Lebanon, Al-Shabaab in Somalia, Boko Haram in Nigeria, the Tamil Tigers in Sri Lanka, and a borderless Al- Qaeda, and lately the Islamic State, as some of the crucial participants in the fourth generation warfare. Krasner (1994) also asserts that the new emerging tasks of intelligence services are becoming more complex, volatile and fluid. It is on the basis of these threats that the need for intelligence tradecraft cannot be over-emphasised. Africa (2012) concludes that overseeing members of the intelligence services in a collaborative and cooperative manner cannot be ignored.

In addition, the Geneva Centre of the DCAF (2006:2) states that the “rapid emergence of international terrorist groups, drug lords and trans-national criminal syndicates remains a major security threat among democratic societies”. On the basis of these threats, it is clear that governments require timely intelligence to counter and/or mitigate the impact of the threats. In his book titled: From Secrets to Policy, Lowenthal (2003:7) emphasises that “intelligence services should provide warnings about imminent strategic threats to national peace and security”. This is also referenced in section 3(2)(4) of the White Paper on Intelligence of 1995, which argues that “the South African intelligence services should provide government with evaluated information in order to safeguard the Constitution and the people”.

It is therefore evident that governments require intelligence advice for the maintenance of national security and the well-being of its citizens. However, intelligence tradecraft requires the creation of a robust intelligence oversight mechanism in order to ensure that intelligence services do not undermine the constitutional order and increase the complexity of the threats. In addition, these checks and balances require the development of sound cooperation mechanisms to ensure that the work of the intelligence oversight structures contribute to the realisation of good governance results.

133

4.3.2 Different intelligence oversight mandates

Of the 28 interviewed officials, five (18%) cited the allocation of varied oversight mandates to various structures as one of the underlying reasons necessitating the need for the cooperation of the various oversight structures. Further, this similar reason was also raised by the interviewed external experts. This is not new and Africa (2012) had earlier indicated that the allocation of various mandates to various structures calls for the cooperation of such structures. This was also emphasised by Den Boer (2012) who argued that for as long as oversight structures are expected to observe certain pieces of legislation when doing their oversight work, the need for their cooperation cannot be over-emphasised. In addition, Respondent 8 of the officials indicated that “….. for as long as the founding legislation of the oversight structures outline their investigative mandates and relevant limitations for access to classified information, the need for them to cooperate with one another remains highly significant”. This reason was further cited by Respondent 2 of the external experts who indicated that the allocation of varied mandates and powers to the structures simply imply the need for their cooperation.

As already reflected in both Chapter 1 and Chapter 4 above, Cwele (2012) emphasised the fact that members of the intelligence services should continue to operate within their relevant legal parameters in doing their work. The post-1994 dispensation separated the various areas of intelligence operation and oversight according to the adopted principle of the separation of powers which is referred to as ‘trias politica’. It is therefore important for the various structures to cooperate with each other in a complementary manner in order to ensure their effective contribution to the realisation of good governance results. In the US, the government legislated the co-existence of institutions responsible for ensuring the legal compliance of the intelligence services to those entrusted with ensuring the robust institutional governance of the services (Gill: 2012; Weller: 1997 and Born and Caparini: 2007).

134

As emphasised by Den Boer (2012), the systematic cooperation of intelligence oversight structures can only be meaningful when legislative mandates clearly allocate powers, methods and required resources to enable the effective overseeing of members of the intelligence services. Despite the need to allocate different mandates to different structures, a coherent legislative mandate should authorise oversight structures to “monitor the legality, efficacy and effectiveness of the intelligence services” (Dube, 2013:109). These include “their budgeting and accounting practices, compliance with human rights standards and other administrative aspects” (Den Boer, 2012:13). In light of the varied legislative mandates as given to various structures, Respondent 2 of the external experts cited the need for coherence and an integrated approach to intelligence oversight as the precondition for the cooperation of intelligence oversight structures. This was also expressed by Respondent 7 of the external experts when he indicated that, “…Human Rights Commission is mandated to deal with any activity which undermines people’s human rights regardless of the perpetrators, but if the violence includes the mismanagement of financial resources then the Auditor- General should be brought into light” (HRC Act 54 of 1994).

In view of the above discussion, it is clear that the cooperation of the intelligence oversight structures remain critical. Therefore, should the perpetrator in this context be a member of the intelligence services, the HRC will still deal with the matter when brought to its attention. However, in such an intervention it is the view of all seven (100%) external experts that the HRC should approach the IGI which is mandated to oversee the functionalities of members of the intelligence services. However, most officials and external experts cited the absence of clear legal instruments to regulate such engagements as the key problematic matter. Therefore, the focus should be on the development of such legal instruments to facilitate the cooperation of the oversight structures.

135

4.3.3 Varied intelligence oversight powers

Given the allocation of varied operational mandates coupled with associated special powers for the fulfilment of certain legislative mandates, two of the interviewed officials cited varied intelligence oversight powers as one of the reasons for the various oversight structures to cooperate with each other. This represented 6% of the interviewed officials. Respondent 5 of the officials indicated that the conscious splitting of powers across structures necessitates the cooperation of such structures. Dlomo (2004) agrees that, the allocation of these powers depends on the nature of the objective to be accomplished by the respective oversight structure. This reason was also cited by one of the external experts representing 14% of the interviewed experts. This was emphasised by Den Boer (2012) when he indicated that oversight structure can be given the power of referral, which means that the structure can refer a particular matter to another body for further processing. On the other hand, an oversight structure can be given the power of exposure, which mainly concentrates on exposing non-compliance, error of judgement or the violation of a legislative framework by members of the intelligence services (Den Boer: 2012). In this case, such a structure might not necessarily have the power to investigate and correct the situation, but only to refer the matter to another structure for correcting the situation.

Based on the discussion, it is clear that the decision to split the mandates and powers across oversight structures aimed to reduce the possible abuse of the allocated powers. This is probably the reason why the US created the overall IGI responsible for the overall integrity and efficiency of the various IGs. Regarding the separation of powers, a typical example is the Nkandla Constitutional Court ruling which found President Zuma to have failed to uphold constitutional provisions. It does not have the power to remove a sitting Head of State, but the NA has that power. All 28 (100%) interviewed officials agreed that the allocation of various powers to the various structures serves as

136 one of the key reasons justifying the need for the cooperation of the various intelligence oversight structures.

Based on the global benchmarking exercise, it is clear that many countries of the world also split power across the various organs of state. This is crucial to reduce the possible risk of the abuse of power. As Caparini (2002) cautions, allowing members of a particular oversight structure to have unlimited access to information with no regard to matters of national security could threaten the existence of the same democratic dispensation they seek to protect. However, it should be noted that there are certain countries which gave their oversight structures the power to access all manner of classified information. However, such countries do have robust systems on information security to counter any possible leaks which could have a negative impact on the country’s national security. As Den Boer (2012) argues, it should be noted that some restrictions could be reasonable at times, especially when auditors want the identities and physical location of intelligence sources for the verification of financial expenditure. Instead, we should use these arguments as the basis for intelligence oversight structures to cooperate with each other to effectively fulfil their respective allocated mandates and powers.

4.3.4 Application of varied intelligence oversight methods

In as much as varied legislative mandates and oversight powers inform the need for the cooperation of intelligence oversight structures, three (11%) of 28 interviewed officials cited the application of varied oversight methods as one of the other reasons justifying the need for the cooperation of oversight structures. In most cases, the adopted oversight method is informed by the description of the mandate outlining the investigation modalities. Of the seven interviewed external experts, one (14%) cited the application of varied oversight methods as rationale for the cooperation of intelligence oversight structures. According to Den Boer (2012), the most popular oversight

137 methods include inspections, hearings, document analysis, interviews, witness statements and access to databases. In fact, the selected oversight method plays a crucial role in determining the nature of information to be sourced. For instance, the parliamentary intelligence oversight committee of Norway conducts annual inspections on members of the intelligence services, while the IG in New Zealand can enter the premises of the services only with prior notice to the Director of the Intelligence Agency (Den Boer: 2012). Since various structures use varied oversight methods for sourcing different information, it is important for them to systematically cooperate with one another in order to complement each other’s work.

In addition to the oversight methods, Respondent 3 of the external experts argued that the legislative mandate provides the investigative methods of the oversight structures. According to Den Boer (2012), an investigation can be initiated by members of Parliament, the executive or the oversight structure on its own volition or following a complaint from members of the public. As already discussed earlier in this chapter, the South African legislative framework allows members of the public to approach the PP, HRC or the IGI should they wish to report any violation perpetrated by intelligence operatives (RSA Constitution: 1996).

This is also the case in other democratic states such as New Zealand, Australia, Belgium, Canada, the US and the UK (Flood: 2004). Further, in South Africa a particular violation can be investigated by various intelligence oversight structures, hence the need for their systematic cooperation in performing their varied legislative mandates.

4.3.5 Different intelligence oversight intervals

According to Born and Mesevage (2012), the legislative mandate also outlines the time frame within which oversight work ought to take place. For instance, in certain cases

138 oversight structures are given the mandate to authorise certain intelligence operations before they take place (RICA: 2002). This is popularly known as ex ante oversight which is mainly done in order to proactively address the possible abuse of intelligence machinery for personal or political gain. This is evidenced in the mandate of the special judge who has to authorise intelligence operations of an intrusive nature given their potential to infringe on people’s rights to privacy as provided for in section 16(1) of the RICA Act of 2002. As Weller (1999) confirmed, the UK uses commissioners to pre- authorise certain intelligence operations before implementation. Further, Roach (2012) confirms that ex ante was seriously recommended in the UNs special report on the promotion and protection of human rights as a proactive tool to stop members of the intelligence services from violating human rights. The question of oversight intervals was cited by only one (4%) official among the 28 interviewed officials as one of the reasons justifying the need for the cooperation of oversight structures.

Further, some oversight structures are given the mandate to oversee the work of intelligence practitioners during their intelligence operations (Den Boer: 2012). This is considered problematic by various operatives, as the oversight structures could easily interfere with the ongoing intelligence operations. On the other hand, the most popular form of intelligence oversight is ex post (after the event). According to Den Boer (2012), ex post oversight allows the structures to review and scrutinise past intelligence operations to determine any non-compliance with the legislative framework. This is the common approach that is mostly used for intelligence oversight purposes. According to Section 188 (1) of the Public Audit Act of 2004 the work of the AG: “in auditing and reporting on the accounts and financial statements of the Intelligence services is a case in point” (Section 188 (1) of the Public Audit Act of 2004).

Further, some statutory frameworks in other counties allow for the oversight work to be conducted at certain intervals. For example, Den Boer (2012) indicates that the report of the ad hoc inquiry into the Australian Intelligence services recommended five to

139 seven-years reviews of the intelligence services. In South Africa, the IGI prepares annual compliance certificates for the JSCI, as does the AG on financial compliance. During the implementation of their work, various structures operate at different intervals to extract varied data for verification. Therefore, it is on the basis of these differentiated approaches that each oversight structure cooperates with others in conducting their tasks at hand. The issue of varied oversight intervals was raised by 17 of the 35 interviewed respondents as one of the key reasons to ensure the cooperation of the different oversight structures. For instance, the IGI should be able to verify with the appointed judge whether certain interceptions of communication were pre-authorised or not. The information should therefore be used during the IG’s compilation of the intelligence service’s compliance certificate for the JSCI.

4.3.6 Different reporting modalities

During empirical data collection, the issue of different reporting modalities was cited by three (11%) of the 28 officials as one of the key reasons justifying the need for the cooperation of oversight structures. Further, one (14%) of the external experts out of the seven interviewed also cited it as one of the key reasons for the cooperation of oversight structures. As part of international best practice, intelligence oversight structures have the responsibility to report their findings to the relevant governing authority. According to Den Boer (2012) such reports include recommendations on what needs to be done by the governing authority or the service itself to address the identified challenges. This can include suggestions on how to improve “service accountability, transparency, legality and effectiveness” (Den Boer, 2012: 80). Further, it is also important for the oversight structures to proactively report suspicious conduct of intelligence operatives to the relevant authority on time. Given the classified nature of information handled by the intelligence services, oversight structures can, according to Wills in Den Boer (2012) produce different reports for different clients. For instance, a classified report can be produced for members of the executive followed by a sanitised

140 report for release into the public domain. In New Zealand, the Prime Minister publishes the IGI reports in national libraries for public consumption.

In citing the underlying reasons justifying the need for the cooperation of various oversight structures, 33 (94%) of the 35 interviewed respondents raised the question of reporting modalities. And, this matter was found to be closely linked with the question of oversight intervals. This confirms that the nature of allocated mandates, powers and reporting intervals mainly determines the nature of the report to be compiled. In terms of the global benchmarking exercise, it is clear that different countries use different reporting methods. Despite the question of reporting methods, it is clear that the compilation of a robust report requires engagements, consultations and sourcing inputs from other stakeholders. Therefore, the need for the cooperation of the various intelligence oversight structures in the compilation of their respective oversight reports cannot be overemphasised.

In the South African oversight model, the primary client of the intelligence oversight structures is the JSCI (parliament oversight) as the custodian of public accountability. In terms of reporting, Respondent 6 of the oversight officers from the OIGI confirmed that they prepare compliance certificates for ratification by the Minister of State Security before their presentation to the JSCI. Similarly, the AG submits its audit reports on intelligence services to the JSCI for consideration. This is done to enable the parliamentary oversight committee to oversee the spending patterns of the intelligence services as provided by section 4(6) of the AG Act 12 of 1995.

Further, reporting modalities are also provided for in the legislative mandates of the respective oversight structures. According to Den Boer (2012), the Netherlands oversight structures submit their annual reports to a member of the executive who should transmit these to parliament within 15 days. Therefore, in view of these developments it is important to ensure the systematic cooperation of various

141 intelligence oversight structures to improve the quality of their reports. As Den Boer (2012) puts it, an oversight report is meant to generate discussions among the relevant stakeholders and this is crucial to facilitate the implementation of the proposed recommendations.

4.3.7 Potential for human rights violations

During the interviews, four (4%) of the 28 interviewed officials emphasised the need for the cooperation of oversight structures to mitigate the possible violations of human rights by members of the intelligence services. Further, two (29%) of the seven external experts also cited human rights violations as one of the reasons justifying the cooperation of these structures. According to Article 12 of the 1948 Universal Declaration of Human Rights: “no human being shall be subjected to arbitrary interference with his privacy, family, home, or correspondence, or to attacks upon his honour and reputation”. The article further provides that “everyone has the right to be protected by the law against such interference or attacks”. Section 7 of the RSA Constitution of 1996 declares that “the Bill of Rights is the cornerstone of the South African democracy and enshrines the rights of all people of the country and affirms the democratic values of human dignity, equality, and freedom”.

Further, the intelligence services are authorised and mandated “to acquire confidential information through surveillance, infiltration of organisations and the interception of communication media” as provided for in the Intelligence Services Act 65 of 2002 (Nathan, 2010:99). In doing this work, members of the services could in the process infringe on citizens’ rights, dignity, equality and freedoms. This prompted scholars like Nathan (2010:99) to ask a pertinent question: “(h)ow can the state guarantee the accountability of the intelligence services members as they conduct their work in inherent secrecy”.

142

Globally, the issue of human rights remains an emotive matter in many states. The modern law of international human rights is based on the Preamble and Articles 1(3), 55 and 56 of the UN Charter, “supplemented by the Universal Declaration of Human Rights, the International Covenant on Civil and Political Rights, and the International Covenant on Economic, Social and Cultural Rights” (Gill, 2012:217). Therefore, even though members of the intelligence services provide proactive advice to policy-makers, there is evidence of human rights abuses perpetrated by intelligence operatives across the globe. In addition, Gill (2012:217) argued that the absence of a coordinated intelligence oversight mechanism does not only threaten human rights, but also threatens national security itself. As a result, proposals by various scholars to create intelligence oversight mechanisms were aimed, among others, at addressing the perpetual human rights violations in various countries of the world. This is crucial for restraining intelligence operatives and ensuring their compliance with the country’s legislative framework.

According to Cole, Eppert and Kinzelback (2008:78), “when the intelligence oversight regime fails to ensure the adequate protection of human rights, communities have the leeway to identify deficiencies and bring to bear international pressure through the use of international human rights complaints and protection mechanisms”. This is evidenced by the pending case of AFRIFORUM at the human rights tribunal complaining about the continuous application of affirmative action which seeks to prioritise the employment of previously disadvantaged individuals in South Africa.

As Wills and Vermeulen point out in Leigh (2012), the view that intelligence services may violate human rights is rooted in their ability to access certain individuals’ personal data. This ability emerges as a result of giving special powers to intelligence services to use intrusive measures to access certain private data. In the application of these powers, some intelligence operatives could easily use this data to infringe on people’s rights to privacy. In the past, members of the US intelligence services used intrusive

143 covert operations to aggressively deal with members of civil rights movements such as Dr Martin Luther King (Den Boer: 2012). As outlined in the problem statement, the absence of a cooperation model for intelligence oversight structures allows the intelligence services to use without detection intrusive measures to conduct surveillance operations for personal or selfish political ends (OIGI: 2006).

4.3.8 Need for transparency and public accountability

The global benchmarking exercise revealed that the main reason for the need to oversee members of the intelligence services is to ensure their transparency and accountability in the performance of their legislative mandates. During collection of empirical data, five (18%) of the 28 interviewed officials cited the need for transparency and the accountability of the intelligence services as justification for the need for cooperation among various oversight structures. In addition, two (29%) of the seven interviewed external experts also cited the same reason as justification for the need of the cooperation of these structures. Since 1994, South Africa experienced inherent tensions between the work of the intelligence services and the democratic values of transparency and accountability. This is not a new matter, and Den Boer (2012) had already indicated that one of the critical reasons for the development of the vibrant oversight mechanisms is to enforce transparency and the accountability of the intelligence services. In further justifying the need for intelligence oversight and the need for the cooperation of various role players, Den Boer (2012) argues that oversight structures have the potential to deter members of the intelligence services from engaging in abusive, opportunistic or illegal behaviour.

This is done through the provision of relevant information on the operations of intelligence structures to both Members of Parliament (MPs) and Members of the Executive. Various structures are given suitable responsibilities for the advancement of the transparency and accountability of the intelligence operations. For instance, it

144 would be unusual to expect the AG to investigate human rights abuses perpetrated by intelligence services, or the HRC conducting financial audits on the utilisation of public funds. Instead the AG conducts audit functions (Public Audit Act 25 of 2004) while the HRC has to investigate human rights abuses (Human Rights Commission Act 54 of 1994).

Considering that various oversight structures possess certain legislative mandates for the fulfilment of certain objectives, Respondent 11 indicated that the quality of intelligence oversight depends on the nature of cooperation among the intelligence oversight structures. This was also emphasised by Gill (2012: 220) when he argued that “when the state has more than one oversight body, they should use ‘statutory gateways’ to enable their effective cooperation during the application of their inquiries”. Interestingly, Oleszek (2010) argues that, it is the complexities of intelligence tradecraft and its associated dynamism that underpin the need for the cooperation of intelligence oversight structures.

4.4 THE CURRENT STATE OF COOPERATION AMONG INTELLIGENCE OVERSIGHT STRUCTURES

In seeking to determine the current state of cooperation among intelligence oversight structures, it is important to take note of the issues highlighted by the Project Avani scandal.

4.4.1 Cooperation gaps exposed by the Project Avani scandal

As already discussed in Chapter 1 in section 1.4, members of the intelligence services unlawfully intercepted people’s private calls, undertook unauthorised surveillance of individuals and circulated damning e-mails to advance factional political interests (OIGI: 2006). During the scandal, members of the intelligence services who were aligned to

145 the then expelled South African Deputy President, Mr Jacob Zuma created bogus e- mails accusing the then sitting President, Mr Thabo Mbeki of purging his deputy in order to gain an upper hand in the 2007 elective conference of the ANC in Polokwane. The Project Avani scandal (OIGI: 2006) exposed the absence of a systematic cooperation oversight model among the various intelligence oversight structures.

These activities occurred despite the existence of a comprehensive intelligence oversight model involving the legislature, executive, judiciary, civil society and the chapter 9 institutions, including broader society and the media. Despite all these, the scandal evolved undetected over a number of months characterised by the abuse of state resources for personal and or factional political interests. These oversight structures were fragmented, un-coordinated and non-cooperative in the performance of their respective legislative mandates. The scandal exposed the fundamental gaps in the operational modalities of the country’s oversight model. Similar sentiments were raised by the ISS in its submission to the MRC (2008) (ISS: 2008:6). It raised the following matters:

 “A lack of sufficient preventative and proactive controls on politically sensitive operations.  A lack of sufficient control on the use of intrusive methods of investigation.  Insufficient control of operational protocols.  The continued politicisation of domestic intelligence operations and the potential misuse of authority in the control of political intelligence operations.  Flaws in the practice of having a senior intelligence official serving in two different posts.  Failure to acquire judicial authorisation for the interception of communication media.  The misuse of state resources for the advancement of personal or factional political interests”.

146

As the scandal unfolded, the fragmentation of the oversight structures revealed their inability to recognise and address the flouting of judicial, financial and civil obligations (ISS: 2007). However, the scandal presented the government with an opportunity to review the effectiveness of the country’s intelligence oversight model, particularly the manner in which the various oversight structures cooperate with each other. As a result, Mr Ronnie Kasrils (2005), the Minister of Intelligence services, established the MRC to assess the mechanisms of controlling or overseeing members of the civilian intelligence agencies. This was emphasised by Dr Frene Ginwala who was appointed chairperson of the MRC (2008:9) when she indicated that the commission was to “strengthen control mechanisms of the civilian intelligence structures in order to ensure their full compliance with the constitutional principles and the rule of law”. This was crucial to develop key interventions to improve the manner in which oversight structures operated. The commission sought to minimise the potential for the intelligence services to be involved in illegal activities and the abuse of power. Unfortunately, the commission overlooked the fundamental issue of the cooperation modalities of the intelligence oversight structures. After extensive work, the findings and recommendations of the Ginwala commission were never implemented.

Against the background of the Project Avani scandal, I asked the selected intelligence interviewees for this research project to assess the current state of cooperation among the various oversight structures. In response, 18 (64%) of the 28 officials described the current state of cooperation as poor. A further eight of the remaining officials (29%) described it as very poor, while only one (4%) saw it as good. However, one (4%) official was not sure about the state of cooperation of the oversight structures. When asked the same question, six (86%) of the seven external experts indicated that the current state of cooperation is very poor. Further, only one expert (14%) regarded the current state of cooperation as poor. Respondent 2 of the external experts indicated that whatever cooperation mechanisms being used are informal, unpredictable and not articulated in the public discourse. Rogers (2012) warned of the need to understand

147 the changing actors and the national threat picture of various countries in order to properly reflect on the current state of cooperation. He also emphasised the significance of considering the dynamics associated with the changing strategic environment around intelligence tradecraft and the subsequent implications for oversight. Figure 4.5 summarises the interview findings about the current state of cooperation among intelligence oversight structures.

148

Figure 4.5: Current state of cooperation among intelligence oversight structures in South Africa

Notwithstanding the need for intelligence oversight, literature warned of the possible disruption of the crucial role of “members of the intelligence services in the evolving policy agenda of the democratic state” (Rogers, 2012:67). Respondent 5 of the external experts indicated that in the democratic space the efficacy of the intelligence services depends on the coherence of oversight structures and the ability of the intelligence services to comply with the country’s legislative framework. The latter assertion is based on the view that effective oversight enhances the legitimacy of intelligence service structures and their acceptance by members of the public.

Although members of the intelligence services are neither responsible for law enforcement nor for policy implementation, their ability to utilise a variety of non- traditional policy tools for resolving governance problems remain critical (Intelligence Services Act of 1994). It is for these reasons that the capacitation and development of cooperation mechanisms for intelligence oversight remain essential. To fully determine 149 the state of cooperation among intelligence oversight structures, I also asked all respondents questions about the impact of the existing legislative framework on the cooperation of oversight structures, the nature of the existing cooperation framework, case referral mechanism, oversight processes and the interface between chapter 9 institutions and other oversight structures. The interview findings on these issues are reflected in the attached Appendices A and B.

4.4.2 Statutory framework for the cooperation of oversight structures

In trying to further determine the state of cooperation among intelligence oversight structures, I asked the 28 officials whether their existing legislative and/or regulatory framework requires its structures to cooperate with other oversight institutions. In response, 21 (75%) of the interviewed officials responded no, three officials (11%) indicated yes while four (14%) were not sure. Of the external experts, six (86%) indicated no and only one (14%) indicated yes. On the other hand, Respondent 5 of the external experts confirmed the existence of various types of ad hoc arrangements between various oversight structures. This was confirmed by Respondent 16 of the officials who indicated that MoUs are signed with various other structures to facilitate their cooperation. Despite these efforts, Respondent 4 of the external experts stated that any coherent government action requires a clear legislative and/or regulatory framework for empowering its implementation processes.

150

Figure 4.6: Whether current legislative and/or regulatory frameworks of the oversight structures require its cooperation with other structures

Section 210(b) of the Constitution of 1996 and section 7(2) of the Intelligence Services Oversight Act of 1994 which underpins the existence of the OIGI, are silent on the need or cooperation of intelligence oversight structures. Due to this silence, the implementation or utilisation of the inter-governmental clause is crucial. The Intelligence Services Oversight Act of 1994 refers to the need for the IGI to account operationally to the JSCI and administratively to the Minister of State Security. However, this provision does not assist in the operational cooperation of the relevant structures but relates to compliance with the Public Finance Management Act 1 of 1999 regarding financial allocations and the need to account for such allocations.

Respondent 8 of the officials indicated that the OIGI has prepared a revised draft regulatory framework to guide the cooperation of the IG’s office and other structures, and is awaiting approval by the Minister of State Security and the JSCI. This is not 151 new; in 2008 the MRC (2008:118) had already recommended the creation of such regulations through its Legislative Review Task Team. The commission recommended the urgent compilation and adoption of “ministerial regulations on the manner in which the Inspector-General for Intelligence conducts his/her investigations and inspections” (Ministerial Review Commission, 2008:118). However, this matter has been delayed for many years and is now compounded by delays in the appointment of the new IG after the expiry of Ambassador Faith Radebe’s contract in March 2015. Given the absence of approved regulatory framework, the current operational modalities of the IG’s office remain guided by section 7(7) of the Intelligence Oversight Act of 1994.

Despite the plethora of legislative powers given to the IGI, Nathan (2010) indicates that the IG does not necessarily account to parliament or the people of South Africa as the office continues to operate in the thrall of secrecy and no longer prepares annual reports for public consumption (www.dailymaverick.co.za). Although the Regulation for the Interception of Communication and Provision of Communication Related Information Act of 2002 compels the intelligence services to obtain judicial authorisation before conducting the interception of communication media, there is no guideline as to how the judge is supposed to cooperate with other intelligence oversight structures on this matter. However, Respondent 12 of the officials referenced the intention of the OIGI to sign a MoU with the appointed judge in order to structure their interface in future. Given the absence of the IG, the MoU with the appointed judge has not been signed up to now. It is therefore believed that the newly appointed IG would resume discussions with the appointed judge once he resumes office during 2017.

In terms of the chapter 9 institutions, Respondent 19 of the officials re-emphasised the legislative mandates of the chapter 9 institutions and their pursuit to protect the country’s constitutional democracy. In this context, most of the officials agree that the existing legislative framework does not provide clarity or direction about cooperation between the chapter 9 institutions and the mainstream intelligence oversight structures.

152

As a result, the intelligence oversight structures remain fragmented, uncoordinated and fail to cooperate adequately when overseeing the intelligence services. Their contribution in holding accountable members of the intelligence services remains minimal and is in certain cases non-existent. This situation does not conform to the provisions of systems theory as articulated by von Bertalanffy (1968) which emphasises the interrelated nature of objects from one to another.

In comparison, there is a clear statutory framework compelling the cooperation and coordination of the South African intelligence structures when dealing with strategic intelligence work. For example, section 4 of the National Strategic Intelligence Act of 1994 as amended by section 4(1) of the General Intelligence Laws Amendment Act 11 of 2013, provides for the establishment of NICOC. It is established to “coordinate strategic intelligence as supplied by members of the intelligence structures for guiding government in its governance processes”. Similarly, one would have thought that with the existence of multiple oversight structures with varied legislative mandates, there was going to be some legislative clarity on the cooperation of these structures when performing their various mandates. Unfortunately, there is no clear legislative framework for guiding the work of the intelligence oversight structures or even compelling them to cooperate and coordinate their work as is the case with the intelligence structures.

4.4.3 Existing institutional mechanisms for intelligence oversight cooperation

In an attempt to solicit further clarity on the current state of cooperation among oversight structures, I asked the 28 officials to indicate if their respective structures have a specific institutional mechanism or a specific platform within which their engagement with other structures occurs. In response, 10 (36%) of the officials indicated yes while 13 (46%) indicated no with only five officials (18%) indicating that they are not sure. Despite the general acceptance of the legislative silence on the

153 cooperation of intelligence oversight structures, the officials cited the existence of up-to- date MoUs as important mechanisms guiding their engagements with other structures. Respondent 4 of the officials specifically cited the existence of the MoU between the OIGI and the office of the AG. This was corroborated by Respondent 23 from the Auditor-General’s office who indicated that the MoU is mainly used for the purposes of quality assurance particularly during the financial auditing of the books for SSA.

Figure 4.7: Existence of institutional mechanisms within which the current cooperation of intelligence oversight structures is taking place

Considering the legislative silence on cooperation, all interviewed officials and external experts concur that the MoU mechanism is necessary given the SSA’s continued utilisation of the Intelligence Services Special Account Act 81 of 1969, and the Secret Services Act 56 of 1978. These accounts allow the service to use undisclosed sums of money in the running of covert operations. During the hearings at the MRC (2008:221), the National Treasury recommended the SSA repeal these pieces of legislation, but to no avail. For the determination of efficacy, I further asked the officials the frequency in

154 which the oversight structures engage with one another. In response, 17 officials (61%) indicated that their engagement is undefined. Only three officials (11%) indicated that their structures meet twice a month while eight officials (28%) confirmed that their structures do not meet with other structures at all. Figure 4.8 presents the schematic summary of the findings in response to the above-mentioned question as posed.

Figure 4.8: Frequency of engagements between the various intelligence oversight structures

After determining the frequency of engagements between the various intelligence oversight structures, I then asked the officials to what extent the frequency of the engagements shaped the agenda of such interaction, or not. A total of 19 (68%) officials indicated that it poorly shaped the agenda. However, six officials (21%) were not sure while two (7%) officials indicated that it partially shaped the agenda. Only one (4%) said it fully shaped the agenda.

155

Figure 4.9: Extent to which the frequency of engagements shaped the agenda of the interaction among oversight structures

Respondent 14 of the officials indicated that the OIGI had in the past signed a MoU with the office of the PP but that it has since lapsed. This was complicated by the untimely departure of the former IG, Advocate Faith Radebe at the end of March 2015, followed by delays in the appointment of the new IG. All 28 interviewed officials agreed that the absence of a formal or even informal forum to discuss matters of common interest remains a daunting challenge for their work. Despite the above MoUs, all the respondents also concurred with the non-existence of other arrangements for engaging other structures such as the HRC, the appointed judge, the Minister of State Security and the JSCI. However, the IGI continues to provide its annual oversight certificates to both the Minister of State Security and the JSCI for their consideration.

Respondent 9 of the officials confirmed that their Strategic Plans and Annual Performance Plans (APP) are submitted to both the Minister of State Security and the

156

JSCI. This is done as part of an administrative obligation and the need to account. However, this assertion does not translate to the formal cooperation framework, but rather illustrates compliance with a supervisory obligation as provided for in the National Strategic Intelligence Services Act of 1994. Therefore, inasmuch as the OIGI is supposed to cooperate with the JSCI, they are actually supposed to account to the committee.

During the interviews, it was confirmed that there is no formal or even informal mechanism for engaging the HRC. Respondent 12 of the officials indicated that the OIGI prepared the MoU for signing with the HRC. However, their initiative caused divisions among the commissioners within the HRC. Anyway, both structures agreed that they will informally cooperate with each other as the need arises. In addition, Respondent 4 of the external experts indicated that the cooperation of the IGI with chapter 9 institutions is mainly associated with case management matters. Therefore, as the HRC receives a very limited number of complaints there was no compelling circumstances for the MoU to be signed.

Further, when asked about the nature of cooperation between the OIGI and the appointed judge, Respondent 13 of the officials indicated that the OIGI engages the judge to verify the number of issued authorisations for the interception of communication media. This is mainly done to empower the IG to compare the number of provided authorisations in relation to the number of applications made by members of the intelligence services. This is crucial for the effective authentication of provided oversight information. However, this arrangement remains unstructured, uncoordinated and informal. Therefore, should members of the intelligence services collude with the appointed judge it would not be possible for the IG to uncover such collusion. Learning from Nzongola-Ntalaja’s (2002) assertion, effective governance reflects effective rules, institutions and good democratic governments. Therefore, all the interviewed respondents concurred with the need to create a robust and coherent cooperation

157 framework accompanied by the enactment of relevant legislative frameworks for guiding its application.

4.4.4 Case referral mechanism

When asked whether their oversight institutions have a formal or informal mechanism for referring cases requiring investigations by other structures, nine officials (32%) confirmed the existence of the formal mechanism, while 12 officials (43%) confirmed the existence of informal mechanisms in their structures. However, seven of the officials (25%) indicated that they are not sure whether there is any case referral mechanism or not at their structures. Generally, the officials concurred that some formal and/or informal mechanisms do exist from structure to structure albeit with few or little follow-up mechanisms. Figure 4.10 below provides the summary of the findings regarding the above-mentioned question. Further, Respondent 11 of the officials cited section 7 of the Intelligence Oversight Act of 1994 which authorises their office to receive complaints from members of the public, the JSCI, the HRC, the Independent Police Investigative Directorate (IPID), and many other structures. The section further indicates that should any member of the public lodge any complaint with the IG which falls outside its mandate, the office is supposed to direct the complainant to the relevant office or structure responsible for handling such cases.

158

Figure 4.10: Whether oversight structures have a formal or informal mechanism for referring cases to other structures

Despite the latter provision by section 7(7) of the Intelligence Oversight Act of 1994, Respondent 11 of the officials indicated that regulations for guiding such a case referral mechanism have not been signed yet by the Minister of State Security and the JSCI as required. As a result, the office lacks a mechanism or framework to conduct necessary follow-ups to determine whether the complainant did eventually lodge the case with the relevant referred structure or not. Further, Respondent 11 of the officials again indicated that it is incumbent on the complainant to locate the relevant structure, lodge the case and make the necessary follow-ups.

As already discussed in Chapter 1 in section 1.4, the absence of a clear case referral mechanism with well-articulated follow-throughs reflect a classic case of fragmentation, lack of cooperation and non-existent coordination, among the intelligence oversight structures. This was confirmed by all 28 (100%) interviewed officials who unanimously

159 agreed that the absence of a coherent and well-structured case referral mechanism continues to hamper the effective follow-through of the referred cases to ensure their logical conclusion. As in the case of the creation of a cooperation framework, all 28 (100%) respondents concurred that in order to create a case referral mechanism, a legislative or at minimum a regulatory framework should be formulated to guide such an arrangement. As Agere (2000) concluded, the establishment of a formal case referral mechanism is crucial for ensuring the realisation of the key tenets of good governance such as transparent, accountable and effective decision-making in resolving the case. To determine efficacy, I further asked the officials whether the mechanism or agreement functions optimally to enable easy referrals and timely responses. In response, 18 officials (64%) indicated that the mechanism is less optimal. However, five officials (18%) were not sure, while three (11%) saw it operating optimally and only two (7%) considered it to be fully optimal. Figure 4.11 presents the schematic representation of the above-mentioned question as posed.

160

Figure 4.11: Whether the case referral mechanism functions optimally to enable speedy referrals and timely responses

4.4.5 Intelligence oversight processes

During the interview process, Respondent 6 of the external experts indicated that the over-reliance of the IGI on the information provided by the overseen structures is very problematic. This is compounded by the absence of a mechanism to triangulate the provided data with any other structure which performs similar oversight functions. However, in terms of the pre-authorised intrusive operations, the IGI is at least able to juxtapose the information as provided by the intelligence services with the appointed judge who is responsible for the pre-authorisation of such operations. This is crucial for picking-up inaccuracies or misrepresentation of facts in the provided information by the affected structure. Further, Respondent 15 of the officials indicated that in their provincial oversight work, they start by engaging intelligence practitioners at the lower

161 levels of the structures and then triangulate their findings with the information provided by the provincial management. This is not only limited to the provincial level, even at the national level oversight officers engage people at the lowest levels of the value chain and then triangulate the findings when engaging members of management.

