The Effectiveness of Influence Activities in Information Warfare

Cassandra Lee Brooker

A thesis submitted in fulfilment of the requirements for the degree of Master of Research

School of Business May 2020 Thesis/Dissertation Sheet

Surname : BROOKER Given Name/s : CASSANDRA LEE Abbreviation for degree : MRes Faculty : UNSW Canberra School : School of Business Thesis Title : The Effectiveness of Influence Activities in Information Warfare

Abstract

Rapid, globalised power shifts, technological advances, and increasingly interconnected, ungoverned communications networks have resulted in the rise of asymmetric grey zone threats. The lines are now blurred between political, civil, and military information environments. The rise of influence activities is the new ‘sharp power’ in information warfare (the iWar). Western democracies are already at war in the information domain and are being out-communicated by their adversaries.

Building on the commentary surrounding this contemporary threat, and based on a review of the literature across three academic disciplines of: Systems Thinking, Influence, and Cognitive Theory; this study aimed to investigate solutions for improving Australia’s influence effectiveness in the iWar. This study asked how systems thinking can offer an effective approach to holistically understanding complex social systems in the iWar; as well as asking why understanding both successful influencing strategies and psychological cognitive theories is central to analysing those system behaviours.

To answer the aim, a systems thinking methodology was employed to compare two contrasting case studies to determine their respective influencing effectiveness. The successful case system comprising the terrorist group ISIS was compared and contrasted with the unsuccessful case system of Hillary Clinton’s 2016 election campaign – using a single stock of influence to determine relevant reinforcing and balancing feedback.

The findings validated the utility of systems thinking analysis for holistically understanding complex iWar systems, and revealed why the case systems were effective or not in raising influence stocks, dominating the iWar, and manipulating cognitive behaviour. The results highlighted the configurational, behavioural, and causal factors contributing to influence effectiveness and were summarised into key themes for each of the research disciplines, e.g. a successful system has control, resilience, a centrality of focus, and strong communications links.

Based on the results, this study concluded with a number of recommendations including: having a resonant strategic narrative and cohesive communications strategy, turning democratic vulnerabilities into strengths, adopting systems thinking approaches, enhancing critical thinking, exploiting civilian capabilities, and regaining control over the media. Further research is required across all three academic disciplines to enhance understanding and resilience, refine approaches, and improve the effectiveness of Australia’s future iWar strategy.

Declaration relating to disposition of project thesis/dissertation I hereby grant to the University of New South Wales or its agents a non-exclusive licence to archive and to make available (including to members of the public) my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known. I acknowledge that I retain all intellectual property rights which subsist in my thesis or dissertation, such as copyright and patent rights, subject to applicable law. I also retain the right to use all or part of my thesis or dissertation in future works (such as articles or books).

………………………………………...... …30 May 2020…... Signature Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years can be made when submitting the final copies of your thesis to the UNSW Library. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

2 | P a g e 5249168 Cassandra Brooker Masters of Research ORIGINALITY STATEMENT

‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed ……………………………………………......

Date …………...... 30 May 2020..……………………......

3 | P a g e 5249168 Cassandra Brooker Masters of Research INCLUSION OF PUBLICATIONS STATEMENT

UNSW is supportive of candidates publishing their research results during their candidature as detailed in the UNSW Thesis Examination Procedure.

Publications can be used in their thesis in lieu of a Chapter if:  The candidate contributed greater than 50% of the content in the publication and is the “primary author”, ie. the candidate was responsible primarily for the planning, execution and preparation of the work for publication  The candidate has approval to include the publication in their thesis in lieu of a Chapter from their supervisor and Postgraduate Coordinator.  The publication is not subject to any obligations or contractual agreements with a third party that would constrain its inclusion in the thesis

Please indicate whether this thesis contains published material or not:

☒ This thesis contains no publications, either published or submitted for publication

Some of the work described in this thesis has been published and it has been ☐ documented in the relevant Chapters with acknowledgement

This thesis has publications (either published or submitted for publication) ☐ incorporated into it in lieu of a chapter and the details are presented below

CANDIDATE’S DECLARATION I declare that:  I have complied with the UNSW Thesis Examination Procedure  Where I have used a publication in lieu of a Chapter, the listed publication(s) below meet(s) the requirements to be included in the thesis. Candidate’s Name Signature Date (dd/mm/yy) Cassandra Lee Brooker 30/05/20

4 | P a g e 5249168 Cassandra Brooker Masters of Research COPYRIGHT STATEMENT

‘I hereby grant the University of New South Wales or its agents a non-exclusive licence to archive and to make available (including to members of the public) my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known. I acknowledge that I retain all intellectual property rights which subsist in my thesis or dissertation, such as copyright and patent rights, subject to applicable law. I also retain the right to use all or part of my thesis or dissertation in future works (such as articles or books).’

‘For any substantial portions of copyright material used in this thesis, written permission for use has been obtained, or the copyright material is removed from the final public version of the thesis.’

AUTHENTICITY STATEMENT ‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis.’ Table of Contents

THESIS PART ONE – THE RESEARCH PROBLEM ...... 9 CHAPTER 1 – INTRODUCTION ...... 10 Introduction ...... 10 Information Warfare (iWar) Contextualised ...... 11 The Problem ...... 13 Research Questions ...... 14 Significance of Research ...... 15 Scope of Research ...... 15 Thesis Overview ...... 16 Conclusion ...... 17 CHAPTER 2 – LITERATURE REVIEW ...... 19 Introduction ...... 19 Part One: Systems approaches to the iWar ...... 20 Systems thinking theory ...... 21 Systems thinking about complex social case studies ...... 23 The benefit of thinking in systems for the iWar ...... 27 Systems Summary ...... 30 Part Two: Analysing the effectiveness of influence activities ...... 31 Why certain influence activities are effective ...... 32 Limitations of existing influence theory research and counter-IO approaches ...... 38 How influence activities impact on system behaviour ...... 44 Effectiveness Summary ...... 47 Part Three: Behavioural theory impacts on stocks and flows of influence ...... 47 Behavioural theories on influencing audiences ...... 48 How behavioural theories apply to systems analysis...... 53 Behavioural Summary ...... 56 Conclusion ...... 56 CHAPTER 3 – METHODOLOGY AND CASE STUDY SELECTION ...... 59 Introduction ...... 59 Part One: Case Study Approach ...... 61 Case Study Selection ...... 63 Part Two: Research Methodology ...... 67 Systems Analysis ...... 68 Methods of data collection and analysis ...... 70 Internal vs External Validity ...... 74 Practical Demonstration of Methodology ...... 75 Conclusion ...... 76

5 | P a g e 5249168 Cassandra Brooker Masters of Research THESIS PART TWO – COMPARATIVE ANALYSIS OF SYSTEM EFFECTIVENESS ...... 78 Introduction ...... 78 CHAPTER 4 – SYSTEM ANALYSIS OF INFLUENCE EFFECTIVENESS: ISIS ...... 80 Introduction ...... 80 ISIS: Antifragile, anticipatory system ...... 80 System Elements ...... 82 Internal Sub-systems ...... 84 External Systems ...... 87 Linkages / Relationships...... 89 Reinforcing feedback loops ...... 94 Sense of Community (R1) ...... 94 Sense of Purpose / Belief in Cause (R2) ...... 97 Success-to-the-Successful (R3) ...... 99 Exposure (R4)...... 102 Balancing feedback loops...... 104 Counter-Influence (B1) ...... 104 Loss of authority / control (B2) ...... 108 Drift to Low Performance (B3) ...... 110 Delayed Feedback ...... 112 Conclusion ...... 112 CHAPTER 5 – SYSTEM ANALYSIS OF INFLUENCE INEFFECTIVENESS: CLINTON CAMPAIGN ...... 115 Introduction ...... 115 Clinton Campaign: Robust system suffering some fragility ...... 116 System Elements ...... 117 Internal Sub-systems ...... 118 External Systems ...... 120 Linkages / Relationships...... 128 Reinforcing feedback loops ...... 131 Counter-Influence (R1) ...... 131 Loss of Control and Authority (R2) ...... 135 Drift to Low Performance (R3) ...... 139 Balancing feedback loops...... 142 Sense of Purpose (B1) ...... 142 Credibility (B2) ...... 145 Sense of Community (B3) ...... 148 Delayed Feedback ...... 150 Conclusion ...... 151

6 | P a g e 5249168 Cassandra Brooker Masters of Research CHAPTER 6 – COMPARATIVE ANALYTICAL FINDINGS ...... 153 Introduction ...... 153 Part One: Comparative Systems Analysis Findings ...... 153 System Characteristics...... 154 System Flows ...... 155 Control ...... 157 Centrality of Focus ...... 159 Part Two: Findings Related to Influence Effectiveness ...... 161 Narrative ...... 161 Emotion ...... 163 Authenticity ...... 164 Community ...... 165 Inundation ...... 167 Conclusion ...... 169 THESIS PART THREE – ANALYTICAL CONCLUSIONS ...... 171 Introduction ...... 171 CHAPTER 7 – DISCUSSION ...... 173 Introduction ...... 173 Part One: How much do we know? ...... 173 Influence Lessons ...... 175 Cognitive Behaviours Lessons ...... 178 Complex Social Systems Lessons ...... 180 Part Two: How has the research answered the gaps? ...... 181 Theoretical ...... 182 Methodological – Systems Thinking ...... 185 Practical ...... 187 Part Three: What does this mean for Australia? ...... 194 Challenges ...... 196 Future Research ...... 203 Conclusion ...... 206 CHAPTER 8 – CONCLUSION ...... 208 Introduction ...... 208 Conclusions ...... 209 Influence Effectiveness Conclusions ...... 209 Behavioural Economics Conclusions ...... 210 Systems Thinking Conclusions ...... 211 Recommendations ...... 213 The Contribution of this Research...... 215 Conclusion ...... 217

7 | P a g e 5249168 Cassandra Brooker Masters of Research REFERENCES ...... 219 APPENDICES ...... 230 Appendix One: Glossary and Acronyms ...... 231 Appendix Two: Key concepts from systems thinking literature ...... 236 Appendix Three: Key findings from systems literature relating to case studies ...... 237 Appendix Four: Key findings from systems literature relating to the iWar ...... 238 Appendix Five: Linkages between the three research disciplines’ literature ...... 239

List of Figures: Figure 1: The Iceberg Model of Systems’ Leverage (Source: Academy for Systems Change) Figure 2: Example of a one stock system with feedback loops – Army PSYOPs Figure 3: Indicative characteristics of systems targeted by influence activities Figure 4: Six Sources of Influence (Source: Patterson et al. 2007:78) Figure 5: Overlapping theory between the three literature disciplines reviewed Figure 6: Flow diagram of the Methodological Process Figure 7: System Model demonstrating the methodology Figure 8: Example of feedback loops maintaining stock level (Source: Systems and Us 2014) Figure 9: ISIS’ Influence Centre of Gravity (COG) Analysis Figure 10: ISIS Baseline System Model – key elements and concepts Figure 11: ISIS Reinforcing Feedback Loop (R1): Sense of Community Figure 12: ISIS Reinforcing Feedback Loop (R2): Sense of Purpose Figure 13: ISIS Reinforcing Feedback Loop (R3): Success to the Successful Figure 14: ISIS Reinforcing Feedback Loop (R4): Exposure Figure 15: ISIS Balancing Feedback Loop (B1): Counter-Influence Figure 16: ISIS Balancing Feedback Loop (B2): Loss of Control Figure 17: ISIS Balancing Feedback Loop (B3): Drift to Low Performance Figure 18: Holistic ISIS System Diagram Figure 19: Clinton Campaign Centre of Gravity (COG) Analysis Figure 20: Clinton Campaign Baseline System Model – key elements and concepts Figure 21: Clinton Campaign Reinforcing Feedback Loop (R1): Counter-Influence Figure 22: Clinton Campaign Reinforcing Feedback Loop (R2): Loss of Control Figure 23: Clinton Campaign Reinforcing Feedback Loop (R3): Drift to Low Performance Figure 24: Clinton Campaign Balancing Feedback Loop (B1): Sense of Purpose Figure 25: Clinton Campaign Balancing Feedback Loop (B2): Credibility Figure 26: Clinton Campaign Balancing Feedback Loop (B3): Sense of Community Figure 27: Holistic Clinton Campaign System Diagram Figure 28: Links and findings of the three research disciplines

List of Tables: Table 1: Key Systems Terms and Definitions (Source: Meadows 2008) Table 2: Comparison of systems thinking with current linear thinking approaches in the iWar Table 3: Summary of Key Findings in the Influence Literature of Part Two Table 4: Comparison of Western democratic limitations with adversary freedom of action in IO Table 5: System 1 and System 2 Heuristics Table 6: Linkages between the Influence literature of Part 2 and Behavioural literature of Part 3 Table 7: Definitions of key methodological terms used in this thesis Table 8: Significant systems characteristics of the two case studies Table 9: Definitions of analytical techniques (Source: Hall & Citrenbaum 2010) Table 10: ISIS Reinforcing Feedback Loops Table 11: ISIS Balancing Feedback Loops Table 12: Trump’s iWar Tactics Table 13: Clinton Campaign Reinforcing Feedback Loops Table 14: Clinton Campaign Balancing Feedback Loops 8 | P a g e 5249168 Cassandra Brooker Masters of Research Thesis Part One – The Research Problem

Intentionally Blank

9 | P a g e 5249168 Cassandra Brooker Masters of Research Chapter 1 – Introduction

Superiority in the physical environment is of little value - unless it can be translated into an advantage in the information environment.