Although triangulation demands the application of comparative analysis of information or data sourced from different sources, in this case it is mainly used for the purposes of combined assurance. Further, as part of the current intelligence oversight processes, the OIGI do have the updated MoU with the office of the AG mainly for combined assurance purposes particularly on issues of financial auditing. When assessing the intelligence oversight processes with specific reference to the work of the IGI, the MRC (2008:246) found that, “the OIGI is sometimes bogged down into addressing human resources grievances and disputes among members of the intelligence services”. Respondent 16 of the officials lamented the continuous stifling of their oversight work by members of the intelligence services who drag them into addressing their petty labour relations concerns. This is despite the dispensation having provided the internal labour relations regime through the establishment of Constituency Forms, Staff Councils and the Intelligence Council on Conditions of Service (ICCS).

All 28 (100%) officials and seven (100%) external experts agreed that the absence of legislative direction on the cooperation of intelligence oversight structures allow the various structures to apply different, uncoordinated mechanisms in their intelligence oversight processes. This situation resembles a chaotic environment with no clear set guidelines for the fulfilment of organisational objectives. This was later emphasised by Respondent 4 of the external experts who indicated that the application of these varied mechanisms create no space for common approaches towards the achievement of effective cooperation among oversight structures. In an effort to improve the intelligence oversight processes particularly in the auditing of covert operations, Gill

162

(2012) suggests that electronic auditing methods can be used particularly in view of the global move towards the implementation of electronic intelligence operations.

However, there was general consensus among all 35 respondents that the current intelligence oversight processes should not be abandoned as the country continues to seek better ways to facilitate the cooperation of intelligence oversight structures. To enhance cooperation, concerted efforts should be made to try to understand and improve the actual operational processes involved in the intelligence oversight processes as undertaken by various structures in performing their allocated mandates.

4.4.6 Cooperation of chapter 9 institutions

The International Ombudsman Institute (2010:23) emphasised the fact that various countries established chapter 9-type institutions “to protect their people against violations of their human rights, abuses of power, errors, negligence, unfair decisions and maladministration”. As the Institute puts it, this is done in order “to improve public administration and make government’s actions more open and its public servants more accountable to members of the public” (International Ombudsman Institute, 2010:23). Globally, chapter 9-type institutions “emerged as a widely established avenue for the lodging of individual complaints against the actions of public authorities” (International Ombudsman Institute, 2010: 24).

Apart from being complaint lodging platforms as earlier discussed in this chapter in section 4.2.5, chapter 9 institutions in South Africa possess persuasive powers for negotiation or mediation purposes to resolve grievances. Their mandate includes the drafting of recommendations which are popularly called ‘remedial actions’. All 35 (100%) respondents again confirmed the absence of a guiding framework to ensure cooperation of conventional oversight structures including chapter 9 institutions. Despite the absence of the cooperation framework for guiding cooperation among

163 chapter 9 institutions, it is relatively easy to accord the office of the PP real independence as the custodian of citizens’ rights complementing the judiciary in administering justice. Respondent 4 of the external experts raised a fundamental point. He indicated that they are not sure whether there is a need for the various oversight structures to coordinate their efforts particularly given the specific obligations they have to fulfil. Respondent 5 of the external experts raised the following fundamental question: what was the intention of the legislature for creating multiple oversight institutions as stand-alone structures and not leaving room for the creation of a cooperation model? In responding to his own question, Respondent 5 indicated that the legislature might have been influenced by the bad experiences of the apartheid legacy wherein too much power was concentrated in one institution such as the then Office of the Ombudsman. In addition, Respondent 3 of the external experts believes that this was done to strengthen the principle of checks and balances wherein various institutions do part of the work to be completed by another structure.

As discussed in Chapter 4 in section 4.2.3, most of the chapter 9 institutions are given the power to access any classified documentation held by any individuals or organs of state, including the power to subpoena witnesses. With these kinds of powers, it is very important for chapter 9 institutions to cooperate with one another during the performance of their varied mandates towards the realisation of good governance results. In terms of the intelligence oversight work of the AG, its audit reports are only presented to the JSCI with confidential and secret classifications. As a result, Respondent 5 of the external experts described this situation as unconstitutional and contradictory to the principle of transparency and openness espoused in the country’s constitutional dispensation. In his submission to the Ministerial Review Commission (2008), the then AG re-emphasised the fact that the audit reports of the SSA should be presented to parliament after consideration by the JSCI.

164

According to the AG, this should be possible as such reports contain no sensitive information that “would prejudice the intelligence services or compromise national security” (AG Annual Report 2014/15:34). A similar view was expressed by Respondent 7 of the external experts who emphasised the centrality of openness in the management of public resources. As additional support, the AG committed to exclude classified information from its audit reports of the intelligence services, and to then specify the fact that sensitive details had been excluded from the report. To illustrate the need for cooperation among chapter 9 institutions, it is impossible for the AG’s staff to obtain financial evidence on the expenditure incurred during the running of covert operations. Further, the AG’s staff cannot interview or access informants’ databases, or confirm the use of covert surveillance equipment.

Operationally, the AG does not have the expertise to verify the accuracy or correctness of any amount paid to contacts or registered sources in relation to the quality of received information. It is for these reasons that cooperation between the AG’s office and the IGI cannot be overemphasised. In this regard, it would be expected that the IG will provide expert advice and assure the AG that all payments to sources correspond with the quality of provided information in accordance with the country’s National Intelligence Priorities (NIPs). This is crucial for the advancement of combined assurance. Currently, the AG does audit the work of the intelligence services and reach conclusions on the existence of internal control measures over intelligence covert operations, but cannot test the effectiveness of such controls. Therefore, the auditing of covert operations will always receive the lowest level of audit assurance from the AG, hence the constant issuance of qualified audits to the intelligence services.

4.5 SUMMARY AND CONCLUSIONS

From the above discussion, it is clear that the need for, and importance of the cooperation of the intelligence oversight structures cannot be overemphasised. This

165 was indicated by the responses from all 35 interviewed respondents from oversight officers to external experts. Based on the summarised findings above, it is clear that the current state of cooperation among intelligence oversight structures falls far short of satisfying the requirements for systematic cooperation as advanced by Rogers (2012). It confirms the assertion that any attempt of these structures to cooperate had remained ad hoc, uncoordinated and in certain cases non-existent. However, this is not surprising. Born (2013: 23) had already indicated that, “the inherent secretive character of intelligence creates tension between the need to maintain the secret and the need to start sharing it”. In this chapter, the research, legislative, policy and empirical evidence confirm the need for the development of a more systematic cooperation model to improve the current state of largely informal cooperation among intelligence oversight structures.

The creation of proper capacity and adequate resource allocation among oversight structures will ensure differentiated service-centric specialisation and facilitate the realisation of good governance results. In terms of the oversight functions, it is important for the appointed overseers to “perform their functions impartially and independently, in good faith, without fear, favour or prejudice” as provided for by section 109 of the RSA Constitution of 1996 and emphasised by other relevant pieces of legislation. Despite the need for cooperation, it should be noted that the success of any oversight structure also depends on the integrity of the incumbent, its independence and professionalism. Based on the 1994 transitional period, the reform of the intelligence services raised high expectations in terms of “intelligence quality, transparency, accountability, control and oversight”, as provided for in the White Paper for Intelligence of 1995. Therefore, despite the creation of a commendable legislative oversight mechanism, the main problem was failure to develop a legislative framework to provide for the cooperation of the intelligence oversight structures when discharging their varied legislative mandates.

166

Considering the responses to the posed questions on the rationale for, and state of cooperation among intelligence oversight structures, there is general consensus among the 35 respondents that developing a systematic cooperation model could help demystify intelligence tradecraft and build trust between communities and members of the intelligence services. There is general agreement that developing a firm criteria and mechanisms to guarantee coherent, cooperative and coordinated intelligence oversight work would help the country in the realisation of good governance results. However, it is also noted that the successful cooperation among intelligence oversight structures would require constant vigilance from numerous actors such as the state, citizens, and some civil society institutions.

In fact, successful cooperation among intelligence oversight structures can reduce duplication and possible competition among them. Whereas this chapter presented data on the rationale for, and state of cooperation among the intelligence oversight structures, the following chapter presents findings on the nature of both structural and operational barriers experienced by members of the oversight structures when engaging each other during the performance of their oversight work.

167

CHAPTER 5 BARRIERS TO IMPROVED COOPERATION AMONG INTELLIGENCE OVERSIGHT STRUCTURES IN SOUTH AFRICA

5.1 INTRODUCTION

While the previous chapter presented research findings on the rationale for effective cooperation among intelligence services, and the current state of cooperation among intelligence oversight structures in South Africa, this chapter summarises the research findings on the nature of the main barriers hampering the effective cooperation of intelligence oversight structures. The adduced data aims to answer the second research sub-question: RQ 2: What are the barriers for the effective cooperation of democratic intelligence oversight structures for good governance results?

As in the case of the previous chapter, the answers to this question emerged from documentary reviews and interviews conducted with the sample of intelligence oversight officers and external experts. In this study, barriers are contextualised as those factors hindering the effective cooperation of various intelligence oversight structures when overseeing members of the intelligence services (Caparini: 2002). This investigation is predicated on the fact that the 1994 transitional period re-tasked members of the intelligence services from safeguarding the apartheid system to guaranteeing human security, rights and freedoms (White Paper for Intelligence: 1995). The new dispensation was necessitated by the emergent constitutional order which became supreme and enshrined democratic principles, culture, and values of a free society (RSA Constitution: 1996). In addition, the new dispensation meant that intelligence services were no longer supposed to suppress public opinion but to protect people’s rights and their well-being within set constitutional parameters.

168

Therefore, the new dispensation called for the creation of multifaceted intelligence oversight machinery which was meant to oversee members of the intelligence services. This was to ensure their sustained subjection and compliance to the country’s new statutory and regulatory frameworks. Further, the new dispensation expected the multiple oversight structures to coordinate their oversight work in compliance with the principle of corporative governance. However, as years went by the fragmentation, non-cooperation and disintegration of the various intelligence oversight structures became apparent. The absence of a clear legislative and regulatory framework for guiding the cooperation of these structures worsened the situation. As a result, silo mentalities, territorialism and turf-wars emerged resulting in the weakened oversight machinery. This chapter determines the nature of the structural and operational barriers experienced by intelligence oversight structures in their quest to cooperate with each other.

5.2 BARRIERS TO IMPROVED COOPERATION OF OVERSIGHT STRUCTURES

In summarising the findings of international benchmarking exercises in Chapter 2 in section 2.4.2, as well as assessing the views of intelligence oversight officials and external experts in Chapter 4 against this background, it became clear that members of the intelligence oversight structures confront a multiplicity of challenges in their attempt to cooperate with each other. Caparini (2002:17-18) highlights five key factors which at the global level play a critical role in barring the cooperation of intelligence oversight structures. These are:

 “The preservation of excessive secrecy coupled with the hiring of non- intelligence officers to conduct intelligence oversight work.  The use of excessive secrecy undermines the relationship of trust between the intelligence services, their oversight structures, the policy-makers and members of the general public.

169

 The granting of discretionary powers to intelligence practitioners presents them with wide ranging authority in a secretive environment.  The utilisation of ‘plausible deniability’ by members of the executive frustrates intelligence oversight efforts given their allowance to easily deny having knowledge of any operation should it back-fire. Further, it insulates members of the executive from dealing with the consequences of controversial intelligence operations; and  The undue utilisation of ‘national security’ narrative compromises accountability and undermines transparency as one of the key tenets of good governance”.

According to Lowenthal (2003), the cooperation of intelligence oversight structures is also hampered by the nature of intelligence tradecraft which is predicated on the fact that secrecy is the lifeblood of the intelligence practice. However, this narration does not necessarily consider the counter-argument that says transparency is the oxygen of democracy. Therefore, it should be understood that the need for the cooperation of intelligence oversight structures seeks to ensure the realisation of balanced acts between the two contradictions. In addition, Bruneau and Dombroski (2001) indicated that the problem of intelligence tradecraft is based on the fact that its processes are secret, its products are secret, and the people who carry out the task are also secret. The MRC (2008:57) added that “even the ministerial regulations governing intelligence services are secret, the Auditor-General’s reports on the intelligence services are secret, their budgets are secret, and their annual reports are also kept secret”. This is an illustration of a situation wherein too little information is released on the activities of the intelligence services and this threatens transparency and the ability of the masses to hold the services accountable. This is demonstrated by the non-publication of annual reports of the intelligence services including the IG’s compliance reports on the various intelligence service structures.

170

Further, Bruneau and Dombroski (2001) indicated that the nature of intelligence tradecraft presents limitations on certain human rights on the grounds of immediate threats to national security and this is supported by international law. Therefore, these limitations bring some element of distrust between the various intelligence oversight structures, hence the absence of effective cooperation. However, the primary concern arises when such an allowance is used for concealing nefarious activities and undermining the rule of law. The relevance of intelligence oversight and the need for the cooperation of oversight structures is necessitated by the complexities associated with intelligence tradecraft (Bruneau and Caparini: 2001). When the legal framework is not explicit enough regarding oversight modalities, members of the intelligence services could exploit the situation and pursue nefarious personal interests as revealed in the Project Avani scandal. It was for these reasons that Born et al (2007) emphasised the need for the adoption of systematic oversight mechanisms in order to preserve people’s constitutional democracy.

Before attempting to solicit views on the nature of barriers to the effective cooperation of intelligence oversight structures, I started by asking both oversight officers and the external experts whether they think there are certain impediments to the effective cooperation of various intelligence oversight structures. In response, 25 (89%) oversight officers responded yes, two (7%) were not sure and only one (4%) indicated that there are no impediments for the cooperation of these structures. In addition, five (71%) of the seven interviewed external experts indicated yes, one (14%) was not sure and one (14%) indicated no, implying the absence of specific impediments to the cooperation of intelligence oversight structures. Figure 5.1 presents the schematic summary of the response to the abovementioned question.

171

Figure 5.1: Whether there are certain impediments or barriers to the effective cooperation of intelligence oversight structures

After getting confirmation from the oversight officers and the external experts on the existence of impediments or barriers for the effective cooperation of oversight officers, I solicited further views on the nature of such impediments. I asked the 28 oversight officers and seven external experts to cite one factor which they consider to be a critical impediment or barrier to the required cooperation. In response, a number of impediments were cited and the most prominent ones include: lack of common oversight doctrine (four) (14%), unclear information classification methods (three) (11%), excessive secrecy and lack of trust (four) (14%), principle of need to know (three) (11%), demand for excessive transparency (one) (4%) and undue utilisation of ‘national security’ (three) (11%). In addition, the lack of a common doctrine was further cited by (two) (29%) external experts, excessive secrecy and lack of trust (one) (14%) and demand for excessive transparency by another (one) (14%) external expert.

172

For analytical and assessment purposes I have distinguished structural barriers from operational barriers, and conceived the difference as follows: structural barriers encompass contextual factors which exacerbate vulnerability and impede access to a particular situation (Gruskin, Ferguson, Alfven, Rugg and Peersman 2013). In addition, Blankenship, Bray and Merson (2000) had earlier described structural barriers as super-structural factors such as economic development which could influence the environment which could in turn shape the experiences of individuals. In this context, structural barriers embrace the fundamental obstacles inhibiting the effective cooperation of oversight structures and are mainly located outside the victim’s area of influence. On the other hand, operational barriers are described by Oie (2014) as the reflection of measures or activities with an explicit purpose to prevent hazard or an occurrence from being realised. Further, Tuset (2014:6) describes operational barriers as something that either “(1) completely prevents something from happening, (2) makes it less likely to happen, (3) completely eliminates the consequences, or (4) reduce the severity of a particular consequence”. Based on these assertions, operational barriers embrace localised obstacles which hamper the occurrence of a particular matter and are mainly located within one’s area of influence. Figure 5.2 presents the schematic representation of the cited impediments and barriers to the cooperation of intelligence oversight structures as raised by oversight officers and the external experts.

173

Figure 5.2: An outline of critical impediments or barriers to the effective cooperation of intelligence oversight structures

174

The following discussion presents the findings of this study on the nature of barriers or impediments for the effective cooperation of intelligence oversight structures. The discussion starts by presenting structural barriers.

5.2.1 Structural barriers to the cooperation of oversight structures

Based on the above findings, the following discussion presents the structural impediments to the effective cooperation of intelligence oversight structures.

5.2.1.1 Lack of a common oversight doctrine

As indicated above, the lack of a common oversight doctrine was cited by four officials as one of the key barriers to the effective cooperation of intelligence oversight structures representing 14% of the interviewed officials. It was also cited by two external experts representing 29% of the interviewed experts. However, this is not unique to South Africa. Desai (2005) indicates that in the US the law enforcement agencies and intelligence services who deal with crime and terrorism also lack common doctrine to facilitate their cooperation to effectively deal with the threats they face.

This is similar to Carter’s (2003) assertion that, lack of common understanding on crucial concepts such as intelligence, national security, and/or oversight present difficulties for the cooperation of the various oversight structures. For instance, during interviews Respondent 7 of the officials indicated that, “it is not part of my job description to engage, cooperate or coordinate my efforts with the other oversight structures when conducting my investigations on a particular complaint at hand”. This assertion confirms the arguments in Chapter 5 under section 5.2.2.2 which indicated that the absence of a clear legislative framework compelling the oversight structures to cooperate with each other discourages the actual cooperation of these structures when doing their oversight work.

175

McGarrel and Schlegel (1993), Walker and Katz (2001) earlier cited the absence of a common doctrine as one of the key challenges to the broader challenge of inter-agency cooperation. This is associated with problems of “information sharing among the law enforcement and intelligence agencies” (Walker and Katz, 2001:45). Some of these issues borders on the question of trust among the various agencies. For instance, Respondent 13 of the officials indicated that, “as the intelligence oversight officer working on inherent secret environment, I cannot be explicit in dealing with my colleagues from the chapter 9 institutions given their inherent transparency and openness as this could compromise the security of the information I will be sharing with them”. This assertion reflects the manner in which the oversight officers at one oversight structure see themselves vis-á-vis their counterparts in the other structures.

In the US, the problem of cooperation has been clearly cited in the findings of the 9/11 Commission Report and the Katrina Lessons Learned Review Group after which the government undertook external and internal organisational reforms for addressing the various barriers to inter-agency cooperation (Jackson and Brown: 2007). Respondent 7 of the external experts suggested that “the South African government should urgently develop a national doctrine which the intelligence oversight structures can refer to when coordinating their intelligence oversight work with other structures”.

5.2.1.2 Unclear information classification methods

In citing impediments for the effective cooperation of oversight structures, three officials raised the question of unclear information classification as one of the barriers. This represented 11% of the interviewed officials. Further, Respondent 3 of the external experts indicated that the discrepancies in the manner in which information gets classified serves as a deterrent on the effective cooperation between intelligence oversight structures. As a result, Lowenthal (2003:58) presented the three main classification levels being used globally and those are confidential, secret, and top

176 secret information. These three levels are globally used within intelligence tradecraft and are described as follows:

 “Confidential classification is given to information that, with unauthorised disclosure, could be expected to cause damage to national security.  Secret classification is given to information that, with unauthorised disclosure, could be expected to cause serious damage to national security.  Top secret classification is given to information that, with unauthorised disclosure, could be expected to cause exceptionally serious damage to national security”.

As Africa (2009) indicated, in South Africa information classification is also done on the same three levels as provided for by the MISS as approved by cabinet. According to Currie and Klaaren (2002) in Africa (2009: 19), “the officials are expected to classify sensitive information as confidential, secret and top secret depending on the degree of harm to national security the information would cause should it be unofficially disclosed”. According to Fraleigh et al (2011:95), “the classified government information should not be made available to the media or general public, but to only government officials but on a ‘need to know’ basis”. In terms of MISS, section 15 (1)(a) of the Protection of State Information Bill asserts that:

“state information may be classified as confidential if sensitive and its unlawful disclosure may be harmful to the security or national interest of the Republic, or could prejudice the Republic in its international relations”.

Section 15 (2)(a) of the same Bill states that state information may be “classified secret if sensitive and/or its disclosure may endanger the security or national interest of the Republic, or it could jeopardise its international relations”. Section 15 (3)(a) of the same Bill states that state information may be “classified top secret if sensitive and its

177 disclosure may cause serious or irreparable harm to the national interest of the Republic or may cause other states to sever diplomatic relations with the Republic”. Similarly, article 12(1) of the Canadian Access to Information Act, 1980 states that:

“access to information can be denied should disclosure of such information detract the government from its ability to conduct international affairs, interfere unduly in the government’s capacity to defend the country, detect, prevent, or suppress subversive or hostile activities”.

In line with these assertions, Respondent 6 of the external experts concedes that in some cases keeping a secret and denying people access to certain information can be justifiable, particularly in view of national security imperatives. Further, section 5(1) of the Protection of State Information Bill (POSIB) defines classified and protected information as “state information which requires protection against unlawful alteration, destruction, or loss and is referred to as valuable information”. According to section 7(1)(a) of the Bill, the Minister of State Security would “prescribe broad categories and subcategories of information that may be classified, downgraded, declassified and protected against destruction, alteration and loss”. Subsequently, section 8(1) of the Bill allows the “head of the organ of state to establish departmental policies, directives and categories for classifying, downgrading and declassifying information and protection against loss, destruction and unlawful disclosure of information created, acquired or received by the organ of state”.

Furthermore, section 49 of POSIB criminalises the classification of “information for the purposes of concealing breaches of the law, furthering an unlawful act, hiding inefficiency or administrative error, preventing embarrassment to a person or organisation, or any other purpose ulterior to the Act”. In its media statement dated 26 October 2010, the Ministry of State Security states that the officials who will be classifying information under the POSIB (2010) will have to follow specific guidelines

178

(www.ssa.gov.za). As a result, for any classified information there will be clear, justifiable, and legitimate national security reasons.

From the global perspective, Fraleigh and Tuman (2011:95) indicated that classified information generally includes “military plans, intelligence activities, scientific and technical material, including information about the strengths and weaknesses of programmes relating to national security”. Considering these assertions, it is clear that the promulgation of POSIB will address the question of unclear information classification as one of the structural barriers for the cooperation of intelligence oversight structures.

However, following the increased public resentment towards the Bill the then Minister of State Security, Dr Siyabonga Cwele committed to removing concepts of national interest and commercial information as the basis for information classification (Media statement dated 26 October 2010). At the time, Dimba (2011: 23) of the “AVAAZ advocacy group had collected over 40 000 signatures for a petition against the Bill”. In addition, the Right2Know campaign had also collected over 900 individual signatures and 350 institutional sympathisers while the DA gathered over 30 000 signatures against the Bill (opendemocracy.org.za). The objection was strengthened by the factual criticism of the Bill by the former Minister of Intelligence, Mr Ronnie Kasrils. He had earlier initiated the 2008 Bill which was again defeated by strong organised community resistance against it. Section 13 (b) of the POSIB states that classified information must be protected from unlawful disclosure “according to the degree of harm that could result from its unlawful disclosure”.

In addition, section 13 (c) of the Bill states that classified information “may be made accessible only to those holding appropriate security clearance in order to enable them to perform their day-to-day work”. A security clearance certificate is awarded to an individual who has successfully undergone a vetting process.

179

According to the Cambridge Advanced Learner’s Dictionary (2006:23), “vetting refers to the examination of something or someone in order to carefully ascertain whether they are acceptable or suitable for a particular purpose or job”. However, Masiapata (2008: 103) argues that “although the vetting process can determine the suitability or unsuitability of an individual for a certain job, it can never guarantee that the accepted individual will be perfectly reliable”. He indicated that the vetting process just minimises the chance of hiring an extremely unreliable/unsuitable person. In this regard, section 18 of the POSIB asserts that a “person who is in possession of a classified record which has been unlawfully communicated or delivered should return such record to any member of the police service or the State Security Agency”. This was the most contagious and rejected part of the Bill.

According to the joint media statement between the Ministry of State Security and the South African National Editor’s Forum (SANEF) the Minister had agreed to, through section 38, harmonise the POSIB with the Protected Disclosures Act 26 of 2000 to protect the rights of whistle-blowers (Media briefing data 12 November 2010). Section 4 of the Protected Disclosures Act of 2000 describes a whistle-blower as a “public official or any person who on their own accord informs the relevant authorities on an act or inaction which is likely to defeat the stated purpose of a specific, or any government office”. Lowenthal (2003:58) cautions that, “any improper classification of information could undermine the very constitutional democracy the Intelligence services seeks to protect”. With improper classification, incriminating documents get concealed and the general requests to access them get turned down. In this regard, similar difficulties are felt by members of the intelligence oversight structures, let alone their quest to cooperate with each other.

Based on the literature review and the findings of empirical data, it is clear that developing a coherent classification system would facilitate the cooperation of intelligence oversight structures. In this regard, such a system could centralise the final

180 authorisation in determining the classification and/or declassification of state secrets. This should be done to enhance uniformity and standardise both classification and declassification systems across all government structures. Ultimately, information classification should focus on what should be published instead of what should be withheld from the public. In the end, the approach will facilitate the work of oversight structures and their cooperation during the performance of their oversight duties.

5.2.1.3 Excessive secrecy and lack of trust

Of the 28 interviewees, four (14%) of the oversight officers cited excessive secrecy and lack of trust as one of the underlying impediments for the cooperation of intelligence oversight structures. A similar matter was also raised by one (14%) external expert as one of the barriers to effective cooperation. Excessive secrecy and lack of trust remains one of the major concerns among intelligence oversight structures and the intelligence services themselves. Inasmuch as intelligence services do not trust each other, Respondent 28 of the officials indicates that even the oversight structures do not trust one another and this makes their cooperation difficult if not impossible. In his examination of inter-agency cooperation, Desai (2005) indicated that the major obstacle for inter-agency cooperation is the absence of common doctrine, different cultural postures and lack of trust among the various agencies. However, Kocaoglu et al (2006: 8) indicate that for defusing lack of trust, any decision not to disclose information must be substantiated with clear reasons for withholding such information.

Furthermore, the MRC (2008:45), states that “the less secrecy and the greater provision of information about the intelligence services would be of benefit to the services themselves”. Nathan (2009: 3) warns that, “any state that over-classifies information lacks credibility, remains difficult to maintain and becomes costly to manage”. Further, Respondent 4 of the external experts indicates that excessive secrecy causes lack of trust among the intelligence services and by implication the efficacy of intelligence

181 oversight structures. According to Jackson and Brown (2007), some of the structures continue to avoid coordinating their efforts given their fear of exposing their internal shortcomings and their elements of mismanagement.

Bok (1982) in Africa (2009) raised concerns about the possible abuse of state secrecy, hence his recommendations for members of the intelligence services to limit their use of secrecy and its associated powers. Although secrecy does not necessarily lead to unaccountability, Brin (1998:210) cautions that excessive secrecy creates the breeding ground for the emergence of corrupt tendencies. Further, Respondent 4 of the external experts added that unnecessary secrecy causes stigma, instils fear and reduces public support for the work of the intelligence services. Therefore, it suffices to say that information should be made available to the public when the national interest to do it is more important than security issues. From the interviews, there was general consensus particularly among the seven external experts, that although it is not wrong to keep legitimate secrets as part of intelligence tradecraft, excessive secrecy has the potential to undermine transparency and compromises accountability.

When asked whether their respective structures were able to establish the relationship of trust with other oversight structures, 20 (71%) officials indicated not yet, three (11%) said no, another three (11%) were not sure and only two (7%) said yes. Milward and Provan (2006) warned that governance networks can easily function where there is high density of trust, few participants, and high levels of goal consensus. Figure 5.3 below shows the schematic picture of the findings of the above question.

182

Figure 5.3: Whether the oversight structures have been able to create the relationship of trust with other intelligence oversight structures

During the discussions, it became clear that the notion of excessive secrecy inherent among the intelligence services complicated attempts to foster cooperation among intelligence oversight structures. This was shared by Respondent 13 of the officials who indicated that, “…due to the high levels of excessive secrecy within the intelligence services they are meant to oversee, the work of the oversight structures also get embroiled into such secrecy”. It is this secrecy which complicates the work of practitioners and thereby undermines the levels of trust among role players.

In his article titled: Is all military information necessarily critical to national security, Nick Sendall (1997:1) “questioned the rigidity of information classification methods for national security reasons”. This criticism gained momentum during the post-Cold War era given the intensification of intelligence oversight mechanisms. Unchecked and unaccountable security apparatus preys on vulnerable community members and trample their basic human freedoms. In Canada, the McDonald Commission indicated

183 that security should not be seen as the most important element of democracy hence, the purpose of the former is about the preservation of the latter.

According to Mathews (1978), excessive secrecy creates a situation wherein the trust between government and its officials gets strained which is followed by distrust between government and its citizens. This distrust transcends employees of the intelligence services and members of the intelligence oversight structures themselves. These kinds of situations predicate lack of cooperation and coordination between the oversight structures. In addition, the undemocratic governance of intelligence services has the potential to distort security priorities and compromise oversight efforts.

According to Africa (2009), excessive secrecy not only manifests at the higher echelons but often results in a lack of transparency and accountability particularly with issues of budgeting and resource allocation. This was earlier raised by Shane, Podesta and Leone (2004:3) who argued that, “excessive secrecy undermines the public’s effort of holding the intelligence services accountable on its utilisation of public funds and the implementation of its legislative mandate”. It is on the basis of these difficulties that members of the intelligence oversight structures struggle to effectively discharge their work and thus fail to coordinate their efforts.

However, it should be noted that ensuring the effective accountability of the intelligence services will never materialise as long as the oversight institutions are not supported to cooperate and coordinate their efforts. As Sendall (1997) puts it, intelligence oversight can be easily applied when members of the intelligence services are committed to balance the need for secrecy with the greater need for transparency. Further, defining clear limits of secrecy and the timeframes for classification and declassification would enhance intelligence acceptance by communities and thus avert the possible utilisation of intelligence in the advancement of personal or group political interests.

184

5.2.1.4 Principle of ‘need to know’

The issue of ‘need to know’ is closely linked to the question of secrecy. During the interviews, three of the officials cited the protection of information on the basis of ‘need to know’ as one of the key barriers to the effective cooperation of intelligence oversight structures, representing 11% of the interviewed officials. According to Warner in Gill and Phythian (2006:6), “an intelligence tradecraft is a secret activity which is mostly undertaken by the state in order to understand and influence foreign entities”. As a result, it is incumbent on the intelligence services to guarantee the protection of the identities of sources as well as the origin of classified information in its possession. Unfortunately, the actual protection of potential sources and the originator of classified information can only be realised through enhanced secrecy. As a result, any complete withdrawal of secrecy in the intelligence tradecraft would frustrate sources and discourage their enthusiasm for future cooperation with members of the intelligence services (Gill and Phythian 2006). Ironically, Respondent 23 of the officials indicated that, all acquired information requires full protection particularly if disclosure could reveal intentions, targets of collection efforts as well as the capabilities of the collection systems.

As Gill and Phythian (2006) put it, there is general consensus among the respondents that unauthorised disclosure of classified information could in certain cases lead to precautionary countermeasures, disruption of operations and the denial of future access. Therefore, the application of the ‘need to know’ principle in this regard is very justifiable particularly when the origin and the content of such information were acquired through an intelligence exchange programme with a Foreign Intelligence Service (FIS). This is critical for the preservation and maintenance of future relationships with such counterparts.

185

According to Brin (1998:315), secrets play an important role particularly in military related matters. In this case, there is little reason to deny the military forces the advantage of keeping their enemies in the dark. However, it is also worthwhile to send selected information to the enemy camp for the purposes of deceiving them (Gill and Phythian: 2006). Democracy or not, these tactics will remain important on the battlefield, as the side with better technical abilities may choose to conceal some forces, while creating ghost units to present a credible threat elsewhere (Brin: 1998). Born and Caparini (2007:200) argue that the application of the ‘need to know’ principle remains critical for two reasons:

 “First, it seeks to ensure that state officials will have access to information only if they have been cleared by security vetting for access at the appropriate level of classification.  Second, a compartmentalisation which seeks to ensure that circulation of knowledge with respect to particular techniques, operations, or targets should be minimised in the interest of security”.

As articulated by Born et al (2007), individuals will only have access to the information that they ‘need to know’ which is relevant to their area of responsibility. According to Africa (2009: 31), the principle of ‘need to know’ “arises in various social contexts, presenting a range of political, legal and ethical dilemmas”. Some of these dilemmas are universal and have been in existence for a long time. For instance, medical doctors are expected to treat the medical records of their patients with high confidentiality. The legal professionals too, are expected to contend with the dilemma of secrecy and the principle of ‘need to know’. In this regard, inability to observe these tenets could result in disciplinary action being taken against such a professional which in extreme cases could result in the withdrawal of their qualification.

186

Furthermore, Brin (1998:210) argues that in a dangerous world, states will continue striving to conceal vital technical, tactical, and strategic information from their potential foes. In her book entitled: Secrets on the ethics of concealment and revelation, Bok (1982:33) describes secrecy as “intentional concealment involving the deliberate withholding, hiding, or concealing of information in order to prevent someone else from uncovering it”. According to Bok (1982) in Africa (2009:31), secrecy can arise in various social contexts such as “medical research and practice, secret societies, trade negotiations, military and state activity”. Furthermore, even the provisions of international law permit governments to legislate in favour of protecting state secrets. Section 40 of PAIA empowers the information officer to “refuse request for access to a record of the body if the record is privileged from production in legal proceedings, unless the person entitled to the privilege has waived it”. Furthermore, Gill and Phythian (2006:34) state that, information security requires the protection of gathered information, method used, and the identity of the sources that provided the information. This should be done by the development of concerted effort to create mechanisms in which information will only be sent to the relevant officials with the requisite clearance and the right to receive such information.

However, it is acknowledged that the imposition of restrictions in the flow of information has the potential to limit discussions and stifle debate among analysts. According to Esther Dyson in Brin (1998:331), the challenge “is not to keep everything secret, but to limit the misuse of information”. It implies that trust, and accessing more information remains critical and this includes an understanding on how that information is being used. In fact, people become more comfortable when other people’s flaws are brought to the surface. However, in view of what has been argued thus far, it is evident that the principle of ‘need to know’ in the intelligence tradecraft cannot be ignored. Further, its application should not hinder the work of intelligence oversight structures and their quest to cooperate with one another. This is important, more so that, various respondents reflected on the fact that the principle of ‘need to know’ can be easily

187 abused and used to hide vital information needed by members of the intelligence oversight structures.

5.2.1.5 Demand for excessive transparency

Despite excessive secrecy, one (4%) of the officials cited the demand for excessive transparency as an impediment for the effective cooperation of intelligence oversight structures. This was also raised by one (14%) of the interviewed external experts. A similar view was expressed by Lowenthal (2003), who argued that while recognising the significance of transparency in good governance, the demand for its excessive application can also hinder effective cooperation in the same manner as excessive secrecy does. For instance, Respondent 5 of the external experts indicated that, “lack of security awareness in some of the oversight structures serves as the source of distrust among these structures, hence the lack of commitment to effectively cooperate with each other”. As indicated earlier, the oversight officers from the IGI’s office do not feel comfortable discussing classified information with their colleagues from the chapter 9 institutions whose operational work is conducted in full transparency/openness. Therefore, these realities hamper the effective cooperation of the various intelligence oversight structures during the implementation of their respective mandates.

As Lowenthal (2003:58) put it, acceptable secrecy applies when the activities and the performance of intelligence services is not “to be as transparent as those of other government structures, nor subject to the same degree of public scrutiny and debate”. For example, Lowenthal, (2003:58) indicated that, “publishing information on the allocation of resources to the intelligence practitioners could risk revealing their capabilities and targets and, in so doing, seriously compromise their effectiveness”. Therefore, transparency cannot be demanded on information pertaining to sources, operations, methods, procedures, and means of data collection (Lowenthal: 2003). It is this kind of information that requires proper classification and due protection.

188

In addition, Respondent 3 of the officials indicated that, any move to force excessive transparency within intelligence tradecraft will have a negative impact on the cooperation of oversight structures thereby compromising national security. Ultimately there is no contest with the need for the anonymity of intelligence covert operatives and the protection of their sources as part of the intelligence tradecraft. Instead, the primary concern is the conscious decision to advance excessive secrecy at the expense of transparency. Brin (1998:331) indicates that the need for transparency could be just as bad as excessive secrecy particularly if taken to extremes, or if applied unevenly or too soon. Like secrecy, the need for excessive transparency could trigger defensive postures where members of the intelligence services do not comply with intelligence oversight practitioners. In this regard, it should be noted that excessive transparency may cause intelligence operatives to withhold information should they realise that their entire input towards oversight processes will be leaked to the public.