Sir Lawrence Freedman The Transformation of Strategic Affairs (2006)

Introduction A combination of rapid globalised change, technological advances in communications and international networks; alongside urbanisation, transnational movement, and population growth has seen an increasingly interconnected, accessible world; a concerning rise in grey zone threats and power being devolved to lesser nations, fringe groups, and individuals. This has led to the rise of influence activities as the new ‘sharp power’ in information warfare (iWar) and a blurring of lines between political, societal, and military information environments (Walker et al. 2018). As Clapper (2018:81) explains, internet and global telecommunications companies have rapidly erased the lines between East and West and between innocent civilians and government agent.

Western democracies are already at war in the information domain and, in being slow to react, are being thoroughly out-communicated by their adversaries (IWD 2018). While Russia’s ‘active measures’ are not new, it, along with tech-savvy groups such as the Islamic State of Iraq and al- Sham (ISIS), have managed to exploit the opportunities presented by this new world order and hybrid battlespace. Adversary information warfare tactics are difficult to counter or even track effectively. This is due to their unconventional methods, rapid widespread dissemination, and highly networked, less hierarchical, and less institutional iWar system structures. Adversary influence activities impact on democratic processes and sovereignty, disrupt populations’ expression of political will, and undermine collective wisdom. The iWar is cheaper than conventional war, is real, effective, and its impacts are immediate.

The West has been complacent, reactive, and risk averse in dealing with this new frontier of hybrid political warfare. Bienvenue et al. (2018b) explain that Australia’s strategic environment is rapidly changing and demonstrating fundamental shifts in power. The way conflicts are now contested has led to an era of hyper-competition in a multi-domain, social media-saturated environment. The West’s adversaries know that to “compete on the traditional playing field of conventional warfare is disadvantageous” (Bienvenue et al. 2018b) and therefore, have asymmetrically dislocated Western nations’ dominant military and political power through the use of cognitive warfare tactics.

10 | P a g e 5249168 Cassandra Brooker Masters of Research Chapter 1 introduces this research into the reasons why certain influence activities prove effective in the iWar. Part One of this chapter sets the context of the iWar problem for Western democratic states like Australia, by providing a brief history and background into contemporary threats facing Australian society in the global information environment (IE). Part Two outlines the problem as it pertains to Australia, details the aim and scope of this research, and the significance of the research gaps and associated findings. Part Three concludes with an overview of the thesis by chapter, before linking with the literature review into the three research disciplines of: systems thinking, influencing activities and behavioural economics contained in Chapter 2.

Information Warfare (iWar) Contextualised Imagine a world where hate wins, where authoritarian regimes and terrorists control the narrative, and where one no longer knows what is true and who to trust. Unfortunately, such circumstances are already occurring across the globe, Australia is unwittingly at war right now. Fear, lies, and outrage now permeate all aspects of society – creating hate, division and conflict. Disinformation already dominates all forms of media and effectively ruins economies, undermines democracy, dumbs-down populations, and erodes national security. Every internet-connected individual is unconsciously embroiled in global information warfare. Western democratic nations are already losing in this iWar, being out-communicated by their adversaries in cyberspace (Brown 2019; Tucker 2016).

Australia has not faced a threat of such proportions since World War II. Information age threats, now bypass physical sovereign borders, poison public debate, and attack government, industry and society (IWD 2018:7). These threats are seemingly unstoppable, usually completely legal, and have blurred the lines between politics and war. Information age threats are enhanced and accompanied by sophisticated manipulation of strategic communications and public opinion. Australian voters, businesses and political discourse have already been targeted. For its part, the Australian Defence Force (ADF) has been largely left to deal with this problem in isolation. The ADF’s stove-piped, doctrinal, linear approaches are unsuited to addressing the unstructured, evolving, complex social problems of the iWar.

As this goes, most Western influencing efforts are led by military units in a bounded battlespace and often lack coordination with other government or civilian agencies, despite doctrine urging such cooperation (US DoD 2017:104-105). Information warfare is today no longer an enabling function supporting military operations in physical domains. The iWar is a multi-disciplinary, cognitive function in its own right - a whole new geopolitical realm. Furthermore, sociocultural systems analysis and intelligence knowledge are not transferred into the contemporary civilian 11 | P a g e 5249168 Cassandra Brooker Masters of Research context for effective democratic social influencing. Contrast this with Russia and China, where the lines between military, government, social media, and business influencing activities are completely blurred and hold a centrality of focus.

There are more than 3.8 billion people online, 2.9 billion on social media; Western nations must reorientate their national security paradigms and view this complexity as an opportunity to exploit to their advantage (Canadian SIS 2018:61). Australia currently lacks effective influence across social, political, economic, and military spheres. The character of the contemporary, high tech, multi-dimensional battlespace highlights the prominence of information warfare capabilities in modern conflict (Bienvenue et al. 2018a; IWD 2018). The blurring of military and civilian media and communications capabilities particularly illustrates the urgency with which the ADF must develop and integrate effective iWar capabilities as part of its combat power – not only to meet future security challenges but to also provide leadership and guidance to other national, political, economic, and civilian agencies in the influence space.

Information warfare is defined as the “conflict or struggle between two or more groups in the information environment” (IWD 2018:10) and closely aligns with the ideas of Sun Tzu and Clausewitz; where the imposition of will or winning the battle of minds can be achieved through the control of information (Ibid:10). The contemporary iWar information environment (IE) comprises a complex web of both interlinking and competing complex social systems. However, this research identifies that current gaps in approaches of Western democratic nations means that systems thinking is not being used to full potential, particularly in a military context, for influencing effectiveness in the iWar. This research will demonstrate that systems thinking is an invaluable approach for not only holistically understanding the IE and addressing own-system vulnerabilities in the conduct of information operations (IO), but also for enabling the effective influencing of audiences and the targeting of adversary systems in the iWar.

Furthermore, much is written about military influence activities (IA), battlefield lessons learnt, psychological operations (PSYOPs) and wartime propaganda techniques, however there are gaps in the literature analysing the actual effectiveness of such activities. Admittedly, it is difficult to obtain metrics in warzones and collecting against measures of effectiveness is resource intensive, particularly among populations with different languages and cultures. However, this thesis argues that by thinking critically about iWar problems, breaking down target audiences into social systems, and using research into human nature and behavioural economics during target systems analysis provides enhanced solutions. Despite rapidly changing technology, shifting geo-political paradigms, asymmetric power imbalances, globalisation, and ever-evolving hyper-connectivity; 12 | P a g e 5249168 Cassandra Brooker Masters of Research human nature, even in the global interconnected IE, is inherently predictable and able to be manipulated. This research asserts that it is possible for Australia to regain the initiative in the iWar, using systems thinking approaches, effective influencing tactics, and cognitive science to overcome the obstacles it currently faces in dominating the IE.

The Problem For many years, governments, corporations, analysts, strategists, and academics have investigated how to better influence target audiences and exploit cognitive human behaviour for control, power and profit. Additionally, the literature review shows evidence of a long and colourful history associated with the evolution of media and advertising to better persuade consumers, whilst the conduct of propaganda and controlling the narrative are established influencing activities as old as warfare itself. Influencing activities to support battlefield operations, such as IO, PSYOPs, deception, and disinformation, have been used in wars throughout history, with the fundamental goal of making the adversary comply with your will, whilst protecting own-force secrets. In the 6th century, the military genius Sun Tzu stated that “the clever combatant imposes his will on the enemy, but does not allow the enemy’s will to be imposed on him” (Giles 2017:48). Sun Tzu also championed deception, influencing, and intelligence tactics, which, along with other writings in his treatise: ‘The Art of War’, have stood the test of time and provide important principles for modern strategists operating in the contemporary iWar (Giles 2017:11,17,48,129).

However, in more recent Western military history, influencing activities have been increasingly planned and executed in isolation to kinetic warfare operations. Kinetic means involving the use of forces of dynamic motion or energy to achieve an effect (ADF 2018), for example: targeting an enemy asset with a ballistic missile. Current ADF information operations doctrine lacks a definition of ‘non-kinetic effects’, for example: targeting enemy audience cognitive biases using communication methods to exert influence. Furthermore, systems thinking has emerged as a viable methodology for understanding and leveraging tangible economic, ecological or weapons systems, and is now increasingly being adopted for analysing complex social systems operating in the iWar (Hall & Citrenbaum 2010; Gharajedaghi 2011:59-66; J. Mark Services Inc. 2018). The research problem, therefore, is that the benefits, knowledge and valuable lessons provided from these historical experiences and the three academic research disciplines have been overlooked, stove-piped into silos of excellence, or have not yet been fused by Western democracies to create holistic, complementary approaches for better influencing strategies in the contemporary iWar.

13 | P a g e 5249168 Cassandra Brooker Masters of Research However, following the results of Russian interference in the 2016 US election and the Brexit vote in the UK, the collective interest in influence activities and cognitive warfare has recently increased. As Jensen and Sear (2019) explain, human consumption of cyberspace content – with all our cognitive biases – has combined with rapid global dissemination and a more tribal, adversarial democratic system, which have blurred the line between politics and war, and created a new type of information warfare. Research into systems thinking continues to evolve from analysing tangible systems to the intangible, for example Stroh’s ‘Systems Thinking for Social Change’ (2015) enables readers to contribute more effectively to society by helping them understand what systems thinking is and why it is so important in their work. For this research and contemporary iWar problems, these new paradigmatic approaches provide a deeper appreciation of Australia’s complex social systems. This enables improved output efficiencies, and a holistic understanding of the mental models and system structures that underwrite Australia’s participation in the contemporary interconnected global environment.

There are key gaps in the literature pertaining to the use of systems analysis for improving non- kinetic targeting and in combining that knowledge with influencing theory and behavioural science. Researching all three disciplines as they relate to each other, adversary systems, and the contemporary iWar; as well as analysing what those findings specifically mean for Australia, has not previously been attempted. Because influencing activities, cognitive behaviours, and systems thinking have all been impacted upon by the interconnected global information environment of cyberspace; research into the results of such modern effects is required to address the gaps in the accepted knowledge. As Jensen and Sear (2019) explain, cyberspace, especially social media, enables the strategic use of information to pursue a competitive advantage in amplifying, influencing, and degrading the capacity of adversaries to make decisions, even at a societal level.

Research Questions The academic aim of this research is to investigate solutions for improving Australia’s influence effectiveness in the iWar. This will be achieved by addressing the research questions arising from the gaps identified from a review of the literature: 1. How does systems thinking offer an effective approach to understanding complex social systems of our adversaries and provide a methodology for developing effective influence and counter-influence strategies for Western nations in the iWar? 2. Why is understanding the effectiveness of influencing activities imperative to appreciating how complex social systems adapt their behaviours based on balancing feedback?

14 | P a g e 5249168 Cassandra Brooker Masters of Research 3. Why is understanding psychological cognitive theories, with respect to influence effectiveness, central to analysing how external pressures impact on a social system’s behaviours and mental models?

Significance of Research The significance of this research is its contribution to addressing gaps in the current literature and the lack of innovative practices where systems approaches to the iWar are misunderstood and under-utilised by the ADF and Australia’s national security organisations. By fusing three disparate research disciplines, this study also contributes to the development of original theories surrounding what makes certain influence activities effective. Finally, this research offers practical solutions, in both the military and civilian contexts, to address Australia’s weaknesses in dominating the contemporary, complex, cognitive iWar environment.

To answer the academic aim and the research questions, this thesis analyses two contrasting case systems to determine the effectiveness of their respective influence in the iWar. The successful influencing case system of the terrorist group ISIS is compared and contrasted with the unsuccessful case system of Hillary Clinton’s election campaign in 2016, using a single system stock of influence to determine flows and delays to reinforcing and balancing feedback loops.

Scope of Research The scope of this research as a Masters’ thesis was limited to the capacity of a sole researcher within a given time frame. Therefore, researching more than two case studies comprising complex social systems in the iWar was beyond the capacity of this thesis, particularly in developing a certain degree of “intimacy” with each of the cases under consideration (Rihoux & Ragin 2009:23). Using two contrasting case systems enabled the scope to be limited to analysis within a specified time period, with a single system stock to be modelled. This allowed research into the systems’ feedback cycles and interconnections related to the relevant narrative data to answer the research questions, rather than attempting to analyse the entire holistic organisation’s historical activities and outputs. The two case studies provided important contextual conditions highly pertinent to this research for examining contemporary iWar events, and where the past behaviours contributing to system flows could not be manipulated (Yin 1994:8,13). Additionally, the two case systems are of comparable size, structure and had similar aims, which enabled an appropriate qualitative comparative analysis; as opposed to modelling a large state-based adversary system, such as Russia, or the massive internet ecosystem, which were both beyond the scope of a single researcher and may have diplomatic or privacy complications.

15 | P a g e 5249168 Cassandra Brooker Masters of Research

This thesis contains three arguments to highlight the gaps in the literature as they pertain to the iWar. The first argument is that systems thinking offers an effective approach to understanding complex adaptive social systems of our adversaries and provides a methodology for developing effective influence and counter-influence strategies for Western militaries and societies. The second argument is that understanding the effectiveness of influencing activities is imperative to appreciating how a complex social system adapts its behaviour based on its reinforcing or balancing feedback loops. The third argument is that understanding psychological behavioural theories, and how they contribute to influence effectiveness, is central to analysing how external pressures impact on a social system’s behaviours and mental models.