However, Respondent 1 of the external experts indicated that information classification and its subsequent protection should remain guided by transparency with exception rather than secrecy. To support this assertion, Respondent 5 of the external experts indicated that members of the public would understand secrecy if only confined to those issues whose disclosure could cause significant harm to national security. These include to the lives of the individuals, members of the intelligence services and the state as a whole. In addition, it was agreed that the determination of such harm should be weighed against the level of public interest on disclosure. However, the Shadow Minister of State Security of the DA, Mr Stubbe (2015a:16) indicated that:

“while it is widely accepted that official communication and operations can only be transparent to a limited extent, to ensure that the relevant operations, sources or assets are not compromised, within this modified system of accountability the need for rigorous control and oversight is greater, not less”.

189

During the interviews, all 35 (100%) respondents indicated that although transparency raises the cost of delinquency it does not render unethical behaviour impossible. Therefore, effective intelligence oversight requires the new kind of power devolution, not from central to local government, but from government to civil society (Brin: 1998). The question of excessive transparency is a global concern among various governments of the world especially within the glare of scrutiny. Although literature reflects the dangers of excessive transparency, empirical data indicates that proper transparency provides the basis for highly democratic and open governance based on the growing contribution of global civil society.

Although only one (4%) official and one (14%) expert raised the issue of excessive transparency as one of the impediments for the effective cooperation of oversight structures, there was general consensus that the absence of universally shared, or at least mutually compatible societal norms and greater reliance on transparency can cause societal conflict. It was in this context that Respondent 8 of the officials argued that some secrets are worth protecting, as transparency only reveals behaviour, not intent. For instance, if the US does not worry about the UK building nuclear weapons it might be because they believe that the latter harbours no hostile intent. However, the US remains anxious about the possibility of a ‘nuclear Iran’ because they cannot be sure of Iran’s intent (Brin: 1998). Based on this example, it is clear that even if all conditions are right, transparency does not always work. This was emphasised by Respondent 8 who argued that knowing the fact that you are being watched does not necessarily make you change your behaviour.

5.2.1.6 Undue utilisation of ‘national security’

During empirical data collection, the question of undue utilisation of ‘national security’ was cited by three (11%) officials as one of the impediments to the effective cooperation of intelligence oversight structures. As Caparini (2002) puts it, international

190 law allows governments to limit certain human rights on the grounds of immediate threats to national security. This assertion is further supported by the International Covenant on Civil and Political Rights; in South Africa, the RSA Constitution of 1996 supports situations wherein national security can be used as the basis for restricting public access to certain classified information particularly the information handled by members of the intelligence services.

The Constitution of 1996 indicates that, “national security seeks to advance and reflect the resolve of South Africans, as individuals and as a nation, to live as equals, to live in peace and harmony, to be free from fear and want and to seek a better life”. In this case, there is general consensus among the various role players that national security is the justifiable ground for limiting certain rights and distributing information on the ‘need to know’ basis. Despite the consensus, other scholars such as Coliver, Hoffman, Fitzpatrick and Bowen (1999:18) argue that, the term ‘national security’ is not “sufficiently precise to be used as a basis for the limitation or restriction of the exercise of certain rights and freedoms of individuals”. It is a term which can have “very broad meaning and application and can thus be used by certain States to justify their imposition of unreasonable limitations or restrictions” (Coliver et al., 1999:18).

Further, Coliver et al (1999) argue that “national security may be invoked to justify measures limiting certain rights only when they are taken to protect the existence of the nation or its territorial integrity, or political independence against force or threat of force”. However, this justification on national security grounds should be responding to a threat to the country as a whole and not just some section of the community.

However, Respondent 7 of the external experts indicated that when ‘national security’ is being used as an excuse to avoid accountability it then stifles the work of intelligence oversight structures and their cooperation efforts. In fact, there is no objection to Herman’s (2001:5) assertion that, “secrecy is the trademark of intelligence tradecraft

191 and remains the basis of its relationship with government and its own self-image”. Nhlanhla (1992:70) asserts that “an intelligence activity is, by its very nature, characterised by secrecy and stealth; an understandable and often necessary feature as the defence of national security often requires the withholding of information that might be used against a country by would-be aggressors”. As Sendall (1997:2) puts it, “there is a need to protect the country’s strategic intelligence and its ability to conduct foreign relations”. These include “military operational plans, strategy, tactics and deployments as well as equipment performance capabilities, all of which maximise the country’s ability to provide its citizens with adequate defence” (Sendall, 1997:2). However, Nathan (2009) warns of the clear dangers associated with the undue utilisation of ‘national security’ to advance excessive secrecy, lack of accountability, power abuse and the infringement of human rights.

In his article entitled Lighting up the Intelligence Services: A Democratic Approach to Intelligence Secrecy and Openness, Nathan (2009:2) argues that national security “encompasses the security of the country, its systems of governance, its values and its entire population”. As a result, he argues that it then provides a compelling basis for state transparency rather than secrecy. This similar approach formed the basis of the South African constitutional order which proclaims that national security must reflect the resolve of its nationals which is to live as equals before the law. Despite the obsession on traditional intelligence secrets, Zikhali (2005:90) argues that “even ordinary government information is classified as sensitive under the guise of national security”. However, there is no question that information relating to pending tax increases, land rezoning and the retrenchment of staff requires due protection. In this regard, “unauthorised disclosure of such information can lead to chaos, corrupt practices or, for some individuals, improper monetary gain” (Zikhali, 2005:90).

Releasing official information before it is publically announced is contrary to prevailing law. In practice, one’s personal interpretation of what the national interest is might in

192 some cases also be different from the official view. Fleishman and Payne (1980:39) argue that, “personal views on the public good may be directly opposed to judgements on public interest”. The unintended consequences will be that public officials may come to believe that the nature of the official obligation requires them to act against their conscience.

Based on the collected primary data, it is clear that not everything ought to be protected and kept secret, but only to the extent that disclosure would compromise or degrade sensitive intelligence sources and methods, thereby becoming self-defeating. Respondent 2 of the external experts indicated that even though publicity about intelligence, either factual or fictional, is rampant, members of the public still believe, particularly in liberal democracies, that governments continue to classify too many documents and keep too many secrets. Therefore, it should be noted that the most effective manner of achieving public education on this matter is by ensuring that members of the public directly benefit from the products of intelligence, its information, and assessments. Most intelligence services of the world have internet websites wherein established codes of ethics for intelligence practitioners are published. As a result, members of the public are able to foster accountability among the intelligence practitioners in accordance with the published code of ethics.

Further, it is clear that the maintenance of national security is an inherent requirement of intelligence tradecraft and without some version of it valuable intelligence and its methods can easily be compromised and jeopardised. Within the intelligence tradecraft, a secret is both a virtue and a necessity. Lastly, the international benchmarking exercise revealed that secrecy cannot be separated from intelligence tradecraft. Therefore, information on the sources and the methods used to source such information should remain unknown to the targets of intelligence. Having understood the structural barriers hampering the effective cooperation of intelligence oversight structures, let us now look at the impact of operational barriers in this regard.

193

5.2.2 Operational barriers to the cooperation of oversight structures

In addition to the structural barriers discussed above, both the literature review and the interviews conducted with oversight officers and external experts raised other barriers which for the purposes of this study are grouped as operational barriers. However, some of the cited barriers do in certain cases qualify as both structural and operational barriers. For instance, Desai (2005) describes the absence of common doctrine as both a structural and operation barrier to the cooperation of oversight structures. This was supported by Jackson and Brown (2007:35) who argued that, “lack of common doctrine among the oversight structures perpetuate silo mentality, fragmentation and introduces undue competition among the various intelligence oversight structures”. This was further supported by Respondent 4 of the external experts that, the absence of concerted efforts to educate, train and develop oversight officers on the need and significance of their cooperation; sustains the problem.

On the other hand, Schwarz (2007) argues that, in certain cases effective cooperation is hindered by the closeness of some oversight officers to members of the intelligence services. In fact, Respondent 2 of the external experts indicated that effective cooperation can be achieved when there is an appropriate level of distance between the overseer and the overseen. This was emphasised by Desai (2005) who argued that, certain structures train their staff members to only focus on their key performance areas with little or no regard to the nature of work that is done by other sister structures. As Carter (2003) puts it, this is exacerbated by the different understanding and interpretation of what intelligence does and how it should account for its work. However, the international benchmarking exercise revealed that the absence of effective cooperation among intelligence oversight structures is not unique.

McGarrel and Schlegel (1993); Walker and Katz (2001) and Egger (1990), reflected on the general lack of cooperation among the various intelligence services or law

194 enforcement agencies during the performance of their unique legislative mandates in various countries of the world. These scholars emphasised that, the differences in decision-making, resource allocation and self-interests play a crucial role in deterring any inter-agency cooperation. To illustrate this point, Walker and Katz (2001) indicated that, police detectives can withhold important information from the visible policing team in order to deter them from getting credit for arresting certain criminals. This is evidenced by what Egger (1990:23) calls undue competition among services or agencies in relation to the “percentage of the crimes solved and the number of perpetrators arrested”. Based on the discussions in this introductory section, it is clear that the nature of barriers facing intelligence oversight structures can be structural, operational or both.

As with the previous section, both oversight officers and external experts were requested to cite impediments hampering the effective cooperation of intelligence oversight structures. Various issues were cited and got grouped as operational barriers. Those issues include; the absence of a clear legislative framework which was raised by four (14%) officials, absence of clear regulatory framework (two) (7%), unclear financial accounting practices (one) (4%), political interference and ‘plausible deniability’ (two) (7%) and the general misunderstanding of the concept of oversight (one) (4%). Further, the absence of legislative framework two (29%) and the issues of political interference and denialism one (14%) were also raised by external experts. The following discussion unpacks each of these operational barriers.

5.2.2.1 Absence of a clear statutory and regulatory framework

During the interviews, four (14%) of the officials raised the issue of the absence of clear legislative framework as one of the critical impediments hindering the effective cooperation of intelligence oversight structures. Interestingly, this was also raised by two (29%) external experts. Respondent 3 of the external experts lamented the silence

195 of the legislative framework in guiding the cooperation of intelligence oversight structures when conducting their work. Despite the silence, section 7 of the Intelligence Services Oversight Act of 1994 allows for the interface of the IGI and the JSCI in overseeing members of the intelligence services. However, the absence of a clear legislative provision causes members to find it unnecessary to cooperate when performing their intelligence oversight work. According to Africa (2009:23), facilitating the cooperation of these structures is crucial for averting the relapsing of the intelligence services into powerful forces that are characterised by the abuse of delegated authority for personal or group political interests.

Despite the absence of any legislative direction on the cooperation of intelligence oversight structures, the IG facilitated the drafting of the MoU with various other structures on the basis of cooperative governance as provided for in section 41 (2) of the RSA Constitution of 1996. They derive their initiatives from section 1(g) of the Intergovernmental Relations Framework Act 13 of 2005 which defines intergovernmental relations as the relationships between the governments or organs of states from different governments in the conduct of their affairs. Based on the interview findings, there is general consensus that the absence of a clear legislation framework discourages the cooperation of these structures during the performance of their varied legislative mandates.

As already indicated earlier, Respondent 11 of the officials indicated that since the promulgation of the Intelligence Services Oversight Act of 1994, the IG had not managed to get approval of its regulations for guiding its operational oversight work in the fulfilment of its legislative mandate. He further indicated that, the regulations have been prepared but not approved as yet by the Minster of State Security and the JSCI. However, the regulations provide for the signing of MoUs with the relevant structures to enhance their cooperation, coordination and the integration of intelligence oversight efforts (OIGI: 2016). It is for this reason that MoUs have been signed between the IG

196 and other intelligence oversight structures. Given the absence of the legislative framework for regulating the cooperation of various intelligence oversight structures, people were and/or are forced to improvise and develop alternative mechanisms.

Intertwined with the absence of clear legislative framework, two (7%) officials cited the absence of a regulatory framework as one of the critical impediments or barriers to the cooperation of oversight structures. The combination of the absence of a regulator and regulatory frameworks remains key operational barriers especially given different operational mandates. For instance, the HRC possesses “wide-ranging mandates which encompasses disregard for civil, political, economic, and people’s social rights” (Human Rights Act 54 of 1994). However, this broad mandate requires the HRC to undergo an intensive prioritisation process in order to concentrate on its programmed areas of work. In addition, one of the critical constraints relates to the availability of human resources to fulfil its broad-ranging mandates. Although the SAHRC does not operate as the court of law, it has strong powers to sue, either in its own name or on behalf of an aggrieved party. According to the Human Rights Act 54 of 1994, the HRC’s jurisdiction extends to and encompasses the state, individuals, and the private sector.

Based on the reviewed literature and the collected empirical data, it is clear that chapter 9 institutions play a vital role in the promotion and protection of democracy (RSA Constitution: 1996). However, their effectiveness is mainly hampered by resource constrains and the weak levels of cooperation with other state institutions. This situation is exacerbated by uncooperative and hostile executives coupled with parliament’s haphazard oversight mechanism on these chapter 9 institutions. Given this situation, the South African Human Rights Commission (SAHRC) had to, on some occasions, “subpoena cabinet ministers and request parliament’s intervention in dealing with the members of the executive” (Human Rights Act 54 of 1994). Further, the findings of the 2007 parliamentary review on chapter 9 institutions emphasised the need for the

197 allocation of additional resources to capacitate and improve the functionalities of these institutions. In addition, the review recommended closer collaboration between these institutions, especially those with overlapping mandates. In this regard, it is critical for the HRC to conduct outreach programmes in order to maintain a good profile among members of the public. The review also recommended that it will be helpful for parliament to consider establishing dedicated structures to oversee the work of the chapter 9 institutions and to debate their submitted reports.

During the interviews, Respondent 6 of the external experts indicated that even though South Africa had successfully established chapter 9 institutions, there have been gaps in the implementation of their mandates, characterised by lack of cooperation. Practically speaking, when the oversight structures are weak and their cooperation non- functional, the political elite can manipulate processes to advance their factional political interest (Human Rights Act 54 of 1994). These events can even occur irrespective of existing controls to protect the institutions from political interference. For instance, the irregular capture of state institutions before the 2009 elections reflects the internal contest for political expediency within the ruling party.

Based on the above discussion, the following example can be cited as an illustration of the undue utilisation of state institutions for advancing people’s political factional interest. In handling down judgement in one of the cases, a judge of the Supreme Court of Appeal indicated that the National Prosecution Authority (NPA) case against the then Deputy President, Mr Jacob Zuma was subjected to a ‘baleful’ of political interference. Further, the Economist (2009:34) indicated that “the charges of corruption, racketeering, money laundering, fraud and tax evasion were dropped given the evidence of a political conspiracy between NPA officials and former President Mbeki”. This and other examples, such as the dismissal of a National Director of Public Prosecutions, the corruption charges against the late Police Commissioner and the ongoing dispute between Constitutional Court justices and the current Judge

198

President of the Cape High Court, illustrate the erosion of institutional integrity. These developments raise questions about the professionalism of these institutions and their ability to survive attempts by influential individuals to subvert them for personal gain.

5.2.2.2 Unclear financial accounting practices

In responding to the question of impediments for the cooperation of intelligence oversight structures, one (4%) of the interviewed officials cited the issue of unclear financial accounting practices as one of the barriers for the effective cooperation of oversight structures. Similar sentiments were directly reflected by the MRC (2008) by indicating that, the utilisation of the slush funds in the financing of covert operations makes financial accounting difficult. During the engagements, Respondent 3 of the external experts reflected on the need for the effective control and oversight of the funds and assets allocated to members of the intelligence services. This is considered critical for ensuring the proactive detection and prevention of the abuse of state resources before it happens.

Accordingly, the current oversight model empowers the AG to oversee the utilisation of financial resources by all organs of state. The empowerment is derived from section 188(1) of the RSA Constitution of 1996 which states that “the AG must audit and report on the accounts, financial statements and financial management of all national and provincial state departments and administration”. This empowerment does not exclude overseeing the financial management by members of the intelligence services. Further, Chapter 11 of the Constitution provides for the principles of transparency, accountability and multi-stakeholder oversight of all the security services including their financial matters. Respondent 6 of the external experts emphasised the need to review the utilisation of ‘grey’ financial channels by intelligence services which are covert in nature.

199

Further, section 3(a) of the Public Audit Act 25 of 2004 declares the AG as the supreme audit institution of the Republic of South Africa. As a result, all institutions which are required “to maintain secrecy or confidentiality, or not to disclose information”, would be required to comply with the directives of the AG. This is applicable even though such a person would otherwise be in breach of their obligation of secrecy, confidentiality or the problem of non-disclosure. Further, section 3(b) of the same Act states that such compliance “is not a breach of any applicable legislation imposing the relevant obligation of secrecy or confidentiality or non-disclosure”. Section 3 gives the AG some super auditing powers which include the auditing of the financials of intelligence services.

However, the Public Audit Act acknowledges the sensitivities associated with the operations of the intelligence services as reflected in section 22(1)(a) of the same Act. Based on these discussions, it is the responsibility of the intelligence services to be transparent in their utilisation of the tax payers’ monies. In the 21st century, the effective cooperation of the oversight structures is also hampered by the fact that intelligence sharing is conducted electronically. This is complicated by the continued existence of the Secret Services Act 56 of 1978 which provided for the establishment of the secret services account to finance covert operations. As cited by Respondent 33, the AG has been struggling to effectively audit the secret account as required by law.

However, the Public Audit Act provides for the establishment of the Secret Services Evaluation Committee to monitor the utilisation of this account to avoid any abuse. Ironically, Respondent 5 of the external experts indicated that the said committee does not exist at the present moment within the current oversight framework. In fact, when conducting the public audit function the AG primarily audits the Security Services Special Account which was created by the Security Services Special Account Act 81 of 1969. Efforts to build robust auditing mechanisms require enhanced cooperation among the various intelligence oversight structures in doing their work. As Gill

200

(2012:17) indicates, “this will facilitate the systematic cooperation of various intelligence oversight structures in the performance of their varied legislative mandates”. Further, the various respondents indicated that the continued fragmentation and non- cooperation of oversight structures contributes to their continued inefficiencies in overseeing members of the intelligence services.

This is despite the Intelligence Oversight Act 40 of 1994 authorising the JSCI to receive annual audit reports such as financial statements and reports from the AG on the financial management of the intelligence services. However, the MRC (2008) found that all this work is reduced into a single line under the National Treasury allocation. This development is in direct contradiction to the provisions of the Public Financial Management Act, 1999 which requires transparent financial management in the allocated funds. These kinds of developments make the oversight of the finances of the intelligence services very daunting and undermine the effective cooperation of the oversight structures.

As Kocaoglu and Figari (2006:8) confirm, the state institution can withhold information “only for a narrow set of legitimate reasons set forth in international law and codified by national law”. In this regard, government may “withhold information from public access only if disclosure would cause demonstrable harm to legitimate interests such as national security or one’s privacy” (Kocaoglu et al., 2006:8). However, these exceptions must be clearly defined by law. Despite the various efforts of encouraging inter- structure cooperation, Houston and Parsons (1998) together with Pressman and Wildarsky (1984), cited distrust in financial management as one of the key operational barriers towards effective cooperation of intelligence oversight structures.

201

5.2.2.3 Political interference and ‘plausible deniability’

The issue of political interference and ‘plausible deniability’ was cited by two oversight officers as one of the key impediments for the effective cooperation of intelligence oversight structures. This represents 7% of the interviewed oversight officials. Further, the same matter was raised by one of the external experts representing 14% of the interviewed experts. To expand the assertion, Oleszek (2010) indicated that intelligence oversight work is conducted in a politically charged environment whereby parliament’s relationship with the oversight structures can easily change from that of cooperation to that of conflict within a short space of time. As Oleszek (2010:46) puts it, the intelligence oversight work can generate favourable “publicity for lawmakers, wins the electoral support at certain constituents and/or rebuts criticisms on favourite programs”.

Just to illustrate the point, it is important to reflect on the findings of the PP’s investigations contained in the ‘secure in comfort report’. The report focused on the security upgrades at President Jacob Zuma’s private residence at Nkandla in northern KwaZulu-Natal. The PP found the President to have unduly benefitted from the non- security updates to his private residence (Public Protector Report on Nkandla: 2014). She then recommended that the President pay back a portion of the money spent on the non-security features at the premises (compound). These findings and the suggested remedial actions caused serious political wrangling among members of the ruling party and the opposition parties in parliament (Public Protector Report on Nkandla: 2014). After the publication of her report, the PP was accused of being partial, political and representing the sentiments of foreign forces with ulterior motives. As a result, she was labelled an agent of the CIA in the US seeking to advance American interests at the expense of South African national interests. It became clear that the PP’s report had raised serious emotions among various role players forcing her to individually hold the fort. However, should the various oversight structures have

202 cooperated and backed the report; the President would not have ridiculed her outlined remedial actions.

Judgement came two years after intensive discussions in parliament when the EFF took the matter to the Constitutional Court. At the time, the EFF’s primary objective was to seek clarity on the binding nature of the remedial actions of the PP whose mandate is to protect constitutional democracy. In delivering his court judgement, the Chief Justice of the Constitutional Court, Justice Mogoeng Mogoeng indicated that the President failed to uphold the Constitution and to protect the office of the PP by failing to implement her remedial actions. It is on the basis of these developments that Respondent 6 of the external experts emphasised the high prevalence of political interference and ‘plausible deniability’ on issues of governance and intelligence tradecraft in particular. Despite the reprisals, the intelligence oversight work enables parliament to strengthen civil society, re-evaluate its activities and allows members of the executive to ensure the effective accountability of the public servants. This is important for ensuring the full observation of democratic tenets such as accountability, transparency, rule of law, and good governance through public participation. Further, Oleszek (2010:25) indicates that, intelligence oversight can serve to achieve three crucial institutional purposes: “checking the power of the executive branch, investigating the manner in which a particular law is being applied and informing members of the public about such developments”.

To correct the question of “plausible deniability”, Caparini (2002:21) suggested that “it might be in the interest of the state that responsible leaders are not informed about all activities of the intelligence services in order to counter the prevalence of ‘plausible deniability’”. This is crucial as some of the “necessary operations can be conducted without the official consent of the democratically elected government” (Caparini, 2002:21). This breaks the chain of command and also the oversight chain, causing supervisors and government to lose control over the intelligence services. Thus,

203

Caparini (2002:21) indicates that:

“Politicians are responsible for making sure not to let agencies withhold information, and the intelligence services must not allow politicians to protect themselves by purposefully avoiding controversial or crucial intelligence from reaching their offices”.

The issue of ‘plausible deniability’ allows government to instruct intelligence services to take specific actions in such a way that they can later deny any recollection of such events. This is critical especially when such actions fail or get exposed in such a way that the revelation damages the integrity of government. In this case, ‘plausible deniability’ remains applicable through the issuance of orders in the long chain of command, through which the original instructions cannot be traced should something go wrong. This is one of the direct challenges of parliamentary oversight given their continued reliance on either the IGI and/or the Minister of State Security.

5.2.2.4 Misunderstanding the concept of oversight

Despite government’s efforts to allocate oversight functions to various structures, one of the oversight officers indicated that the misunderstanding of the concept of oversight plays a crucial role as an operational barrier to the effective cooperation of intelligence oversight structures. During the interviews, it became clear that indeed various oversight officers from the various oversight structures expressed different understandings on the concept of oversight. These varied understandings manifested during the discussions on the rationale for, and the state of cooperation among the intelligence oversight structures, as already indicated in Chapter 4 in sections 4.3 and 4.4 respectively. For instance, 18 (64%) of the oversight officers considered the cooperation of oversight structures to be poor, eight (29%) found it to be very poor, while one (4%) found it to be good and another one (4%) was not sure. During the

204 interviews, Respondent 11 of the oversight officers indicated that oversight is crucial for the improvement of performance and efficiency. Further, Respondent 3 of the external experts indicated that, “the absence of an effective oversight mechanism can give intelligence operatives the freedom to negate its legislative framework and ultimately trample on people’s rights”.

Despite this assertion, Respondent 22 of the oversight officers indicated that, “the cooperation of intelligence oversight structures will be the bringing together of the structures to gang against members of the intelligence services to disrupt and disorganise the operational trajectory of the intelligence services”. According to Dirk Stubbe (2015:11), the Shadow Minister of State Security in the DA, “the South African intelligence oversight model might appear strong on the surface, but upon further examination it is very weak and problematic at best” (Democratic Alliance: 2015). In emphasising the significance of intelligence oversight, Stubbe (2015) indicated that, the cooperation of the intelligence oversight structures is vital for ending the abuse of members of the intelligence services by political parties for their own selfish political ends.

According to Agranoff (2012: 12), “any institution that is not exposed to critical scrutiny, be it direct through competition, indirect through transparency or formally by rigorous performance measurements, will lose its edge”. Therefore, lack of intelligence oversight and control could lead to the poor performance; or worse the emergence of anarchy or lawlessness, in the democratic state (Agranoff: 2012). As reflected by various respondents, the varied understanding of the concept of oversight among oversight structures serves to impede the effective cooperation of different intelligence oversight structures. This reality serves as a deterrent to the various interventions undertaken to enhance the cooperation of oversight structures for good governance results.

205

5.3 SUMMARY AND CONCLUSIONS

Based on the above analysis and assessment, it is clear that members of the intelligence oversight structures confront a multiplicity of barriers when attempting to cooperate with one another during the performance of their varied legislative mandates. Further, it also became clear that the nature of barriers range from structural to operational barriers. Given the absence of clear legislative and regulatory frameworks for guiding the cooperation of these structures, empirical data indicates that some institutions resorted to signing ad hoc MoUs for the purposes of opening communication channels with each other. However, MoUs are short-term solutions which are mainly dependent on inter-personal relationships. Further, empirical data confirms that the post-1994 intelligence oversight model lacks coherence resulting in many shortfalls. These shortfalls result in various challenges and barriers discouraging collective efforts and teamwork.

Despite these barriers and different understandings of oversight, the majority of respondents remain committed to cooperate with each other for the purposes of realising their respective professional mandates. This is necessitated by an understanding that should cooperation fail, the work of the oversight structures will remain fragmented and consequently less effective. However, the empirical data indicates that such cooperation does not come without difficulties. As Gill (2012) puts it, some of the complexities are associated with the loyalties of agency staff, IGs, and parliamentarians. The significance of the cooperation of intelligence oversight structures is underpinned by the problem of excessive secrecy which causes overseers to depend on the information provided by overseen structures.

Based on the arguments thus far, it is clear that the need for the cooperation of intelligence oversight structures is premised on the fact that although intelligence services are meant to protect the Constitution they have the full potential to threaten the

206 same constitutional order. Therefore, the ultimate idea of intelligence oversight and the cooperation of various structures; is predicated on the intention to create a mechanism to prevent the political abuse of intelligence services. According to Agranoff (2012:14), “comprehensive oversight embraces the provision of effective governance of the agencies, upholding the rule of law, and ensuring the proportionate use of their exceptional powers towards the protection of human rights”.

Chapter 6 summarises and assesses the main improvement strategies to achieve more effective coordinated oversight cooperation among intelligence services in South Africa.

207

CHAPTER 6 STRATEGIES TO IMPROVE COOPERATION OF INTELLIGENCE OVERSIGHT STRUCTURES IN SOUTH AFRICA

6.1 INTRODUCTION

Having assessed the rationale for, and the state of cooperation among intelligence oversight structures together with the nature of barriers for their effective cooperation, this chapter presents strategies to improve the systematic cooperation of these structures for good governance results. The presentation is based on the findings in response to the third research sub-question which is: RQ 3: What strategies can be developed to ensure the systematic cooperation of democratic intelligence oversight structures for good governance results?

The strategies to be presented emerged in response to the above-mentioned question from the systematic review of available literature and interviews with the oversight officers and external experts on issues of intelligence tradecraft and oversight. As the preceding chapters determined the status quo and the problems of cooperation among oversight structures, this chapter moves into the assessment of potential solutions to the already identified problems (Caparini: 2002; O’Brien: 2001 and the Ministerial Review Commission: 2008). As already discussed, the significance of these strategies is to minimise the possible abuse of the discretionary powers allocated to members of the intelligence services.

Before soliciting strategies to enhance the cooperation of intelligence oversight officers, I started by asking oversight officers whether there is a need to review the mandates and the roles of the various intelligence oversight structures to facilitate their cooperation. As such, 22 (79%) officers indicated yes, three (11%) indicated no and only two (7%) officers were not sure. In addition, a similar question was posed to the

208 external experts and six (86%) confirmed yes and only one (14%) person was not sure. Figure 6.1 presents the schematic representation of the findings on the above- mentioned question.

Figure 6.1: Whether there is a need to review the mandates and roles of the various intelligence oversight structures to facilitate their cooperation with each other

Respondent 15 of the oversight officers cited the need to improve the operational modalities of the various oversight structures. This is centred on the need for the promulgation of various legislative prescripts for changing the current oversight model. The chapter starts by tabling proposals on reshaping the key oversight structures as presented in Chapter 4 under section 4.1 titled: South African democratic intelligence oversight architecture.

209

6.2 RESHAPING THE KEY INTELLIGENCE OVERSIGHT STRUCTURES

As a follow-up from the above-mentioned question, I asked both oversight officers and the external experts in which structure they would want to see its mandates, roles and functions reviewed to improve its cooperation with other structures. In response, 17 (61%) of the oversight officers cited the IGI, six (21%) cited the JSCI, three (11%) cited civil society and the media, and two (7%) cited the Ministry of State Security. Interestingly, a similar trend was observed with external experts where three (43%) cited the IGI, two (29%) cited civil society and the media, one (14%) cited the JSCI and another one (14%) cited the Ministry of State Security. Figure 6.2 presents the schematic representation of the findings on the above-mentioned interview question.

Figure 6.2: Tabulation of particular structures whose mandates, roles and functions requires reviewing to improve their cooperation with other structures

210

Before presenting proposals on reshaping the intelligence oversight structures, Desai (2005:58) advises that any attempt to improve the cooperation of intelligence oversight structures should start by addressing the following four strategic matters:

“Firstly; the fact that intelligence oversight structures lack a formal overarching concept of cooperation or ‘doctrine’ for coordination. Secondly; that intelligence oversight structures lack an independent authority responsible for the development and training of personnel in such a doctrine. Thirdly; individual structures use different regional structures to organise their work. Fourthly; the fact that personnel policies within most, if not all, structures develop personnel who are primarily dedicated to their own structure rather than the inter-structure community”.

Given Desai’s (2005) assertion, I found it prudent to consider these factors during the development of the envisaged systematic cooperation model reflected as Figure 7.2 in Chapter 7. This is more so given the findings of empirical data which indicates that oversight structures are not individually equipped and/or capable to effectively hold intelligence operatives accountable. Therefore, it remains critical for these structures to join hands and collectively hold members of the intelligence services accountable. As Agranoff (2012:722) puts it, “any attempt to strengthen the protection of human rights from intelligence operatives requires the collective auditing and reviewing of the operation of the intelligence services at all times”. According to Dietrich (2015), those who talk about intelligence oversight should also reflect on the significance of the cooperation of the intelligence oversight structures. In fact, the effective cooperation of the structures is crucial for the effective implementation of the civilian intelligence oversight and control architecture.

As part of reshaping the intelligence oversight structures, the following discussion starts with the JSCI, followed by the Ministry of State Security, the IGI and civil society and

211 the media. Naturally, the discussions cover both the suggested structural refinements and the move to improve certain operational trajectories in their oversight functionalities.

6.2.1 Refocusing the Joint Standing Committee on Intelligence (JSCI)

Based on Figure 6.2 above, it is clear that the JSCI was raised by six (21%) of the oversight officers and one (14%) of the external experts as one of the critical structures requiring mandate refinement or operational changes to facilitate its cooperation with other oversight structures. However, Dietrich (2015) says that it should be noted that parliamentarians do not generally have substantive knowledge about intelligence activities, and are therefore, prone to misrepresenting facts in their deliberations. To support this assertion, Dlomo (2004) had earlier found a concerning disjuncture between the work of the JSCI and that of the IGI with the latter only providing yearly compliance certificates to the former. His findings were based on his study of the parliamentary intelligence oversight mechanism in the post-1994 dispensation. This was also reflected by Respondent 3 of the external experts who indicated that under the fifth administration, the IG engaged the JSCI only twice in the whole year. The first time was when the IG was presenting its strategic plan and the APPs to the committee. The second time was when the IG officials were orientating newly appointed members of the JSCI after the 2014 general elections.

During the orientation the IGI was mainly apprising the committee on the IG’s roles and functions in the broader intelligence dispensation. David Maynier, the then shadow Minister of Intelligence of the DA (2011) raised concerns on the complacency of the JSCI in its oversight work on members of the intelligence services. He accused the commission of using a ‘laissez faire’ approach in its oversight responsibilities. The following discussion proposes certain amendments to some operational approaches of the JSCI. These suggestions seek to facilitate the operational reconfiguration of this

212 committee for enhancing its cooperation efforts with members of other intelligence oversight structures.

6.2.1.1 Open meetings on matters of national interest

Despite its handling of classified information, Respondent 7 of the external experts indicated that, the JSCI should conduct public hearings on matters of national interest as this will help to demystify the work of the intelligence services and educate members of the general public on their role in upholding national security. In this regard, such matters can either be led by members of the intelligence services or experts from academia, research institutions and/or members of civil society. As part of parliament, members of the JSCI have the responsibility to effectively oversee the activities of the intelligence service on behalf of the ordinary members of the public (www.parliament.gov.za). Therefore, addressing matters of national interest in the public domain would serve to counter-balance the current criticism of the committee’s secrecy and absence of assertiveness in doing its work. In this case, Respondent 4 of the external experts indicated that, “when engaging members of the public the committee can explore their views on the respective mandates allocated to the various intelligence services vis-à-vis that of the JSCI”. For instance, certain national security threats such as cyber-crime, espionage and subversion require close partnerships between members of the intelligence services and civil society. It is for these reasons that members of civil society should be apprised on crucial national security matters.

In addition, Respondent 5 of the external experts indicated that, “cyber security threats require that any user of an Information Communication Technology (ICT) linked system should ensure that their devices are safe, secure and not compromised by worms, viruses or other malware”. He also added that preventing terrorism requires members of the public to serve as the eyes and ears of the government in supporting members of the intelligence services. Therefore, robust oversight can only happen when members

213 of the public have access to certain basic information on the work of the intelligence services and that enlightenment can be achieved through the oversight work of the JSCI. This consideration would allow the JSCI to promote and support the country’s national security effort while demystifying intelligence tradecraft. The then NIA indicated that the launch of its annual Golf Day, its website, and the ‘Vukuzenzele’ social responsibility programmes were attempts to reach out to members of the public and demystify the intelligence practice (www.nia.gov.za).

According to the Shadow Minister of State Security of the DA, Mr Dirk Stubbe (2015b), the JSCI has consistently failed to advise members of the public on the progress of complaints reported to them and dismally failed to effectively use its powers to define the steps to be taken to address the various problems facing the country. Given the failure of the JSCI to decisively advance its parliamentary oversight responsibilities, it remains practically impossible for the other oversight structures to consider effective cooperation with the committee except to submit the required oversight paperwork as provided for in law for compliance purposes. For instance, the IGI submits the annual compliance certificates on the work of the intelligence services to the JSCI.

6.2.1.2 Tabling and debating JSCI reports in NA and NCOP

Contrary to the views of securocrats, Respondent 3 of the external experts emphasised the need for the JSCI to prepare for and lead special discussions in the NA and the NCOP on matters of national security. This would serve to educate members of the public on the work of the committee and demonstrate their seriousness as the custodians of parliamentary oversight towards upholding national security in collaboration with members of the national executive. For instance, in the UK, intelligence based questions are tabled in parliament for as long as the questions do not seek information about individuals and/or intelligence operations. According to the DA’s Shadow Minister of State Security Mr Dirk Stubbe (2015a), since the formation of

214 the JSCI and the enactment of the Intelligence Services Oversight Act 40 of 1994, the annual reports of the committee have always been tabled late in direct contravention of the country’s legislative and regulatory prescriptions.