Thesis Overview This thesis comprises three parts encompassing three different research disciplines of: systems thinking, influence effectiveness, and behavioural science to achieve the study aim. This thesis argues systems thinking approaches enable a holistic understanding and improved targeting of the complex adaptive social systems of our adversaries, and provide methods for developing effective influencing and counter-influence strategies. Systems thinking also enables better understanding of own-system fragility and vulnerabilities, across all aspects of democratic society, which facilitates increased resilience, cognition and effectiveness in the iWar. This research also provides understanding into the effectiveness of influencing activities used throughout history and how these lessons apply to the contemporary global battlespace of the internet commons. Both the systems and influence research are linked with related psychological and behavioural theories to better understand mental models that can be leveraged or manipulated. The fusion of these three research disciplines is unique and enables a deeper understanding of ‘enemy’ systems, their influencing tactics, and their aims of achieving maximum psychological effect through exploitation of heuristics and cognitive biases.

Part One comprises Chapters 2 and 3. Chapter 2 is a literature review which identifies the research questions and highlights the gaps in the three literary fields as they pertain to the iWar. Chapter 3 outlines the methodology, comprising a pragmatic epistemology and systems modelling approach. Chapter 3 also includes justification of employing a qualitative comparative case study analysis, using narrative data, as well as rationalising the selection of the two case systems. It concludes with a discussion on the internal and external validity of the methodological approach and demonstrates the steps for conducting the narrative analysis and creating the systems’ models.

16 | P a g e 5249168 Cassandra Brooker Masters of Research Part Two of this thesis, comprising Chapters 4, 5 and 6, demonstrates the process of conducting the pragmatic narrative analysis of the two cases’ respective qualitative data to consequently map each system’s model surrounding their influence stocks and present the associated findings. Chapter 4 comprises a system analysis of influence effectiveness using ISIS as a case study. This model provides a holistic understanding of the components, structures, feedback loops, and behaviours contributing to an anti-fragile, anticipatory adversary system in the iWar. Chapter 5 mirrors Chapter 4 in conducting a system analysis of an ineffective influencing system, using Hillary Clinton’s election campaign in 2016 as the case study. This research assessed Clinton’s campaign to be an open system with a robust level of resilience that suffered increased fragility throughout the bounded time period of the 2016 Election, due to an asymmetry of balancing and reinforcing feedback. Chapter 6 concludes Part Two by presenting the findings arising from the modelling of the two case systems and subsequent comparative analysis, to highlight the configurational, behavioural, and causal factors contributing to their respective levels of influence.

Part Three of this thesis, comprising Chapters 7 and 8, discusses and concludes the results of the research as they relate to the gaps, before making key recommendations. Chapter 7 discusses the implications of the findings of the literature review in Part One and case system analyses in Part Two of this thesis to achieve the aim of investigating solutions for improving Australia’s influence effectiveness in the iWar. The discussion in Chapter 7 revolves around responding to questions surrounding: how much is known, how has the research answered the gaps, and what does this research mean for Australia? Chapter 8 concludes the thesis with a summary of the key arguments, recommendations and in highlighting the contribution of this research to Australia’s iWar efforts as well as to the body of literature surrounding this study.

Conclusion There is a requirement for new approaches in researching and fusing theories for discovering practical solutions to enhance Australia’s influencing effectiveness in the iWar, as well as contributing to the body of literature surrounding contemporary IA. The research in this thesis will realise the aim in demonstrating that Australia can achieve effective influence over population, military, and government systems in the iWar from non-technical information operations. This will be achieved by fusing the three research disciplines of systems thinking, influencing activities, and behavioural science to highlight the gaps in the literature (Chapter 2) and this then informs the research methodology (Chapter 3). This research will compare and contrast two case studies using a systems modelling approach to demonstrate how and why Western democracies have proven ineffective in the iWar (Chapters 4, 5 & 6). The two complex 17 | P a g e 5249168 Cassandra Brooker Masters of Research social case studies selected for this research comprise: a successful influencing system of ISIS and an unsuccessful influencing campaign of Hillary Clinton in the 2016 US Election. The utility of adopting a systems thinking analytical approach will be validated for understanding both the modern, complex social systems of key adversaries and own-force iWar systems; as well as determining why the case studies were effective or not in raising their system influence stock, dominating the IE, and manipulating cognitive behaviours.

18 | P a g e 5249168 Cassandra Brooker Masters of Research Chapter 2 – Literature Review If a revolution destroys a government, but the systematic patterns of thought that produced that government are left intact, then those patterns will repeat themselves… There’s so much talk about the system and so little understanding.

Robert Pirsig Zen and the Art of Motorcycle Maintenance (2005)

Introduction Australia is already losing the iWar and Western democratic nations are being out-communicated by their adversaries. A key gap is that psychological operations (PSYOPs) and information warfare are spoken about extensively in the grey literature, however those discussions are largely omitted from the corpus of academic knowledge. The contribution of this research will address gaps in the current literature where systems thinking approaches to the iWar are misunderstood and under-utilised by the ADF; and to provide analysis on what makes certain influence activities effective in the contemporary, complex, globalised iWar environment.

This thesis contains three arguments to highlight the gaps in the literature pertaining to the iWar. The first argument is that systems thinking offers an effective approach to understanding complex adaptive social systems of our adversaries and provides a methodology for developing effective influence and counter-influence strategies for Western militaries and societies. The second argument is that understanding the effectiveness of influencing activities is imperative to appreciating how a complex social system adapts its behaviour based on its reinforcing or balancing feedback loops. The third argument is that understanding psychological behavioural theories, and how they contribute to influence effectiveness, is central to analysing how external pressures impact on a social system’s behaviours and mental models.

Based on the three arguments, the chapter is divided into three parts. Part One reviews the academic literature surrounding systems thinking and justifies a systems-based analytical approach as an effective method for Australia to conceptualise complex social systems for future iWar targeting. Part Two examines the literature regarding the effectiveness of various influence activities studied throughout history in marketing, media, elections, and propaganda. This thesis will attempt to expand the academic knowledge to address the gap in the literature surrounding Australia’s strategic-level influence activities. Part Three reviews the literature surrounding psychological concepts and cognitive traits relating to behavioural economics, to determine what types of influence activities are most effective on system behaviours and outputs.

19 | P a g e 5249168 Cassandra Brooker Masters of Research This review captures the main concepts of each part, shows relationships between those concepts, and identifies deficiencies in the literature to establish the theoretical basis for further academic research. The literature reviewed also highlights key gaps and limitations in Western democratic approaches to the contemporary iWar problem.

Part One: Systems approaches to the iWar While certain niche sections operating in Australia’s intelligence agencies adopt systems thinking approaches to analysing and targeting adversary complex adaptive social systems; a systems approach to the non-kinetic iWar by the broader ADF, and by extension the Australian Government, is misunderstood, underutilised and overlooked in favour of traditional, stove-piped, doctrinal processes (J. Mark Services Inc. 2018; US DoD 2014, 2017; Australian Army 2018; ADF 2018). Whilst general systems theory is increasingly being adopted within some units, current intelligence analysis and Information Operations (IO) planning efforts are siloed, follow divergent doctrines, and have different objectives, definitions, outputs, and levels of risk acceptance (US DoD 2017, 2014). Systems thinking offers an extremely effective approach to understanding both friendly and adversary complex, adaptive, social systems; and provides a sound methodology for this research to analyse the effectiveness of influencing activities and recommend resulting iWar strategies.

A key gap in the literature exists regarding the use of systems analysis for non-kinetic information warfare. A definition of non-kinetic effects, from a targeting perspective, is contained in Appendix 1 (ADF 2018). Whilst, PSYOPs and IO are discussed extensively in the grey literature, including commentary on the importance of effective influence; those considerations are often omitted from the corpus of academic knowledge – due to reasons of secrecy or security, or to focus instead on lessons learnt. Some academic literature covers kinetic target systems analysis (TSA) to inform effective targeting, and Hall and Citrenbaum (2010) recommend systems-based analytical approaches to complement other intelligence analysis techniques. Goble (2002) was an early advocate of applying systems thinking to complex IO and improving PSYOPs activities. Goble (2002:31-32) contends that ‘complexity science’, should be used to predict and measure the effects of IO, in the same manner TSA models predict and measure the effects of kinetic weapons. Despite these champions of systems thinking, the application of proper systems approaches for strategic influence is lacking and thereby undermines own-force ability to conceptualise complex social systems for effective targeting (Albino et al. 2016), i.e. where influence is the adversaries’ stock and their effectiveness shapes their system’s flows, linkages and feedback loops (see Table 1 for definitions).

20 | P a g e 5249168 Cassandra Brooker Masters of Research The purpose of Part One in this chapter is to review the academic literature related to systems thinking and how it applies to the iWar. Section One provides an overview of systems thinking concepts, outlining the theoretical knowledge to be applied in this thesis. These theories link with the Section Two review of the literature regarding systems thinking about complex sociocultural systems. Section Three reviews the literature surrounding the benefits of thinking in systems for the iWar; in that, linear analytic approaches are not suited to complex environments and unstructured problems, and tend to over-simplify the system being examined.

Systems thinking theory The literature surrounding systems thinking provides the theory and associated methodology to understanding a system holistically, however, gaps are revealed when applying these methods to complex socio-cultural systems. A leading authority in systems thinking is Meadows (2008), who explains the decomposition of a system into key components to better understand structures and behaviours. This analysis into system workings assists in determining what type of leverage can be applied to improve underlying behaviours. Meadows (2008:2) contends that as the world becomes increasingly complex, a systems thinking lens assists in solving contemporary problems. The decomposition of a system into its elements, sub-systems, interconnections, feedback loops, flows, and stocks; identifies opportunities for understanding behavioural patterns, common traps, and applying leverage for change. Definitions of these key system features are contained in Table 1 (below) and at Appendix 1.

However, a gap in Meadows’ (2008) literature is a predominant use of functional economic systems to demonstrate the theories used; rather than intangible, complex social systems with non-physical resources, stocks, sources or sinks. Meadows’ systems theory can certainly be applied to complex adaptive systems, as Stroh (2015) demonstrates, however, Meadows admitted a lack of confidence in providing solutions to intervene in social systems, because finding leverage points in complex, dynamic systems is difficult (2008:146).

Social systems comprise the most complex class of systems (Kim 1999:4); it is this class of adversary system being targeted in the iWar and, which, presents an associated gap in the literature. Stroh (2015) supplements Meadows’ research in providing systems thinking theory for social change. Stroh (2015:36-37) also expands on Meadows’ iceberg theory (2008:88) for complex social systems, exposing the hidden part of the iceberg alluded to by Meadows, comprising: patterns of behaviour, the system structure (relationships between parts), mental models shaping the system, and outputs which could be leveraged (Figure 1). Stroh (2015:37) contends that studying these hidden elements unveils the root causes of chronic, complex 21 | P a g e 5249168 Cassandra Brooker Masters of Research problems; and therefore this thinking was applied to this research to determine how to effectively influence iWar systems.

Key Systems Definition Terms

System A set of elements or parts that is coherently organised and interconnected in a pattern or structure that produces a characteristic set of behaviours, often classified as its “function” or “purpose.”

Archetypes Common system structures that produce characteristic patterns of behaviour. Stock An accumulation of material or information that has built up in a system over time. Flow Material or information that enters or leaves a stock over a period of time via feedback loops. Flow direction is depicted using feedback ‘link’ arrows. Feedback Loop The mechanism (rule or information flow or signal) that allows a change in a stock to affect a flow into or out of that same stock. A closed chain of causal connections from a stock, through a set of decisions and actions dependent on the level of the stock, and back again through a flow to change the stock. Can be a balancing or reinforcing feedback loop. Balancing A stabilising, goal-seeking, regulating feedback loop, also known as a “negative feedback loop” feedback loop because it opposes, or reverses, whatever direction of change is imposed on the system. Reinforcing An amplifying or enhancing feedback loop, also known as a “positive feedback loop” because it feedback loop reinforces the direction of change. These can be either vicious or virtuous cycles. Interconnections Interconnections are the relationships that hold the system elements together. They can be / Relationships physical, informational, emotional etc. Relationships can be linear or non-linear. Linear A relationship between two elements in a system that has constant proportion between cause relationship and effect and so can be drawn with a straight line on a graph. The effect is additive. Nonlinear A relationship between two elements in a system where the cause does not produce a relationship proportional (straight-line) effect. Source and Sink Stocks at the beginnings and ends of flows are called sources and sinks, respectively. They mark the boundary of the system diagram but rarely mark a real boundary, because systems rarely have real boundaries. Resilience The ability of a system to recover from perturbation; the ability to restore, repair or bounce back after a change due to an outside force. Self-organisation The ability of a system to structure itself, to create new structure, to learn, or diversify.

Table 1: Key Systems Terms and Definitions (Meadows 2008:4,13,95,187-188)

Stroh (2015) predominantly applied systems thinking retrospectively to understand why social systems were successful or not. Stroh’s approach provides insights and solutions for contemporary social problems, encourages change within open, accessible systems, and enables prospective analysis of lessons learnt within an organisation. However, while Stroh’s guidance was valuable in analysing faults in Clinton’s unsuccessful election campaign and in ADF approaches to the iWar, it is not as easily applied to externally analysing an adversary system,

22 | P a g e 5249168 Cassandra Brooker Masters of Research such as ISIS. This is because adversary systems, such as terrorist groups and authoritarian regimes, are closed systems, meaning they are less transparent about their structures, flows, and behaviours and remain largely inaccessible to Western researchers.

Gharajedaghi (2011:59-60) complements the aforementioned literature by analysing the components and relationships characterising sociocultural systems, comprising elements linked almost entirely by interconnections of information and shared image (emergent culture). Sociocultural systems, such as ISIS, are an “organisation of meanings emerging from a network of interactions among individuals”, with their shared image constituting the principal bond among members (Gharajedaghi 2011:60-61). Analysing the behaviour of information-bonded sociocultural systems presents a different proposition from open social or functional systems studied by Meadows (2008) and Stroh (2015). The integration of components, with individual thoughts and behaviours, into a cohesive whole within a sociocultural system is an ongoing challenge (Gharajedaghi 2011:59) and links with the literature reviewed in Parts Two and Three. Albino, Figure 1: The Iceberg Model of Systems’ Leverage Friedman, Bar-Yam and Glenney (2016) also (Source: Academy for Systems Change) championed systems theory to address improving military strategies against complex adversary systems, albeit from kinetic targeting perspectives.