In fact, Respondent 1 of the external experts indicated that, “since the resumption of power of the fourth and the fifth administrations, the tabling of the JSCI’s reports to the NA and NCOP had always been late after which most of the issues would have been already resolved”. However, this late reporting revives old controversies and invites undue criticisms on members of the intelligence services as the national debate ensued on obsolete matters being discussed out of context. Therefore, consistent tabling and debating of the committee’s reports in the NA and NCOP would serve to contextualise issues, report findings and observations. This move would demonstrate to the entire parliament that the JSCI is serious about intelligence oversight.

According to Stubbe (2015a), a democratic dispensation should not have ‘no go’ zones for the legislature as its involvement gives legitimacy and direct democratic accountability. Therefore, the work of the JSCI should garner public confidence by ensuring that members of the intelligence services serve the state as a whole and protect the Constitution rather than narrow political or sectorial interests. Given the continuous non-tabling of their oversight reports to the NA and NCOP, some of the intelligence oversight structures were discouraged to put in more effort particularly in terms of cooperating with other structures (Reflections of Respondent 1 of the external experts).

6.2.1.3 Undoing the JSCI model for a select or portfolio committee system

According to the DA’s shadow Minister of State Security Mr Dirk Stubbe (2015b:13), “the JSCI has over the years failed to achieve the higher levels of legitimacy given its continued placement of party political interests above public interest under the guise of

215 national security imperatives”. During the interviews, Respondent 3 of the external experts started questioning the utilisation of the joint standing committee model instead of select or portfolio committee approaches. In this regard, he associated the inefficiencies of the JSCI to its operational model which restricts frank discussions on matters of national security. In addition, Respondent 4 of the external experts indicated that, “the best operational model that has proven effective based on the international best practice is the select or portfolio committee model”. The portfolio committee system is regarded as the best approach and is used by other security cluster committees such as that of the police and/or defence. Respondent 13 of the oversight officers defended the JSCI model as an all-inclusive approach wherein the police, military and civilian intelligence structures, accounts for their intelligence operations under a classified platform.

On the other hand, Respondent 7 of the external experts indicated that, “switching to portfolio approach will help parliamentarians from the pains of establishing ad hoc committees for processing pieces of intelligence related legislation”. This approach will streamline the work of parliament on matters of intelligence oversight as members of the NA will get to specialise and learn about how best to handle intelligence oversight while balancing it with the needed transparency (openness). With the select or portfolio committee systems, NCOP will be able to invite either the Minister or Deputy Minister of State Security to discuss national security matters in cases where there are provincial implications. To strengthen their argument, Respondent 7 of the external experts indicated that the portfolio committee approach will facilitate intelligence based debates towards addressing issues of provincial concern on a non-partisan basis. This point was supplemented by Respondent 1 of the external experts who indicated that, “such a move would assist the government in driving its national and provincial debates and consensus on balancing national security from national interest”.

216

In addition, Respondent 11 of the oversight officers supported the adoption of the select or portfolio committee system and indicated that the suggestion is also informed by the fact that intelligence is a national competence. Respondent 11 further argued that this move would serve as confidence building measures on parliamentary intelligence oversight in the eyes of members of the public. Judging from the collected empirical data, it became clear that most of the interviewees supported the idea of moving towards the select or portfolio system of the parliamentary committee for intelligence oversight. Based on the discussion thus far, it is also argued that a select or portfolio committee will cooperate with ease with these various other intelligence oversight structures.

In his article on the oversight mechanism for members of the intelligence services, Mr Stubble (2005a) indicated that, “if one was to grade South Africa in terms of the strength of its intelligence accountability and oversight mechanisms, one would struggle to give it a passing grade”. A similar concern was raised by Dlomo (2004) about the effectiveness of parliamentary intelligence oversight mechanisms. He discovered poor levels of interaction between the JSCI and the IGI when doing their oversight work, hence the suggestion to reshape the joint standing committee approach to a select or portfolio committee mechanism.

6.2.2 Refocusing the Ministry of State Security

As indicated earlier, the issue of refocusing the Ministry of State Security was raised by two (7%) of the oversight officers and one (14%) of the external experts as one of the critical structures whose mandate, roles and functions should be reviewed to enhance its cooperation with other oversight structures. According to its website (2016), the Ministry of State Security is responsible for supporting its Minister and the Deputy Minister in their superintendence and control of members of the intelligence services. As already discussed, section 2(3) of the Intelligence Services Act of 1994 empowers

217 the Minister to oversee and control the functioning of state security apparatus to advance national security imperatives. As a result, the efficacy of the ministry in the fulfilment of these functions requires the development of sound internal control measures to be implemented in concurrence with the JSCI. However, as the interviews unfolded Respondent 4 of the external experts started questioning the rationale and the relevance of having the Ministry of State Security in the country.

During the interviews, the most prominent suggestion on refocusing the Ministry of State Security borders on the de-establishment of the ministry and the allocation of intelligence services to the Presidency or the Ministry of Home Affairs after its renaming to the Ministry of Interior. Respondent 17 of the oversight officers was of the view that intelligence tradecraft does not require direct political guidance while Respondent 12 of the oversight officers felt the opposite. The proponents of the former assertion indicated that political principals more often than not orchestrate the abuse of state machinery for personal political ends. Those who support the retaining of the ministry indicated that moving intelligence services to the presidency or even Ministry of Interior would not exonerate the respective principals from abusing the state machinery for personal and/or political ends.

Interestingly, Respondent 3 of the external experts juxtaposed the significance of NICOC in supporting the Ministry of State Security through coordinated strategic intelligence pictures vis-à-vis the absence of such mechanism in the oversight space. As part of the suggestions on refocusing the ministry, Respondent 9 of the oversight officers bemoaned the overstretched nature of the ministry staff and thus suggested undertaking concerted efforts to capacitate the office to be able to fully discharge its control function on members of the intelligence services. Being the custodian of national security, the Minister of State Security is further empowered by law to engage the entire public service towards addressing national security imperatives. Therefore, matters of capacity remain very critical. In fact, the ministry should be sufficiently

218 staffed to be able to discharge its administrative functions and bolster effective internal control mechanisms on members of the intelligence services. As the executive overseer and primary controller of the work of the intelligence services, the Minister of State Security has the responsibility to ensure that staff integrity remains at the highest point at all times.

Given the fragmentation of the intelligence oversight structures as reflected in Chapter 4 in section 4.2, the Minister of State Security together with the JSCI should be demanding coordinated reports from the oversight structures. Respondent 8 of the oversight officers indicated that, “the Minister should be able to encourage the various intelligence oversight structures to engage each other in overseeing members of the intelligence services”. This move would encourage joint planning, synergise efforts, avoid duplication and create the culture of complementarity when discharging varied legislative frameworks. Inasmuch as the Minister assumes accountability for the production and distribution of integrated strategic intelligence assessments coordinated through NICOC, he should also assume executive responsibility for integrated oversight reports through the IGI in which the overall reporting and the issuance of compliance certificates is done. To amplify the point, Respondent 5 of the external experts emphasised the need for the Minister of State Security to continuously engage with the Minister of Police and the Minister of Defence under the auspices of the Inter-Ministerial Security Committee (IMSC) on intelligence and oversight matters. This assertion is premised on the fact that both DI and CI reside under the superintendence of the Ministers of Defence and the Police respectively (Intelligence Services Act: 1994).

However, Respondent 21 of the oversight officers cautioned that the refocused Ministry of State Security should be able to exercise restraint when dealing with members of the chapter 9 institutions. This is predicated on the legislative imperatives empowering such institutions to discharge their functions without fear, favour or prejudice (RSA Constitution: 1996; Public Protector Act of 1994; Human Rights Act of 1994; Public

219

Audit Act of 2004). During the interviews, there was general understanding that the ability of the government to effectively manage oversight work in a collaborative and coordinated manner would advance combined assurance and inspire public confidence in the work of the intelligence services. Once refocused, the Ministry of State Security would be able to empower intelligence oversight officers, thus enabling them to exude confidence and become proactive in dealing with abuses of national security apparatus (OIGI: 2006).

6.2.3 Refocused IGI and other structures

As part of reshaping the intelligence oversight structures, the need for refocusing the IGI was raised by 17 (61%) of the oversight officers and three (43%) of the external experts. However, most of the discussions and the subsequent suggestions focused on the OIGI’s operational modalities on intelligence oversight processes. As already reflected, the OIGI signed, or is about to sign, MoUs with various other intelligence oversight structures. Up to this point, it was indicated that MoUs are already signed with the Office of the PP and the AG while engagements are underway to finalise same with the appointed judge. This is mainly done for the purposes of advancing combined assurance. During the interviews, both Respondents 4 and 7 of the external experts cited the signing of these MoUs as a clear indication of the need for the cooperation of intelligence oversight structures in doing their work. As a result, the suggestion for refocusing the IGI remains critical.

Firstly, Respondent 21 of the oversight officers emphasised that the IG should also be reviewing how the intelligence services operationalise the NIPs. To support the assertion, Respondent 12 of the oversight officers indicated that, “it is part of the oversight responsibilities of the IG to check the alignment of the services’ identified projects and determine whether the projects addresses the country’s national security priorities as expressed in the NIPs”. This approach would complement the

220 government’s outcomes model, optimise relationships and promote the alignment of the oversight cycle of the IG to the AG’s auditing timelines. In other words, the IG’s operational model should be proactive and not be too focused on historical issues of concern. When properly applied, the respondents wish for intelligence oversight to reach a stage wherein the reliance of one structure on the other is placed at the centre of the oversight process.

Secondly, Respondent 1 of the external experts prefers the refocused IGI to be able to advance the collaborative, coordinated and combined intelligence oversight approach with systematic and well thought out engagements. This will simplify the parliamentary NA. This approach will enable the JSCI to champion its annual report including that of the intelligence services, the certificates from the IGI and the audit reports from the AG, as a single business process. As a result, each of the role players would know when to deliver and table their respective reports to the NA and NCOP. This move will demonstrate the intensive nature of the intelligence oversight mechanism which cannot be manipulated.

In this regard, Respondent 8 of the oversight officers indicated that a refocused IGI would restore the public confidence in the country’s intelligence services and the broader intelligence oversight model. Although parliament plays a leading role in monitoring budgetary expenditure through the AG’s office, the inputs of civil society and the media are essential. As Gill (2007) noted, it is important for members of civil society and the media to monitor the allocation of intelligence budgets and how such monies are spent in order to hold them accountable.

6.2.4 Strengthen the role of civil society and the media

Of the 28 interviewees, three (11%) of the oversight officers cited the need for strengthening civil society and members of the media as one of the key non-statutory

221 intelligence oversight instruments. This was further cited by two (29%) of the external experts as one of the components to be strengthened to enhance its cooperation with other intelligence oversight structures. This is predicated on the fact that the operational trajectory of the intelligence services deserves to be overseen by the cooperating multiple stakeholders. Various respondents indicated that civil society can report their discoveries through whistle-blowing channels while members of the media conduct public debates on intelligence-related matters. In this case, civil society is conceptualised by Cole et al (2008) as the other centre of power, consisting of aggregate non-governmental organisations, people and institutions which advance the shared interests, purposes and values of the communities. Cole et al (2008) indicated that civil society includes the manner in which people associate with one another in order to manage their lives, voice opinions, pursue interests and exchange information.

During the engagement, all seven (100%) external experts supported the involvement of members of the public in overseeing members of the intelligence services in order to improve the accountability and transparency of public officials. However, Respondent 22 of the oversight structures cautioned that the contribution of civil society can only be achieved when they possess some kind of knowledge and expertise on the workings of members of the intelligence services. As a result, he emphasised the need for the intelligence services to educate members of the public on the nature of the work they do and the possible areas for abuse.

As Cole et al (2008) put it, civil society organisations have the ability to form an agenda and put pressure on the government to ensure the speedy resolution of the issue at hand. However, of serious concern is the tendency of such organisations dissolving or becoming incapacitated once the issue has been addressed. For instance, the Treatment Action Campaign (TAC) championed the distribution of anti-retroviral (ARV) drugs to people living with Human Immunodeficiency Virus/Acquired Immune Deficiency Syndrome (HIV/AIDS) and once the government started rolling out the

222

ARVs, the organisation started losing most of its funding and had to change focus. Although this is understandable, civil society should create resilient organisations with the ability to address a multiplicity of issues at the same time. Despite these realities, most of the respondents still supported the involvement of civil society in overseeing members of the intelligence services. This should be done for ensuring the responsiveness, transparency and compliance of intelligence services to the country’s legislative framework. Further, as the White Paper on Intelligence of 1994 professes the participation of civil society will contribute towards strengthening the policies governing intelligence oversight.

According to Hillebrand (2012), the media serve as the transmitters and scrutinisers of information on government, private and individual activities on any important matter. As such, members of the intelligence services are no exception. As Hillebrand (2012) puts it, the media has the ability to draw public and political attention to human rights infringements, potential abuse of power and lack of accountability by members of the intelligence services. For instance, it was a newspaper article which revealed the circulation of e-mails within the then NIA outlining the possible conspiracy against the then ANC deputy president, Mr Jacob Zuma. Further, the so-called Browse ‘Mole’ Report was also revealed by the media implicating the former ANC deputy president, Mr Jacob Zuma, Congress of South African Trade Unions (COSATU), the South African Communist Party (SACP), the ANC Youth League and the Youth Communist League, for planning a street-based revolution to oust the then president of the ANC and of the country, Mr Thabo Mbeki. This was purported to have been authored from within the NPA.

It was the media which revealed that the NIA was caught conducting surveillance on Mr Saki Macozoma and his family. The complaint was later reported to the then Minister of Intelligence, Mr Ronnie Kasrils for intervention (OIGI: 2006). It was these kinds of publications in the media which raised public interest in the quality of intelligence

223 oversight in South Africa. In his writing on ‘shock theory’, Lock Johnson argued in Hillebrand (2012) that, “law-makers rarely engage in intensive oversight, unless a major scandal or failure forces them to pay attention to the dark side of government”.

The publication of the findings of the Khampepe Commission of enquiry which revealed the NPA’s pseudo-intelligence operations which were conducted by the former Scorpions without any intelligence oversight; generated public interest on intelligence oversight discipline (JCPS Report, 2007:2). Further, Marina Caparini in Den Boer (2012:4) argued that, “the media constitutes an inter-connective medium linking individuals and groups with government and play a critical role in conveying information about shifting public opinion and policy preference”. In fact, it is through press freedom that members of the public are informed to be able to hold their government representatives accountable for the non-delivery of public services as promised. In many cases, it is investigative journalists who bring the inappropriate activities of the intelligence services to the surface for the government and members of the public to scrutinise.

Although Hillebrand (2012) acknowledges the role of the media in intelligence oversight, he also confesses that the ability of the media to scrutinise intelligence requires of them to have some basic knowledge of intelligence tradecraft and its oversight. However, he agrees that if the formal intelligence oversight bodies are unwilling, or incapable of conducting robust intelligence oversight, the media coverage will be able to put pressure on oversight structures to do well (Hillebrand: 2012). The immediate problem associated with this approach is that members of the opposition parties or some factions can use the media to advance their selfish political goals. In addition to reshaping the key intelligence oversight structures, Respondent 26 of the oversight officers cited the need to improve the operation cooperation modalities of the intelligence oversight structures and the following discussion focuses on that matter.

224

6.3 IMPROVING THE OPERATIONAL COOPERATION OF INTELLIGENCE OVERSIGHT STRUCTURES

In his writing on the improvement of cooperation among police agencies, Martin (2004) argues that, the cooperation of the various policing agencies should embrace highly technological approaches which will assist governments in addressing current policing challenges. This is crucial for facilitating the shift from a traditional law enforcement model to contemporary approaches in dealing with people’s safety concerns (Martin: 2004). Similarly, the development of the systematic cooperation of intelligence oversight structures should pay special attention to the emergence of newer approaches of oversight by revelation. For instance, the emergence of new revelations such as WikiLeaks and Afrileaks which were characterised by the leakage of intelligence documents to members of the public requires new interventions.

According to Gill (2012:18), “the intelligence oversight structures should be able to engage all intelligence services in order to hold all responsible entities and/or individuals accountable”. These developments challenged the conventional oversight model and enjoined structures to consider the emergence of both official and unofficial revelations. Despite the sources, oversight structures should be able to proactively address all kinds of data that comes their way. In the 21st century, the systematic cooperation of intelligence oversight structures can be achieved when various structures foster greater cooperation with each other (Gill 2012:18). These include the creation of wider interest among community members, researchers, academics, journalists and wider civil society on the intelligence oversight subject.

In addition to the identification of critical structures requiring review, I also asked both oversight officers and the external experts to present suggestions on what can be done to improve the cooperation of various intelligence oversight structures. In response, 13 (46%) of the oversight officers cited the need to empower the oversight officers

225 themselves, while six (21%) cited the need to streamline intelligence oversight legislation, five (17%) cited the need to improve trust among structures, three (10%) cited the need to balance secrecy with transparency and two (7%) cited the need to streamline the operational regulation directing issues of intelligence oversight. In terms of the external experts, two (29%) cited the need to streamline legislation while another two (29%) cited the need to streamline regulation. Another one (14%) suggested the empowerment of officers, one (14%) cited balancing secrecy with transparency and one (14%) cited the need to improve trust among structures. Figure 6.3 presents the summary of findings in response to the above-mentioned question.

Figure 6.3: Tabulation of suggestions on what needs to be done to improve the cooperation of the various intelligence oversight structures

The following discussion unpacks the suggested interventions in detail.

226

6.3.1 Streamline the statutory and regulatory frameworks

During the interviews, the need for streamlining the country’s oversight statutory framework was cited by 21% of the oversight officers and 29% of the external experts as one of the essential steps towards improving the operational cooperation of intelligence oversight structures. A similar view was earlier expressed by Gill (2012) who indicated that, the intelligence oversight structures should be able to use statutory gateways in order to collaboratively address enquiries at hand. This is the important approach to facilitate the effective achievement of the various legislative mandates allocated to the various intelligence oversight structures. According to Den Boer (2012), the legislative mandates of intelligence oversight structures remain effective when overlapping each other instead of being just complementary. This is an important approach as no single oversight structure could be generally sufficient to oversee members of the intelligence services.

As already expressed by scholars like Caparini (2002), Lowenthal (2003), Dlomo (2004), it is always important to ensure the allocation of a wider mandate to respective oversight structures as this serves to avoid linearity which could result in an imperfect oversight model. Therefore, the rationale for the cooperation of intelligence oversight structures arises in support of this approach. Respondent 1 of the external experts indicated that, “given the apparent non-cooperation of intelligence oversight structures with one another there is a need to review the existing intelligence oversight Act and insert a section that will force these structures to cooperate with each other”. She further highlighted the need for the inclusion of the FIC in the list of intelligence services to be overseen by the various intelligence oversight structures.

To illustrate the need for streamlining the statutory framework, Respondent 1 further indicated that given the mandate and level of access afforded to the IG’s office by the intelligence services, the AG can use the IG’s work to support its audit opinion.

227

Therefore, it is essential for the IG’s office to provide the necessary assistance to the AG for ensuring the satisfactory auditing of the intelligence expenditure associated with covert operations. Concerning the JSCI and the AG, both institutions signed a MoU in order to enhance their operational cooperation in their oversight work over the intelligence services. The MoU has started bearing fruit. For instance, “at the invitation of the JSCI, (the AG) has attended budget (presentations) by the intelligence services to the Committee, briefed the Committee on the audit report findings, served as an expert witness when the services have met with the JSCI to discuss the audit reports”, and also investigated specific irregularities.

As already discussed, the AG investigated the escalation of expenditure on the construction of the new COMSEC building at Musanda (Parliamentary Monitoring Group: 2010). However, their investigation could not avail much given their isolation and frustration in attempting to solicit relevant information on the matter. In this regard, Respondent 1 suggested that, “inasmuchas there is the NICOC to coordinate the strategic intelligence picture by members of the intelligence services, there is a need to have a similar structure for coordinating the work of the intelligence oversight structures”. In the interim, the various heads of the respective structures can rotate in their chairing of such meetings on a quarterly basis as they seek to coordinate their efforts towards ensuring the effective oversight of members of the intelligence services.

Although, section 7(7) (c) of the Intelligence Services Oversight Act 40 of 1994, empowers the PP to refer intelligence-related complaints to the IG, this is not enough to ensure the effective cooperation of these structures in doing their work. Despite the signing of the MoU between Advocate Madonsela (PP) and Advocate Radebe (IG) on 12 November 2010, there is a need for the legislative provision to guide the cooperation modalities of these structures. However, this assertion does not discount the fact that the signed MoU covers the key issue of case referral mechanism. Furthermore, the MoU was extended to include the sharing of information, best practice, and undertaking

228 joint outreach programmes. From this experience, it is clear that complaints can be misplaced, hence the significance of institutional cooperation in the implementation of oversight and control. All this is done in order to easily direct complaints to the relevant structure. Better oversight will improve results in all intelligence structures.

However, Respondent 18 of the oversight officers cautioned that any move to streamline the oversight legislative framework should not compromise the quality of the control and oversight work on the intelligence services. On the other hand, Dube (2013) indicated that for as long as the intelligence services continue to enjoy the right to secrecy coupled with the right to invade people’s privacy, the need for intelligence oversight will remain relevant. As a result, Caparini (2002:67) concluded that:

“Given the power and the inherent risk of the abuse of power by the intelligence services, they should remain subject to extensive controls and rigorous oversight by the elected and duly appointed civil authorities”.

Based on the discussions above, it is clear that streamlining oversight legislation will force members of the various intelligence oversight structures to cooperate with each other when overseeing members of the intelligence services. This is crucial for their contribution towards the improvement of the realisation of good governance results.

In addition to streamlining the legislative framework, two (7%) of the oversight officers and two (29%) of the external experts cited the need to streamline the regulatory framework. According to Africa (2009: 45), there is no evidence which suggests that access to information alone can result in significant shifts in power relations and the ability to hold someone accountable. Therefore, any consideration of remedies aimed at improving the cooperation of various intelligence oversight structures should include the drafting of concerted regulatory framework for guiding the operational engagement of the various structures with each other in completion of the refined legislative

229 framework. To support this point, Respondent 9 of the oversight officers emphasised that, “the absence of clear regulatory framework makes it difficult for the oversight officers to engage their counter parts and freely share information as required”. In his speech at the launch of the new archives building at Musanda, the then Minister of Intelligence, Mr Kasrils (27/05/2005), quoted the ancient Roman maxim ‘Quis custodiet ipsos custodies?’ Which means, who will guard the guardians? This maxim can also be interpreted to mean, who will supervise the spies?

Based on this maxim, it can be argued that the OIGI and all other oversight structures have been established to guard the guardians and supervise the spies. However, there is common understanding that the OIGI is not able to properly hold the entire membership of the intelligence services accountable without the help of other structures sharing a similar oversight mandate.

Further, Africa (2009:20) indicates that, “any intelligence oversight mechanism has to ensure that the classification and declassification of information is done in the public interest”. In practical terms, for transparency to be given meaningful effect there should be fully-weighted countervailing systems in place, including strengthened oversight mechanisms. In order to ensure the relative balance between transparency and secrecy, the government should consider the establishment of an Independent Review Mechanism. This would be done in order to avoid a situation where the person who makes a decision to classify information becomes the same person who determines the appeal against such a decision. This is similar to the arrangement in which the intelligence oversight officers tend to rely on the information supplied by the people they are meant to oversee. This anomaly is already happening with some requests to access information through PAIA, leaving freedom of information advocates to conclude that the PAIA-related internal appeals procedures are superfluous.

230

Despite the expansion of the OIGI mandate, civil society groups have proposed the establishment of the Information Commissioner or Regulator as the appeals authority in terms of POSIB. However, during deliberations, the Ad Hoc Committee and the government proposed the establishment of an Independent Board which would report to the Minister of State Security (Parliamentary Monitoring Group: 2010). At the same time, the ANC has tabled an amendment that would authorise the Minister of State Security to establish a mechanism to assist him with both the review and the appeals process.

Effectively, this could suggest a Committee of Advisers on Appeals. Otherwise, all 35 (100%) interviewed respondents concur that the absence of regulatory guidance on their cooperation makes it difficult to solicit the necessary finances to fund certain collaborative operations with other sister oversight structures. Irrespective of what might be agreed upon, the ultimate goal is that there should be an oversight mechanism which would ensure that information classification is done in the interest of the nation as a whole. Ultimately, the oversight mechanisms are necessary preconditions to ensure that intelligence is always conducted in the best interests of democracy. It should however be noted that the suggestion to streamline the regulatory framework on the cooperation of intelligence oversight structures is informed by the fact that pieces of legislation require the development of regulations to ensure their successful implementation. Therefore, in this case the need to ensure that the streamlining of both the statutory and regulatory frameworks is highly interlinked and needs to complement each other.

6.3.2 Empower intelligence oversight officers

As part of deliberations to improve the systematic cooperation of intelligence oversight structures, 13 (46%) of the oversight officers and one (14%) of the external experts cited the need to empower the intelligence oversight officers themselves. As part of the

231 international benchmarking exercise it became clear that the most common way of empowering human resources for the improvement of any systems is through training. According to Cole et al (2008), the most fundamental commitment that is required of any organisation is its willingness and ability to train its staff complement to ensure their adaptability to change. This will facilitate greater understanding among the members of the intelligence oversight structures and reinforce their legitimacy and credibility as the custodians of the vibrant oversight mechanism. As Gill (2012) puts it, each and every responsible government should make it their virtue of necessity to encourage all forms of oversight and the cooperation of such structures in order to counter any possible disregard for the country’s legislative framework. As Respondent 3 of the external experts indicated, the need for this virtue was confirmed by the emergence of the global WikiLeaks which challenged the manner in which conventional intelligence tradecraft was being practiced.

According to Cole et al (2008:56), “developing the capacity to assert the primacy of human rights, often in difficult circumstances, is a particular challenge in some transition and many post authoritarian or post conflict countries”. Following the discussion on the need for legislative and regulatory guidance for the cooperation of intelligence oversight structures, a sizable number of respondents (as indicated earlier) agreed on the fact that oversight officers should be adequately trained to be able to professionally effect such engagement with the ultimate intention of holding people accountable. This was also emphasised by Jackson and Brown (2007) who indicated that, the ability to improve the cooperation of intelligence oversight structures demands the continuous training of personnel with clear emphasis on those cooperation modalities. Further, oversight officers must actively seek to acquire more skills to improve their stakeholder management capacities. Without such skills, oversight officers would risk being marginalised and unable to promote their legislative mandate and affect appropriate change. Given the intention of this study to enhance the cooperation of the intelligence oversight structures, the need for the capacitation of the oversight officers remains

232 paramount. Respondent 1 of the oversight officers indicated that, at the current juncture there are no explicit courses, modules or programmes that are run by the IA or any other institution to train the intelligence oversight officers on intelligence oversight let alone the cooperation modalities of the oversight structures.

Therefore, the move towards the improvement of the cooperation of the intelligence oversight officers requires the development of modules to empower the officers to enhance their operational cooperation with each other. Respondent 19 of the oversight officers indicated that, “special symposiums, conferences, seminars and information sharing gatherings can be organised among intelligence oversight structures from the various structures to discuss best practices and lessons learned”.

6.3.3 Balance secrecy with the need for transparency

As part of the interventions to improve cooperation among oversight structures, the need to balance secrecy and transparency was raised by three (10%) of the oversight officers and one (14%) of the external experts. However, it has been observed in the recent past that despite the provisions of the statutory frameworks, tenets of good governance, and global move to balance the secrecy of intelligence tradecraft with the requirement for transparency, various democratic states still confront a wide range of competing pressures (Levy: 2004). It is on the basis of these pressures that critical questions emerge which require an objective consideration in order to reconcile the two concepts. In this regard, Levy (2004:34) asks the following question: how does one preserve secrecy without undermining the essential tenets of good governance and democracy? However, the converse question is equally valid, and that is: how does one preserve democracy and human rights without undermining national security? In this case, a most vexing problem emerges: what should be done when releasing confidential information harms national security but promotes good governance?

233

According to the White Paper on Intelligence of 1994, the Sub-Committee of the Transitional Executive Council (TEC) on Intelligence argued that the principle of balancing secrecy and transparency is one of the bases for transforming and democratising the intelligence services. However, 21 years after the transition the question of balancing secrecy and transparency is still a concern. This was illustrated by the latest sentiments expressed by members of the JSCI in their Annual Report (2015/2016:14) when they indicated that:

“efforts should be made to find a balance between the need for accountability and good financial governance and access to information that has national security and intelligence implications”.

In this context, various democratic countries confront the paradox that agencies that are established to ensure national security might in fact undermine it as well as democracy. Considering the latter assertion, Africa (2009:15) asks the following critical questions:

“How should this paradox be managed? What powers should the intelligence services have? What limits should be placed on those powers? How can the abuse of these powers be prevented in conditions of secrecy? Is there too much secrecy? Furthermore, should the intelligence services monitor politicians and political parties which threaten violence? In what circumstances and subject to what safeguards can the services infringe on constitutional rights in the interests of national security? What is meant by national security? How should national security and intelligence priorities be determined?”

In this regard, Kocaoglu and Figari (2006:3) enquire whether the promulgation of the transparency laws narrows opportunity for mismanagement and the diversion of public funds, and should such laws permit public scrutiny of budgets and administrative decision-making? However, South Africa has already legislated on these matters, but

234 the competing pressures persist. It is to address these realities that this study advocates the advancement of systematic cooperation of the oversight structures to address these questions. When engaging the respondents, a sizable number of respondents concur that the question of balancing secrecy with transparency is critical for the simplification of the cooperation of the intelligence oversight structures when doing their oversight work.

Furthermore, Stone (2008:3) gave the full account of a situation and gave the following narration in terms of posed questions:

“Suppose a government official conducts a study on the effectiveness of the security measures at various national nuclear power plants and then concludes that most of the plants are vulnerable to terrorist attack. In this case, the critical question is, should this study be kept secret or should it be disclosed to the public? The reality is that publishing the report might endanger the nation by revealing its vulnerability to terrorists; while on the other hand, its publication might alert the public to the situation and thus enable them to hold accountable the government officials who have failed to keep the nation safe” (Stone: 2008:3).

In this regard, it is clear that the publication of such information could either harm and/or benefit good governance. Therefore, the critical question remains: should the study be made public or not? In an attempt to answer the latter question, it should be determined whether the benefits of transparency are more than the costs. In other words, the question is whether the benefits for democracy outweigh the costs for national security? (Stone: 2008:3). A follow-up is who answers the first question? Senior intelligence bureaucrats with vested interests in the outcome of the decision, junior officers worried about irregular motives of their bosses, the media who already has the information, or judges as part of their judicial functions (Stone: 2008:3).

235

Considering all these questions, it can be argued that it is an objective response to these questions which would allow for effective management of the competing pressures. In this regard, “it is important to note that all decisions involve both risk and uncertainty and the best solutions grow from the effective evaluation of many alternatives against defined criteria” (Stone: 2008:3). Objective and transparent criteria for such decisions are crucial. Therefore, in order to enhance an objective response to the above-mentioned questions, the authorities should consider the following criteria as suggested by Stone (2008:59):

 “determine whether the information constitutes a state secret whose publication would constitute a threat to national security;  determine whether the publication of such information would not violate human rights and the general social fabric;  determine whether the non-publication of such information would have a negative impact on the public interest; and  determine whether the non-publication of the information would compromise the principles of transparency and accountability”.

The primary aim of these criteria is to enable the authorities to assess information systematically and rigorously before taking action. Alvin Toffler provided the following suggestions contained in Brin (1998:203):

“Though I lean toward a libertarian view on the free availability of information, I cannot simply brush aside the concerns of the FBI and Interpol that the inability to control the spread of certain information can lead to the mass murder of innocent people. The stakes are larger than they have ever been. It is one thing to say we don’t like big government and we don’t want Big Brother looking over our shoulder. But do we like the alternative of a world where the most ruthless and most

236

depraved can gain access to the means of mass destruction, apply them and get away with it?”

In this regard, there is no doubt that every sensible person needs some element of security and a peaceful world. Following these guidelines, one can reduce the potential for abuse. Despite calls to balance secrecy with the need for transparency, it should be noted that there are sacred foundations of intelligence tradecraft requiring protection at all times. This is predicated on the fact that intelligence secrecy will always be necessary for as long as nation states exist. However, there are aspects of intelligence work which are not necessarily secret and those can be discussed in the public space. But, the craft itself should remain secret and protected at all times. For instance, during World War II there were reams of documents containing intelligence methods and tactics which were only declassified and released 60 years later.

Therefore, the ability to balance secrecy and transparency has the potential to facilitate the cooperation of the various intelligence oversight structures. As already discussed, the fact that various oversight structures handle various types of information that is differently classified has a negative impact on the level of cooperation among the various structures.

6.3.4 Improvement of trust among intelligence oversight structures

To improve the cooperation of intelligence oversight structures, five (17%) of the oversight officers and one (14%) of the external experts proposed the development of mechanisms to improve levels of trust among intelligence oversight structures. As earlier expressed by Schwarz Jr (2007) he indicated that, the effective cooperation of intelligence oversight structures can only be achieved when the relevant structures exchange coherent data on a range of intelligence work. This can only be achievable through an open information sharing mechanism that is grounded on the foundational

237 principle of trust. However, the ability to possess wide ranging data is only possible through the effective cooperation of the various oversight structures with varied legislative mandates. This is supported by Carter (2003) who argued that “interagency cooperation and information sharing are the cornerstones” of successful intelligence management.

This is further supported by research which discovered that an absence of such joint activities led to the government’s failure to pre-empt a number of disasters (Jackson and Brown: 2007). The failure of the intelligence oversight structures to cooperate with each other during their intelligence oversight work fragments their effort thereby allowing members of the intelligence services to do as they please.

Similar findings were made by the US 9/11 Commission which indicated that if the cooperation and information among the law enforcement agencies had been better, 9/11 could have been avoided (Jackson and Brown: 2007). As described by Desai (2005), the absence of interagency or inter-structure cooperation is mainly caused by lack of trust or common culture among the various agencies. Therefore, it is important to work towards the improvement of inherent trust among the various intelligence oversight structures. This is critical for the improvement of cooperation among the various structures. In order to improve the level of trust among members of the oversight structures, concerted effort is required to demystify intelligence tradecraft and this has remained a primary concern of the various democratic governments. Although in South Africa much has been done to address this phenomenon, considerable challenges still persist. Excessive secrecy and poorly resourced oversight mechanisms complicate the entire trajectory.

Therefore, the unequivocal adoption of transparency, openness and accountability remains the most effective demystification mechanism of intelligence tradecraft. This demystification agenda could be further enhanced by the adoption of a complaints

238 helpline under the management of the IG. However, literature indicates that the issue of trust remains elusive as it cannot be imposed but should be earned and this is similar to the issue of respect.

6.4 SUMMARY AND CONCLUSIONS

Based on the above analysis and assessment, it is clear that a number of interventions can be undertaken to improve the cooperation of intelligence oversight structures during the application of their varied legislative mandates. During the research, it became clear that a concerted effort is required to improve the capacity of the intelligence oversight structures to enable them to conduct their oversight work without fear or favour. This is crucial given the pressure to improve their intelligence oversight approach. In this regard, it is important to entrench the culture of coordination and cooperation among the various intelligence oversight structures. If not, the guardians will remain blind and continue to operate in silos, thus failing to have a positive impact on governance results. During the interview process, it became apparent that the realisation of a coherent intelligence oversight mechanism would demand the participation of reshaped intelligence oversight structures. These include refocusing of the JSCI, the Ministry of State Security, the IGI and strengthening the role of civil society and the media.

Considering the outlined barriers for the effective cooperation of oversight structures, this chapter highlighted the need for the improvement of the operational intelligence oversight mechanism. This includes the streamlining of the statutory and regulatory frameworks. In any democratic dispensation, there is a need to ensure the creation of coherent legislative and regulatory mechanisms for regulating the work of government and the private sector. In any work environment, the value and significance of human resources cannot be overemphasised. For any change to be realised there is an inherent need for the empowerment of human capital through addressing the

239 complexities of change management. Further, the discussion emphasised the need to balance secrecy as the tenet of intelligence with the requirement for transparency as one of the key tenets of good governance. Lastly, the previous discussion emphasised the need for the intelligence oversight structures to trust each other in order to achieve cooperation and facilitate information sharing.

240

CHAPTER 7 PROPOSED IMPROVED COOPERATION MODEL FOR INTELLIGENCE OVERSIGHT STRUCTURES IN SOUTH AFRICA

7.1 INTRODUCTION

While the preceding chapters presented data and findings in response to the three research questions, this chapter integrates the implications of the findings and assessments so far in the form of a more formal cooperation model for intelligence oversight structures in South Africa, and summarises the findings, conclusions and recommendations of the research project.