A summary of the reviewed systems literature and the key concepts and gaps identified, which apply to this research, is contained at Appendix 2. The application of systems thinking tools provided by Meadows and Stroh, such as common archetypes, balancing feedback, and the concept of stocks and flows; informs this thesis’ comparative case study analysis. The utility of applying systems theory to case studies such as adversarial sociocultural systems, like ISIS, or complex social systems, such as Clinton’s election campaign, is reviewed in the next section.

Systems thinking about complex social case studies This section reviews the literature in support of the argument that systems thinking is more effective than traditional analytic approaches for understanding complex sociocultural systems of 23 | P a g e 5249168 Cassandra Brooker Masters of Research the two case studies explored in this thesis. That is, an adversary operating in the contemporary iWar environment and an intangible social system of an election campaign; in order to determine the effectiveness of each in maintaining their respective influence stocks.

Both Meadows (2008) and Stroh (2015) provide systems tools and behavioural definitions that could be applied to the complex case studies compared in this thesis, and the mental models they describe also link into the behavioural literature of Part Three. Both contend there must be at least one reinforcing loop driving growth and one balancing loop constraining growth, because no system can grow forever in a finite environment. Figure 2 demonstrates this concept with an example of a single stock, Army PSYOPs system’s components, flows and feedback loops.

Figure 2: Example of a one stock system with feedback loops – Army PSYOPs

Meadows (2008) predominantly analysed physical systems with functional goals and increasing tangible stocks with balancing feedback, where visible capping or checking occurred. Stroh (2015:50) discusses this reinforcing feedback within system cycles as being either ‘virtuous’ or ‘vicious’, which correlates to the cycles observed for ISIS and the Clinton campaign, respectively. However, while feedback loop theory can be applied to the virtuous cycle of ISIS’s rise in influence, it is more difficult to apply neatly to the declining influence stock of the Clinton campaign, with multiple, complicated external pressures and often misaligned, competing sub- systems.

Meadows’ (2008) explanation of time lags in stock flow changes, feedback mechanisms, hierarchies, behaviours, and system resilience all comprehensively assist the case study analysis. Her discussions surrounding the characteristics of complex systems to learn, diversify, and evolve

24 | P a g e 5249168 Cassandra Brooker Masters of Research (2008:4), the ‘Success to the Successful’ archetype (2008:127), and higher meta-resilience achieved by self-organising systems (2008:79), all provide valuable insights into how ISIS was able to maintain system effectiveness for so long. These aspects contributing to ISIS’ influencing success are demonstrated in Chapter 4. Conversely, Meadows’ recognition of psychological and political tendencies to blame external influences for the cause of system problems (2008:4), competing goals of sub-systems (2008:15), and her ‘Drift to Low Performance’ archetype (2008:122), highlighted the intrinsic systems problems, undesirable perceptions, and behavioural characteristics of system structures that produced them. These were all problems which beleaguered the Clinton campaign, as will be demonstrated in Chapters 5 and 6.

Stroh’s approach (2015) complements Meadows’ definitions (2008) of system archetypes to enhance understanding of the complex social systems’ behaviours observed in the case studies. Reviewing the systems literature and understanding the associated concepts strengthened the justification of the case studies’ selection and enabled the comparative system analysis methodology, as discussed in Chapter 3. For example: Stroh’s discussion of ‘quick fixes’ and ‘fixes that backfire’ (2015:56); highlighted how, in the rapid tempo of the 2016 US Election, the Clinton campaign increasingly fell into such traps when reacting to disinformation or scandals, instead of investing in long-term, core solutions. Stroh also discusses ‘Accidental Adversaries’, where systems and sub-systems inadvertently obstruct each other’s performance in a vicious cycle (2015:61-62). This phenomenon was also evident during the 2016 Election; with the Clinton campaign, media, Obama administration, and Bernie Sanders’ supporters all having competing goals and conflicting messages; which essentially undermined them all (Clinton 2018). The system analysis of the Clinton campaign in Chapter 5 explains these archetypes in further detail.

Stroh (2015:64) details specific social problems of systems, which directly correlate to behaviours observed in the case studies, particularly the Clinton campaign. These issues include: ‘Competing Goals’ within a system, unintended ‘Escalation’ with an adversary, self-created limits on ‘Growth or Investment’, and ‘Drifting Goals’ where tolerance is developed for low performance. External system events impacting on both cases included the exploitation of social media and undermining the integrity of the internet, which links with ‘Tragedy of the Commons’ discussions (Stroh 2015:65; Meadows 2008:116). All of these key systems concepts were applied to the case systems’ analysis in Part Two of this thesis and demonstrate the utility of systems thinking as an effective approach to understanding complex adaptive social systems of adversaries and a methodology for developing effective influence and counter-influence strategies for Western militaries and societies.

25 | P a g e 5249168 Cassandra Brooker Masters of Research Hall and Citrenbaum (2010) provide systems approaches to intelligence analysis, which can be readily applied to the case studies. Their advocacy of conducting ‘decomposition’ (2010:77-92) at the beginning of the intelligence process to discern meaning is a key systems approach. Decomposition enables insight into the relevant attributes, characteristics, interactions, and behaviours of an adversary system and links to the literature reviewed in Parts Two and Three (Hall & Citrenbaum 2010:77-92). ‘Recomposing’ and fusing the data with intelligence, allows analysts to synthesise it into a whole-of-system knowledge to drive future collection and effective targeting. Hall and Citrenbaum (2010) promote systems thinking to enhance intelligence analysis, however, they diverge from other systems scholars to focus on critical thinking and other complementary analytical techniques.

Gharajedaghi’s (2011) paper into sociocultural systems assisted in the analysis of the ISIS case study, particularly in understanding their differences from moderate Muslim society. Gharajedaghi (2011:62) contends that traditional sociocultural systems manifest greater inertia and resistance to change, but that inertia also makes cultures resilient and sustainable. So, whilst ISIS is an Islamic-based organisation, it remained viable by questioning assumptions, being self- organising and self-evolving. Gharajedaghi (2011:65) explains sociocultural systems such as ISIS, must co-evolve with their surrounding environment and change the rules of interaction over time. This can be compared with members of traditional societies who lack the freedom to question sacred cultural codes. Gharajedaghi (2011) states that: “disturbance to any system will be resisted at first, but if it survives initial attempts at suppression and resonates with pre-existing frustrations, an iterative process of deviation amplification begins” (Ibid:66), which allowed ISIS to produce rapid structural changes in its less developed social system. Conversely, extreme ideologies have proven to be major obstructions to the viability of a social system, and therefore, Gharajedaghi’s (2011) insights provided understanding for analysing ISIS’ balancing feedback loops relating to counter-influence actions and a drift-to-low-performance working against them.

Albino et al. (2016) supplemented the literature justifying systems approaches to analysing complex social case studies and discussed alternative system responses to attack - depending on whether a system was fragile, robust, resilient, anti-fragile or anticipatory on a scale of strength (Albino et al. 2016:4,11). They contend that ‘robustness’ and ‘resilience’ characterised systems able to tolerate stress and recover from shock, therefore, targeting them was a wasted effort. A ‘fragile’ system deteriorates when stressed, while an ‘anti-fragile’ system grows stronger (Albino et al. 2016:12-13). ‘Anticipatory’ responses are a key capability of complex systems, which is the ideal Western democracies should be aiming for in the iWar.

26 | P a g e 5249168 Cassandra Brooker Masters of Research These system response characterisations provided understanding into the behaviours and structures of the case studies. For example, Albino et al. (2016) state that an anti-fragile system “must be opposed by means of other dimensions” (2016:12); i.e. dislocating a military anti-fragile system’s power through its political, social, or economic fragility. Albino et al. (2016:12) explain this anti-fragile archetype using examples of the Taliban and Al-Qaeda, which gained members, power and support as a result of the conditions of conflict (Figure 3). In these cases, as with ISIS, the actions of the Coalition fuelled their systems’ will, cohesion, organisation, and provided lessons in responding to stressors and adapting accordingly (Albino et al. 2016:14). Although Albino et al. (2016) were concerned primarily with addressing weaknesses in military strategic thinking as it applied to kinetic targeting; their systems analysis applies equally to the iWar.

This section has highlighted how systems thinking concepts are invaluable for analysing complex social systems and supported the argument that systems thinking approaches are more effective than traditional analytic approaches for understanding the case studies to be analysed in this thesis. Understanding the systems tools, behavioural responses and archetypes described in the literature enabled the holistic analysis of complex sociocultural systems of adversaries operating in contemporary iWar environments, and of a seemingly intangible social system of an election campaign. Appendix 3 contains further detail on how the systems literature specifically relates to the analysis of two case studies conducted in Part Two of this thesis. The literature reviewed throughout the previous two sections highlights the benefits of thinking in systems for analysing complex adversaries encountered in the iWar and these benefits will be explored more comprehensively in the next section.

The benefit of thinking in systems for the iWar Linear or reductionist analytical approaches are not suited to complex environments and unstructured problems. When applied to complex problems, current doctrinal analytic approaches tend to over-simplify the system (Byrne 2005:102; Goble 2002:1; Givens 2012:63). Meadows (2008) contends that a “diverse system with multiple pathways is more stable and less vulnerable to external shock” (Ibid:4); however, such systems are also more complex, particularly in the iWar. Therefore, a methodical approach is required to decompose adversary systems to truly understand and target those network pathways and redundancies. Systems thinking approaches are already used by Western militaries for kinetic TSA of tangible, functional weapon systems but, as evidenced by the lack of literature, is not effectively applied to non-kinetic, complex socio- cultural systems’ analysis. A summary of the benefits of systems thinking compared with traditional linear approaches, as analysed through this research, is contained at Table 2.

27 | P a g e 5249168 Cassandra Brooker Masters of Research Linear Doctrinal Approaches Systems thinking

• Kinetic, functional target systems analysis focused Can be applied to complex sociocultural systems of our adversaries, • Siloed within Military Joint Effects / Targeting space neutral populations & target audiences in the iWar • Not considered at the National-Political level for the iWar

Network, stakeholder and link analyses all largely focused on Analyses linear and non-linear relationships (links) between nodes targeting the nodes to determine cause and effect, and apply leverage

• Systems thinkers set own boundaries on limits of system to be Set doctrinal processes, checklists, and templates for Intelligence analysed and problems to address. and Planning discourages alternative critical thinking approaches • Scalable methodology enables exploration of various aspects & branches

Ad-hoc, siloed approaches to target system analysis by different Holistic approach to understanding the whole system, including all military branches, focused on specific sub-systems, rather than its sub-systems, flows, and external pressures the whole

Focus on reacting to the outputs or events produced by an Uncovers underlying behaviours, structures and mental models adversary system driving a system’s outputs, which can be leveraged

Enables the measurement of system changes to processes (rather Difficult to measure effectiveness of non-kinetic targeting / than outputs or behaviours), following the application of leverage influence activities, as MoE focuses on tangible system outputs or nudges

Limited literature related to iWar systems problems. Complexity science & systems approaches being adopted by niche elements Extensive academic research surrounding systems thinking of the intelligence community, largely misunderstood by majority approaches to complex, sociocultural systems’ problems of senior leaders in Defence and at the strategic level

Intelligence analysis and planning outputs limited by time and Researching, understanding and solving problems within complex resource constraints systems is very time intensive and continually evolving Table 2: Comparison of systems thinking with current linear thinking approaches in the iWar

Goble (2002:32), an early champion of using systems thinking for IO, argues that because scientists applied complexity science to solve real problems across other disciplines, it could work for military PSYOPs. Goble (2002:31) contends that ‘complexity science’ (systems analysis) should be used to predict and measure the effects of IO, in the same manner mathematical models measured and predicted effects of kinetic weapons. He argued that since TSA models were an integral part of the targeting process, models of adversary decision systems were integral to effective IO targeting (2002:45). Blackmore (2003:17-18) agrees in following a system-of- systems analysis to enable effective attacks on all aspects of an enemy's capabilities and value system; with the goal of changing attitudes, decisions, and behaviours of a target audience.

Goble (2002) was ahead of his time in recommending emerging systems analysis approaches to IO and associated intelligence gathering. He highlighted that measures of effectiveness (MoE) could be established for sociocultural systems targeted by PSYOPs, by collecting intelligence on the specific links and nodes within a system to determine impacts on functionality (2002:14). Goble understood the complexity of the IE, where military forces must “conduct information operations against one of the most complex, adaptive systems – the human mind” (2002:1). He

28 | P a g e 5249168 Cassandra Brooker Masters of Research argued that whilst linear thought processes, prevalent in the military, fully understand MoE applied to lethal fires, the same thought processes lacked understanding of the non-linear effects of non-kinetic influencing (Goble 2002:1,28,31).

However, Goble’s recommendation (2002:43-46) of establishing military relationships with complexity scientists or civilian think tanks to support IO, was ill-advised for two reasons. The first is the ethical issues raised by the conduct of human terrain analysis by civilian scientists for targeting purposes, even if non-kinetic (Buckle 2018). The second is that by outsourcing systems thinking to civilian agencies, specialist corporate knowledge is lost within Department of Defence organisations and further siloed. Although, one benefit of Goble’s recommendation would be the generation of interest and academic literature on the subject, which is presently an identified gap.