It is important to contextualise this chapter against the background of the submission of Brigadier Cronje, a former member of the apartheid security forces, to the Truth and Reconciliation Commission in 1996. He indicated that during the apartheid years members of the security forces were allowed to act outside the confines of the legal (law) system for as long as they were able to protect the state, the government and minority white population (Submission to the Truth and Reconciliation Commission (TRC): 1996).

During apartheid times, the security services could do anything to protect the apartheid system with no input from any of the then oversight structures. However, in the new post-apartheid democratic dispensation the emergence of the White Paper on Intelligence (1994:13) pronounced a totally new dispensation for the security services. It emphasised that the new “democratic dispensation seeks to establish an effective, integrated and responsive intelligence machinery that can serve the Constitution and the government of the day through the provision of relevant, credible and reliable intelligence”. These provisions still remain the founding principles of the new intelligence dispensation in the post-apartheid democratic South Africa. During the

241 transition, there was explicit reference to the continued need for intelligence tradecraft, its oversight and the need to balance secrecy with the democratic requirements for transparency, accountability and good governance.

Ironically, the White Paper on Intelligence (1994) did not mention the need for the cooperation of intelligence oversight structures but instead emphasised the cooperation of intelligence services, hence the creation of NICOC. The non-articulation of the need for the cooperation of intelligence oversight structures persisted despite the creation of multiple oversight structures with varied legislative mandates.

The main rationale for a more systematic cooperation model is the improvement of the working modalities of the intelligence oversight structures in conformance with the provisions of the White Paper on Transforming Public Service Delivery (1997:6) (also known as the Batho Pele White Paper). It indicates that, “the government is committed to continually improve the lives of the people of South Africa by a transformed public service which is representative, coherent, transparent, efficient, effective, accountable and responsive to the needs of all through the creation of a people-centred and people- driven public service, which is characterised by equity, quality, time and a strong code of ethics”. The realisation of this commitment requires the support of a balanced Intelligence functionality in government that is overseen and controlled by better resourced and functioning intelligence oversight structures within a clear and coherent cooperation model.

A primary purpose of this research is therefore to improve the cooperation of the intelligence oversight structures in order to enhance their contribution towards the promotion of good governance. To achieve this, the research critically analysed and assessed the state of cooperation among intelligence oversight structures to gain better insight into the nature of the barriers they face in order to decide what can be done to improve the cooperation and eventually hopefully the results of these structures.

242

This concluding chapter starts by providing a synopsis of the preceding six chapters in order to highlight the main conclusions from each chapter. It then proceeds with the presentation of an improved model of cooperation among the various intelligence structures, based on the summarised findings, and also recommends additional strategies to reduce current barriers to better cooperation and to improve the outputs and outcomes of the activities of these structures. The chapter concludes with a reflection on whether the research questions has been answered and the research objectives therefore have been achieved, and also about the extent to which the research has contributed original knowledge to the theory and practice of the governance of intelligence, as well as public governance as a specialised field of study.

7.2 SUMMARY OF RESEARCH FINDINGS AND RESULTS

The initial problem statement for the research described the intelligence oversight structures as fragmented, uncoordinated and non-cooperative when doing their oversight work.

The objectives of this study as reflected in Chapter one, section 1.7 were to:

 Establish the rationale for, and current state of cooperation among the South African democratic intelligence oversight structures.  Identify the nature of barriers for the systematic cooperation of democratic intelligence oversight structures.  Establish the nature of strategies to be adopted to improve the cooperation of democratic intelligence oversight structures to produce better governance results.

In pursuit of the above objectives, the study was packaged into seven chapters. The first three chapters are foundational and covered the study context, review of good international oversight systems, and research methodology. The rest of the chapters

243 presented data on the three research sub-questions, ie the rationale for, and state of cooperation, barriers to cooperation, and strategies to improve cooperation. This last chapter assesses the presented data and presents conclusions and recommendations (see section 1.12 in Chapter 1).

7.2.1 Chapter 1: General background and scientific orientation to the study

This chapter started by providing an overview of the South African intelligence architecture starting with the way that the pre-1994 dispensation looked like. After the presentation, it became clear that the pre-1994 intelligence dispensation did not have clear operational separation among the civilian, police and or military Intelligence services. It mainly focused on the gathering of domestic intelligence on national security threats mainly caused by the then liberation movements (Dombroski in Bruneau and Boraz: 2007). Its primary objective was to ensure the protection of the apartheid regime and white minority population at the expense of the black majority. Further, the chapter described the modalities around the 1994 transitional dynamics of the intelligence services. This was mainly characterised by the enactment of new statutory and regulatory frameworks including White Papers on intelligence, defence, safety and security. This was followed by a summative assessment of the integration processes involving the various statutory and non-statutory intelligence structures.

The chapter then provided a short, critical summary and assessment of the current South African intelligence architecture that covered the civilian, defence and police intelligence machinery. This assessment indicated that democratic South Africa moved to operationally separate the various intelligence groups within their specific legislative mandates (Cawthra: 2005) but that a lack of cooperation among these operational units seriously hampered the effectiveness and efficiency of these structures. The chapter also summarised the problem statement, the purpose, the research questions, objectives, design, methodology, ethics and the envisaged significance of the research.

244

7.2.2 Chapter 2: Contextual and theoretical variables influencing good international intelligence oversight systems

This chapter started by presenting the main findings of the global benchmarking exercise seeking to check what other countries are doing in the implementation of their intelligence oversight mechanisms. As such, special focus was placed on advanced democracies such as Australia, Canada, New Zealand, the UK and the US. The main reason for studying these countries was to develop the global picture on intelligence oversight models and specifically to understand the manner in which intelligence oversight structures cooperate with each other. In Australia, it was found that the establishment of the NSC served as the central mechanism for intelligence oversight (Flood: 2004). As for Canada, it was found that there is no statutory framework to mandate the cooperation of the various intelligence oversight structures (Den Boer: 2012).

In New Zealand, intelligence oversight work was found to be concentrated in the Office of the Inspector-General of Intelligence and Security (Weller: 1999). As for the UK, it was discovered that the ISC is located at the centre of intelligence oversight and thus serves as the mechanism to enhance the cooperation of the various oversight structures (Weller: 1999). As for the US, it was found that there is a lack of an institutional mechanism to facilitate the cooperation of the various intelligence oversight structures (Gill: 2012).

Further, this chapter critically analysed and assessed the two conceptual and theoretical frameworks that were used in the interpretation of the research findings. Good governance theory was used to guide the study towards the realisation of intended objectives. On the other hand, systems theory was adopted to shape the proposed systematic cooperation model to improve the operations and results of the

245 intelligence structures concerned. In conclusion, this chapter provided the conceptual framework which guided the research approach to the posed research questions.

7.2.3 Chapter 3: Methodological approach to the study

This chapter provided the study’s research methodology and started by describing a qualitative paradigm as the chosen paradigm, followed by the presentation of reasons as to why I chose it. In the reasoning, it was indicated that given the kind of problem, the research questions and objectives, the situation demanded the utilisation of a qualitative approach. This was based on its ability to allow me to spend many hours in the field collecting extensive data to gain an insider perspective on the cooperation modalities of intelligence oversight structures (Creswell: 1998). Further, this chapter described the context of a qualitative case study as the chosen research method. This was informed by its ability to allow me to empirically inquire and investigate the phenomenon in its real life context (Leedy and Ormrod: 2005). In terms of research procedure, the chapter outlined the data collection instruments that entailed interviews for primary data collection and a documentary and literature review for secondary data collection. Data was compiled through an initial critical, comparative literature assessment of scholarly, statutory, policy, technical and popular documentary sources and publications on various relevant issues. This literature review and assessment was supplemented by the researcher’s personal knowledge and experience of a number of relevant issues as well as by specialised, purposive interviews with two categories of interviewees to gather primary data on the topic. The two interview categories consisted of intelligence oversight officers and external experts. The chapter concluded with a summary of the requirements for reliability and validity of the research results. This assessment indicated that internal validity, construct validity, external validity and reliability can be promoted by following Yin’s (1994) conclusions that fairness, truthfulness and transparency should remain underlying principles for conducting

246 coherent social sciences research. The extent to which these goals have been achieved in this thesis is assessed below in section 7.6.

7.2.4 Chapter 4: Structural cooperation among democratic oversight structures in South Africa

This chapter started by providing a critical analysis and assessment of the South African democratic intelligence oversight architecture which ranges from legislative to executive, judicial, civil and chapter 9 institutions. It then proceeded to present the findings of the study in response to the first research sub-question seeking to understand the rationale for, and state of cooperation among intelligence oversight structures. The chapter assessed the cooperation gaps exposed by the Project Avani scandal, which inter alia include the failure to acquire judicial authorisation before intercepting communication media, and the continued politicisation of domestic intelligence operations (OIGI: 2006). The chapter also assessed the main themes of the general rationale for a more effective cooperative approach to intelligence oversight, which included different intelligence oversight mandates, varied allocated intelligence oversight powers, application of varied intelligence oversight methods, different intelligence oversight intervals, different reporting modalities, emergence of fourth generation warfare, potential for human rights violations and the need for public accountability (Den Boer: 2012).

The chapter further presented data on the state of cooperation among intelligence oversight structures. It included themes on the legislative framework, the cooperation of intelligence oversight structures, the existing intelligence oversight cooperation model, the care referral mechanism, intelligence oversight processes and the cooperation of the chapter 9 institutions. The data confirmed the absence of concerted efforts to ensure the effective cooperation of the intelligence oversight structures when

247 overseeing members of the intelligence services (Strauss and Corbin: 1990; Denzin and Lincoln: 1994 and Leedy and Ormrod: 2005).

7.2.5 Chapter 5: Barriers to improved cooperation of intelligence oversight structures in South Africa

This chapter presented data in response to the second research sub-question, seeking to understand the nature and implications of barriers experienced by intelligence oversight structures when cooperating with one another during the performance of their varied legislative mandates. The data indicated that intelligence oversight structures confront both structural and operational barriers in their quest to cooperate with one another (Caparini: 2002). Structural barriers include: the lack of a common doctrine, unclear information classification methods, excessive secrecy and lack of trust, principle of need to know, demand for excessive transparency and undue utilisation of national security. The development of the current post-apartheid intelligence oversight model did not adequately address these fundamental structural challenges to facilitate the cooperation of intelligence oversight structures (Ministerial Review Commission: 2008). This is complicated by the absence of a common doctrine on intelligence oversight among the intelligence oversight structures.

The structural barriers are reinforced by the existence of operational barriers which continues to frustrate practical efforts of oversight structures to engage one another for good governance results (Ministerial Review Commission: 2008). The operational barriers in this case include the absence of clear legislative and other regulatory frameworks, unclear financial accounting practices, political interference and plausible deniability, and varied institutional reporting lines. The data confirms the non- availability of legal instruments to promote or enforce the cooperation of intelligence oversight structures. It is for these reasons that various structures concentrate on their sole mandates as provided for in their founding legislative prescripts with no regard to

248 the work of other structures. As a result, the findings in this chapter confirmed that the work of the intelligence oversight structures remain fragmented, uncoordinated and non-cooperative; as earlier referenced in the problem statement contained in Chapter 1 in section 1.4.

7.2.6 Chapter 6: Strategies to improve cooperation among intelligence oversight structures in South Africa

This chapter presented data in response to the third research sub-question seeking to solicit suggestions on strategies to improve the systematic cooperation of intelligence oversight structures when overseeing members of the intelligence services. It integrates the implications of the previous chapters. The suggested strategies advocate for two broad approaches, namely: the need to reshape the key intelligence oversight structures; and to improve the operational modalities of the intelligence oversight structures (Ministerial Review Commission: 2008). Regarding the need to reshape key intelligence oversight structures, the research findings identified the need to reshape the JSCI, the Ministry of State Security, the IGI and civil society and the media. In terms of the JSCI, the main themes emphasise the need to open certain committee meetings to members of the public based on the national interest, to table and debate certain committee reports in the NA and the NCOP; and undoing the joint committee approach for a select or portfolio committee system.

Regarding the need to improve the operational modalities of the intelligence oversight mechanism, the emerging themes centred on the need to streamline the legislative and regulatory frameworks, empower intelligence oversight officers, and balance secrecy with the need for transparency and improving trust among the intelligence oversight structures (Africa: 2009). Operationally, it remains crucial to ensure the enactment of strict legislative and regulatory platforms to enhance the cooperation of intelligence oversight structures while implementing their varied legislative mandates.

249

7.2.7 Chapter 7: Proposed improved cooperation model for intelligence oversight structures in South Africa

This chapter starts off by providing summaries of all chapters starting from Chapter 1 to Chapter 7 to indicate and summarise the integrated logical flow of arguments causally linking the problem statement, research questions, objectives, design and methodology to the research data, findings, conclusions and recommendations. The chapter then presents the systematic model to improve the cooperation of intelligence oversight structures, emanating from the above findings and conclusions. It also reflects on the contribution of this research to the current scholarly body of knowledge on the topic as well as on the contribution of the research to the improvement of the profession of intelligence services and their results. Finally, this chapter identifies constraints encountered during the research and also emerging trends, problems and gaps requiring further investigation in future research projects. Based on this above summary, the following section presents the key conclusions and recommendations of the research.

7.3 KEY RESEARCH CONCLUSIONS AND RECOMMENDATIONS

The main conclusions and recommendations for improvement based on the six critical thematic areas which emerged during the research, are the following:

7.3.1 Theme 1: Absence of a legislative regulatory framework to guide and maximise the cooperation of intelligence oversight structures

Based on the research outcomes contained in Chapter 5, sections 5.3.1 and 5.3.2, it can be concluded that the South African intelligence oversight model currently does not provide for the clear legislative regulatory frameworks required for guiding the cooperation of intelligence oversight structures. This need is illustrated by individual

250 structures having to enter into bilateral agreements through MoUs to regulate their engagements with each other. As such, it is recommended that the Intelligence Oversight Act 1994 be amended to include a legal instrument for guiding the cooperation of intelligence oversight structures when overseeing members of the intelligence services. Such legal instrument should include the interaction among key intelligence oversight structures and the chapter 9 institutions.

7.3.2 Theme 2: Lack of a common doctrine in the implementation of the intelligence oversight model by intelligence oversight structures

Given the research outcomes as contained in Chapter 5, section 5.2.1, it can be concluded that the South African intelligence oversight model does not have a common doctrine on the implementation of intelligence oversight on members of the intelligence services by intelligence oversight structures. As a result, various structures conduct intelligence oversight based on their narrow legislative mandates with no regard for the work of other structures. In this regard, concerted efforts should be made to develop a common doctrine guided by the principles of cooperation, coordination and consultations with each other. This is necessitated by the existence of intelligence oversight structures in all three governance branches (legislature, executive and judiciary).

Further, it is also clear that working together is important for the accomplishment of legislative mandates. Therefore, it is important to establish formal cooperation mechanisms to ensure the successful realisation of the effective oversight mechanisms. It is further recommended that these institutions conduct joint public awareness programmes in order to educate members of the public about their existence and interrelationships. This is critical for avoiding working in silos, creating duplication of efforts and undue competition among various oversight structures. In practice, the effective functioning of the IGI will ensure that intelligence operatives obey the

251 legislative framework of the country, thereby, reducing the number of people seeking recourse from the PP and the HRC.

7.3.3 Theme 3: Existence of undue political interference and the application of ‘plausible deniability’

Given the research outcomes as contained in Chapter 5, section 5.3.4, it can be concluded that political interference continues to hamper and skew the work of the intelligence services as the ruling elite manipulates them for advancing their own political agendas. As already indicated through the problem statement in Chapter 1, section 1.4, Project Avani, though legally constituted, was abused to fight political factional battles within the ruling ANC. Therefore, the tendencies of political interference frustrate any effort to ensure the effective oversight of members of the intelligence services. Further, it can be concluded that the direct application of ‘plausible deniability’ to issues of intelligence tradecraft, remains problematic. This is characterised by the denial of knowledge and/or having authorised sensitive or covert activities by the policy-makers and members of the executive, especially when things go wrong. This is done through the executive’s pronouncement, with some plausibility, that activities were carried out by intelligence operatives without their knowledge or approval. This approach remains problematic and defeats the basic tenets of good governance such as openness, transparency and accountability.

‘Plausible deniability’ increases the risk of misunderstandings between senior officials and their subordinates. However, when ‘plausible deniability’ fails, it has the potential to seriously discredit the political office bearer or official who would have invoked it as a defence mechanism. However, when it succeeds, it creates an impression that the government is not in control of the state. Based on the above finding, it can be concluded that the failure of ‘plausible deniability’ can cause serious harm to the country’s democratic system. In South Africa, various legislative prescripts continue to

252 advise subordinates to obey lawful commands at all times. This is legislated in order to discourage the application of ‘plausible deniability’ in the execution of a legislative mandate. Therefore, it is recommended that all relevant authorities, be they senior or junior public officials, should commit to avoid ‘plausible deniability’ at all times. As for members of the intelligence services, it is always important to discard this principle given its nature of frustrating the entire oversight effort.

7.3.4 Theme 4: Continued emphasis on excessive secrecy at the expense of transparency and public accountability

Based on the research outcomes as stipulated in Chapter 5, section 5.2.3, it can be concluded that democratic South Africa continues to emphasise excessive secrecy at the expense of transparency and public accountability. The issue of intelligence oversight is consistently challenged by inherent secrecy perpetrated by members of the intelligence services. In practice, it is accepted that secrecy is a trademark of intelligence tradecraft and remains the basis of its relationship with government and its own self-image. Further, the International Covenant on Civil and Political Rights allows the state to use secrecy for the protection of national security as legitimate grounds for limiting the rights to freedom of expression and access to information. Mr Joe Nhlanhla, the late former Minister of Intelligence confirmed the legitimacy of this governance approach.

Secrecy is an inherent requirement of intelligence tradecraft and without some degree of it, valuable intelligence and its methods can easily be jeopardised (Africa: 2006). Therefore, within the intelligence tradecraft, a secret is both a virtue and a necessity. However, inasmuch as secrecy is the lifeblood of intelligence, it should also be noted that transparency is the oxygen of democracy. It is therefore clear that not everything ought to be protected and kept secret, but only those things whose disclosure would compromise the maintenance of national security. It is recommended that the

253 intelligence services should maintain an up-to-date website with credible information that is useful to members of the public. It becomes worthwhile and value for money when members of the public have access to certain intelligence products covering topics such as food, energy, and water, security assessments. This is critical to the populace for the effective projection of soft power and the promotion of intelligence work as well as for the demystification of intelligence work and the enhancement of the legitimacy and acceptability of the Intelligence services.

Ordinary people still associate intelligence tradecraft with the capturing and elimination of government opponents. In fact, foreign intelligence practice for example, is all about the first line of defence in diplomatic matters for the benefit of the entire country. In the 21st century, almost 98% of intelligence information comes from overt sources and its main function lies in its ability to separate the wheat from the chaff. In terms of the challenges associated with information classification and declassification, it is recommended that a coherent declassification system should be developed to enhance the tenets of openness, transparency, and public accountability.

In order to achieve this, an inter-agency National Declassification Unit (NDU) can be established. The NDU would better coordinate and manage the national declassification effort within the intelligence services. To improve public confidence in the work of the intelligence services, the composition of the NDU could include experts from within intelligence services; members from government, opposition parties, labour, academia; and representatives of civil society.

7.3.5 Theme 5: Impact of the proliferation of intelligence control and oversight structures on members of intelligence services

As presented in Chapter 4 in section 4.2 the post-apartheid South Africa created a multiplicity of intelligence oversight structures located in various levels of governance.

254

This ranges from legislative to executive, judicial, civil including state institutions, supporting constitutional democracy. Although the Intelligence Oversight Act 1994 places the intelligence oversight mandate with the IGI and the JSCI, there are a number of structures such as the chapter 9 institutions which have broad mandates that have the potential to impact the work of members of the intelligence services.

Therefore, the proliferation of these structures must be accompanied by a coherent cooperation mechanism to facilitate their engagement with each other. However, the available data indicates that the country does not have such cooperation mechanisms for these structures. It is therefore recommended that the proposed systematic cooperation model presented below in section 7.5 be adopted for implementation by the intelligence oversight structures.

7.3.6 Theme 6: Unclear structural governance challenges of key intelligence oversight structures

Based on the available data, it can be concluded that the South African intelligence oversight model is characterised by unclear structural governance challenges particularly within the key intelligence oversight structures. In South Africa, the IG is appointed by the State President and approved by the National Assembly with a supporting vote of a two-thirds majority. Given this process, there is no doubt that the political nature of this appointment compromises the independence and flexibility of the IG on the implementation of its legislative mandate. It goes without saying that presidential appointments are not conducive to the independence of the supposedly independent entities charged with oversight responsibilities.

Based on the above discussion, it is recommended that the appointment of the IGI should be made by bodies such as the judiciary, civil society or any other independent body. In this case, such a body should be responsible for the appointment, the

255 supervision and if necessary, the discipline and determination of the necessary budgetary requirements. Currently, the IGI is operationally preoccupied with the handling of human resources disputes of members of the intelligence services. To advance the cooperation of intelligence oversight structures, the US government established the IGI office for each intelligence service. Further, they established the CIGIE to ensure that the IGI operates efficiently, effectively and without reproach.

Furthermore, the available data also shows the wrong bureaucratic positioning of the office of the IGI. It is located under the DG of the SSA which is one of the primary intelligence services it is supposed to oversee. In fact, the SSA DG has the ultimate authority on the administrative decisions of the IG and this has the potential to undermine the IG’s independence during the performance of its work. This situation is not unique to the IG, but applies to the state institutions supporting constitutional democracy. The PP, for instance, relies on the Department of Justice for financial allocation, while it is one of the departments the PP is meant to monitor and investigate should the need arise. It is for this reason that the IG should be removed from under the SSA to enable it to secure its own budget through the parliamentary process as an independent entity. In an attempt to strengthen its independence, a highly qualified individual should be appointed to the post of the IG. For instance, in New Zealand, an IG can only be appointed if that person has served as a judge of the High Court. This is an important arrangement South Africa can learn from to smooth the process of appointing the IGI.

Based on the above discussion, it is recommended that government should ensure the proper capacitation of the intelligence oversight officers to enable them to easily interface with all other officers from the other structures. Given this assertion there is a need to appoint competent oversight officers with adequate financial support to enable them to discharge their functions with diligence. Further, the IGI staff complement should include all relevant competencies, ranging from legal, investigating, and strong

256 administrative support with relevant skills and expertise. Staff members should regularly be sent for training and refresher courses to enhance their acuity in the execution of their work. Based on these conclusions, it is now important to consider their implications for the theoretical framework.

7.4 PROPOSED IMPROVED COOPERATION MODEL FOR INTELLIGENCE OVERSIGHT STRUCTURES

After considering all inputs, I found it necessary to ask both oversight officers and external experts about the need to create a new model to help remedy the situation. That is, to ensure the systematic cooperation of oversight structures when overseeing members of the intelligence services. In response, 26 (93%) of the oversight officers said yes, one (3.5%) said no and one (3.5%) was not sure. As for external experts, all seven (100%) said yes to the creation of a new model. Figure 7.1 illustrates the schematic depiction of the responses from the respondents.

257

Figure 7.1: Whether there is a need for the creation of a new model to redefine the South African architecture to include the systematic approach in their endeavour to cooperate with each other

Based on the suggested strategies and overwhelming support among the research respondents for the creation of a new model to improve the effective cooperation of intelligence oversight structures, the following model is proposed to be used by oversight structures in their quest to cooperate better with each other. This proposed model seeks to promote the effective cooperation of intelligence oversight structures and ensure the realisation of the underlying tenets of good governance which according to Adejemboi (1998) include public accountability, transparency in governance procedures, abiding to the rule of law, and effective and efficient public service. Therefore, this proposed model conforms to the fundamental principles of systems theory which according to Laszlo and Krippner in Jordan (1998), denotes the complexity of interacting components together with the relationship between them that identifies the boundary-maintaining entity or a process.

258

Of all 35 respondents, 34 (97%) concurred that South Africa needs a systematic improved cooperation mechanism to improve the efficacy of the multiple intelligence oversight structures with varied powers and operational mandates. Figure 7.2 presents the proposed model to be used by intelligence oversight structures to fully cooperate with one another.

Figure 7.2: Improved systematic cooperation model for intelligence oversight structures

Source: (www.parliament.gov.za)

This proposed model places the OIGI at the centre of the intelligence oversight model. As already discussed in Chapter 4, section 4.2.4; section 7 (7) of the Intelligence Services Oversight Act 1994 empowers the IGI to monitor compliance of members of the intelligence services with the Constitution, applicable laws and relevant policies on intelligence and counter-intelligence. Further, section 8 (a) of the same Act empowers the IGI to have access to intelligence, information or premises under the control of any intelligence services. Therefore, the IGI can play a central role in coordinating the work of the various intelligence oversight structures. As demonstrated in Figure 7.2 above, 259 all the linkages to the IGI are represented by the solid line and this means that all structures have to cooperate with the office while addressing any matters with members of the intelligence services. However, it is noted that the OIGI does not have the capacity to effectively deliver on this mandate. It is for this reason that the above recommendations call for the urgent capacitation of the office. As with the IGI, the linkages to the JSCI are also reflected in the solid line to emphasise that all structures can be summoned to brief the committee on the work they do with members of the intelligence services.

Further, the envisaged cooperation model allows all other structures to engage each other on any matter of common concern. Therefore, such linkages are represented by dotted lines that reflect the flexibility of such engagements. Based on the envisaged model, the engagements of the other structures with the IGI and the JSCI remain solid and compulsory. In this proposal, the IGI should be given the power to convene all other structures to discuss matters of common interest. This effort seeks to align government’s work, reduce duplications and ensure coherence in the work of various intelligence oversight structures. This modality will also assist the intelligence oversight structures to contribute towards the promotion of good governance in South Africa. The final accountability lies with the parliamentary platform under the auspices of the JSCI. As Fisher puts it in Duke (2002:323), “the American Congress statutorily requires the executive branch to provide reports on its activities”. This model empowers the IG to coordinate the functionalities of the oversight structures and the JSCI to receive briefings on the functionalities of the intelligence services, including their compliance to the country’s legislative frameworks.

7.5 IMPLICATIONS OF THE RESEARCH FOR THEORY AND PRACTICE

As already discussed in Chapter 2 in section 2.4 this study used both systems theory and good governance theory. Systems theory was used for two main purposes. First,

260 it was used to evaluate the state of cooperation among intelligence oversight structures through the application of a process evaluation technique. Second, it was used to shape the proposed improved systematic cooperation model to be used by the intelligence oversight structures when cooperating with each other. As for good governance theory, it was mainly used as a yardstick for the determination of whether the intelligence oversight structures contribute towards the promotion of good governance results in South Africa. The network theory assisted in understanding the interactions and relationships among the various structures with varied mandates, though working towards one goal. It allowed me to understand the disparity between cooperation, coordination and collaboration. From the network framework, it became clear that communities today are facing a multiplicity of challenges which could not be solved by an individual structure operating in silos but by a multiplicity of structures which keep on cooperating, coordinating and collaborating, their efforts. It is because of these realities that the operations of intelligence oversight structures should observe the complexities of governance network processes which emphasise the complex interaction of interdependent structures for the realisation of common goals.

Given these two theoretical considerations, the primary contribution of this study to theory and practice is anchored on its presentation of the new systematic cooperation model for the intelligence oversight structures to effectively hold members of the intelligence services accountable and then contribute to the promotion of good governance in the country. To properly expand this study’s original contribution to both theory and the professional practice of good governance and the effective oversight of intelligence management in South Africa, I would like to revisit the critical question posed by Respondent 5 in Chapter 4 under section 4.4.6. He posed the question: what was the intention of the legislature for creating multiple oversight institutions as stand- alone structures and not leaving room for the creation of a cooperation model?

261

Theoretically, the legislature’s intention was to spread oversight powers across various structures with the objective to enhance the principle of checks and balances on each other. But now, literature and operational experience shows persistent failure of such a spread as clearly articulated in the problem statement in Chapter 1, section 1.4. It is for this reason that this research presents a systematic cooperation model for application by intelligence oversight structures when cooperating with each other.

Operationally, Nzongola-Ntalaja (2002), Dlomo (2004), Netshitenzhe (2005) and Africa (2009) support this theoretical shift from the original intention of the legislature. They concur that good governance can be realised when the oversight functions of various structures are coordinated, formalised, systematised, and cooperate with the work done by other structures. This operational shift is also supported by Agere (2000:8) who indicated that, “the absence of an effective intelligence oversight mechanism could breed corruption, which could later breed poor governance”. As already articulated in Chapter 1 the existence of poor governance serves as a major constraint towards economic development and national security. Sendall (1997) and Africa (2009) advise that sustaining democratic governance is dependent on the ability to balance secrecy with the need for transparency and accountability, as critical tenets of good governance. This is regarded as an important benchmark by scholars like Fitzgerald et al (1995), Fox and Meyer 1996) and Waldo (1994), for sustaining the emerging democracies of the world.

7.6 CONSTRAINTS ON THE RESEARCH AND DIRECTIONS FOR FUTURE RESEARCH

As already hinted at in both Chapters 1 and 2, this research confronted a few constraints. 1) The literature review confirmed that issues of intelligence tradecraft, its oversight or general operational modalities, are the least researched subjects in many parts of the world and South Africa is no exception. This situation is mainly caused by

262 inherent secrecy which underpins the fundamental principles of the discipline and guides the operational conduct of intelligence operatives and/or practitioners. This research was even more constrained as it took a step further by looking into the cooperation modalities of intelligence oversight structures when overseeing members of the intelligence services. 2) Given sensitivities associated with the intelligence discipline, some interviews started slowly when respondents displayed some sense of being uncomfortable. This was illustrated by some respondents opting not to sign the consent form. However, with time and guarantees of confidentiality the situation improved significantly. 3) Lastly, the spread of the research context which focused on structures located across the legislative, executive, judicial, and administrative and chapter 9 institutions; constrained my ability to drill deeper into some of the issues, hence my identification of areas requiring future research discussed below.

Despite the constraints summarised above, I did a number of things to ensure the realisation of internal, construct, and external validity of the research results, including their reliability. 1) Despite my working in the environment, I ensured internal validity by removing myself from the research context and mainly focused on the causal relationships between variables as they emerged through the research process. 2) Despite the utilisation of multiple concepts with multiple meanings, I ensured the construction of a clear interface between the concepts and the theoretical and conceptual frameworks to guarantee construct validity. 3) The clearly crafted conceptual and theoretical frameworks assisted the process of making empirical observations. Using cross-case analysis across the various oversight structures allowed me to conduct good analytical generalisation as reflected in the earlier summaries of this chapter. 4) For the overall reliability of the research results, I kept solid case study databases with notes, documents and narratives in easily retrievable format for any possible future use. In sum, research unfolded in a transparent, cordial and frank manner under a clearly articulated research methodology to guarantee successful replication should the need arise.

263

However, the research uncovered a multiplicity of issues requiring further investigations by future researchers. Those issues are presented in accordance with the six research conclusions summarised in the previous section.

First, given the conclusion that the South African intelligence oversight model lacks the legislative and regulatory frameworks for guiding the cooperation of intelligence oversight structures, there is a need to investigate the impact of fragmented, uncoordinated and non-cooperating intelligence oversight mechanisms to the accountability of members of the intelligence services.

Second, given the confirmation of the lack of common doctrine in the implementation of an intelligence oversight model, there is a need to investigate the modalities for the development of a common doctrine for the work of the intelligence oversight structures.

Third, given the discovery of the existence of undue political influence and the application of ‘plausible deniability’, the impact of political interference into the credibility of the intelligence services can be investigated. The impact of ‘plausible deniability’ on the issue of accountability can also be investigated.

Four, considering the confirmation that there is excessive secrecy at the expense of accountability, there is a need to conduct further investigations into the feasibility of balancing secrecy and transparency.

Five, after the confirmation of the proliferation of intelligence control and oversight structures on members of the intelligence services, there is a need to do further investigations on the impact of proliferated oversight structures on the quality of intelligence oversight.

264

Six, following the discovery of unclear structural governance challenges of key intelligence oversight structures, there is a need to conduct further investigations into the impact of the administrative location of the IGI vis-à-vis its independence and efficacy.

7.7 SUMMARY AND CONCLUSIONS

Based on the above analyses and assessments, it is evident that the cooperation of the intelligence oversight structures remained ad hoc and in certain cases non-existent. It also became clear that the implementation of an intelligence oversight mechanism was fragmented, uncoordinated and non-cooperative and based on a silo mentality. The discussion of results confirmed the fact that the failure of the intelligence oversight structures to coordinate their efforts remains the critical factor in their failure to effectively contribute towards the promotion of good governance. As a result, it became imperative to advance legislative and regulatory amendments to ensure the effective cooperation among these structures.

Further, members of the intelligence services have to observe the constitutional provisions and the entire legislative framework by the nature of the work they do. However, the research discovered a multiplicity of intelligence oversight challenges which have to be addressed. Following the confirmation of the interrelationship between governance and intelligence tradecraft, it can be concluded that both governance and systems theories need each other for the realisation of the envisaged democratic order. Based on the findings, it was concluded that an unclear legislative mandate and lack of appropriate oversight mechanism can present a self-generated national security threat. Furthermore, given the establishment of the multiplicity of institutions with overlapping mandates, it can be concluded that developing formal channels of interaction and cooperation remains significant. Although the research discovered the impossibility of an empirical balance between transparency and secrecy,

265 it can be concluded that transparency enhances accountability which is important for good governance, while excessive secrecy breeds corruption and bad governance.

Ultimately, it can be concluded that the responsibility to protect our democracy does not exclusively lie within the oversight institutions, but with the entire population. This assertion is based on the reality that, although the intelligence structures of government cannot always be fully transparent, they can be governed in a democratic good governance approach. Therefore, it is important to ensure that oversight of such institutions extends beyond legislative mechanisms, but includes being overseen by the community at large. However, this can only be achieved when members of the community are well educated and know exactly what to look for. Therefore, the significance of outreach programmes aimed at educating members of the public cannot be overemphasised. The primary outcome of this chapter was the presentation of a proposed improved systematic cooperation model to be used by intelligence oversight structures. The model proposes the centrality of the IGI as the primary coordinator and custodian of the cooperation modalities. With the model, the IGI should, together with other structures, make representations to the JSCI on the work of members of the intelligence services.

266

APPENDIX A:

QUALITATIVE INTERVIEW GUIDE AND RESULTS OF INTERVIEWS WITH 28 OVERSIGHT, AUDIT AND OMBUD OFFICERS

My name is Mr Nakampe Michael Masiapato. I am a doctoral student at the University of Johannesburg within the School of Public Management and Governance. I am conducting an academic research to understand the state of cooperation among the various intelligence oversight structures and explore the nature of barriers experienced in such cooperation. Once determined, I would like to solicit proposed strategies to ensure the systematic cooperation of democratic intelligence oversight structures for good governance results. Therefore, the main research question is posed as follow: How can the cooperation of the South African democratic intelligence oversight structures be improved for good governance results? Therefore, in order to respond to this main question, I would like to ask you some questions and your responses shall remain anonymous. To this end, the interview will take about 45 minutes of your time.

The interview questions are grouped into three categories in accordance with the three research sub-questions.

1. Research sub-question 1: Determination of the rationale for, and the state of cooperation among the intelligence oversight structures for good governance results.

1.1 In your view, do you think it is necessary for the various intelligence oversight structures to cooperate when overseeing members of the intelligence services?

Yes No Not sure 25 2 1 89% 7% 4%

267

1.2 If so, what could you cite as the underlying reason justifying the need for the cooperation of the various oversight structures?

Reasons for cooperation Number of Percentage respondents 1. 4th generation warfare 3 11% 2. Oversight mandates 5 18% 3. Oversight powers 2 6% 4. Oversight methods 3 11% 5. Oversight intervals 1 4% 6. Reporting modalities 3 11% 7. Human rights violations 4 14% 8. Transparency and accountability 5 18% 9. Uncertain of reasons (not sure) 2 6%

1.3 How would you describe the current state of cooperation among intelligence oversight structures in South Africa?

Very poor Poor Not sure Good Very good 8 18 1 1 0 29% 64% 4% 4% 0%

1.4 Does your establishing legislative and/or regulatory framework requires your structure to cooperate with other intelligence oversight institutions?