Whilst Goble’s arguments were valid for applying predictive modelling to the sociocultural adversary systems in the iWar, (2002:12-13), and while he acknowledged the increasing complexity of warfare requiring systems thinking approaches (2002:1); it could be argued his experience of the information environment in 2002 was not the multifaceted, interconnected, three-dimensional, media-saturated battlespace of today. At the time Goble (2002) recognised a systems thinking approach was not a “panacea for explaining the infinite number of interrelated complex adaptive systems of the world” (Ibid:32), and that the practicality of using systems analysis in everyday life or iWar planning would be a continued struggle due to the complexity.

Albino et al.’s (2016:3) more contemporary literature complements Goble’s (2002) argument that military planners must think beyond traditional capabilities and consider non-conventional contributions to centre of gravity, such as adversary system robustness, structures, and resilience. Albino et al. (2016) criticise current doctrinal approaches focused on applying force, and champion systems thinking for targeting adversaries non-kinetically. They (2016:2) point out that “traditional perspectives of military strategy implicitly assume fragility”, which undermines Western military planning efforts to target enemy systems that prove to be robust or anti-fragile, and thereby are strengthened by attacks against them.

Both Stroh (2015:83) and Albino et al. (2016:15) demonstrated how complex social systems in the iWar, such as ISIS, were able to strengthen themselves through feedback mechanisms, using setbacks as a source of inspiration for future responses. Albino et al. (2016:15) add that these systemic evolutions and adaptations were infrequently adopted by Western strategic planners to enable better responses to the dynamic and unpredictable nature of the iWar. The system archetypes (Stroh 2015:67; Meadows 2008:111-140) and responses (Albino et al. 2016:4) 29 | P a g e 5249168 Cassandra Brooker Masters of Research described in the literature provide valuable insights into the complex issues underpinning adversary systems in the iWar; as well as essential self-awareness required to improve dysfunctional dynamics in own force systems.

Since Goble’s (2002) paper, the United States Department of Defense (DoD) has updated its ‘Joint Operations’ publication (2017) to include systems perspectives of adversary, friendly and neutral systems in the operational environment (OE), with the information environment (IE) contained within (2017:102). The ‘Joint Operations’ publication (US DoD 2017:104) encourages practitioners to view the OE as a “set of complex and constantly interacting political, military, economic, social, information, and infrastructure systems”, and that the interaction of these systems affects the conduct of operations. The doctrine advises systems thinking perspectives help focus intelligence, facilitate operational design, and enable efficient, detailed planning (US DoD 2017:105). This demonstrates the US DoD understands the benefits of using systems thinking approaches in operational planning, however gaps remain in its iWar doctrine with respect to applying an appropriate methodology (US DoD 2018; 2014). Appendix 4 provides a summary of where key systems theories arising from the literature apply specifically to benefit iWar activities.

Systems thinking in the IE remains broad, unspecific and stove-piped from established systems approaches in Operational (2017) and Planning (2011) doctrines (US DoD 2018; 2014:63). The deficiency in systems approaches to the unconventional and internet-centric iWar is also evident in Australian doctrine (Australian Army 2018). Whilst doctrine is continually reviewed, and grey literature and niche training expound the importance of effective IO and systems approaches; there remains a literary gap commending the benefits of systems approaches in achieving effective influence in the iWar.

Systems Summary Part One of this chapter reviewed the literature related to employing systems approaches to iWar problems. Section One covered contemporary systems thinking theory, which highlighted the utility of using systems methods in enhancing critical thinking and gaining deeper understanding of complex, contemporary problems. The literature demonstrated that a systems-thinking lens can be applied to complex adaptive systems and sociocultural systems encountered in the iWar. A review of the systems thinking theories underscored the gap between academia embracing systems approaches to research; with that of Western democracies who largely overlook systems thinking for influencing, in favour of traditional processes and risk-averse practices (Table 2).

30 | P a g e 5249168 Cassandra Brooker Masters of Research The literature reviewed in Section Two supports the argument that a systems thinking approach was more effective than traditional analytic approaches for understanding the case studies to be analysed in this thesis. Understanding systems tools, behavioural responses and archetypes enables the holistic analysis of complex sociocultural systems of adversaries operating in contemporary iWar environments, and of a seemingly intangible social system of an election campaign.

The systems literature reviewed in Section Three highlights the benefits of applying a systems lens to specifically analyse complex socio-cultural systems encountered in the iWar, and which Western militaries already apply kinetically. The references emphasise that linear analytic approaches are not suited to complex environments and unstructured human problems. When applied to complex problems, current doctrinal approaches over-simplify the system, as demonstrated in the comparison of approaches in Table 2. Once complex adversary systems in the iWar are understood, then the application of effective influencing techniques and behavioural theories can be facilitated, as discussed in Parts Two and Three respectively.

Part Two: Analysing the effectiveness of influence activities The literature regarding the effectiveness of influencing activities across other disciplines, such as advertising and election campaigning, provides valuable insights for addressing deficiencies of Western democratic approaches to the iWar and for holistically analysing the systems being targeted to influence. It is not enough to rely on military IO in the modern globalised iWar environment, where the lines between civilian, political and military communications are blurred.

Understanding the effectiveness of influence activities is imperative to appreciating how complex social systems adapt behaviours in response, based on balancing feedback loops – i.e. either declining towards systemic fragility or strengthening through adaptation. Part Two of this chapter reviews influence-related literature, with section one examining the reasons why certain influence activities have been effective throughout history across various disciplines. However, the literature does not specifically address the effectiveness of influence in the contemporary iWar. There are limitations in both influence theory research and of certain approaches adopted by Western democratic nations, or lack thereof, in today’s internet age, as highlighted in section two. Section three reviews the literature regarding how influence activities impact on system behaviour; and links to systems theory explored in Part One and behavioural economics examined in Part Three. The academic knowledge explored in Part Two should be expanded to address gaps surrounding the level of effectiveness of Australia’s strategic influence, outside of niche stovepipes. 31 | P a g e 5249168 Cassandra Brooker Masters of Research

Why certain influence activities are effective There exists extensive research into which advertising, marketing, media biases, propaganda, and electioneering activities have proven most effective at persuading target audiences throughout history. Part Two will review a selection of this literature to demonstrate which influencing activities have proven effective and how these lessons apply to complex system understanding and the iWar. The behavioural economics theories that contribute to the effectiveness of these persuasion methodologies will be examined separately in Part Three.

In 2007, DellaVigna and Kaplan researched the ‘Fox News Effect’ of media bias. They contended that while newspapers provided informed commentary of current events; they were unable to respond to unfolding situations in a timely manner and with the same reach of cable news. Nowadays, with the global, instantaneous reach of the internet, this study is somewhat redundant, however they did uncover valuable insights regarding influence. They showed (2007:1228) Fox News’ media bias altered voter behaviour, political beliefs, and significantly impacted on the 2000 US election. Fox News modestly increased the Republican vote share (2007:1188), which may have contributed to Bush’s victory in the unusually close election. DellaVigna and Kaplan forecasted this influence would increase over time as Fox News’ audiences and diffusion grew (2007:1212). DellaVigna and Kaplan (2007:1228) estimated that exposure to Fox News induced a substantial percentage of non-Republican viewers to vote Republican. Their results suggested media could have a sizeable political impact, which nowadays has intensified with the more extreme biases contained within the echo chambers and filter bubbles of social media.

Gerber, Karlan and Bergan (2009:36) cited DellaVigna and Kaplan’s study into the effects of newspapers on voting behaviour and political opinions. Whilst they found no effect of left or right leaning papers on political knowledge, opinions, or turnout; they did deduce that voters receiving either paper increasingly supported the Democratic candidate, suggesting that media slant mattered less than exposure and journalistic content choices (Gerber et al. 2009:35). They also found even short exposures to daily newspapers appeared to influence voting and possibly affected turnout behaviour (2009:47).

The effect of exposure, particularly repeated exposures, in influencing audiences is considerably persuasive, with researchers frequently documenting strong associations between media exposure with a distinctive slant and viewers’ political attitudes (Gerber et al. 2009:36). DeMarzo, Vayanos and Zwiebel (2003:912) contend that repeated exposures were a form of “persuasion bias” and 32 | P a g e 5249168 Cassandra Brooker Masters of Research highlighted how unfair, excessive airtime given to one side in political campaigns or court trials, and which contained repeated arguments, had a significant influencing effect. DeMarzo et al. (2003:912) added that marketing, propaganda, and censorship strategies all effectively employed repeated exposures to an idea for greater persuasive effect. Berger and Fitzsimons (2008:1) confirmed these findings, in that repeated incidental exposure to features of the everyday environment could influence product evaluation and choice. They contend that repeated exposures to environmental cues prime related product representations in the memory and, when activated, were known to affect judgment, preference, and decision-making (Berger & Fitzsimons 2008:2).

In their research into the ‘Search Engine Manipulation Effect’ (SEME) and its possible impact on the outcomes of elections, Epstein and Robertson (2015:1) built on the aforementioned persuasion bias created from exposure to show the ordering of search results had a powerful and persuasive impact on subjects’ recollection and evaluation of that item. Kahneman (2011:83), whose behavioural research is also reviewed in Part Three, agrees that sequence matters, because a ‘halo effect’ increases the weight of first impressions or in this case the first entry on a list, to the point that subsequent information was discarded. Kahneman (2011:86) explains this mental short- cut as a combination of the coherence-seeking ‘System 1 thought process’ generating intuitive impressions, which a lazy ‘System 2’ then endorses and believes (see Table 5 in Part Three). For this reason, Epstein and Robertson (2015:1) explain, internet search rankings have a significant impact on consumer decision-making, because users trust and choose higher-ranked results. Epstein & Robertson (2015:1) add that this type of influence could impact on election results, like the ranking of names on a ballot paper (Thaler & Sunstein 2009:249), and search engine companies having the power to influence elections with impunity.

Additionally, as observed during the 2016 US Election, algorithms sorting search results and news feeds can be manipulated by adversaries, external to tech companies, for their own persuasive purposes. Epstein and Robertson (2015:9) concluded that although audiences were subjected to a wide variety of influences during political campaigns, the manipulation of search rankings exerted a disproportionate influence over voters; due to: the digital bandwagon effect, skewed exposure towards certain candidates, and because voter attention has shifted from traditional information sources towards the internet – with all its biases, disinformation, and filter bubbles.

Social media has changed the face of warfare, it is global, complex and invasive of civilian lives and businesses. Singer and Brooking (2018) researched the effectiveness of influencing activities in the contemporary, interconnected world of the global internet community, that is, the 33 | P a g e 5249168 Cassandra Brooker Masters of Research modern iWar battlespace. They argue (2018:18) that military propaganda has proven historically ineffective, and while modern warfare attempts to undermine enemy morale, it is almost never successful. In one decade, social media significantly changed traditional IO approaches and the face of warfare. Nowadays, attacking an enemy’s “most important centre of gravity – the spirit of its people” (Singer & Brooking 2018:18) no longer requires conventional attacks or traditional propaganda; rather, anyone with an internet connection can create strategic effects within seconds. Those most effective at influencing in the contemporary iWar battlespace have mastered key elements of traditional PSYOPs and adapted them for the internet. Singer and Brooking (2018:154,158-179) explain the key IO concepts ensuring the success of adversaries in the iWar, that is, having an effective narrative that resonates, using emotion as an arousal – particularly anger as an emotional contagion, being authentic to create brand engagement and online tribalism, and inundation of persuasive messaging (‘exposure effect’).

These groups had successfully fused foundational PSYOPs tactics with new techniques peculiar to the social media age, which Western democracies and traditional media outlets have been restricted from employing in response. Large institutions too bureaucratic or hesitant to weave an effective narrative are losing control in the iWar (Singer & Brooking 2018:159). Russian trolls and extremist groups, such as ISIS, successfully manipulate social media, out-publish, and out- manoeuvre traditional forms of journalism. The Canadian SIS (2018:25) explain that unlike our adversaries, Western mainstream media (MSM) is largely hamstrung by tone, fairness, allegiance to facts, and context over conclusions. Fake stories are a new form of political activism, revenue, addictive entertainment, and fuel dopamine cycles of ‘likes’ and ‘shares’ among internet partisans (Singer & Brooking 2018:134).

These tech-savvy groups also harness the simplistic power of evocative imagery through videos and humorous memes, which deliver their point rapidly, enjoy longevity (even if proven false), are easily shared, and impact on waning attention spans – by 2015 the average attention span of internet users was measured at eight seconds (Singer & Brooking 2018:159). The effectiveness of imagery to persuade is confirmed by Hall and Citrenbaum’s research into semiotics (2010:255), where cultural signs, symbols, and imagery represent cultural identity and how people communicate. Berger and Fitzsimons (2008) agree that memes, symbology, and colours are more powerful than slogans or words for effective advertising. Berger and Fitzsimons (2008:12) warn that certain imagery and environmental cues can even undermine program implementation, trigger addictive behaviours, and influence the salience of cultural identities; which can lead to the prevalence of certain behaviours within specific environments.

34 | P a g e 5249168 Cassandra Brooker Masters of Research In their analysis of ‘inundation’ and ‘personalisation’ as effective influencing tactics, Singer and Brooking (2018:173-174) contrasted the success of Trump’s election campaign with Clinton’s weaknesses. Trump had the most social media followers and deployed his network on a massive scale, “pushing out the most messages, on the most platforms, to the most people” (Singer & Brooking 2018:173; Clapper 2018:395; Benkler et al. 2018:7,17,19). Like the Russians, Trump’s campaign exploited social media tools to amplify messages, expand support bases, generate multiple story lines and micro-target voters with tailored personalised messaging – utilising data stolen by Cambridge Analytica (Singer & Brooking 2018:176-179; Benkler et al. 2018:96-97). These effective tactics resulted in “Team Trump” controlling the national conversation and created “perfect” messages for dynamically and simultaneously engaging different groups of voters (Singer & Brooking 2018:176-177; Benkler et al. 2018:19,37). The integration of human factors, cultural understanding, accurate linguistics, and personalised tailoring messages cannot be overstated for attaining maximum influence over target audiences (Blackmore 2003:21; Hiscox 2018). Berger and Fitzsimons’ (2008:12) agree that customised slogans and tailored advertising, targeted towards specific geographic regions or demographic groups, prove extremely successful – as they benefit from links with everyday environmental cues. These tactics were quickly understood and applied by both the Trump campaign and the Russians, not only for generating support and reinforcing their narrative but also for widening social fissures and creating conflict (Clapper 2018:395; Wardle 2019; Reynolds 2018; Benkler et al. 2018:132,269; The Facebook Dilemma 2018).

Following Russia’s interference in the 2016 US election, the Canadian Security Intelligence Service (SIS 2018) analysed the effectiveness of disinformation in the contemporary environment and highlighted the deficiencies of Western democracies to counter such operations. The SIS (2018:7) agreed with Singer and Brooking (2018) that the reach and speed of social media escalated the impacts of disinformation. It has become a highly effective tool for skilled purveyors of falsehoods to influence political views, exacerbate divisions and undermine the ability to mediate the quality of public information – subsequently threatening the integrity of democratic discourse (SIS 2018:7; Wallis 2019; Laity 2015:23; Uren 2018). Russia’s disinformation campaign is so effective because it employs an extensive, industrial-level network of trolls, sock puppets and bots, whose activities are intensified and corroborated by a multi- pronged state-owned media machine, diplomatic support, forgeries, and de-facto alliances with organisations such as WikiLeaks (SIS 2018:7,20,25; Paul & Matthews 2016:1; Donovan 2019). Russian influence is not bound by borders, is often untraceable, and unconcerned by democratic ethics, values or legalities (SIS 2018:33-34; Donovan 2019).

35 | P a g e 5249168 Cassandra Brooker Masters of Research Paul and Matthews (2016) characterised the contemporary Russian propaganda model as a “firehose of falsehood” (2016:1) because of the inundation of numerous messages across multiple channels. They explained Russia’s “shameless willingness to disseminate partial truths or outright fictions” (2016:1) was successful in its design to entertain, confuse and overwhelm target audiences. The rapid, continuous, repetitive nature of Russian IO, and their lack of truth and consistency, actually runs counter to conventional wisdom on effective influencing, where the veracity and consistency of information was always deemed of high importance (Paul & Matthews 2016:9). Russia has proven successful in direct persuasion and achieving obfuscation, confusion, and disruption using unconventional tactics (Paul & Matthews 2016:1,9; Wallis 2019; Watts 2018:227; Clapper 2018:2-4).

Enhancing Russia’s effectiveness is their propagandists’ remarkable responsiveness to events. They are not hamstrung, as traditional journalists are, by fact-checking or source verification, and they repeat disinformation and plagiarise stories, meaning they are first to publish and their articles rebroadcast by legitimate news outlets (Hellman & Wagnsson 2017:156). In accordance with the ‘halo effect’ discussed earlier (Kahneman 2011:83), Paul and Matthews (2016:4) explain that stories published by Russia’s propaganda machine leave a resilient first impression, the information of which audiences are more likely to accept and favour when later faced with conflicting messages. Russia’s use of repetition also creates an ‘illusory truth effect’, where repetition leads to familiarity, and familiarity leads to acceptance as truth (Paul & Matthews 2016:4; Jensen & Sear 2019; DeMarzo et al. 2003:909). This ‘illusory truth effect’ is related to the ‘exposure effect’ and the ‘frequency heuristic’ discussed in Part Three. Psychological research suggests that Russia’s IO strategy has the potential to be highly effective due to behavioural reasons explored in the literature in Part Three (Paul & Matthews 2016:9).

An effective psychological warfare plan adopted by Hezbollah against Israel in Lebanon (1982– 2000), allowed a relatively small guerrilla force, with no conventional capabilities, to exert its will on a regional power and ultimately force Israel’s withdrawal from Southern Lebanon. Hezbollah employed traditional PSYOPs archetypal messages directed at target audiences of the home front, the enemy, and neutrals (Schleifer 2006:3). Hezbollah enhanced their effectiveness by exploiting events for maximum propaganda effect, employing an excellent hearts and minds campaign, undermining Israeli Force morale, and using symbolism and emotionally evocative visual media as an effective weapon (Schleifer 2006:4,6,7; Afzal & Wallis 2016:83).

Hezbollah’s tactics were later adopted by Al Qaeda and ISIS – particularly the use of well- presented videos, which allowed control over a very selective view of frontline reality and 36 | P a g e 5249168 Cassandra Brooker Masters of Research allocated footage a significance well above its battlefield worth (Schleifer 2006:6). ISIS also learnt from Hezbollah the power of instigating events, with full control over the narrative and visual media, to gain powerful propaganda wins and force the Coalition into reacting. Hezbollah regarded the capture of symbolic events on video as their main mission and, with simple equipment and creative thinking, could net huge military and psychological dividends (Schleifer 2006:6; Afzal & Wallis 2016:83; Ruthven 2007:131). ISIS also learnt the importance of viciously imposing control over local civilian populations and projecting power to accentuate its military and religious dominance (Schleifer 2006:7-8; Afzal & Wallis 2016:75; Maltby & Thornham 2012:34).

When comparing Hezbollah’s planning of objectives designed for maximum psychological effect, with Western military IO planning, it highlights the lack of flexibility, opportunities, and imagination available in our non-conventional operational processes. Hezbollah took a systems analysis approach in familiarising itself with Israeli culture, its psyche, and in allocating the Israeli public and military into several functional or ideological sub-groups (Schleifer 2006:8-12). This systems analysis paid off, as it enabled tailored messaging against targeted audiences and increased their influence stock. They were also masterful at balancing local vicious anti-West rhetoric with credible messaging along human rights themes, to appeal to Western liberal mores (Schleifer 2006:13). This ensured their messaging and footage was aired on mainstream media, including in Israel, which was another key lesson ISIS learnt in creating global reach and a persuasive belief in its military prowess (Schleifer 2006:12-13; Lieber & Reiley 2016:50). Hezbollah displayed creative military thinking in its systems analysis approaches and astute employment of PSYOPs tactics, resulting in the successful combination of IO, guerrilla warfare, terrorism, and religious ideology to overcome the asymmetry between it and a major regional power (Schleifer 2006:15). Hezbollah provided key lessons surrounding the limitations of conventional Western tactics in the iWar, which enemies of the West have heeded. These limitations are discussed in the next section. Table 3 summarises the most effective influencing techniques arising from the literature reviewed in Part Two.

37 | P a g e 5249168 Cassandra Brooker Masters of Research Part Two – Influence Effectiveness Concepts

MEDIA BIAS & AUDIENCE SUSCEPTIBILITY • Fox News’ media bias, journalistic content choices & unfair airtime • Use of covert, invisible sources of influence manipulates & changes behaviours = hard to counter, especially if audience impressionable • Different populations as audiences (systems) have varied susceptibility to influence

EXPOSURE, INUNDATION & REPETITION • Repeated exposures are considerably persuasive • Amplifies messages, expands support bases, generates multiple story lines • Repeated exposures to environmental cues prime related representations in the memory • Known to affect judgment, preference and decision making • “Super-spreaders” spread disinformation like a virus, difficult to counter. • Inundation of numerous messages across multiple channels • Disinformation spreads ~six times faster than real news • Contributes to echo chambers, algorithmic tailored newsfeeds, biased search rankings • Sophisticated influence activities in uncontrolled, unbounded commons of the internet • Used in marketing, propaganda & censorship, with charismatic or authoritative messengers

DISINFORMATION / FAKE NEWS / FORGERIES / BOT ACCOUNTS • Extensive, industrial level, multi-pronged, multi-domain, multi-platform, 24/7, never ending • Fire-hose of falsehood (Russia) • Reach & speed of social media escalated the impacts of disinformation - highly effective in influencing • Remarkable responsiveness of propaganda to events / Contain element of truth = highly effective • First to publish, with false articles rebroadcast by legitimate news outlets • Rapidly evolving trial and error of various iWar tactics by adversaries / Unbound by ethics, legalities or journalistic standards

ORDERING & SEQUENCING • Ordering of search results has a powerful and persuasive impact • Users trust and choose higher-ranked results • Digital bandwagon effect, unequal skewing of exposure towards certain candidates • Gaming of algorithms to skew search results’ rankings

CONTROLLING THE NARRATIVE • An effective narrative must be simple, novel and resonate • Use of visual media to enhance and stir emotions • Out communicate adversaries / competition narratives • Harness power of instigating events, full control over the narrative = powerful propaganda wins

VISUAL IMAGERY & ENTERTAINMENT • Memes, symbology, and colours are more powerful than slogans or words as an effective advertising • Raw entertainment, humour, visual media, videos – content exploiting biases and stereotypes • Evocative imagery, slick videos/humorous memes deliver the point rapidly, enjoy longevity, easily shared, exploit waning attention spans

EMOTIONAL CONTAGION • Using emotion – particularly anger / outrage - captures the most attention on social media • Trolling / inciting societal divisions, riots, protests • Arouse audiences, evoke sympathy, provide humour, feed fury, and alter experiences

SOCIAL CONTAGION • Being authentic & build sense of community / social in-groups / Echo Chambers / Filter bubbles • Integration of human factors, cultural understanding, accurate linguistics, personalised tailoring of messages for maximum influence • Contributes to virtuous cycles in ‘disinformation system’

PERSONALISATION & TAILORING • Creation of tailored messages, micro-targeting & simultaneously engaging different groups of voters = effective influencing tactic • Customised slogans and tailored advertising, targeted towards specific geographic regions or demographic groups, prove extremely successful, benefit from links with everyday environmental cues

ADDICTIVE CONTENT & FEEDBACK MECHANISMS • Attention from receipt of likes / notifications and other feedback provides dopamine hit • Imagery & environmental cues can undermine program implementation, trigger addictive behaviours, influence cultural identities • Fake stories / memes / click bait fuel dopamine cycles of likes and shares among internet partisans, enabled by mobile internet Table 3: Summary of key findings in the influence literature of Part Two

Limitations of existing influence theory research and counter-IO approaches In reviewing the literature into why certain influence activities are effective, gaps in both the research and the limitations of Western approaches to countering effective IO campaigns becomes 38 | P a g e 5249168 Cassandra Brooker Masters of Research apparent. While Yaworsky (2009:652) contend there is a distinct lack of scholarly research into PSYOPs, despite being practised for decades, the modern iWar has moved beyond the military and, as Singer and Brooking (2018) explain: “war and politics have never been so intertwined – politics has taken on elements of information warfare, while violent conflict is increasingly influenced by the tug-of-war for online opinion” (Ibid:262).

The previous section showed that while much academic research exists into the effectiveness of certain media persuasion techniques, advertising, marketing and historic propaganda successes, less is known about how Western nations can effectively influence audiences in this contemporary, complex, iWar environment. The West is hamstrung by democratic values, ethics, credibility concerns and transparency requirements and the lines are blurred in the iWar between civil society, politics and the military. Therefore, this literature review relies upon a limited number of key resources in a nascent field from which to draw conclusions, as they relate to the contemporary iWar.

The more dated literature surrounding the effectiveness of PSYOPS (Schleifer 2006; Yaworsky 2009; Blackmore 2003) and electioneering (DellaVigna & Kaplan 2007; Gerber et al. 2009), which relates to the case studies examined in this thesis, neglects to address the contemporary dynamics of the uncontrolled, unrestricted frontier of the global internet commons – specifically social media – with its rapid growth, extensive reach, malicious content, echo chamber dynamics, and security deficiencies. This new dynamic battlefield is undermining traditional thinking and doctrinal approaches to IO. Nations are not just at war in the information environment, but so are their citizens, businesses and communities; and many Western democracies do not even realise it. Leaving the iWar for the military or government to contend with is no longer an option, it will take the collective effort of the entire nation to counter adversary activities intent on threatening societal structures and undermining the integrity of public discourse and democratic institutions (Watts 2018:227; Jensen & Sear 2019; Canadian SIS 2018:7,11; Clapper 2018:2-3).

Information Age Challenges Blackmore (2003) foretold the impacts of the information age, and associated global communications, regarding the ease and speed at which data could be transmitted, manipulated, and influence exerted. Despite significant improvements, he highlighted IO capabilities were still under-resourced and misunderstood, despite predictions that PSYOPs would likely become the weapon of first choice (Blackmore 2003:10-12). This gap, he argued, could be improved through exploiting the internet as a potential force multiplier, and when synchronised with kinetic effects, could have profound mission success (Blackmore 2003:10). Recognising the complexity of this 39 | P a g e 5249168 Cassandra Brooker Masters of Research future iWar battlespace, Blackmore (2003:17-18,23) also recommended TSA approaches be applied to non-kinetic IO targeting and that qualitative measures of effectiveness (MoE) be developed accordingly.

Blackmore (2003:12) also recognised the difficulties Western nations faced in the iWar: the complexity of deterring or coercing rogue states and non-state actors (with irrational, immoral leadership and proxy borders); exacerbated by the constant requirement to maintain international and domestic support. Western democracies find themselves at a distinct disadvantage – having been shaped by the Enlightenment, the West seeks to be logical, consistent, transparent and accountable, however “these are not the values of a good troll” (Singer & Brooking 2018:211) nor indeed an effective influencer. The Canadian SIS (2018:86) adds that, ironically, the greatest threats to democracy are now enabled by US companies: Facebook, Google, Twitter and YouTube.