Yes No Not sure 3 21 4 11% 75% 14%

1.5 Does your oversight institution have a model or set of guidelines to facilitate your engagement with other oversight structures?

268

Yes No Not sure 0 24 4 0% 86% 14%

1.6 If so, how does this model or these guidelines assist to facilitate your cooperation and coordination with other oversight structures?

None

1.7 Do you have a specific institutional mechanism such as a committee, Memorandum of Understanding (MoU), Service Level Agreement (SLA) or a specific platform within which the engagement occurs?

Yes No Not sure 10 13 5 36% 46% 18%

1.8 How frequent does your oversight institutions engage with other oversight structures?

Twice a week Twice a month Not at all Undefined 0 3 8 17 0% 11% 28% 61%

1.9 To what extent does the frequency of the engagements shape the agenda of such interaction?

To what extend frequency shape Number of respondents Percentage engagements Fully shape 1 4% Partially shape 2 7% Poorly shape 19 68%

269

Not sure 6 21%

1.10 Does this framework function optimally in facilitating your engagement with other oversight structures?

Yes No Not sure 9 15 4 32% 54% 14%

1.11 Does your oversight institution have a formal or informal mechanism for referring cases requiring investigations to other oversight structures?

Formal Mechanism Informal Mechanism Not sure 9 12 7 32% 43% 25%

1.12 Does this mechanism or agreement function optimally to enable referrals and timely responses?

Does case referral mechanism Number of respondents Percentage function optimally Fully optimal 2 7% Optimal 3 11% Less optimal 18 64% Not sure 5 18%

2. Research sub-question 2: Nature of barriers to the effective cooperation of democratic intelligence oversight structures for good governance results.

2.1 Do you think there are certain factors or things which serve as impediments for the effective cooperation, of various intelligence oversight structures?

270

Yes No Not sure 25 1 2 89% 4% 7%

2.2 If so, what would you describe as one of the critical impediments or barriers for the effective cooperation of intelligence oversight structures?

Structural barriers Number of Percentage respondents 1. Lack of common doctrine 4 14% 2. Unclear information classification 3 11% 3. Excessive secrecy and lack of trust 4 14% 4. Principle of ‘need to know’ 3 11% 5. Demand for excessive transparency 1 4% 6. Undue utilisation of national security 3 11% Operational barriers Number of Percentage respondents 1. Absence of legislative framework 4 14% 2. Absence of regulatory framework 2 7% 3. Unclear financial accounting 1 4% 4. Political interference and denialism 2 7% 5. Misunderstanding oversight 1 4%

2.3 Does your institution allocate specific funds to facilitate the cooperation of the various intelligence oversight structures?

Yes No Not sure 0 23 5 0% 82% 18%

271

2.4 If so, are the allocated funds adequate to ensure the realisation of the task at hand?

None

2.5 Does your post establishment provide for a specific job type to specifically facilitate your organisational cooperation with other oversight structures?

Yes No Not sure 0 25 3 0% 89% 11%

2.6 If not, how is this function performed in your oversight institution?

Most respondents cited MoUs and SLAs as the interim measures currently used by oversight officers/auditors and/or ‘ombuds’ officials for engaging other structures.

2.7 Do the oversight officers have sufficient time at their disposal to engage with other oversight structures? More time Less time Sufficient time Not sure 0 17 7 4 0% 61% 25% 14

2.8 Has your oversight structure been able to create the relationship of trust with other intelligence oversight structures?

Yes No Not yet Not sure 2 3 20 3 7% 11% 71% 11%

2.9 If not, to what extent has this contributed to the poor state of cooperation between your structure and other intelligence oversight institutions?

272

Badly Very badly Not at all Not sure 0 17 7 4 0% 61% 25% 14%

2.10 To what extent do individual personalities hamper the effective cooperation of intelligence oversight structures with each other?

Badly Very badly Not at all Not sure 0 17 7 4 0% 61% 25% 14

2.11 Does your institution experience undue political interference in its operations thereby having a negative impact on its quest to cooperate with other structures?

Yes No Not sure 2 8 18 7% 29% 64%

3. Research sub-question 3: Strategies to ensure the systematic cooperation of democratic intelligence oversight structures for good governance results.

3.1 In your opinion, is there a need to review the mandates and roles of the various intelligence oversight structures to facilitate their cooperation with each other?

Yes No Not sure 22 3 2 79% 11% 7%

3.2 If yes, which structure in particular would you like to see their mandates, roles and functions reviewed to improve their cooperation with other structures?

273

Structures to review mandates Number of Percentage respondents 1. Joint Standing Committee for Int. 6 21% 2. Ministry of State Security 2 7% 3. Inspector-General for Intelligence 17 61% 4. Strengthen civil society and media 3 11%

3.3 Based on the identified structures above, what kind of review would you like to see for the improvement of cooperation among oversight structures?

Various respondents gave various suggestions as contained in chapter 6 under section 6.2 thesis.

3.4 In addition to reviewing the mandates, roles and functions of certain structures, what else do you suggest needs to be done to improve the cooperation of the various intelligence oversight structures?

Suggestions improve Number of Percentage cooperation respondents 1. Streamline legislation 6 21% 2. Streamline regulations 2 7% 3. Empower oversight officers 13 46% 4. Balance secrets with transparency 3 10% 5. Improve trust among structures 5 17%

3.5 Based on the suggestions above, how would you like them refined to improve the cooperation of intelligence oversight structures?

Various respondents gave various suggestions as contained in chapter 6 under section 6.3 in the thesis.

3.6 In your view, do you think there is a need for the creation of a new model to redefine the South African intelligence oversight architecture to include the need for the systematic 274

cooperation of yourselves when overseeing members of the intelligence services?

Yes No Not sure 26 1 1 93% 3.5% 3.5%

3.7 In your view what should be the key components of such a model?

Various respondents gave various inputs as contained in chapter 7 under section 7.3 in the thesis. -end-

275

APPENDIX B:

QUALITATIVE INTERVIEW GUIDE AND RESULTS OF INTERVIEWS WITH 7 EXTERNAL EXPERTS

Interview questions were grouped into three categories in accordance with the three research sub-questions.

Research sub-question 1: Determination of the rationale for, and the state of cooperation among the intelligence oversight structures for good governance results.

1.1 In your view, do you think it is necessary for the various intelligence oversight structures to cooperate when overseeing members of the intelligence services?

Yes No Not sure 7 0 0 100% 0% 0%

276

1.2 If so, what would you cite as the underlying reason justifying the need for the cooperation of these oversight structures with each other?

Reasons for cooperation Number of respondents Percentage 1.Varied oversight mandates and 1 14% powers 2.Varied oversight methods and 1 14% modalities 3. Reporting modalities 1 14% 4. Human rights violations 2 29% 5. Transparency and accountability 2 29%

1.3 How would you describe the current state of cooperation among the South African intelligence oversight structures as constituted?

Very poor Poor Not sure Good Very good 6 1 0 0 0 86% 14% 0% 0% 0%

1.4 Do you think their establishing legislative and/or regulatory framework requires of them to cooperate with the other intelligence oversight structures?

Yes No Not sure 1 6 0 14% 86% 0%

1.5 Do you think they have a model or set of guidelines which facilitate their engagement with other oversight structures?

I think so I don’t think so I am not sure 0 5 2

277

0% 71% 29%

1.6 If they have, do you think such model or guidelines do assist them in ensuring their effective cooperation with other oversight structures?

None

Research sub-question 2: Nature of barriers to the effective cooperation of democratic intelligence oversight structures for good governance results.

2.1 Do you think there are certain factors or things which serve as impediments for the effective cooperation of the South African intelligence oversight structures?

Yes No Not sure 5 1 1 71% 14% 14%

2.2 If so, what would you describe as one of the critical impediments or barriers for the effective cooperation of the South African intelligence oversight structures?

Structural barriers Number of Percentage respondents 1. Lack of common doctrine 2 29% 2. Excessive secrecy and lack of trust 1 14% 3. Demand for excessive 1 14% transparency

Operational barriers Number of Percentage respondents 4. Absence of legislative framework 2 29% 5. Political interference and denialism 1 14%

278

2.3 In your view, do you think oversight structures allocate funds or even human resources to specifically drive consultations, engagements and cooperation with other oversight structures?

Yes No Not sure 0 23 5 0% 82% 18%

2.4 To what extent do you think individual personalities hamper the effective cooperation of intelligence oversight structures with each other?

Badly Very badly Not at all Not sure 0 5 1 1 0% 71% 14% 14%

2.5 Do you think these oversight structures experience undue political interference in their operations thereby having a negative impact on their quest to cooperate with other structures?

Yes No Not sure 3 3 1 43% 43% 14%

Research sub-question 3: Strategies to ensure the systematic cooperation of democratic intelligence oversight structures for good governance results.

3.1 In your opinion, is there a need to review the mandates and roles of the various intelligence oversight structures to facilitate their cooperation with each other? Yes No Not sure 6 0 1 86% 0% 14%

279

3.2 If yes, which structure in particular would you like to see their mandates, roles and functions reviewed to improve their cooperation with other structures?

Structures to review mandates Number of Percentage respondents 1. Joint Standing Committee for Int. 1 14% 2. Ministry of State Security 1 14% 3. Inspector-General for Intelligence 3 43% 4. Strengthen civil society and media 2 29%

3.3 Based on the identified structures above, what kind of review would you like to see for the improvement of cooperation among oversight structures?

Various respondents gave various suggestions as contained in chapter 6 under section 6.2 in the thesis.

Suggestions improve Number of Percentage cooperation respondents 1. Streamline legislation 2 29% 2. Streamline regulations 2 29% 3. Empower oversight officers 1 14% 4. Balance secrets with transparency 1 14% 5. Improve trust among structures 1 14%

3.5 Based on the suggestions above, how would you like them refined to improve the cooperation of intelligence oversight structures?

Various respondents gave various suggestions as contained in chapter 6 under section 6.3 in the thesis. 280

3.6 In your view, do you think there is a need for the creation of a new model to redefine the South African intelligence oversight architecture to include the need for the systematic cooperation of yourselves when overseeing members of the intelligence services?

Yes No Not sure 7 0 0 100% 0% 0%

3.7 In your view what should be the key components of such a model?

Various respondents gave various inputs as contained in chapter 7 under section 7.3 in the thesis.

-end-

281

APPENDIX C CONSENT FORM FOR RESEARCH PARTICIPANTS

I, (full names) ______hereby agree to participate in the research project on ‘Cooperation of South African intelligence oversight structures for good governance’.

I understand that my participation in the project include being interviewed with the possibility of being recorded. To this end, I hereby agree to voluntarily participate in this process with the full understanding that my responses will be used in the project.

However, I do understand that such responses will be treated confidentially and that my name will not be quoted or published with the results of the study. I further understand that the collected data will also be treated confidentially and will be stored in password protected electronic devices to be locked in cabinet as promised by the researcher.

I also understand that I can withdraw from the project at any time of the research process without penalty should I choose to do so. Therefore, I concede that my participation in the process is voluntary and I have not been coerced or pressured to sign this consent form.

Signature: ______Date: ______

282

BIBLOGRAPHY Auriacombe, C.J. 2009. Methodological Perspectives for Novice Grounded Theory Researchers in Public Administration and Management Research. Journal of Public Administration. 44(4.2): 824-848.

Abrams, D.B., Boutwell, W.B., Gizzle, J. and Heimendinger, J. 1994. Cancer Control in the Workplace: The Working Well Trial. Preventive Medicine. 23:15-27.

Adejemboi, S. 1998. Africa and the challenges of democracy and good governance in the 21st century. Development Policy Management Forum (DPMF). Proceedings of the DPMF Annual Conference on democracy, civil society and governance in Africa, 7-10 December 1998, Addis Ababa.

Africa, S. 2006. Policy for managing access to intelligence information in post-apartheid South Africa. Unpublished doctoral thesis. Johannesburg: University of Witwatersrand.

Africa, S. 2009. Well-kept Secrets: The right of access to information and the South African Intelligence Services. Johannesburg: Acumen Publishing Solution.

Africa, S. 2012. The policy evolution of the South African Civilian Intelligence services: 1994 to 2009 and beyond. Institute of Strategic and Political Affairs, University of Pretoria. Pp. 97-134.

Africa, S. and Mlombile, S. 2001. Transforming the Intelligence services: Some reflection on the South African experience. Available at: http://www..law.harvard.edu/programs/criminal-justice/south-africa.pdf.

Africa, S. 2002. The Restructuring of the Intelligence Services in South Africa: An Assessment of the Transformation Process. Paper presented at the ASDR Security Sector Reform in Africa Conference. Accra: ASDR.

African Development Bank Group. 2003. Combating Corruption in Africa, Proceedings of the Regional Learning Workshop on Combating Corruption in Africa, Jointly organised by the African Development Bank, African Union, Transparency International, World Bank Institute and the Global Coalition for Africa, 27-30 January 2003: African Union Conference Centre, Addis Ababa. Available at: www.u4.no/helpdesk, Accessed on 30 January 2011.

African National Congress (ANC) Website. 2010. Available at: www.anc.org.za, Accessed on 12 April 2010.

African Union (AU). 2010. Available at: www.au-union.org, Accessed on 14 July 2010.

283

Agere, S. 2000. Promoting Good Governance: Principles, Practices and Perspectives. Management and Training Services Division. London: Commonwealth Secretariat.

Albritton, R.B. 2009. Are Democracy and Good Governance Always Compatible? Competing Values in the Thai Political Arena, Working Paper Series, Issued by International Political Science Association, Chile: Jointly Published by Globalbarometer.

Alford, R.R. and Friedland, R. 1985. Powers of Theory: Capitalism, the State, and Democracy, New York: Cambridge University Press.

ALLREFER.com. 2009. Military Intelligence and Intelligence Coordination. Available at: www.reference.allrefer.com/country-guide-study/south-africa, Accessed on 05/01/2010.

Anderson, R.E., Carter, I., and Lowe, G.R. 1999. Human behaviour in the social environment (5th ed.). New York: Aldine de Gruyter.

Andersen, T. Denrell, J. Bettis, R. 2007. Strategic Responsiveness and Bowman’s Risk Return Paradox. Denmark: Copenhagen Business School. Strategic Management Journal. 28(4): 409-427.

Andrade, A., and Joia, I.A. 2012. Organisational structure and ICT strategies in the Brazilian judiciary system. Government Information Quarterly. 29(1):532-542.

Androniceanu, A. 2007. Defining the clients in the public sector: A social exchange perspective, Public Administration Review. 63(3):337-346.

Angranoff, R. 2012. Navigating the complexities of collaboration, George Washington University Press, Washington DC. US. Public Administration Review. 74(2):281- 283.

Angranoff, R., and McGuire, M. 2004. Collaborative Public Management: New strategies for Local Government. Washington, DC: Georgetown University Press.

Annual Report. (2001/2002). 2012. Intelligence and Security Committee (UK). CM 5542. Intelligence and National Security. 22(11):14-37.

Annual Report (2014/2015). 2015. RICA Appointed Judge. Pretoria: Government Printers.

284

Asante-Darko, K. (Dr). 2007. What is good governance? South African Institute of International Affairs. www.mg.co.za/article/2007-10-29-what-is-good-governance, Accessed on 28 June 2011.

Asian Development Bank (ADB). 2009. Elements of good governance. www.adb.org/documents/policies/governance/gov, Accessed on 23/09/2009.

Aspin-Brown Commission, 2006. Commission on the Roles and Capabilities of the US Intelligence Community. Commissioned by the United States Congress. Washington, DC: United States Congress.

Auditor-General Annual Report 2008/9. www.AGSA.gov.za, Accessed on 15 June 2011.

Auditor-General Annual Report 2010/2011. www.AGSA.gov.za, Accessed on 15 November 2010.

Australian AID. August 2000: Good Governance- Guiding principles for implementation. www.ausaid.gov.au, Accessed on 27 June 2011.

Azanian People’s Organisation (AZAPO). www.azapo.org.za, Accessed on 25 March 2010.

Babbie, E, Mouton, J., Vorster, P. and Prozesky, B. 2001. The practice of social research. Cape Town: Oxford University Press.

Badenhorst, C. 2008. Dissertation writing: A research journey. Pretoria: Van Schaik Publishers.

Baderin, M.A. 2000. Dialogue Among Civilization as a New Paradigm for Universalism in International Human Rights. Oslo: 13th Annual Meeting of the Academic Council of the United Nations System (ACUNS), 16-18 June 2000. Oslo: ACUNS.

Baiochi, G. 2005. Militants and citizens: The politics of participatory democracy in Porto Alerge. Stanford: Stanford University Press.

Baldino, D. (ed). 2010. Democratic Oversight of the Intelligence Services. Annandale: Federation Press.

Bambers, M., Rugh, J., and Mabry, L.S. 2006. Real World Evaluation. London: Sage Publishers.

Barber, B.R. 1984. Strong Democracy: Participatory Politics for a New Age. San Francisco: University of California. 285

Bartle, P. 2007. Community empowerment: what does it takes to strengthen a community. www.cec.vcn.bc.ca/emp/modules/emp-el.htm, Accessed on 04 March 2014.

Basch, C.E. 1985. Avoiding Type III Errors in Health Education Program Evaluations: A Case Study. Health Education Quarterly. 12(4):315-331.

Baxter, P., and Jack, S. 2008. Qualitative study methodology: Study design and implementation for novice Researchers. The Qualitative Report. 13(4):544-559. www.nova.edu/ssss/QR/QR13-14/baxter/pdf, Accessed on 02 February 2014.

Baynham, S. 1990. Journal of Modern African Studies. Cambridge: Cambridge University Press. Pp.429.

Beauchamp, T. and Childress, J. 1983. Principles of biomedical ethics. (2nd ed.). Oxford: Oxford University Press.

Becker, H.S. 1996. The epistemology of qualitative research. First appeared in Jessor, R., Colby, A., and Shweder, R. 1996. Ethnography and human development: Context and meaning in Social Inquiry. Chicago: University of Chicago Press.

Behling, O. 1980. The case for the natural sciences model for research in organisational behaviour and organisational theory. Academy of Management Review. 5(4):483-490.

Bergh, D.D., Perry, J. and Hanke, R. 2006. Some predictors of SMJ article impact. Journal of Strategic Management. 27:81-100.

Berman, J. and Flint, L. 2003. Guiding Lights: Intelligence Oversight and Control for the Challenge of Terrorism, Commentary, Centre for Democracy and Technology. www.cdt.org/pulications, Accessed on 09 July 2010.

Bernard, H. 2000. Social research methods: Qualitative and quantitative approach. New York: Sage Publications.

Bernhard, D.C. 2003. Competitive Intelligence: Acquiring and Using Corporate Intelligence and Counter-Intelligence. London: Prentice Hall, Financial Times.

Bernhardt, W. 2003. A qualitative conceptual framework for conducting risk and threat assessment in the South African domestic intelligence environment. Unpublished Doctoral Thesis. Pretoria: University of Pretoria.

286

Bertaux, D. 1981. Pathways to social class: A qualitative approach to social mobility. New York: Clarendon Press.

Biggs, S. and Smith, S. 2003. A paradox of learning in project cycle management and the role of organisational culture. World Development. 31(10):1743-1757.

Blank, R. 1993. Developing a system of education indicators: Selecting implementing, and reporting indicators. Educational Evaluation and Policy Analysis. 15(1):65-80.

Blankenship, K.M., Bray, S.I., and Merson, M.H. 2000. Structural intervention in public health. AIDS. 14 (Suppl 1): S11-21.

Blaxter, L., Hughes, C., and Tight, M. 2006. How to research. Maidenhead: Open University Press.

Bless, C. and Higson-Smith, C. 1995. Fundamentals of Social Research: An African Perspective. Kenwyn: Juta Publishers.

Bogdan, R.C., and Biklen, S.K. 1982. Qualitative research for education: An introduction to theory and methods. Boston: Allyn and Bacon Inc.

Bok, S. 1982. Secrets- on the ethics of concealment and revelations, New York: Pantheon Books.

Booth, W., Ebrahim, R., and Morin, R. 2001. Participatory monitoring and evaluation and reporting: An organisational development perspective for South African Non- Governmental Organisations. Braamfontein: Pact Publishers.

Booton, F., and Dearling, A. (eds.). 1980. The 1980s and beyond. New York: Warwichshire.

Bophuthatswana National Security Council Act, 1981 (Act 27 of 1981).

Bophuthatswana Internal Intelligence Services Act, 1982 (Act 25 of 1982).

Born, H. and Caparini, M. 2007. Democratic Control of the Intelligence services: Containing Rogue Elephants. Ashgate: DCAF Publications.

Born, H. and Leigh, L. 2005. Making Intelligence Accountable: Legal Standards and Best Practice for Oversight of Intelligence Agencies. Oslo: Publishing House of the Parliament.

287

Born, H. and Leigh, L. 2005. Making Intelligence Accountable: Legal Standards and Best Practice for Oversight of Intelligence Agencies. Oslo: Publishing House of the Parliament.

Born, H. 2004. Towards effective democratic oversight of intelligence services: Lessons learned from comparing national practices. Quarterly Journal. III(4).

Born, H., and Wills, A. 2012. (ed.). Overseeing intelligence services, a Toolkit. Geneva Centre for the Democratic Control of the Armed Forces. Netherlands, Den Haag: Ministry of Foreign Affairs.

Born, H, and Mdsevage, G.G. 2012. Introducing intelligence oversight. In Born, H, and Wills, A. 2012. Overseeing Intelligence services: A Toolkit. Geneva Centre for the Democratic Control of the Armed Forces. Netherlands, Den Haag: Ministry of Foreign Affairs.

Born, H. 2013. International Intelligence cooperation and accountability. Milton Park: Routledge. Pp.23.

Bornbright, D. 2012. Use of impact evaluation results. The Rockefeller Foundation. www.interaction.org/impact-evaluation.notes, Accessed on 09 May 2014.

Bourdieu, P., and Wacquant, J.D. 1992. An invitation to reflexive sociology. Chicago: University of Chicago Press.

Boyle, B. 2009. Sunday Times. Interview with the former IGI (IGI), Mr Zolile Ngcakani, Sunday Times, 20/05/2009. www.nia.net/OIGI/News, Accessed on 17 November 2010.

Brandell, J.R. 2001. Theory and Practice in Empirical Social Work. New York: Sage.

Brikci, N., and Green, J. 2007. A guide to using qualitative research methodology - London School of Hygiene and Tropical Medicines. London: Medecins Sans Frontiers.

Brin, D. 1998: The Transparent Society: Will Technology Force Us to Choose Between Privacy and Freedom? Massachusetts: Perseus Books.

Brown Commission. 1996. Report of Commission on the Roles and Capabilities of the United States Intelligence Community: Washington DC: Presidential and Congress Publications and Reports.

Brown, D. 2001. Evaluation of political programmes in the Caribbean region: Conceptual problems. Santiago: Chile Publication, Co. 288

Brown, D.C. 1996. Why ask why: patterns and themes of causal attribution in the workplace. Journal of Industrial Teacher Education. 33(4) 47-65.

Bruneau, T.C and Boraz, S.C (ed). 2007. Reforming Intelligence: Obstacles to democratic control and effectiveness. Austin: University of Texas Press.

Bruneau, T.C and Dombroski, K.R. 2001. Reforming Intelligence: The Challenges of Control in New Democracies. International Journal of Intelligence and Counterintelligence. Philadelphia: Taylor and Francis, Inc.

Bryman, A. 2001. Social Research Methods. Oxford: Oxford University Press.

Bryman, A. and Burgess, R.G. (eds.). 2008. Qualitative research. Vol.3. London: Sage. Pp.143-164.

Bryson, J.M., Crosby, B.C. and Middleton Stone, M.M. 2006. The Design and Implementation of Cross-Sector Collaboration: Propositions from the Literature. Special Issue. Public Administration Review. 66:44-45.

Buckmaster, N. 1999. Associations between outcome measurement, accountability and learning for non-profit organisations. International Journal of Public Sector management. 12(2):186-197.

Burgelman, R.A. 1983. A model of the interaction of strategic behaviour, corporate context and the concept of strategy. Academy of Management Review. 8(1):61-70.

Calder, B.J., Phillip, L.W., and Tybout, A.M. 1982. The concept of external validity. Journal of Consumer Research. 9(3):240-244.

Calhoun, C. (ed). 2002. Dictionary of the Social Sciences. Oxford: Oxford University Press.

Cambridge Advanced Learner’s Dictionary. 2006. Cambridge: Cambridge University Press.

Camerer, L. 2001. Protecting Whistle-blowers in South Africa: The Protected Disclosure Act, No 26 of 2000. Occasional Paper No 47-200. Pretoria: Institute of Security Studies.

Campbell, D.T. 1963. From description to experimentation: interpreting trends as quasi- experiments. In problems in measuring change. In Harris, C.W (ed.).1963. Problems in measuring change. Madison: University of Winston Press. Pp.212- 244. 289

Campbell, D.T. 1978. Qualitative knowing in action research. In M. Brenner, P. Marsh, and M. Brenner. (eds.). The social context of method. London: Croom Helm.

Canadian Security Intelligence Service Act of 1984 (Act 1984). www.csis-scrs.gc.ca, Accessed on 26 April 2011.

Canadian Government’s Access to Information Act of 1980 (Act 1980). art.12 (1). www.csis-scrs.gc.ca, Accessed on 26 April 2011.

Caparini, M. 2002. Challenges of control and oversight of intelligence services in a liberal democracy. Geneva Centre for the Democratic Control of the Armed Forces (DCAF): Geneva. October 2002. Geneva: Geneva Centre for the Democratic Control of the Armed Forces (DCAF).

CARE/IIED. 2012. Conference draft: Participatory M&E reflection and learning for community based adaptation: A manual for local practitioners. Available at: www.care.org/sites/default/files/.../CC- 2012-CARE-PMERL-manual.pdf.

Carter, D.L., 2003. Law enforcement intelligence: A guide for state, local, and tribal law enforcement agencies. Washington, DC: USA Department of Justice.

Cavaye, A.L.M. 1996. Case study research: A multi-faceted research approach for information systems. Journal of Information Systems. 6:227-242.

Cawthra, G. 2005. Security Governance in South Africa. Written essay. Johannesburg: Centre for Defence Security Management.

Centre for the Study of Violence and Reconciliation (CSVR). 2010. www.csvr.org.za, Accessed on 26 April 2010.

Chambers, R. 1994. The origins and practice of participatory rural appraisal. World Development. 22: 959-961. Amsterdam: Elsevier.

Chambers, R. 1997. Whose reality counts? Putting the first last. London:Intermediate Technology Publications.

Chandler, A.D. 1962. Strategy and structure: Chapters in the history of the American Industrial Enterprise. Cambridge: MIT Press Publishers.

Charles, C.M. 1988. Introduction to education research. New York: Longman City Press.

290

Charmaz, K. 2006. Constructing grounded theory: A practical guide through qualitative analysis. Thousand Oaks: Sage Publishers.

Chemicals, J., Bayar, S., van der Walt, G. and Fox, W. 1998. The fundamentals of public personnel management. Braamfontein: Juta and Co, Ltd.

Chenail, R.S. 1995. Presenting qualitative data: The qualitative report. 2(3). December 1995. www.nova.edu/ssss/QR/QR2-3/presenting.html, Accessed on 15 January 2014.

Child Fund Zambia. 2012. Participatory M&E- training guide. TEVETA: act cap 288 of the Laws of Zambia. Lusaka: TEVETA.

Christie, K. 2000. The South African Truth Commission. New York: Palgrave

Cilliers, J. 2003. An overview of the Armed Forces of the TBVC countries. Pretoria: Institute for Defence Policy.

Cilliers, J. and Reihardt, M. 1995. About Turn: The transformation of the South African Military and Intelligence. Midrand: Jetline Print.

Cimbala, S.J (ed.). 1987. Intelligence and Intelligence policy in a democratic society. New York: Transnational Publishers Inc.

Citizen-Government Based Monitoring Framework. 2013. Department of Performance, Monitoring and Evaluations, Government Printers. www.presidency-dpme.gov.za, Accessed on 31 August 2013.

CIVICUS- World Alliance for Citizens Participation: Participatory governance. www.civicus.org, Accessed on 22/09/2009.

Cloete, F, Rabie, B and de Coning, C. 2014. Evaluation Management in South Africa. Sun Media, Stellenbosch.

Cloete, F., Moller, V., Dzengwa, S. and Davids, I. 2003. Assessment of Integrated Indicator Framework for Ten Year Policy Review, Report for the Office of the Presidency, Union Buildings, Pretoria: The Presidency. June 2003. Pp. 91.

Cloete, F. 2007. Knowledge management and trust in government: lessons from South Africa, paper presented at the 7th Global Forum on Reinventing Government: Vienna: Building trust in government, 26-29 June 2007.

291

Cloete, F. 2009. Evidence-Based policy analysis in South Africa: Critical assessment of emerging Government-Wide Monitoring and Evaluation System. Journal of Public Administration. 44 (2):293-311.

Cloete, F. 2000. At full speed the tiger cubs stumbled. Pretoria. HSRC.

Cloete, J.J.N. 1994. An Accountable Government and Administration for the Republic of South Africa. Pretoria: Van Schaik.

Cloete, F. and De Coning, C. 2011. Improving Public Policy: from theory to practice. (3rd ed). Pretoria: Van Schaik Publishers.

Cochran, M. 1986. The parental empowerment process: Building on family strengths. In Harris,J. (ed.). Child psychology in action: Linking research and practice. Bookline: Croon Helm Publishers. Pp. 12-33.

Cock, J. 1995. Forging a new army out of old enemies: Women in the South African Military. Women’s studies quarterly. 97-111.

Cole, E., Eppert, K. and Kinzelback, K. 2008. (ed.). Public oversight of the Security Services: A Handbook for civil society organisations. United Nations Development Programme. Also available at http://europeandcis.undp.org/home/list/publications, Accessed on 06 June 2015.

Coliver, S. Hoffman, P. Fitzpatrick, J. and Bowen, S, 1999. Secrecy and Liberty: National Security, Freedom of Expression and Access to Information, Hague: Martinus Nijhoff Publishers.

Combs, S. 2009. Transparency: Crucial to the State’s Financial Well-Being. www.window.state.tx.us/newsinfo/columns/090226-transparency.html, Accessed on 22 September 2009.

Constitution of the Republic of South Africa, 1996. Pretoria: Government Printers.

Cook, T.D. and Campbell, D.T. 1979. Quasi-experimental design: Design and analysis issues for field settings. Skokie: Rand McNally.

Cooley, S., and Mark, K. 2004. Local Development Plan and PMandE: Structure and Methodology: The Urban Institute, Albanian Development Fund Community Work Project II. Available at: www.urban.org/researc-area/economic-development. . Corney, T. 2009. Participation and empowerment: Is youth a form of community development?. Pretoria: University of South Africa Press.

292

Coulibaly, A.K., Keita, M.Y., Konare, R.D. and Aboubacrine, A.A. 2007. Participatory monitoring and evaluation as a means of empowering local government in the region of Mopti. Bamako: Bamako Communications.

Council of Inspectors-Generals on Integrity and Efficiency (CIGIE). 2011. www.ignet.gov/igs, Accessed on 7 April 2011.

Crabtree, B., and Miller, W. (eds.). 1999. Doing qualitative research. (2nd ed). London: Sage Publishers.

Cracknell, B.E. 2000. Evaluating Development Aid: Issues, Problems and Solutions. London: SAGE Publications, Pty. Ltd.

Crate, S.C. 2003. Three Paradigms of Public Administration. Orono: University of Maine.

Crawford, A. 2010. Counting the cost of our descent: Xenophobia in South Africa today, New Routes. 1/2010.

Craythorne, D.L. 2006. Municipal Administration: The Handbook. (6th ed.). Wotton: Juta and Co. Ltd.

Craythorne, D.L. 1997. Municipal administration: A handbook. (4th ed.). Kenwyn: Juta.

Creswell, J.W. 1994. Research design qualitative and quantitative approach. London:Sage Publishers.

Creswell, J.W. 1998. Qualitative inquiry and research design: Choosing among five traditions. Thousand Oaks, CA: Sage Publishers.

Creswell, J.W., and Miller, D.L. 2000. Determining validity in qualitative Inquiry. Theory into practice. 39(3):124-130.

Cronback, L.J. 1975. Beyond the two disciplines of scientific psychology. American Psychologist. 30(2):116-127.

Cronje. 1996. Vlakplaas Security Police Base Commander: Submission to the Truth and Reconciliation Commission: 27/10/1996.

Crouch, M. and Mackenzie, H. 2006. The logic of small samples in interview based qualitative research, Social Science Information. 45(4): 483-499.

293

Crowford, N. 1995. South Africa’s new foreign and military policy. Opportunities and Constraints. Africa Today. 1st and 2nd Quarters: 88-107.

Currie, I. and Klaaren, J. 2002. The Promotion of Access to Information Act, 2002, Cape Town: Benchmark Books, Siber Inc.

Cwele, S. 2008. Address by Dr S Cwele, MP, Minister of Intelligence Service son the Intelligence services Day. Musanda in Pretoria: 28 November 2008.

Cwele, S. 2012. Budget Vote. Pretoria: Department of State Security.

Daily Maverick. 2013. Appreciations of RICA by Intelligence Services. www.dailymaverick, Accessed during August 2013.

Daily Maverick. 2014. Malema and Madonsela are CIA spies. www.dailymaverick.co.za, Accessed during July 2014.

Dassah, M.O., and Uken, E.A. 2005. Stimulating evaluation demand in Africa. Journal of Public Administration. 40(4):733-743.

Datta, P., and Sen, P. 2009. Women in urban local governance: empowerment through participation. Journal of Governance and Public Policy. 4(3): 48-62.

Davies, P. and Mills, A. 1999. “Ethics, empowerment, and ownership’, in J. Quinn and P. Davies (Eds.), Ethics and Empowerment, West Lafayette: Inchnor Business Books.

DCAF Intelligence Working Group. 2002. Intelligence Practice and Democratic Oversight: A Practitioner’s View. Geneva: Geneva Centre for the Democratic Control of the Armed Forces (DCAF), July 2003. Geneva: Geneva Centre for the Democratic Control of the Armed Forces.

DCAF. 2006. Contemporary Challenges for the Intelligence Community. Geneva: Geneva Centre for the Democratic Control of the Armed Forces.

De Jong, B. Platje, W. and Steele, R.D. (ed.). 2003. Peacekeeping Intelligence: Emerging Concepts for the Future, Oakton, Virginia: OSS International Press.

Defence Intelligence Agency (DIA). 2011. www.dia.mil, Accessed on the 7 April 2011.

Denscombe, M. 1998. The Good Research Guide: for small-scale research projects, Buckingham: Open University Press.

294

Den Boer, M. 2012. Conducting oversight. In Born, H, and Wills, A. 2012. Overseeing Intelligence services: A Toolkit. Geneva: Geneva Centre for the Democratic Control of the Armed Forces.

Dentin, N.K and Lincoln, Y.S. (eds.). 1994. Handbook of Qualitative Research. Thousand Oaks: Sage Publishers.

Denzin, N. 1984. The research act. Englewood Cliffs, NJ: Prentice Hall.

Denzin, N.K and Lincoln, Y.S. (eds.). 1994. Handbook of Qualitative Research. California. Thousand Oaks: Sage Publishers.

Desai, S.B. 2005. Solving the interagency puzzle. Policy Review. 129: 57-71.

Department of Co-operative Governance and Traditional Affairs. 2009. State of local government report. Pretoria: Government Printers.

Department of Defence (DOD) Annual Report 2006/7. www.dod.mil.gov.za, Accessed on 09 May 2010.

Department of Performance Monitoring and Evaluations. 2010. Policy Framework on Government-wide Monitoring and Evaluation: Pretoria: Government Printers.

Department of State Security (DSS) Website. www.intelligence.gov.za, Accessed on 30 May 2010.

Dictionary.com. Definition of Intelligence Agency, Available at www.dictionary.reference.com, Accessed on 10 August 2011.

Dietrich, J.H., 2015. Of toothless Windbags, Blind Guardians and Blunt Swords: The Ongoing Controversy about the Reform of Intelligence Services Oversight in Germany. Intelligence and National Security, Routledge Taylor and Francis Group. Also available at: http://dx.doi.org/10.1080/02684527.1017246.

Dimba, M. 2011. Summary on the progress of South Africa’s Protection of Information Bill (2010), an article available in the Open Democracy Advice Centre. www.opendemocracy.org.za, Accessed on 30 March 2011.