The literature highlights the limitations of existing influence research and counter-IO approaches, specifically in the use of multi-platform social media exploited by authoritarian regimes and terrorist organisations. Conventionally weaker adversaries can easily maintain power over populations by controlling information and narratives, generating fear, steering opinion, ‘gas- lighting’ (psychological manipulation to cause doubt in one’s own sanity), bypassing traditional journalism, and quashing dissenters (Singer & Brooking 2018:49,70,95,103,116; Patrikarakos 2017:43). The leading example is Russia’s ‘active measures’ campaign, which employs various experimental, psychological IO methods on a multi-pronged media front, as part of its ‘Gerasimov Doctrine’. This systematic approach to influencing global populations contrasts with the haphazard way Western governments regard the modern iWar battlefield, leaving them struggling to regain the initiative or control the narrative (Singer & Brooking 2018:106; Wu 2016:41; Australian Army 2018). Hellman and Wagnsson (2017:154) agree that democracies, whether out of complacency or wilful ignorance, have not taken seriously the prospect that emboldened adversaries could “reshape the undefended post–Cold War liberal order” and engagement is unavoidable for addressing this undermining of democratic institutions and values.

Addressing Inundation and Inertia To even attempt to counter the sheer volume of only Russia’s influence activities is impossible. The ‘super-spreaders’ employed by the West’s adversaries spread disinformation on the internet like a virus (Singer & Brooking 2018:119). During the 2016 election, the American public’s insatiable demand for political click-bait enabled the spread of disinformation to occur on an unprecedented scale; which diluted the truth, and overwhelmed traditional news outlets and fact 40 | P a g e 5249168 Cassandra Brooker Masters of Research checkers. Disinformation spreads six times faster than real news, contributes to echo chambers, filter bubbles, algorithmic tailored newsfeeds, and biased search rankings (Epstein & Robertson 2015:2; Patrikarakos 2017:239; Canadian SIS 2018:9-10). Singer and Brooking (2018:121) explain that “the internet has created the equivalent of several billion newspapers, tailored to the tastes of every social media user – consequently, there exists a set of ‘facts’ for every conceivable point of view” (Ibid:121).

The sophistication of Russian propaganda and influence activities of adversaries in the uncontrolled, unbounded commons of the internet requires the West to abandon its simplistic understanding of the iWar (Canadian SIS 2018:39). Whilst Western nations forgot the propaganda lessons of World War II, (i.e. keep innovating, massive scale deployment, using all technical means and media available); Russia embraced the mantra: “to succeed, propaganda must be total” (Wu 2016:41), realising early that sporadic efforts would fail and continues to evolve its iWar campaign accordingly. Hellman and Wagnsson (2017) contribute to the literature through their analysis into why Russia has been so effective, while Western democracies have been hamstrung by inertia, ethical issues or stove-piped strategies. They recommend combining various iWar strategies but caution against inadvertently undermining other systems, and to ensure that disinformation aimed at foreign audiences does not unintentionally influence the domestic population (Hellman and Wagnsson 2017:158,164).

Hellman and Wagnsson (2017:153) align with Blackmore (2003) and Singer and Brooking (2018) in recognising the difficulties faced by Western liberal democracies in responding to strategic narratives involving propaganda, and the question of whether matching adversary use of disinformation in peacetime was unethical. Blackmore (2003:19) explains that IO is often incorrectly associated with propaganda and underhanded activities, which contributes at the political level to ineffective counter influence strategies, a loss of credibility, and inappropriate resourcing.

A Gap in Strategic Communication Western democracies have been unwittingly dragged into the iWar and as such, are compelled to act to counter hostile influence activities, respond to unwelcome narratives and safeguard their own narratives and democratic freedoms (Hellman & Wagnsson 2017:155; Watts 2018:89; Clapper 2018:395; Wardle 2019). As this realisation sets in, the gap in the literature relating to regaining the initiative in strategic communications is slowly being addressed. Additionally, NATO is championing the importance of a strong narrative as a strategic necessity for Western nations (Laity 2015:24-25; NATO 2016:5) Whilst Hellman and Wagnsson (2017) attempted to 41 | P a g e 5249168 Cassandra Brooker Masters of Research address the gap regarding Western approaches to the iWar, their recommended strategies were reactive and defensive, further demonstrating the obstacles facing democratic nations in this space. They (2017:166) also recognised the constraints to activating iWar strategies, where political and strategic cultures, policies, historical experiences, and changing threat perceptions all influenced the choice of strategy or more often, inaction. Blackmore (2003:13-14) urges the coordination of information outputs by all government departments, from the outset of a crisis, to ensure appropriate collective messaging. Canada’s SIS (2018:26) highlights the need for enabling strategic counter-influence capabilities, explaining the current “naïveté of technology companies, futurists, the general public, and policy-makers” (Ibid:26) leads to an underestimation of how much damage can be done to Western democracies by unscrupulous adversaries with sophisticated IO campaigns.

Blackmore (2003:20) recognised the importance of tailored, focused messaging for effectiveness in his advocacy of conducting ‘Human Factors Analysis’ on target audiences – as a critical enabler to a systems targeting process focused on the human psyche. However, even though human terrain analysis is still conducted by the ADF at the operational level, it could be argued such detailed analysis and extensive use of intelligence resources would be wasted on the vast target audiences of the internet, with 24/7 information saturation, massive scale disinformation, and the rapidly evolving trial and error iWar tactics of our adversaries. Military IO cannot keep pace in monitoring the modern technological iWar battlespace and, therefore, unified messaging and a strategic narrative must be issued from the national level.

Limitations of Military Influence Other weaknesses and inefficiencies identified in the literature regarding military IO strategies included: the construction of ‘imagined audiences’ in the absence of access to the civilian population, illiteracy and innumeracy in target audiences, colonialist rhetoric, and using cultural assumptions, inappropriate imagery, or language (Maltby & Thornham 2012:35-36,43-44). Additionally, PSYOPs has a poor public reputation, being associated with the ‘black arts’ of deception and disinformation, and often failing in its effectiveness due to political interference or risk aversion of senior leadership (Taylor 2007:96).

There were also limitations raised in the literature regarding applying measures of effectiveness (MoE) to influence activities. Notwithstanding ‘big data’ issues presented by the iWar environment, many MoE remain quantitative rather than qualitative, and intelligence collection is not effectively tasked in response. PSYOPs planning is often compartmented from kinetic operational and intelligence planning, with the MoE development and analysis often left to the 42 | P a g e 5249168 Cassandra Brooker Masters of Research public affairs cell to determine in isolation (Goble 2002:22,45; Schneider 2002:8,16). Table 4 provides a comparison of Western democratic nations’ limitations with adversary freedom of action in conducting IO.

Western Democracies Adversary groups / Authoritarian Regimes

Multi-domain, multi-pronged campaigns involving extensive, industrial IO / PSYOPs siloed into military operational effects & level network of military, government, civilians, businesses, state-owned targeting. Limited cross-agency or WoG approaches, media, local tech companies, hacker groups etc. Have effective narratives strategy or narrative. that resonate.

Extensive use of deception and disinformation. State-owned media is Hamstrung by democratic values, ethics, credibility biased, distributes propaganda, and not held to journalistic standards. concerns and transparency requirements. Need to be Exploit cognitive biases and heuristics. Not bound by national borders, logical, consistent, and accountable to both domestic activities often untraceable and unconcerned by Western democratic and global audiences. ethics, values or legalities.

Linear thinking, attack of nodes (often kinetically) - Employ systematic, creative approaches to analysing and influencing rather than developing IO effects non-kinetically global populations i.e. attack links to create social division rather than targeting links / relationships / cognitive biases nodes, e.g. Hezbollah, Russia.

Complacency or wilful ignorance – reactive to adversary Proactive, rapidly responsive to events, will target everyone, including iWar attacks, risk averse in implementing own iWar own population. No recognised boundaries. Strategic or Political-level activities. Too bureaucratic or hesitant to weave an endorsement. Unconcerned by exposure or consequences. effective narrative.

Focus on technical cyber operations rather than Multi-faceted, multi-domain, complex systems approach to IO influencing. A simplistic understanding of the iWar. campaigns. Theft of personal data for personalised, tailored messaging. Military doctrine does not cover social media. One Use emotion as an arousal – particularly anger as a contagion. message fits all approaches. Exploitation of social contagions. Use of humour and imagery.

Overwhelmed by the open, boundless, unrestricted Super-spreaders ensure the unprecedented spread, inundation, speed, commons of the internet and the associated exposure, repetition of disinformation. Trolls and bots exploit access of uncontrolled big data. Concerned about encroaching on internet to enhance echo chambers, filter bubbles, and emotive content civil liberties & freedom of speech. in Western countries to undermine democratic institutions and elections.

Hold paradigm of being at war; iWar activities therefore top priority and IO capabilities under-resourced and misunderstood, highly resourced to achieve obfuscation, confusion, and disruption within often incorrectly associated with propaganda & perceived enemy populations. Provides traditionally weaker adversaries underhanded activities. disproportionate power over West. Centrality of focus on influence.

Fact checking and monitoring disinformation is a High volumes of state-sponsored or troll factory generated content massive job, which constrains profitability and is overwhelms fact checkers, platform content monitoring, and traditional unpalatable for social media companies and journalism. Ensures fake news is seen first, provides resilient first mainstream media to ally with governments & impression. intelligence agencies.

Traditional mainstream media hamstrung by tone, Out-publish and out-manoeuvre traditional forms of journalism through fairness, allegiance to facts, and context over constant inundation of salacious, addictive clickbait. conclusions. Table 4: Comparison of Western democratic limitations with adversary freedom of action in IO

Whilst these obstacles to effective influencing by Western nations represent both a gap in the literature and in current practical approaches, Singer and Brooking (2018:211) explain the very openness of democracies and the internet could be advantageous in the iWar. The conduct and countering of IO is no longer the purview of the military or even the government, there is now a requirement to engage civilian media and tech companies in the iWar. One of the most successful 43 | P a g e 5249168 Cassandra Brooker Masters of Research strategies for countering hostile influence, recommended by a number of researchers, is the education of citizenry to apply critical thinking to online information and media reports (Canadian SIS 2018:11,49; Singer & Brooking 2018; NATO 2016; Wardle 2019). This enables people to identify disinformation and empowers communities to dynamically counter electronic enemies without political or military restraints, as demonstrated in Lithuania (Singer & Brooking 2018; NATO 2016; Uren 2017; J. Mark Services Inc. 2018). Disinformation poisons public debate and threatens democracy, the SIS (2018:11,49) contend that raised public awareness and online literacy skills are needed to distinguish the truth, counter adversary narratives and deception tactics, and halt the viral spread of disinformation. So, while the challenges from hostile influence activities have reached a new level of complexity in the contemporary hyper-connected iWar battlespace, and whilst more than 3.8 billion people are connected to the internet, 2.9 billion on social media (Canadian SIS 2018:61); Western nations must view this complexity as an opportunity to overcome its limitations and exploit the vast online populace as a resource in countering its enemies. A key step in this process is understanding the research into how influence activities impact on friendly and adversary system behaviours.

How influence activities impact on system behaviour This section links literature related to influence effectiveness with the complex system theories explored in Part One, as well as the behavioural theories examined in Part Three, which all ultimately affect the influence stocks of the case studies analysed in this thesis. From a systems perspective, the definition of an effective influencing activity would be one that forces a complex social system to adapt its behaviour as a result, based on its balancing feedback loops (Figure 2). As demonstrated through the literature reviewed so far, much has been written about the effectiveness of various influencing campaigns and how adversary systems are enabled by the rapidly evolving, contemporary iWar environment. However, there are no systematic or standardised assessment tools provided in the academic nor grey literature for holistically analysing participant iWar system behaviours.

The very nature of PSYOPs is to influence an audience’s emotions, motives, objective reasoning, and ultimately change its behaviour favourably towards the originator’s objectives (Goble 2002:15). Contemporary asymmetric warfare now involves political, civilian, military, and economic spheres; meaning the power of kinetic technologies has been displaced by the need to influence, persuade, and control those audiences (Maltby & Thornham 2012:33; Yaworsky 2009:658; Brown 2019; Bienvenue et al. 2018a). Analysing these audiences through a systems thinking lens, allows a deeper understanding of the conditions in which such complex social systems exist, the vulnerabilities that may be targeted, and the sensitivities, stressors, motivations, 44 | P a g e 5249168 Cassandra Brooker Masters of Research and attitudes providing interconnections and driving system flows. Hellman and Wagnsson (2017:156) emphasize the importance of exploiting linkages and relationships within an audience system, by using cross-media messaging and multiple interconnections in a reinforcing feedback loop.

Understanding system interconnections is also important for understanding the vulnerabilities, weaknesses, and effects of discord within one’s own social system in order to strengthen it from enemy influence. As Blackmore (2003) explains, “discord across an alliance can have a dramatic effect on the will of an adversary, by drastically improving his morale” (Ibid:15), and as discussed in Part One, misguided attacks can actually strengthen enemy anti-fragile systems (Stroh 2015:83; Albino et al. 2016:15). While the effectiveness of influence activities in changing system behaviour is difficult to measure, a good systems analysis of the target audience exposes links, which are the Achilles' heel of the network, because they can be exposed, manipulated and weakened (Blackmore 2003:23; Hall & Citrenbaum 2010:125; Meadows 2008:16,153). Effective systems analysis also determines target system levels of fragility, robustness or anti-fragility, which in turn identifies characteristics or attributes that may be effectively targeted, through a variety of psychological, economic, and political measures (Albino et al. 2016:5-6; Meadows 2008:77-78).