Dimond, L. 2005. Democracy, Development and Good Governance: The Inseparable Links: Paper Delivered at the Maiden Annual Democracy and Governance Lecture of the Ghana Center for Democracy and Development. Accra: British Council Hall. 1st March 2005.

295

Dlomo, T.D. 2004. An analysis of Parliamentary Intelligence Oversight in South Africa with specific reference to the Joint Standing Committee on Intelligence. Pretoria: University of Pretoria.

Dobson, D. and Cook, T.J. 1980. Avoiding Type III Errors in Program Evaluation: Results from a Field Experiment. Evaluation and Programme Planning. 3: 269- 276.

Draft, R. L., and Lewin, A.Y. 1990. Can organisation studies begin to break out of the normal science straighjacket? An editorial essay. Organisational Science. (91): 1-9

Du Toit, D., Knipe, A., Van Niekerk, D., Van der Waldt, G., and Doyle A. 2002. Service excellence in governance. Sandown: Heinemann Publishers (Pty) Ltd.

Du Toit, D.F.P., and Van der Waldt, G. 1999. Public Administration Management: The grassroots. Kenwyn: Juta Publishers.

Dube, F.B. 2013. Accountability and oversight of intelligence services in South Africa post 1994. Unpublished Masters Dissertation submitted at School of Public and Development Management. Johannesburg: University of Witwatersrand.

Duke, J.J. 2002. Identifying Congress’s use of the appointment, appropriations, and impeachment powers and GAO investigations, in addition to congressional subpoenas and contempt proceedings. In Fisher, L. Congressional Access to Information. Available at: scholarship.law.duke.edu/cgi/viewcontent/.cgi?article=1172&content=dlj.

Duncan, J. 2017. South Africa paid little attention to two significant reports released by its Parliament in the dying days of 2016. Available at www.dailmarverick.co.za, Accessed on 11 January 2016.

Duvenhage, A. 2002. Political transformation and management. London: Allyn and Bacon Publishers.

Earp, J.A., and Ennett, S.T. 1991. Conceptual Models for Health Education Research and Practice. Health Education Research. 6(2):163-171.

Economist. 2009. Corruption charges against NPA boss Vusi Pikoli. www.economist.com, Accessed during September 2015.

Egger, S.A. 1990. Serial murder. Westport, CT: Praeger.

Eisenhardt, K.M. 1989. Building theories from case study research. Academy of Management Review. 14(4): 532-550. 296

Eisenhardt, K.M. and Graebner, M.E. 2007. Theory building from case: Opportunities and challenges. Academy of Management Journal. 50(1):5-32.

Eisner, E.W. 1991. The enlightened eye: Qualitative enquiry and the enhancement of education practice. New York: Macmillan Publishing Company.

Ekpo, M.U. 1979. Bureaucratic Corruption in Sub-Saharan Africa, Washington DC: University Press of America.

Elahi, K.Q. 2003. Microfinance, Empowerment and Sudra Women in India, Journal of Development in Practice. 13(5): 570-572. Taylor and Francis Ltd. www.jstor.org/stable/4029946, Accessed on 17 February 2014.

Erikson, E.H. 1963. Childhood and society. New York: Norton

Estrella, M., and Gaventa, J. 1997. Who count reality? Participatory monitoring and evaluation: A literature review. IDS working paper 70. Originally prepared for International Institute for Rural Reconstruction (IIRR) in Philippines in November 1997.

Estrella, M. (ed.). 2000. Learning from change: Issues and Experiences in Participatory Monitoring and Evaluation. Brighton: Institute of Development Studies. Also available at: http://www.idrc.ca/EN/Resources/Publications/Pages/IDRCBookDetails.aspx?Pub licationsID=348, Accessed on 09 May 2014.

Ewen, J. 1981. The education and training of youth workers in Australia. Melbourne: Office of Youth Affairs.

Falconer, D.J., and Mackay, D.R. 1999. Ontological problems of pluralist research methodologies. Available at: http://aisel.isworld.org/article.asp?subject_ID=31andPublication_ID=2, Accessed on 25 September 2013.

Fals-Borda, O. 1985. The challenges of social change. London: Sage Publications, Ltd.

Fawcett, S.B., Paine-Andrews, A., Francisco, J.A., Schultz, J.A., Richter, K.P., Lewis, R.K., Williams, E.L., Harris, K.J., Berkley, J.Y., Fisher, J.L. and Lopez, C.M. 1995. Using empowerment theory in collaborative partnership for community health and development, American Journal of Community Psychology. 23(5).

Feagin, J., Orum, A., and Sjobert, G. (eds.). 1991. A case for case study. Chapel Hills, NC: University of North Carolina Press. 297

Ferlie, E., Pettigrew, A., Ashburner, L. and Fitzgerald, L. (eds.). 1996. The new public management in action. Oxford: Oxford University Press.

Filip, V.F. 2006. The Intelligence Phenomenon in a New Democratic Milieu. Romania case study. Published Masters Thesis. Monterey, California: Naval Post Graduate School.

Financial Intelligence Centre Website. 2010. www.fic.gov.za, Accessed on 03 May 2010.

Fischer, A.M. 2009. Towards Genuine Universalism with Contemporary Development Policy, The Hague, Netherlands: Institute of Social Studies.

Fitzgerald, P. Mclennan, A. and Munslow, B. (ed.). 1995. Managing sustainable development in South Africa. Cape Town: Oxford University Press.

Fjeldstad, O.H. 2003. ‘Fiscal decentralization in Tanzania’. Journal of Public Administration, 38(2):4-419.

Fleishman, J.L. and Payne, B.L. 1980. Ethical dilemmas and the education of policy makers, Hastings Center, Institute of Society, New York: Ethics and Life Sciences.

Fleishman, R. 2009. To participate or not to participate? Incentives and Obstacles. In R. O’Leary and I.B Bingham. (eds.). The Collaboration Public Manager: New Ideas for the 21st Century. Washington DC: Georgetown University Press.

Flood, P. 2004. Report of the Inquiry into Australian Intelligence Agencies to Prime Minister Howard, Canberra: Information Technology and Arts, Department of Communications, July 2004.

Fontana, A. and Frey, J.H. 2000. From structured questions to negotiated text. In N.K Denzin and Y.S. Lincolm (eds.). Hand book on qualitative research. (2nd ed.). Thousand Oaks, CA: Sage Publishers.

Fook, J., Ryan, M., and Hawkins, L. 1997. Towards a theory of social work expertise. British Journal of Social Work. 27:399-417.

Fowler, A. 1997. Striking a balance: A guide to enhancing the effectiveness of Non- Governmental Organisations in International Development. London: Earthscan.

Fox, J. 2007. The uncertain relationship between transparency and accountability, Development in practice. 17(4):663-671. 298

Fox, W. and Meyer I. 1996. Public Management Dictionary. Kenwyn: Juta and Co. Ltd.

Fox, W. and Meyer, I.H. 1995. Public Administration Dictionary. Kenwyn:Juta and Co. Ltd.

Fraleigh, D.M. and Tuman, J.S. 2011: Freedom of Expression in the Marketplace of Ideas, Los Angeles: SAGE.

Frederickson, H.G. 1999. The Repositioning of American Public Administration. Political Science and Politics (32:4):701-711.

Frederickson, H.G. and Smith K.B. 2003. The Public Administration Theory Primer, Colorado: Westview Press.

Freire, P. 1972. Pedagogy of the oppressed. Harmondsworth: Penguin Publishers.

Friedman, B.D. 2009. Philosophy of Responsiveness. Georgia: North Georgia College and State University.

Furubo, J.E., Rist, A.C. and Sandahl, R. (eds.). 2002. International atlas of evaluation. New Brunswick in New York: Transaction Publishers.

Gall, M.D. Borg, W.R and Gall, J.P. 1996. Educational research: An introduction. (6th ed.). White Plains, New York, USA: Longman Publishers.

Gandz, J. and Bird, F.G. 1996. The ethics of empowerment, Journal of Business Ethics, Vol. 15(4):383-392. Springer. www.jstor.org/stable/25072762, Accessed on 17 February 2014.

Geldenhuys, J.S. and Knipe, A. 2006. The organisation of government: An introduction, Pretoria: Van Schaik Publishers.

Gerdes, L.I. 2004. Espionage and Intelligence Gathering: Current Controversies, New York: Greenhaven Press.

Gibbert, M., Ruigrok, W., and Wicki, B. 2004. What passes as a rigorous case study, Strategic Management Journal. 29(13):1465-1474.

Gildenhuys, J.S.H. 2004. The philosophy of Public Administration: A holistic approach: An introduction for undergraduate students. Stellenbosch: Sun Press.

Gill, P. and Phythian, M. 2006. Intelligence in an Insecure World. Cornwall: MPG Books Ltd. 299

Gill, P. 2012. Intelligence threat, risk and the challenges of oversight: Intelligence and National Security. 27(2):206-222. London: Routledge, Taylor and Francis Group.

Gill, P. 2007. Evaluating intelligence oversight committees: The UK Intelligence and Security Committee and the ‘war on terror’. Intelligence and National Security. Also available at http://www.tandonline.com/loi/fintw20, Accessed on 17 July 2015.

Gladden, E.N. 1961. An Introduction to Public Administration. London: Staples Press.

Glaser, B. and Strauss, A. 1967. The discovery of grounded theory: Strategies for qualitative research. New York: Aldine Publishing Company.

Glass, G. V. (ed.). 1976. Evaluation Studies Review Annual. Vol.1 Thousand Oaks, California: Sage.

Glasgow, R.E., Vogt, T.M., and Boles, S.M. 1999. “Evaluating the Public Health Impact of Health Promotion Interventions: The RE-AIM Framework.” American Journal of Public Health. 89(9):1322-1327.

Glanz, K., Lewis, F.M. and Rimer, B.K. 1997. Linking Theory, Research, and Practice. In K. Glanz, F.M. Lewis, and B.K. Rimer. (eds.). Health Behavior and Health Education: Theory Research, and Practice. San Francisco: Jossey-Bass.

Gordillo, G. and Andersson, K. 2004. From policy lessons to policy actions: motivation to take evaluation seriously, Journal of Public Administration and Development. 24(4):305-320.

Government Communications Security Bureau. 2011. New Zealand. www.gcsb.govt.nz/about-us/history.html, Accessed on 26 April 2011.

Graham, P. 1998. A presentation to the Racism and Xenophobia Workshop co-hosted by the South African Human Rights Commission and the Greater Johannesburg Metropolitan Council, Braamfontein: Metropolitan Centre, 15 October 1998.

Green, L.W. (ed.). 1977. Evaluation and Measurement: Some Dilemmas for Health Education. American Journal of Public Health. 67:151-156.

Green, L.W., Kreuter, M.W., Deeds, S.G., and Patridge, K.B. 1980. Health Education Planning: A Diagnostic Approach, Mountain View, California: Mayfield.

300

Gregory, S. 1997. Economic Intelligence in the Post-Cold War Era: Issues for Reform. February 1997. www.fas.org/irp/eprint/snyder/economic.htm, Accessed on 12 November 2008.

Gruber, J. and Trickett, E.J. 1987. Can we empower other? the paradox of empowering in the governing of an alternative public school, American Journal of Community Psychology. 15:53-371.

Gruskin, S., Ferguson, L., Alfven, T., Rugg, D. and Peersman, G. 2013. Identifying structural barriers to an effective HIV response: using the National Composite Policy Index data to evaluate the human rights, legal and policy environment, Journal of the International AIDS Society. 16:18000.

Guba, E.G., and Lincoln, Y.S. 1989. Fourth General Evaluation. London: SAGE Publications, Inc.

Guest, G., Bunce, A., and Johnson, L. 2006. How many interviews are enough? An experiment with data saturation and variability. Interdisciplinary Social Sciences. 49:92.

Guest, G., Bunce, A. and Johnson, R. 1993. How many interviews as enough? An experiment with data saturation and variability. Field methods. 18(1):59-82.

Guttentag, M. and Saar, S. (eds.). 1977. Evaluation Studies Review Annual. Vol.2. Thousand Oaks, California: Sage.

Guijt, I. 1999. Participatory monitoring and evaluation for natural resources management and research. International Institute for Environment and Development. University of Greenwich: National Resource Institute Publishers.

Halchin, L., and Kaiser, F. 2012. Congressional Oversight of Intelligence: Current Structure and Alternatives. Washington: Congressional Research Service.

Hammersly, M. 1989. The dilemma of qualitative method: Herbert Blummer and the Chicago tradition. New York: Routledge Publishers.

Hannah, G. O’Brien, K. and Rathmell, A. 2005. Intelligence and Security Legislation for Security Sector Reform. London: Rand Europe.

Harms Commission. 1990. Commission set up by President De Klerk led by Judge Louis Harms to investigate the operations of the Security Police and the CCB, Pretoria: Government Printers.

301

Haskell, K. 2001. Direct Democracy or Representative Government? Dispelling the Populist Myth. Boulder, Colorado: Westview Press.

Hasset, A.R. 2006. Individual, organisational, community empowerment: Applying a community psychology framework to a school development. Unpublished doctoral thesis. Johannesburg: Faculty of Arts, University of the Witwatersrand.

Hatting, J.J. 1986. Owerheidsverhoudinge-‘n Inleiding. Pretoria: University of South Africa,

Hedley, J.H. 1995. The intelligence community: Is it broken? How to fix it? Centre for the study of intelligence, Washington DC: 96 Unclassified.

Herman, M, 2001. Intelligence Services in the Information Age: Theory and Practice, Cornwall: MPG Books Ltd.

Heywood, A. 2008. Essential of UK Politics. London: Palgrave Macmillan. Also available at www.palgrave.com/politics/essentials, Accessed on 09 November 2010.

Hillebrand, C. 2012. The Role of News Media in Intelligence Oversight. Intelligence and National Security. 27(5):687-706.

Himmelman, A.T. 2001. On Coalition and the Transformation of Power Relations: Collaborative Betterment and Collaborative Empowerment, American Journal of Community Psychology. 29:2:277-84.

Holt, P. 1995. Secret Intelligence and public policy, New York: Congressional Quarterly, Inc, 1995.

Houston, J. and Parsons, W.W. 1998. Criminal Justice and the Policy process. Chicago, IL: Nelson-Hall.

Hood, C. 1991. A public management for all seasons. Public Administration. 69(1):3-19.

HSRC Institutional Review. 2003. Human Sciences Research Council Publishers. Cape Town: South Africa.

Human Rights Commission Submission to the Ministerial Review Commission on Intelligence. 2007. www.intelligence.gov , Accessed on 27 August 2010.

Human Rights Commission Website. 2010. www.sahrc.org.za, Accessed on 27 August 2010.

302

Hunter, M. 2011. The Protection of Information Bill. Pretoria News. 18/03/11.

Hutton, L. 2007. Looking beneath the cloak: An analysis of intelligence governance in South Africa, Pretoria: ISS Paper 154. November 2007. Pretoria: ISS.

Hutton, L. 2010. (Researcher- Security Sector Governance Programme: ISS.) Submission from the Institute of Security Studies (ISS) and the Open Society Justice Initiative to the Ad Hoc Committee considering the Protection of Information Bill (B6-2010). Pretoria: ISS.

Hycner, R.H. 1999. Some guidelines for the phenomenological analysis of interview data. Available at: https://philpapers.org/rec/HYCSF. Accessed on 09 November 2010.

Hyden, G. and Braton, M. (eds.). 1993. Governance and Politics in Africa: Colorado: Lynne Rienner.

Ijeoma, E.O.C. 2010. Mainstreaming Government-wide Monitoring and Evaluation Policy in South Africa: An eye on impact assessment. Journal of Public Administration. 45(2):343-360.

IGI and Security (IGIS). 1986. Australia. www.igis.gov.au, Accessed on 19 April 2011.

IGI and Security (IGIS). Australia: Annual Report 2006/7, Parliamentary and Media Liaison. www.igis.gov.ua, Accessed on 19 April 2011.

Inspector-General Act, 1978 (Act 1101 of 1978), USA Act of Parliament. www.ignet.gov.igs, Accessed on 19 April 2011.

Inspector-General of Intelligence and Security Act, 1996. www.IGIS.NZ. Accessed on 10 April 2015.

Intelligence Services Oversight Act, 1994 (Act 13 of 1994). London: United Kingdom Parliament.

Institute of Security Studies (ISS). 2007. Submission to the Intelligence Ministerial Review Commission. 2007. www.iss.org, Accessed on 25 September 2010.

Intergovernmental Relations Framework Act, 2005 (Act 13 of 2005). Pretoria: Government Printer.

International Centre for Integrated Mountain Development (ICIMOD), 2010. Improving Local Governance in the Hindu Kush-Himalayas: Documenting Good Governance

303

Practices and Lessons Learnt, 14-30 June 2010: Also available at: www.icimod.org. Accessed on 29 June 2011.

International Institute of Sustainable Development (ISSD). 2011. www.issd.org. Accessed on 14 April 2011.

International Ombudsman Institute (IOI). 2010. www.law.ualberta.ca/centre. Accessed on 20 November 2010.

Isett, K.R., Mergel, I.A., LeRoux, K., Mischen, P.A. and Rethemeyer, R.K. 2001. Networks in public administration scholarship: Understanding where we are and where we need to go. Journal of Public Administration Research and Theory. 21(1):157-173.

Isingoma, P. and Reddy, P.S. 2006. Decentralised service delivery in Uganda: Issues: challenges and constraints. Journal of Public Administration. 41(2):90-109.

ISO 17799. 2011. Information Classification. Available at: www.17799-news.the- hamster.com/issue09-news1.htm, Accessed on 12 May 2011.

Jackson, A.L. and Brown, M. 2007. Ensuring efficiency, interagency cooperation, and protection of civil liberties: Shifting from a traditional model of policing to intelligence led policing (ILP) paradigm. Criminal Justice Studies. 20(2):111-129.

Johnson, S.M. 1991. Teachers at work: Achieving success in our schools. New York: Basic Books.

Joint Standing Committee of Intelligence (JSCI) Annual Report 2007/8. www.pmg.org, Accessed on 16 November 20101.

Joint Standing Committee of Intelligence (JSCI) Annual Report 2008/9. www.pmg.org, Accessed on 16 November 2010.

Joint Standing Committee of Intelligence (JSCI) Annual Report 2009/2010. www.pmg.org, Accessed on 24 February 2015.

Joint Standing Committee of Intelligence (JSCI) Annual Report 2015/2016. www.parliament.gov.za, Also available at: www.dailymarverick.co.za, Accessed on 11 January 2017.

Joint Standing Committee of Intelligence (JSCI) Report on Oversight and Accountability, 2007. www.pmg.org, Accessed on 11 November 2010.

304

Jordan, J.S. (ed.). 1998. Systems Theories and A Priori Aspects of Perception. Amsterdam: Elsevier Science. Pp. 47-74.

Jones, H. 2011. A guide to monitoring and evaluating policy influence. Overseas Development Institute. www.odi.org.uk, Accessed on 25 August 2013.

Justice, Crime Prevention and Security (JCPS) Cluster. 2007. Statement on the Investigation into the Document entitled ‘Special Browse Mole Report’ 30 July 2007: Issued by Government Communication on behalf of the JCPS Cluster. Available at: www.intelligence.gov.za/mediastatements, Accessed on 21 November 2010.

Kasper, B.M. 2001. The End of Secrecy? Military Competitiveness in the Age of Transparency, Occasional Paper No. 23, Centre for Strategy and Technology. Montgomery:USA: Air War College, Air University, Maxwell Air Force Base.

Kasrils, R. 2005. A speech at the Inauguration Ceremony for the new Archives Building in Musanda titled ‘Quis Custodient Ipsos Custodes’- Who shall Guard the Guardians, Pretoria: 27 September 2005.

Katsui, H. 2007. What is Participatory Research Method?, www.kumpuvuori.net/research/participatory.htm, Accessed on 02 January 2009.

Keke, J. 1993. The Morality of Pluralism, Princeton. New Jersey. Available at: www.questia.com, Accessed on 02 January 2009.

Khampepe Commission of Enquiry. 2006. Mandate and Location of the Directorate of Special Operations (DSO)- formerly known as Scorpions. www.info.gov.za, Accessed on 11 November 2010.

Kin, P.S. 2009. Building Trust in Government by Improving Governance: Searching for a Feasible Way for Developing Countries, Presentation delivered at Park University on 19 March 2009 by Professor Pan Suk Kim- Director of Public Administration of Yonsei University, Korea. www.un.org/intradoc/groups/public, Accessed on 29 June 2011.

Kitzinger, J. and Barbour, R.S. 1999. Introduction: the challenges and promise of focus group, In Barbour, R.S. and Kitzinger, J. (eds.). Developing focus group research: Politics, theory and practice. Thousand Oaks: Sage Publishers. Pp. 1-20.

Klijn, E.H., and Koppenjan, J. 2012. Governance network theory: past, present and future. Policy and Politics. 40(4):187-206.

305

Kocaoglu, N. and Figari, A. 2006. Using the Right to Information as an Anti-corruption tool. Transparency International. Berlin: Open Society Institute. www.transparency.org, Accessed on 1 February 2011.

Kohlbacher, F. 2006. The use of Qualitative Content Analysis in Case Study Research. Forum: Qualitative Social Research. 7(1).

Koneru, I. 2007. E-Governance: Empowering citizens through e-inclusion. Haderabad: The Icfai University Press.

Kotze, H. 1983. Comparing attitudes towards effective governance. A focus on the Northern Province. An unpublished paper delivered during a conference on governing democracy in a rural province, Northern Province.

Kotze, L. 1998. Promoting participation in environmental decision making through the South African courts: myths or reality? Potchefstroom Electronic Law Journal. Cape Town: Juta Publishers.

Krasner, S.D, 1994. Sovereignty: Organised hypocrisy. Princeton University Press.

Krauss, S.E. 2005. Research Paradigms and Meaning Making: A primer. The Qualitative Report. 10(4):758-770. Selangor, Malaysia: University of Putra.

Kuhn, T. 1970. The Structures of Scientific Revolution (With Postscripts). Chicago: University of Chicago Press.

Kuhn, T.S. 1996. The Structure of scientific revolutions. (3rded.). Chicago: the University of Chicago Press.

Kumar, R. 2011. Research methodology. London: Sage Publications, Ltd.

Kusek, J.Z. and Risk, R.C. 2004. Ten steps to a result-based monitoring and evaluation system. Washington DC: The Work Bank.

Kvale, S. 1996. Interviews: An introduction to qualitative research interviewing. Thousand Oaks, California: Sage Publications Inc.

Labuschagne, R. 2002. On South Africa’s Secret Service: An Undercover Agent’s Story. Alberton: Galago Publishers.

Lan, Z. and Anders, K.K. 2000. A paradigmatic view of contemporary public administration research: An empirical test. Administration and society. 32(2):138- 165.

306

Lange, L. and Luescher, T.M. 2003. A monitoring and evaluation system for South African higher education: conceptual, methodological and practical concerns. Pretoria: Council on Higher Education.

Larbi, G.A. 1999. The new public management approach and crisis state. Geneva: United National Research Institute for Social Development.

Lategan L.O.K. 2005. Research, Monitoring and Evaluation in Service Learning: the distinct characteristics of research into service learning. Acta Academica. (Supplementum). Pp.99-115.

Laurence, P. 2005. Red Ronnie’s faith in the holy trinity. Focus. (39). Johannesburg: The Helen Suzman Foundation.

National Intelligence Agency Annual Report for 2002/03 and 2003/4. www.nia.gov.za.

Leedy, P.D. and Ormrod, J.E. 2005. Practical research. New Jersey: Pearson Education Inc.

Leonard-Barton, D. 1990. A dual methodology for case studies: Synergistic use of a longitudinal single site with replicated multiple sites. Organisation Science. 1(3):248-266.

Levy, N. (ed.). 2004. Balancing Secrecy and Transparency in a Democracy - Hefer Commission: the Case Study. Pretoria: South African National Academy of Intelligence.

Lewin, K. 1990. Data collection and analysis in Malaysia and Sri Lanka. In Vulliamy, G., Lewin, K. and Stephens, D. 1990. Doing educational research in developing countries. London: Falmer Press.

Lewis, J. and Ritchie, J. 2003. Generalising from qualitative research. In Ritchie, J. and Lewis, J. (eds.). Qualitative research practice: A guide for social science students and researchers. London: Sage Publishers.

Lincoln, A. 1809. Definition of democracy. www.democracy-building.info/definition- democracy.html, Accessed on 10 August 2011.

Lincoln, Y.S., and Guba, E.G. 1985, Naturalistic Inquiry. Newbury Park: Sage Publishers.

Lind, W.S. Colonel Nightengale, K. Captain Schmitt, J.F. Colonel Sutton, J.W. and Lieutenant Colonel Wilson, G.I. 1989. The changing Face of War: Into the Fourth

307

Generation. Marine Corps Gazette. USA: October 1989. Available at: www.mca- marines.org.

Lindblom, C.E. 1950. Politics and Markets- Pluralistic, Political Science and Economics. New Haven, USA: Yale University.

Linnan, L. and Stecker, A. 2002. Process evaluation for public health intervention and research: an overview. In Stecker, A, Linnan, L. (eds.). Process evaluation for Public Health Intervention and Research. San Fransisco, CA: Jossey-Bass. Pp.1- 23.

Locke, K. 2002. New Zealand’s Intelligence and Security Agencies lack proper control and oversight. www.greends.org.nz, Accessed on 26 April 2011.

Loe, S.J. 2008. Parliamentary Question 11 for oral reply, Published on 19 February 2008, Replied in the National Assembly on 27 February 2008 by Ronnie Kasrils. www.intelligence.org.za, Accessed on 26 November 2011.

Lowenthal, M.M. 2003. Intelligence: From Secrets to Policy. (2nd ed.). Washington DC: CQ Press.

Lustgarten, L., and Leight, I. 1994. In from the cold: National security and parliamentary democracy. Oxford: Clarendon Press.

MacArthur, J. 2012. TREC, Cooperatives and the future of Renewable Electricity in Canada. Alberta, Canada: The Canadian Social Economy Research Partnership.

McLeroy, K.R., Bibeau, D., Steckler, A. and Glanz, K. 1988. An Econological Perspective on Health Promotion Programs. Health Education Quarterly. 15:351- 377.

Mafunisa, M.J. 2000. Public Service ethics. Cape Town: Juta.

Maier, R., Egger, A., Barth, A., Winker, R., Osterode, W., and Kundi, M. 2006. Effects of short and long-term unemployment in physical work capacity and on serum cortisol. International Archives of Occupational and Environmental Health. 79(3):193-198.

Makananisa, T.D. 2011. An investigation into the state of Municipal Service Delivery: A case study of Vhembe District Municipality. Unpublished Masters Dissertation. Johannesburg: University of Witwatersrand.

March, J.G., Sproull, L.S. and Tamuz, M. 1991. Learning from samples of one or fewer. Organisation Science. 2(1):1-13. 308

Martin, R. and Feldman, E. 1998. Access to Information in Developing Countries. Transparency International Working Paper. www.transparency.org/publications, Accessed on 26 January 2011.

Martin, K. 2004. Domestic intelligence and civil liberties. SAIS Review. 24:7-21.

Masiapata, N.M. 2008. Ethics and ethos in the South African Police Service: An overview of the Johannesburg Central Police Station. Saarbrucken: VDM Verlag- Dr Muller Publishers.

Mason, M. 2010. Sample size and saturation in PhD studies using qualitative interviews. Available at: www.qualitative-research.net.

Masuku, T. 2005. Strengthening democratic policing in South Africa: enhancing and coordinating the internal and external accountability systems of the South Africa Police Service. Pretoria: Centre for the Study of Violence and Reconciliation.

Mathews, A.S. 1978. The darker reaches of government, Access to Information about Public Administration in three Societies. Cape Town: Juta and Co. Ltd.

Maurer, A.C. (ed.). 1985. Intelligence: Policy and Process, London: Westview Press.

Mavhivha, E. 2007. Leadership and governance perspectives in local government administration in South Africa: Limpopo Province. Unpublished PhD thesis. Pretoria: Faculty of Economics and Management Sciences. Pretoria: University of Pretoria.

May, T. 2001. Social research: Issues, methods and process. (3rd ed.). Buckingham: Open University Press.

Maynier, D. 2011. The complacency of the JSCF in its oversight work. www.da.org.za, Accessed during July 2014.

Mbeki, T. 2005. Address of the President of South Africa, Thabo Mbeki, at the Intelligence servicesDay, 10th anniversary Awards Ceremony and inauguration of the wall and Garden of Remembrance, Musanda, Tshwane: 24 November 2005. www.dirco.gov.za/docs/speeches/2005/mbek1125.htm, Accessed on 20 October 2013.

McCreary, L.L., Kaponda, C.P.N., Kafulafula, U.K., Ngalande, R.C., Kumbani, L.C., Jere, D.L.N., Norr, J.L. and Norr, K.F. 2010. Process evaluation of HIV prevention peer group in Malawi: a look inside the black box. Health Education Research. 309

25(6):965-978.

McGarrel, E.F., and Schlegel, K. 1993. The implementation of federally funded multijurisdictional drug task force: Organisational structure and interagency relationship. Journal of Criminal Justice. 21:231-244.

McGrath, J.E., and Bringberg, D. 1983. External validity and the research process: a comment on the Calder/Lynch dialogue. Journal of Consumer Research. 10(1):115-124.

McGraw, S.A., and others. 1989. Methods in Program Evaluation: The Process Evaluation System of the Pawtucket Heart Health Program. Evaluation Review. 13:459-483.

McKenzie, K. 2009. Local leadership in falling apart in delivery: The magazine of local government. Cape Town: ABC Press.

Mcmillan, J.H., and Schumacher, S. 1989. Research in education. A conceptual introduction. New York: Wesley Longman.

Mebrahtu, E. 2002. Perceptions and practices of monitoring and evaluation: International Non-Governmental Organisations experiences in Ethiopia. Development in Practice.12 (3/4): 501-517.

Merriam, S. B. 2002. Introduction to Qualitative Research. San Francisco: John Wiley and Sons, Inc.

Merriam, S. B. 2001. Case studies as qualitative research. In Merriam, S.B. 2001. Qualitative research and case study applications in education. San Francisco: Jossey-Bass.

Merriam, S.B. 1998. Qualitative research and case study applications in education. San Francisco: Jossey-Bass.

Merton, R.K. 1987. The focused interview and focus group. Public Opinion Quarterly. 51:550-556.

Merton, R.K., Fiske, M. and Kendall, P.L. 1956. The focused interview. New York: The Free Press.

Metz, A.J.R. 2007. Part1 in a Series on Practical Evaluation Methods: Why conduct program evaluation. Washington, DC: Child Trends - The Atlantic Philanthropies.

310

Meyer, C.H. 1976. Clinical social work in the eco-systems perspective. New York: Columbia University Press.

Michael, J. 1994. Privacy and Human Rights. Dartmouth: UNESCO Publishing.

Michel, J. 2003. Epistemology of measurement: The relevance of its history for quantification in the Social Sciences. Social Sciences Information. 42:515-535.

Miewald, R.D. 1978. Public Administration: A critical perspective. New Jersey: McGraw- Hill, Inc.

Miles, M.B. 1979. Qualitative data as an attractive nuisance: the problem of analysis. Administrative Science Quarterly. 24(4):590-601.

Miles, M.B., and Heberman, A.M. 1994. Qualitative data analysis: An expanded sourcebook. California, Thousand Oaks: Sage Publishers.

Miller, H.T. and Fox, C.J. 1995. Modern Public Administration: towards discourse. New York: Sage Publications, Inc.

Miller, S. 1999. Noble cause corruption’ in policing. The African Security Review. 8(3).

Milward, B.H. and Provan, K.G. 2006. A Manager’s Guide to Choosing and Using Collaborative Networks, Washington, DC: IBM Centre for the Business of Government.

Ministerial Review Commission. 2008. Conducting Intelligence in a Constitutional Democracy. www.intelligence.gov.za, Accessed on 9 November 2008.

Ministry of State Security, Joint Media Statement with the South African National Editor’s Forum (SANEF) on the Protection of Information Bill (12/11/2010). Available at: www.ssa.net, Accessed on 14 March 2011.

Ministry of State Security, Media Statement on the Protection of Information Bill (26/10/2010). www.ssa.net, Accessed on 14 March 2011.

Mkhonto We Sizwe ‘Spear of the Nation’. www.anc.org.za, Accessed on 10 January 2010.

Mohamed, N.A., Talib, N.B.A., Ahmad, M.F.B., Shah, I.B., Leong, F.A. and Ahmad, M.S. 2012. Role of Community participation in building construct in community development. International Journal of Academic Research. 4(1).

311

Molomo, M. and Somelekae, G. 1998. Making a difference: Non-Governmental good governance and service delivery. Pretoria:Juta.

Monare, M. 2004. Spy monitor has no time for political games. Sunday Independent. 15/02/04. www.iol.co.za.

Monare, M. 2006. Rules of the spying game to change. Pretoria News. 04/05/2006. P.5.

Montesquieu, C.L. 1798. L’esprit des Lois (Principles of Laws). www.highbeam.com/doc/1g1, Accessed on 5 November 2010.

Mook, D.C. 1983. In defence of external validity. American Psychologist. 38:379-387.

Morse, J.M. 1994. The significance of saturation. Qualitative Health Research. 5(3):147-149.

Morse, J.M. 1995. Qualitative nursing research. New York: Sage Publishers.

Morton, R.K. 1989. The focussed interview and focus groups: Continuities and discontinuities. Public Opinion Quarterly. American Association of Public Opinion Research. 51: 550-566.

Mubangizi, B.C. and Tshishonga, N. 2013. Political and community oversight for good governance in South Africa: Journal of Public Administration. 48(2):299-320.

Mukherjee, N. 1995. Participatory Rural Appraisal and Questionnaire survey: Comparative Field experience and methodological innovation, concepts publishing, New Delhi. OECD, 2001 evaluation feedback for effective learning and accountability. Paris. www.oecd.org/dataoecd/10/29/2667326.pdf, Accessed on 14 February 2014.

Municipal Financial Management Act of 2003 (Act 56 of 2003). Also referred to as Local Government: Municipal Financial Management Act of 2003 (Act 56 of 2003). Pretoria: Government Printer.

Municipal Structures Act of 1998 (Act 117 of 1998). Also referred to as Local Government: Municipal Structures Act of 1998 (Act 117 of 1998). Pretoria: Government Printer.

Municipal Supply Chain Management Policy, 2003. Pretoria: The Treasury.

Municipals Systems Act of 2000 (Act No 32 of 2000). Also referred to as Local Government: Municipals Systems Act of 2000 (Act No 32 of 2000). Pretoria: Government Printer. 312

Murden, S. 2007. Staying the Course in ‘Fourth Generation Warfare - Persuasion and Perseverance in the Era of the Asymmetric Bargaining War. Contemporary Security Policy.28(1):197-211.

Muthuma, E.W. 2011. Economic cooperation in Kenyan credit cooperatives: exploring the role of social capital and institutions. Unpublished PhD Thesis. Johannesburg: School of Public and Development Management, Wits University.

Naidoo, G. 2004. Leadership and governance for a sustainable public service: The case for selected South African Public Service Organizations. Unpublished thesis submitted in partial fulfilment of the requirements for the degree PhD (Public Affairs). Pretoria: University of Pretoria.

Nasrullah, A.M. 2005. From public administration to new public management: An analysis, Pakistan Journal for Social Sciences. 3(1):197-204.

Nathan L. 2009. Lighting up the Intelligence Community: A Democratic Approach to Intelligence Secrecy and Openness: Global facilitation network for security sector reform. Available at: www.ssrnetwork.net, Accessed on 19 October 2010.

Nathan, L. 2010. Lighting up the Intelligence Community: An agenda for intelligence reform in South Africa. African Security Review. 18(1):91-104.

National Development Plan Vision 2030. 2011. Pretoria: The Presidency.www.gov.za.

National Intelligence Agency (NIA). www.nia.gov.za. Accessed on 1 August 2009.

National Intelligence Service. 1969-1994. 25 Years Celebration Manual, Pretoria: Government Printers.

National Journal (United States of America). 2011. The Promise Audit: Tracking President Obama’s Progress on Campaign Promises. www.promises.nationaljournal.com/defense/create-information-declassification- center/ .Accessed on 12 May 2011.

National Institute of Health. 2001. Qualitative Methods in Health Research: Opportunities and Consideration in Application and Review. Washington, D.C: National Institute of Health. Office of Behavioural and Social Sciences Research.

New Zealand Government Communication Bureau. 2015. www.GCCB.AU.

Ndimande, B.S. 2012. Decolonising research in the post-apartheid South Africa: The politics of methodology. Qualitative Inquiry. 18(3):215-226. 313

Neill J. 2006. Analysis of Professional Literature Class 6: Qualitative Research I. www.wilderdom.com/OEcourses/PROFLIT/Class6Qualitative1.htm

Netshithenzhe, T.E. 2005: Comparative Analysis of the Roles and Functions of the IGI with specific reference to South Africa. Unpublished Masters Dissertation on Security Studies. Pretoria: University of Pretoria.

Neuman, W.L. 2010. Social research methods: Qualitative and Quantitative approaches. (6th ed.). New York: Allyn and Bacon.

Neuman, W.L. 2011. Social Research Methods: Qualitative and Quantitative Approaches. Boston: Pearson Education Inc.

New Zealand IGI and Security Act, 1996 (Act 47 of 1996). Wellington: New Zealand.

New Zealand Intelligence and Security Committee Act, 1996 (Act 46 of 1996). Wellington: Department of Prime Minister and Cabinet. Date of assent. 1 July 1996.

News 24. 02/10/2009. www.news24.co.za, Accessed on 15 October 2009.

Nhlanhla, J.M. 1992. The SABTVC and Liberation Movements: Modalities of combining their Intelligence services. The Strategic Review of South Africa. XIV(2).

Nhlanhla, J.M. 1998. The price of peace is eternal vigilance. The 1998 Budget Speech of the Deputy Minister JN Nhlanhla. Pretoria: Ministry of Intelligence Services:

NIA Annual Report 2003/4. Available at: http://www.nia.gov.za, Accessed on 7 April 2011.

Nothern Province Determination of Types Municipality Act, 2000. Polokwane: Government Printers.

Ntliziywana, P. 2009. Professionalization of Local Government: Legal avenues for enforcing compliance with competency requirements. Unpublished LLM Dissertation. Cape Town: Faculty of Law, University of Western Cape.

Nussbaum, D. 2006. Dead silence: when journalists are muzzled, corruption goes unchecked. Berlin: Transparency International.

Nylen, W.R. 2002. Testing the empowerment Thesis: The participatory Budget in Belo Horizonte and Betim, Brazil, Comparative Politics. 34(2):127-145.

314

Nzewi, O. 2012. Participatory Monitoring and Evaluation of Governance in Africa: Lessons from the civil society African Peer Review Mechanism Monitoring Project in South Africa. Africa Insight. 41(4):36-56.

Nzongola-Ntalaja, G. 2002. UNDP role in promoting good governance. Seminar for the International Guests at the Congress of the Labour Party of Norway, Oslo. www.undp.org/oslocentre/pub.htm, Accessed on 10 October 2013.

O’Brien, K.A. 2001. International and national security. 16(3):27-59. London: Frank Cass Publishers.

O’Brien, K.A. 2003. Controlling the Hydra: An Historical Analysis of South African Intelligence Oversight, “Making Intelligence Accountable: Executive and Legislative Oversight of Intelligence services”. Oslo: September 2003.

Office of the Inspector-General for Intelligence. Available at: www.ssa.gov.za, Accessed on 14 October 2016.

O’Flynn, J. 2007. From new public management to new value: Paradigmatic change and managerial implications. The Australian Journal of Public Administration. 66 (3):353-366.

Oie, S. 2014. Operational barrier elements: Good practices. DNV.GL. Safer, Smarter, Greener.16 October 2014. Available at: www.sintet.no/globalassets/project/hfc/.../10--operationalbarriers-oie--.pdf.

Oleszek, W.J. 2010. Congressional oversight An Overview. American Congressional Research Service. 75700. Available at www.crs.gov, Accessed on 6 June 2015.

Olowu, Dele and Sako, Soumana 2002. IDASA Democracy Index, Pretoria: IDASA Press.

Omoyefa, P.S. 2008. Public sector reform in Africa: A philosophical re-thinking. Africa development. (4):15-30.

Organization for Economic Co-operation and Development. 2004. Security Systems Reform and Governance: Policy and Good Practice. www.oecd.org/dataoecd/19/37318586.pdf Accessed on 7 October 2009.

Organization for Economic Co-operation and Development. 2009. OECD Guiding Principles for Regulatory Quality and Performance. www.oecd.org/dataoecd/19/37318586.pdf. Accessed on 7 October 2009.

315

Oxford Dictionaries Online. 2011. The World’s most trusted dictionaries. www.oxforddictinaries.com, Accessed on 30 June 2011.

Pan African Congress (PAC) Website. www.pac.org.za, Accessed on 11 January 2010.

Parliament Website. www.parliament.gov.za, Accessed on 25 August 2010.

Parliamentary Monitoring Group (PMG). 2007. Operational problems in the Office of the Public Protector. www.pmg.org/minutes . Accessed on 11 November 2010.

Parliamentary Monitoring Group (PMG). 2010. Minister of State Security, Siyabonga Cwele briefing the Ad Hoc Committee on the Protection of Information Bill (b6- 2010). www.pmg.org, Accessed on 30 March 2011.

Parliamentary Monitoring Group (PMG). 2015. Operational modalities of the Joint Standing Committee on Intelligence (JSCI). www.pmg.org, Accessed on 24 October 2016.

Pateman, C. 1970. Participation and democratic theory. Cambridge: Cambridge University Press.

Patton, M. Q. 2002. Qualitative research and evaluation methods. (3rd ed.). Thousand Oaks, CA.: Sage Publications, Inc.

Patton, M.E. 1990. Qualitative Evaluation and Research Methods. (2nd ed.). Newbury Park: Sage Publishers.

Patton, M.Q. 1997. Utilisation focused evaluation: the new century text. (3rd ed.). London: Sage Publishers.

Paul, S. 2002. Holding the State to Account: Citizen Monitoring in Action. Bangalore: Public Affairs Centre.

Peires, J.B. 1992. The implosion of Transkei and Ciskei, African Affairs. The Royal African Society. 91:365-387.

Penrose, E.T. 1960. The growth of the firm. A case study: The Hercules powder company. Business History Review. 34(1):1-23.

Perrin, B. 2012. Linking monitoring and evaluation to impact evaluation. The Rockefeller Foundation. www.interaction.org/impact-evaluation.notes, Accessed on 09 May 2014.

316

Perry, C.L. 1997. The Child and Adolescent Trial for Cardiovascular Health (CATCH): Intervention, Implementation, and Feasibility for Elementary Schools in the United States. Health Education Behaviour. 24(6):716-735.

Peters, B.G. and Pierre, J. 1998. Governance without Government? Rethinking Public Administration. Journal of Public Administration Research and Theory. 8(2):223- 243.

Peters, B.G. and Savoie, D.J. (ed.). 2000. Governance in the Twenty-first Century: Revitalizing the Public Service. Ottawa: Canadian Centre for Management Development: McGill-Queen’s University Press.

Pettigrew, A. 1973. The politics of organisational decision-making. London: Tavistock.

Piccioto, R. 2003. The logic of mainstreaming: a development evaluation perspective. Evaluation. 8(3):322-339.

Piccioto, R. and Wiesner, E. (eds.). 1998. Evaluation and development, New Brunswick: Transaction Publishers.

Plummer, M.L., Wight, D. and Obasi, A.L. 2007. A process evaluation of a school- based adolescent sexual health intervention in rural Tanzania: The MEMA kwa Vijana programme. Health Education Research. 22: 500-12.

Portney, K.E. Thomson, K. and Berry, J.M. 1993. The Rebirth of Urban Democracy. Washington DC: Brookings Institute.

Premchand, A. 1998. Public expenditure management. Washington DC: IMF.

Presidency. 2007. Policy Framework on Government-wide Monitoring and Evaluation. Pretoria: Government Printers.

Presidential Review Commission. 1998. Developing a culture of good governance. Report on the Presidential Review Commission on the reform and transformation of the Public Service in South Africa. www.dpsa.gov.za/docs/reports/presreview/, Accessed on 30 Jun 2009.

Pressman, J., and Waldavsky, A. 1984. Implementation. Los Angeles, CA: University of California Press.

Promotion of Access to Information Act of 2000 (Act 2 of 2000). Pretoria: Government Printer.

317

Public Protector Annual Report 2008/9. www.publicprotector.gov.za, Accessed on 10 November 2010.

Public Protector Annual Report 2009/10. www.publicprotector.gov.za, Accessed on 10 November 2010.

Public Protector Annual Report 2012/2013. www.publicprotector.gov.za Accessed on 1 July 2017.

Public Protector Report on Nkandla 2014. www.publicprotector.gov.za, Accessed on 1 July 2017.

Public Protector Website 2010. www.publicprotector.gov.za, Accessed during November 2015.

Punch, K. 2005a. Introduction to social research: Qualitative and Quantitative approaches. (2nd ed.) London: Sage.

Punch, K.F. 2005b. Developing effective research proposal. Thousand Oaks: Sage Publication.

Ragin, C.C. 1992. Introduction. In Ragin C.C. and Becker H.S (eds.) What is a case? Exploring the foundation of social inquiry. Cambridge: Cambridge University Press.

Rallis, S.F., and Rossma, G.B. 2000. Dialogue for Learning Evaluator as a Critical Friend. Available at: www.onlinelibrary.wiley.com/doi/10.1002/ev.1Rap1174/full.

Rappaport, J. 1981. In practice of paradox: A social policy of empowerment over prevention. American Journal of Community Psychology. 9:1-25.

Rappaport, J. 1984. Studies in empowerment: A social policy of empower over prevention, Prevention in Human services. 3:1-7

Rapport, J. 1987. Terms of empowerment/exemplars of prevention: Towards a theory of community psychology. American Journal of Community psychology. 15(2):121-148.

Rauch, J. 1998. Police reform and South Africa’s transition: a case study written for the Northern Ireland Programme at the Kennedy School of Government. Cambridge, USA: Harvard University.

318

Rauch, J. 2000. Police reform and South Africa’s transition, Paper presented at the South African Institute for International Relations Conference, CSVR: South Africa.

Renney, A. 1971. The Governing of Men. (3rd ed.). Hinsdale, Illinois: Dryden.

Rezmovic, E.L. 1982. Program Implementation and Evaluation Results: A Re- examination of Type III error in a Field Experiment. Evaluation and Program Planning. 5:111-118.

Rhodes, R.A.W. 1997. Understanding Governance. Buckingham: Open University Press.

Ritchken, E. 1995. The Reconstruction and Development Programme (RDP), Governance and Rural Development. In Fitzgerald, P. Mclennan, A. and Munslow, B. (eds.). 1995. Managing sustainable development in South Africa. Cape Town: Oxford University Press.

Roach, K. 2012. Overseeing information sharing. In Born, H. and Wills, A. 2012. Overseeing Intelligence services: A Toolkit. Geneva: Geneva Centre for the Democratic Control of the Armed Forces.

Roberts, B. 2006. Prerequisites for the Successful Implementation of the Promotion of Access to Information Act 2 of 2000. Journal of Public Administration. 41(2.1). July 2006.

Robbins, S. P., Chatterjee, P. and Canda, E.R. 2006. Contemporary human behaviour theory: A critical perspective for social work. (2nd ed.) Boston: Allyn and Bacon.

Rocha, S. 2004. Has anyone said ethics?, Safety of beneficiaries? Some consideration about info gathering in the field, analysis and advocacy unit. London: Medecins Sans Frontiers.

Rodee, C.C. Anderson, T.J. and Christol, C.Q. 1967. Introduction to political science. (2nd ed.). New York: McGraw-Hill.

Roe, M.J. and Siegel, J.I. 2007. Political Instability and Financial Development, Paper presented at the Seminar in Law and Economics, Harvard Law School: Harvard University. www.law.harvard.edu/corgov, Accessed on 01 July 2011.

Rogers, P.J. 2012. Introduction to impact evaluation. The Rockefeller Foundation. www.interaction.org/impact-evaluation.notes, Accessed on 09 May 2014.

319

Roherty, J.M. 1992. State Security in South Africa: Civil-Military relations under P.W Botha, New York: M.E Sharpe Inc.

Rubin, H. and Rubin, I. 1995. Qualitative Interviewing: The art of hearing data. Thousand Oaks: Sage Publishers.

Rupiya, M. (ed.). 2005. Evolutions and Revolutions: A Contemporary History of Militaries in Southern Africa. Pretoria: Lesedi Litho Printers (ISS).

Sangole, N. 2007. The role of community-driven participatory MandE in empowering communities and improving their decision-making: A case study of the Kari/CIAT collaborative project, Coastal Kenya, Unpublished MA Dissertation. Bellville: Institute for Social Development, Faculty of Arts, University of the Western Cape.

Sangster-Gormley, E. 2013. How case study research can help to explain implementation of the nurse practitioner role. Nurse Researcher. 20(4):6-11.

Sangweni, S. and Balia, D. (ed.). 1999. Fighting corruption: towards a national integrity strategy. Pretoria: University of South Africa.

Scandura, T.A. and Williams, E.A. 2000. Research methodology in management: current practices, trends, and implication for future research. Academy of Management Journal. 43(6):1248-1264.

Schneider B.R., and Heredia, B. (eds.). 2003. Reinventing the Leviathan: Politics of administration reform in developing countries. Boulder, Colorado: Lynne Rienner Publishers.

Schwandt, T.A. 2001. Dictionary of qualitative inquiry. (2nd ed.). California: Thousand Oaks and Sage Publishers.

Schwarz Jr, F.A.O. 2007. The Church Committee and a new era of intelligence oversight. Intelligence and National Security. 22:2: 270-297, DOI, Also available in http://dx.doi.org/10.1080/02684520701303881, Accessed on 16 July 2015.

Schwella, E. 1991. Organisational development in Schwella, E., Fox, W. and Wissink, H. (eds.). Public Management. Kenwyn: Juta Publishers.

Schwella, E. 1996. Public resource management. Cape Town: Juta.

Scriven, M. 1991: Evaluation Thesaurus. (4th ed.). Newbury Park, CA: Sage.

Scriven, M. 1967. The methodology of evaluation. In R.W Tyler, R.M Gagne and M. Scriven. (eds.). Perspective of curriculum evaluation. Chicago, IL: Rand McNally. 320

Searle, J.R. 1995. The Construction of Social Reality. New York: Simon and Schuster, USA: Penguin Group.

Security Services Special Account Act of 1969 (Act 81 of 1969). Pretoria: Government Printers.

Security Intelligence Review Committee of Canada (SIRC). 2010. www.sirc-csars.gc.ca, Accessed on 21 July 2010.

Sekeran, U.M.A. 1992. Research methods for business. Skill building approach. New York: John Wiley and Sons.

Sendall, N. 1997. Is All Military Information Necessarily Critical for National Security. African Security Review. 6(2). 1997. www.iss.co.za, Accessed on 12 August 2011.

Serrano-Garcia, I. 1984. The illusion of empowerment: Community development within a colonial context. Prevention in Human Services. 3:173-200.

Shane, P.M. Podesta, J. and Leone C. 2004. A Little Knowledge: Privacy, Security and Public Information after September 11. New York: The Century Foundation Press.

Shulsky, A.N. 1993. Silent Warfare. Understanding the World of Intelligence. (2nd ed.). Washington: Brassey.

Sidani, S., and Sechrest, L. 1996. Analysis and use of qualitative data. Research Monograph 166, Retrieved from www.drugabuse.gov/pdf/monographs/monograph166/292_309.pdf, Accessed on 26 September 2013.

Siggelkow, N. 2007. Persuasion with case studies. Academy of Management Journal. 50(1): 20-24.

Smit, B. and Wandel, J. 2006. Adaptation, adaptive capacity and vulnerability global environmental changes. 16(3):282-292.

Smith, M. 1988. Developing youth work: Informal education, mutual aid and popular practice. Milton Keynes: Open University Press.

Smith, K.G., Carrol, S.J. and Ashford, S.J. 1995. Intra- and Inter-organizational Cooperation: Towards a Research Agenda. ACAD Manage. 38(1): 7-23.

321

Sole, S. Dawes, N. and Tabane, R. 2006. Why is Billy Masetlha still at large?, Published in Mail and Guardian 31 March 2007, Available at Mail and Guardian Online: www.mg.co.za/articlePage.aspx?articleid=268165andarea=/ins.

Solomon, R.C. 1992. Ethics and Excellence: Cooperation and Integrity in Business, Oxford University Press, London.

Source Watch Website. 2011. www.sourcewatch.org/index.php?title=Fourth- generation_warfare, Accessed on 05 May 2011.

South Africa (Republic). 1957. Criminal Procedure Act of 1957 (Act 51 of 1957). Pretoria: Government Printers.

South Africa (Republic). 1967. Suppression of Terrorism Act of 1967 (Act 83 of 1967). Pretoria: Government Printers.

South Africa (Republic). 1969. Public Service Amendment Act of 1969 (Act 87 of 1969). Pretoria: Government Printers.

South Africa (Republic). 1972. Security Intelligence and State Security Council Act of 1972 (Act 64 of 1972). Pretoria: Government Printers.

South Africa (Republic). 1977. Criminal Procedure Act of 1977 (Act 51 of 1977). Pretoria: Government Printers.

South Africa (Republic). 1978. Bureau for State Security Act of 1978 (Act 104 of 1978). Pretoria: Government Printers.

South Africa (Republic). 1982. Protection of Information Act of 1982 (Act 84 of 1982). Pretoria: Government Printers.

South Africa (Republic). 1992. Interception and Monitoring Prohibition Act of 1992 (Act 127 of 1992). Pretoria: Government Printers.

South Africa (Republic). 1994. Human Rights Commission Act of 1994 (Act 54 of 1994). Pretoria: Government Printers.

South Africa (Republic). 1994. IGI Act of 1994 (Act 40 of 1994). Pretoria: Government Printers.

322

South Africa (Republic). 1994. Intelligence Services Act of 1994 (Act 38 of 1994). Pretoria: Government Printers.

South Africa (Republic). 1994. Intelligence Services Control Act of 1994 (Act 40 of 1994). Pretoria: Government Printers.

South Africa (Republic). 1994. Public Service Act of 1994 (Act 35 of 1994). Pretoria: Government Printers.

South Africa (Republic). 1994. White Paper on Intelligence. Pretoria: Government Printers.

South Africa (Republic). 1994. National Strategic Intelligence Act of 1994 (Act 39 of 1994). Pretoria: Government Printers.

South Africa (Republic). 1994. Prevention and Combating of Corruption Services of 1994 (Act 12 of 1994). Pretoria: Government Printers.

South Africa (Republic). 1994. Public Protector Act of 1994 (Act 23 of 1994). Pretoria: Government Printers.

South Africa (Republic). 1995. Auditor-General Act of 1995 (Act 12 of 1995). Pretoria: Government Printers.

South Africa (Republic). 1995. White Paper on the Transformation of the Public Service. Pretoria: Government Printers.

South Africa (Republic). 1995. Police Services Act of 1995 (Act 68 of 1995). Pretoria: Government Printers.

South Africa (Republic). 1996. Constitution of the Republic of South Africa, 1996. Pretoria: Government Printers.

South Africa (Republic). 1996. National Archives and Records Service of South Africa Act, 1996 (Act 43 of 1996). Pretoria: Government Printers.

South Africa (Republic). 1996: White Paper on Defence. Pretoria: Government Printers.

South Africa (Republic). 1997. White Paper on Transforming Public Service Delivery. (Batho Pele White Paper). Pretoria: Government Printers.

South Africa (Republic). 1998. Executive Members’ Ethics Act of 1998 (Act 82 of 1998). Pretoria: Government Printers. 323

South Africa (Republic). 1999. Public Financial Management Act of 1999 (Act 1 of 1999). Pretoria: Government Printers.

South Africa (Republic). 2000. Promotion of Access to Information Act of 2000 (Act 2 of 2000), Pretoria: Government Printers.

South Africa (Republic). 2000. Promotion of Administrative Justice Act of 2000 (Act 3 of 2000). Pretoria: Government Printers.

South Africa (Republic). 2000. Promotion of Equality and Prevention of Unfair Discrimination Act of 2000 (Act 4 of 2000), Pretoria: Government Printers.

South Africa (Republic). 2001. Financial Intelligence Centre of 2001 (Act 38 of 2001). Pretoria: Government Printers. Also available at: http://www.fic.gov.za.

South Africa (Republic). 2002. Intelligence Services Act of 2002 (Act 65 of 2002). Pretoria: Government Printers.

South Africa (Republic). 2002. Intelligence Services Control Amendment Act of 2002 (Act 66 of 2002). Pretoria: Government Printers.

South Africa (Republic). 2002. Regulation of Interception of Communications and Provision of Communication-Related Information Act of 2002 (Act 70 of 2002). Pretoria: Government Printers.

South Africa (Republic). 2004. Public Audit Act of 2004 (Act 25 of 2004). Pretoria: Government Printers.

South Africa (Republic). 2006. Office of the IGI. Executive summary of the final report on the findings of an investigation into the legality of the surveillance operations carried out by the NIA on Mr S Macozoma. Extended terms of reference report on the authenticity of the allegedly intercepted e-mail. Media Briefing: 23 March 2006.

South Africa (Republic). 2007. Minimum Information Security Standards. Pretoria: Government Printers.

South Africa Monitoring and Evaluation Association (SAMAEA). 2013. www.samaea.org.za/samaea-2/phtml, Accessed on 31 August 2013.

South Africa. 1912. Defence Ac of 1912 (Act 13 of 1912). Pretoria: Government Printers.

324

South Africa. 1950. Suppression of Communism Act of 1950 (Act 44 of 1950). Pretoria: Government Printers.

South Africa. 1951. Bantu Authorities Act of 1951 (Act 68 of 1951). Pretoria: Government Printers.

South Africa. 1953. Criminal Law Amendment Act of 1953 (Act 8 of 1953). Pretoria: Government Printers.

South Africa. 1953. Public Safety Act of 1953 (Act 3 of 1953). Pretoria: Government Printers.

South Africa. 1958. South African Police Act of 1958 (Act 7 of 1958). Pretoria: Government Printers.

South Africa. 1971. National States Constitution of 1971 (Act 21 of 1971). Pretoria: Government Printers.

South Africa. 1995. South African Police Services Act of 1995 (Act 68 of 1995). Pretoria: Government Printers.

South Africa. 2010: Protection of Information Bill (B6-2010). Pretoria: Government Printers.

South African Government Website. www.info.gov.za, Accessed on 21 March 2010.

South African Intelligence Workers Code of Conduct of 1995. www.intelligence.gov.za, Accessed on 26/08/2010.

South African Police Code of Conduct of 1997. www.saps.org.za, Accessed on 23/09/2010.

South African Secret Service Ten Year Review Report 2004. Media Briefing 23 March 2006. www.sass.gov.za, Accessed on 23 April 2010.

South African Secret Services (SASS).1990. www.sass.gov.za, Accessed on 25 October 2009.

South Scan, 1990. Civilian Intelligence in the Apartheid South Africa. 16 November 1990, Pp. 325. Available at: www.sass.gov.za.

Stake, R. 2005. Qualitative Case Study. Thousand Oaks, California: Sage Publishing.

325

Stake, R.E. 1995. The art of case study research. Thousand Oaks, California: Sage Publications Inc.

Steckler, A., McLeroy, K.R., Goodman, R.M., Bird, S.T. and McCormick, C. 1992. Towards Integrating Qualitative and Quantitative Methods: An Introduction. Health Education Quarterly. 19(1):1-8.

Steele, R.D. 2002. The New Craft of Intelligence: Personal, Public and Political- Citizen’s Action Handbook for Fighting Terrorism, Genocide, Disease, Toxic Bombs and Corruption. Oakton. Virginia: OSS International Press.

Stevens, S.S. 1975. Psychophysies: Introduction to its perceptual, neural and social prospects. New Brunswick, USA: Transaction Books.

Stephenson, W. 2000. A Man Called Intrepid, p.XVI: The Globe Pequot Press. www.winstonchurchill.org/i4a/pages/index.cfm, Accessed on 15 May 2010.

Stevenson, W. 2008. Information about Canadian spymaster William Stephenson, who set up Great Britain’s spy operations in the Western Hemisphere. www.winstonchurchill.org/i4a/pages/index.cfm, Accessed on 15 May 2010.

Stewart, D.W. and Shamdasani, P.N. 1990. Focus groups: theory and practice. Thousand Oaks, CA: Sage Publishers.

Stillman, R.J. 1980. Public Administration. Cases and Concepts. (2nd ed.) Boston: Houghton Mifflin Company.

Stoker, G. 1998. Governance theory: Five Propositions. International Social Science Journal. 50(1):17-28.

Stone, G.R. 2008. On Secrecy and Transparency: Thoughts for Congress and a New Administration. New York: American Constitution Society for Law and Policy.

Stone, E.J., McGraw, S.A., Osganian, S.K. and Elder, J.P. 1994. Process Evaluation in the Multicentre Child and Adolescent Trial for Cardiovascular Health (CATCH). Health Education Quarterly. (supp.2):1-148.

Strategic Assessment 1996. Instruments of U.S. Power. ,www.maxwell.af.mil/au/awc/awcgate/sa96/sa96ch06.html. Accessed on 20 March 2010.

Strauss, A. and Corbin, J. 1990. Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park: Sage Publications.

326

Struening, E.L., and Guttentag, M. (eds.). 1975. Handbook of Evaluation Research. Thousand Oaks, California: Sage.

Stubbe, D. 2015a. Oversight of Intelligence Services essential in ending political abuse of spies. Available at: www.da.org.za, Accessed on 29 March 2015.

Stubbe, D. 2015b. More oversight needed on State Security. Available at: www.da.org.za. Accessed on 29 March 2015.

Stufflebeam, D.L., and Shinkfield, A.J. 2007. Evaluation Theory, Models and Applications. New York: Jossey-Bass.

Suchman, E.A. 1967. Evaluative Research: Principles and Practice in Public Service and Social Action Programs. New York: Russell Sage Foundation.

Sudit, E.F. 1996. Effectiveness, Quality and Efficiency: A Management Oriented Approach, New York: Kluwer Academic Publishers.

Swift, C. and Levin, G. 1987. Empowerment: An emerging mental health technology. Journal of Primary Prevention. 8:71-94.

Tashakkori, A. and Teddlie, C. 1998. Mixed methods: Combining Qualitative and Quantitative Approaches. Thousand Oaks, California: Sage.

Taut, S. and Brauns, D. 2003. Resistance to evaluation: A psychological perspective. Evaluation. 9(3):247-263.

Tellis, W. 1997. Application of a case study methodology. The Qualitative Report. 13(3) September, 1997. www.nova.edu/sss/QR/QR3-3/tellis2.html, Accessed on 10 March 2014.

Thelen, E. and Smith, L.B. 1994. Dynamics systems theories: Theoretical Models of Human Development. New York: John Wiley and Sons Inc.

Thamm, M. 2015. IGI wanted Thuli Mandonsela MK II. Available at www.dailymarverick.co.za, Accessed on 14 November 2015.

Thomas, G. 2011. A typology for the case study in social sciences following a review of definition, discourse, and structure. Qualitative enquiry. 17(6):511-521.

Thomson, A.M. and Perry, J.L. 2006. Collaboration Processes: Inside the Black Box. Special issue. Public Administration Review. 66: 20-32.

327

Thornhill, C. 2006. Effective Municipal Government and Administration as Preconditions for Efficient Service Delivery. Journal of Public Administration. Conference Proceedings. September 2006.

Thornhill, C. 2009. Intergovernmental Relations: The Case of Co-operative Local Government. Journal of Public Administration. 44(3).

Thornhill, C. and Van Dijk, G. 2010. Public Administration Theory: Justification for Conceptualization. Journal of Public Administration. 45(1.1).

Tsholofelo, T. 2014. Intelligence Democracy Dilemma: A Critical Evaluation of the Intelligence Oversight Regime in Botswana. Botswana, Gaborone: Publisher Unknown.

Thornton, J. 2007. Benchmarking North Korean Economic Policies: The Lessons from Russia and China. Working Paper, UWEC-2007-35. Washington: University of Washington, Department of Economics.

Tight, M. 2010. The curious case of case study: A viewpoint. International Journal of Social Research Methodology. 13(4):329-339: Routledge: Taylor and Francis Group.

Todd, P. and Bloch, J. 2003. Global Intelligence: The world’s secret services today. London: Zed Books.

Transkei (Independent State). 1987. Transkei Intelligence Service and State Security Council Act of1987 (Act 67 of 1987). Bisho.

TRC Report. 2003. The State outside South Africa between 1960 and 1990: Truth and Reconciliation Commission of South Africa report. Volume 2, Chapter 2, 1993. Available at: www.justice.gov.za/trcreport/finalreport/volume%204.pdf.

Tuset, G.D. 2014. Operational Barriers Elements in Critical Drilling Operations: Understanding Failure Mechanisms in the Primary Well Barrier. Department of Maritime Technology: Denmark: Norwegian University of Science and Technology.

United Kingdom (UK). 1994. The Intelligence Services Act, 1994 (Act c.13 of 1994), London: UK Government Printers.

United Nations Commission on Human Rights (UNCHR). 1996. Johannesburg Principles on National Security, Freedom of Expression and Access to Information, Article 19, London. www.article19.org/pdf, Accessed on 01 February 2011. 328

United Nations Department of Economic and Social Affairs. 2007. Public Governance Indicators: A Literature Review. New York: United Nations Publications.

United Nations Department Programme (UNDP). 2011. Transparency and Accountability, New York: United Nations Publications. www.undp- pogar.org/governance/transparency, Accessed on 12 February 2011.

United Nations Development Programme. 1997. Governance for Sustainable Human Development Policy Paper. www.magnet.undp.org/policy/chapter1.htm, Accessed on 12 January 2014.

United Nations Development Programme. 1997. Governance for Sustainable Human Development Policy Paper. www.hmagnet.undp.org/policy/chapter1.htm. Accessed on 12 March 2010.

United Nations Economic and Social Commission for Asia and the Pacific. 2009. Poverty Reduction Section, Bangkok. www.unescap.org/pdd, Accessed on 25/09/2009.

United Nations Universal Declaration of Human Rights of 1948. Adopted and proclaimed by General Assembly Resolution 217 A (III) of 10 December 1948. www.lexmercatoria.org, Accessed on 09 April 2014.

Van der Merwe, C.G. and Du Plessis, J.E. 2004. Introduction to the Law of South Africa. The Hague: Kluwer Law International.

Van der Merwe, H. 1996. The research process: problem statement and research design. Pretoria: Van Schaik.

Van Niekerk, D. 2001. The State and government. In van Niekerk, D., van der Waldt, G. and Jonker, A. Governance, politics and policy in South Africa. Cape Town: Oxford University Press.

Van Niekerk, D. Van Der Waldt, G. and Jonker, A. 2001. Governance, politics and policy in South Africa. Cape Town: Oxford University Press.

VanWynsberghe, R. and Khan, S. 2007. Redefining case study: International Journal of Qualitative Methods. 6(2):75-89.

Venda (Independent State). 1988. Venda National Intelligence Service Act of 1988 (Act 31 of 1988). Thohoyandou. 329

Vishwanath, T. and Kaufman, D. 1999. Towards Transparency in Finance and Governance, World Bank Policy Research Working Paper, September 1999. www.infor.worldbank.org/etools/docs/library/18291/tarawish.pdf, Accessed on 26 January 2011.

von Bertalanffy, L. 1968. General System Theory: Foundations, Development and Applications. New York: George Braziller.

Vyas-Doorgapersad, S. 2011. Paradigm shift from new public administration to new public management: theory and practice in Africa. The Journal of Trans Disciplinary Research in Southern Africa. 7(2):235-250.

W.K. Kellog Foundation. 2012. Logic Model Development Guide. www.wkkf.org/knowledge-centre/resources/2006/02/WK-kelloggFoundation- Logic-Model-Development-Guide.aspx, Accessed on 09 May 2014.

Wageningen International Centre for Development Innovation. 2010. The Australian process model for sustainable regional development. www.portals.wi.wu.nl/ppme/?process model. Accessed on 21 August 2013.

Wakefield, J.C. 1996. Does social work need the eco-systems perspective? Part 1. Is the perspective clinically useful? Social Service Review. 70(1):1-32.

Waldo, D. 1994. In Rosenbloom, D.H. Goldman, D.D. and Ingraham, P.W. 1994. Contemporary Public Administration. New York: McGraw-Hill Inc.

Walker, S. and Katz, C.M. 2001. The police in America. New York: McGraw-Hill.

Wampler, B. 2007. Participatory budgeting in Brazil: Contestation, cooperation, and accountability. University Park: Pennsylvania State University Press.

Webster, W. 1988. The role of intelligence in a free society. Miami, US: University of Miami.

Webster Online Dictionary. 2014. www.webster-dictionary.org/, Accessed on 05 March 2014.

Weick, K. 2007. The generative properties of richness. Academy of Management Journal. 50(1):4-19.

Weller, R.R. 1997. Comparing Western IGI and Security. International Journal of Intelligence and Counter Intelligence, Pp. 383-406.

330

Weller, G. 1999. Oversight of Australia’s Intelligence services, International Journal of Intelligence and Counter Intelligence. 12(4): 500.

Welman, J.C. and Kruger, S.J. 1999. Research methodology for the business and administrative sciences. Johannesburg: International Thompson.

White Paper on Local Government, 1998. Pretoria: Government Printers.

White Paper on Intelligence, 1994. Pretoria: Government Printer.

White Paper for Defence, 1996. Pretoria: Government Printers.

White Paper on the Transformation of the Public Service, 1997. Pretoria: Government Printers.

White, H. 2008. “Of profits and participation: The use of mixed methods in quantitative impact evaluation”. Working Paper no.7. Network of networks on Impact evaluation (NONE). Available at: www.worldbank.org/ieg/honie/docs/wp7_white.pdf, Accessed on 12 January 2014.

Wholey, J.S.H., Harry, P. and Newcomer, K.E. 2004. Handbook of Practical Program Evaluation. (2nd ed.). San Francisco: Jossey-Bass.

Wikipedia. 2009. Available at: http://en.wikipedia.org/wiki, Accessed on 29 December 2009.

Wikipedia. 2011. Available at: http://en.wikipedia.org/wiki, Accessed on 29 December 2009

William, C.A. 1993. Intelligence support to the UN Peacekeeping Operations. Washington D.C: Industrial College of the Armed Forces. Pp.30.

Windsor, R.A., Baranowski, T., Clark, N. and Cutter, G. 1984. Evaluation of Health Promotion and Education Programs. Mountain View, California: Mayfield.

Wolff, P.H. 1987. The development of behavioural states and the expression of emotions in early infancy: New proposals for investigation. Chicago: University of Chicago Press.

Wood, G. and Mills, G. 1992. The present and future role of the Transkei Defence Force in a changing South Africa. Journal of Contemporary African Studies. (2):225-269.

331

Woosley, T.E. 1996. Intelligence support for counter-proliferation. Research Report. Carlisle, USA: Carlisle Barracks, PA, Army War College. Pp. 40

World Bank. 2009. Good governance. www.worldbank.org, Accessed on 6 October 2009.

World Bank. 2010. Participatory Monitoring and Evaluation. Topics: Participatory and civil engagement. web.worldbank.org, Accessed on 15 July 2013.

World Bank. 1994. Governance: the World Bank Experience. In Born, H. et al. 1994. ‘Parliamentary Oversight’, p.23. Available at: www.worldbank.org, Accessed on 6 October 2009.

www.sssa.gov.za, State Security Agency. Pretoria: Government Printer. Also available at: www.sssa.gov.za.

Wynne, K. 2009. (Canadian Minister of Education). Realizing the Promise of Diversity: Ontario’s Equity and Inclusive Education Strategy. www.edu.gov.on.za/eng, Accessed on 29 June 2011.

Yawa, H, 2001. Northern West MEC for Local Government and Housing: Keynote address at the South African Local Government Association (SALGA) National Assembly, Mafikeng Civic Centre. www.info.gov.za/speeches/index.htm, Accessed on 20-09-2012.

Yin, R.K. 1994. Case study research: Design and methods. (2nd ed.). Thousand Oaks, California: Sage Publishing.

Yin, R.K. 2003. Case study research: Design and methods. (3rd ed.). Thousand Oaks: Sage Publishers.

Yin, R. K. 2009. Case Study Research: Design and Methods. Applied Social Science Research Methods Series - Volume 5. (4th ed.). Thousand Oaks, California: Sage Publishing.

Zikhali, L.J. 2005. The Ethics of Transparency in the Public Sector, Unpublished Masters Dissertation. Mhlathuze: University of Zululand.

332