Even more effective is the use of covert, invisible sources of influence, such as exploiting biases and other psychological heuristics, to manipulate and alter system behaviours. These covert tactics are harder to counter or defend against, especially if the system comprises impressionable people of certain demographics who believe they have made decisions autonomously and without coercion (Epstein & Robertson 2015:5-6,8; NATO 2016:4; Benkler et al. 2018:273). It could be argued that the gaming of algorithms and creation of echo chambers during the 2016 US Election was an example of this covert influence. Epstein and Robertson (2015:8) explain this invisibility is especially dangerous as a means of control, in biasing voter behaviour, and influencing attitudes and beliefs; which can ultimately effect the outcomes of close elections.

Rather than defining Russia’s ‘active measures’ as one system of one state, the Canadian SIS (2018:15-20) characterises all disinformation disseminated online as a complex system in itself, spread through a network of independent actors, supported by an ecosystem of websites. The SIS (2018) conducted a systems analysis of this complex, dynamic ‘disinformation system’ and identified sub-systems of actors interconnected by different motivations, ideologies and incentives. This allowed a deeper understanding into the problem of online disinformation, arguably in more detail that may have been achieved through traditional intelligence approaches 45 | P a g e 5249168 Cassandra Brooker Masters of Research such as network or stakeholder analysis (Canadian SIS 2018:18-20). They argued that by focusing on state actors, while ignoring the roles and motivations of independent actors, risked oversimplifying the complexity of the system and therefore limits effective counter IO solutions (Canadian SIS 2018:21-22). Whilst seeing the big picture is important, analysing all online disinformation as a whole system creates an unwieldy, evolving boundless problem, which may prove overwhelming to policy-makers deciding on solutions or dilute the effects of those solutions. This approach by the SIS (2018) also neglected to consider target audiences as systems and the effects that disinformation and influence had on their behaviours, as researched by NATO’s Strategic Communication Centre of Excellence (2016:4) and Yaworsky (2009:665).

This section has highlighted how influence activities in the literature can impact on own force, adversary, and neutral systems in the iWar, particularly in leveraging the interconnections and relationships within those systems and the associated feedback loops. Understanding a system’s level of resilience and accessibility is also key to employing appropriate and effective influencing tactics to manipulate its structures, mental models and behaviours. Figure 3 provides indicative characteristics of various systems’ resilience and accessibility, which have both been targeted and have conducted iWar tactics throughout history. If this persuasive manipulation can be achieved using covert exploitation of heuristics of elements within the system, it proves even more effective and underlines the importance of holistically understanding participant iWar systems.

Figure 3: Indicative characteristics of systems targeted by influence activities

46 | P a g e 5249168 Cassandra Brooker Masters of Research Effectiveness Summary Part Two of this chapter reviewed the literature related to the effectiveness of various influence activities across a number of disciplines such as political propaganda, advertising, marketing, election campaigning, and PSYOPS. Important gaps were identified in the academic literature regarding the effectiveness of influence activities in the contemporary, multi-faceted, online iWar, particularly in the obstacles facing Western democratic nations and the holistic approaches they should employ instead of traditional, doctrinal, stove-piped methodologies. The academic knowledge explored in Part Two is an area that should be expanded to address the gaps surrounding the level of effectiveness of Australian military and political influencing activities, outside of niche Department of Defence siloes.

Section Three concluded Part Two by reviewing the literature regarding how influence activities impact on system behaviours by linking into the system theories explored in Part One. Section Three highlighted the importance of understanding a system holistically and its associated resilience before developing solutions to change behaviours. The behavioural theories contributing to the effectiveness of influence activities covered in Part Two, will be examined further in Part Three, including how they may impact on the influence stocks and flows of the two case study systems analysed in this thesis.

Part Three: Behavioural theory impacts on stocks and flows of influence There is a vast body of literature relating to ‘Behavioural Economics’ and ‘Nudge’ theory, which is not possible to explore fully in this chapter. Therefore, this abridged review will explore the key psychological concepts relating to why certain influence activities prove more effective and how those theories apply to analysing the associated impacts on a complex social system. Systems thinking enables people to see their behavioural contributions to problems, increasing their ability to develop effective solutions (Stroh 2015:83). Meadows (2008:107) explains that ‘bounded rationality’ impacts on system functioning, as humans misinterpret information, misperceive risk, and focus on current events rather than long-term behaviour (see Glossary, Appendix 1).

There is less a gap in the literature than a gap in contemporary application of behavioural theory to systems problems by Western democracies in the iWar. A review of the literature highlighted key psychological concepts that could assist in determining what types of influence activities are most effective at leveraging complex social system behaviours, and how vulnerabilities could be targeted effectively. It is imperative the West catches up in understanding and exploiting this wisdom. As Paul and Matthews (2016:8-9) explain, the contemporary Russian propaganda model features psychological techniques and peripheral cues that are highly effective, including their 47 | P a g e 5249168 Cassandra Brooker Masters of Research tactics of contradiction and inconsistency, which, as explained in Part Two, are actually counter to conventional knowledge on effective influencing, where the veracity and consistency of information was deemed highly important. Propaganda is a form of communication designed to manipulate a target population by affecting its beliefs, attitudes, or preferences in order to obtain behaviour compliant with political goals of the propagandist (Benkler et al. 2018:29). Therefore, understanding how psychological and cognitive theories enable effective influence and contribute to shifts in behaviours is vital for informing this study.

Behavioural theories for influencing audiences A number of human behavioural traits have been exploited by propagandists, marketers, advertisers, and campaigners to ‘nudge’ or persuade audiences more effectively. Many of these techniques, which capitalise on heuristics and bias, seek to ensure target audiences only superficially analyse information with their ‘System 1’ thought process, that is, they take mental shortcuts and make quick, automatic decisions based on emotion and unconscious processes of perception and memory. One example is Russia’s use of repetition, which can create an ‘illusory truth effect’. As a form of persuasion bias, this leads to familiarity, and familiarity leads to acceptance as truth (Paul & Matthews 2016:4; DeMarzo et al. 2003:909; Kahneman 2011:81). This frequency heuristic also relates to the ‘exposure effect’ previously discussed, as demonstrated by the ‘Search Engine Manipulation Effect’ (SEME) and Kahneman’s ‘halo effect’ (Epstein & Robertson 2015:1; Paul & Matthews 2016; Kahneman 2011:82-83).

Exploiting the intuitive, automatic ‘System 1’ thinking reinforces numerous cognitive biases, argument fallacies and logic errors, which are often exploited by influence industries. Whereas ‘System 2’ enables critical, rational and deductive thought, it also requires cognitive effort and conscious attention; therefore, people usually default to their unconscious, effortless ‘System 1’ thinking, due to limited attention spans and computational capacity (Thaler & Sunstein 2009:19- 20; Facione 2015:16-17). This is why personalised, emotive messaging and narratives framed as stories, which play on biases and social norms, can be so effective. Conversely, this is also why electoral candidates making first impressions with complex arguments and detailed statistics often fail to influence voters (Thaler & Sunstein 2009:20; Patterson et al. 2007:61).

‘System 1’ thinking relies heavily on a number of cognitive heuristics for making rapid judgements, and even good critical thinkers fall for ‘System 1’ thinking errors. The most frequent heuristics operating in ‘System 1’, and exploited by effective influencers, are listed in Table 5 (Facione 2015:18-19; Kahneman 2011:80). These heuristics, especially when exploited, can lead people to jump to conclusions, make inaccurate estimates, unwise decisions, and have mistaken 48 | P a g e 5249168 Cassandra Brooker Masters of Research expectations, unfounded fears or biases. ‘System 1’ is gullible and biased to believe and, in the absence of ‘System 2’ thinking, will generate its own context, discard rejected alternatives, and avoid conscious doubt (Kahneman 2011:80-81; Hiscox 2018). Even a nonsensical statement will evoke initial belief, which is why ‘fake news’ and ‘clickbait’ are so successful with non-critical thinkers (Watts 2018:249; Facione 2015:24-25). Additionally, due to ‘System 1’s speed, ease and dominance over ‘System 2’, people are more likely to be influenced by baseless persuasive messaging, such as commercials and propaganda, when they are tired or depleted (Kahneman 2011:81).

System 1 System 2 Automatic, quick, no effort, relies on heuristics and Deliberate, effortful, slow. Requires conscious attention. Only unconscious processes of perception and memory. Biases mobilised when System 1 doesn’t offer a ready low-effort answer cannot be turned off. Gullible, generates its own context, discards rejected alternatives, does not exercise conscious doubt, limited attention span. ‘Availability’ – assess an event as more or less likely ‘Satisficing’ – choosing the first alternative that is ‘good enough’ depending on how readily examples come to mind or and discarding the rest, without proper critical analysis swayed by recent memories

‘Affect’ or gut reaction – for rapid decision making ‘Risk or loss aversion’ – leads people to avoid taking risks

‘Anchoring with adjustment’ – using incorrect scales (or ‘Association’ – where a word or idea reminds us of anchors), or readily available numbers as a starting point something else resulting in biased probabilistic reasoning, which results in evaluative judgments that are difficult to sway ‘The illusion of control’ – where people overestimate their ‘Simulation’ – in imagining various scenarios playing out, ability to control events, overly confident, overly optimistic, leads to phobias, incorrect expectations especially in their own abilities ‘Similarity’ – comparison of a personality or situation like ‘Hindsight Bias’ – misconstrue / overestimate their personal our own or ‘Representativeness’ – using implicit influence on past events, selective memory stereotype biases or explicit biased judgements Jump to conclusions, make inaccurate estimates, unwise ‘Dominance Structures’ – applied to decisions, bias toward decisions, hold mistaken expectations, unfounded fears chosen option i.e. elevating the merits and diminishing the flaws and biases, bad habits of a chosen option, relative to other options

Table 5: System 1 and System 2 Heuristics

‘System 2’ thinking is not immune to heuristics and biases (Table 5), which can also be exploited by influence activities. These heuristics lead people to avoid taking risks, make evaluative judgments that are difficult to sway, misconstrue their personal influence on past events as a ‘Hindsight Bias’, and to apply ‘Dominance Structures’ to decisions, for example, elevating the merits and diminishing the flaws of a chosen option relative to other options (Facione 2015:20- 21). The dangers of reduced attention spans, cognitive limitation, and a lack of critical thinking in contemporary Western society from adversary influence is highlighted by Kahneman (2011:86), who explained that a lazy ‘System 2’ would endorse many intuitive beliefs arising from the impressions generated by ‘System 1’. 49 | P a g e 5249168 Cassandra Brooker Masters of Research

To effectively influence and ‘nudge’ systems, Hiscox (2018) argues we need to target the psychological roots of behaviour by understanding human cognitive limitations, biases, self- control issues, and social pressures. Influence techniques highlighted in Part Two demonstrated the West’s adversaries readily understood these concepts in their exploitation of popular thinking in the iWar. Argument fallacies, such as ‘straw-man’ tactics or using ‘red herrings’, were commonly employed by Russia and during the Trump US presidential campaign in 2016. Also employed extensively in the iWar are cognitive biases such as: ‘Persuasion Bias’ through exposure, as discussed (DeMarzo et al. 2003); the ‘Appeal to Authority Bias’ utilised by ISIS and conspiracy theorists to enhance perceptions of source credibility and expertise; and ‘Confirmatory Bias’ used extensively by Russia to reinforce discriminatory beliefs, stereotypes and opinions in internet echo chambers to exacerbate social divisions in democratic states.

Adversaries in the iWar combine knowledge of these behavioural traits with other tactics that grab the attention of ‘System 1’ cognition, such as: ‘Present Bias’, repetition, situational or environmental cues, and use of imagery, emotion, colour, stories, and addictive dopamine rewarding feedback mechanisms on social media (Berger & Fitzsimons 2008:2-4; Hiscox 2018; Canadian SIS 2018:26-27; Thaler & Sunstein 2009:3,255). The research demonstrated these approaches influenced humans outside of conscious awareness and created an accumulated conceptual fluency over time (Berger & Fitzsimons 2008:11-12). Gerber et al. (2009) concur, stating “changes in opinion often occur without a subject being able to recall the facts that caused opinions to shift” (Ibid:48), which explains why fact checking lost out against the ‘culture war’ of the 2016 US presidential election.

One of the most powerful tools in the psychological arsenal of the iWar is the effectiveness of social contagion. Bond et al. (2012:295-296) found social messages and close friends’ contagion were the most effective form of influence, particularly when conducted face-to-face and for those messages appealing to social pressures and identity. The power of social norms and group approval ensures a tendency to conform and primes identities to be associated with certain behaviours, as people fear the social costs, such as: stigma, rejection, or ridicule. This relates to the ‘Appeal to the Masses’ logic error, which is often applied by politicians and advertisers to emphasise to what or whom the majority were supposedly choosing or supporting (Hiscox 2018; Thaler & Sunstein 2009:65). Peer comparison has considerable impacts on behavioural change, for example: investment decisions influenced by friends could result in speculative booms, and ‘pluralistic ignorance’ of group-think creates cultural bandwagons or herd behaviour (Datta & Mullainathan 2014:29; Thaler & Sunstein 2009:59,66). Also, the research showed that persuasion 50 | P a g e 5249168 Cassandra Brooker Masters of Research bias was significantly more effective if information were repeatedly conveyed by multiple sources connected through a social network (DeMarzo et al. 2003:911; Bond et al. 2012:295).

The demonstrated power of ‘homophily’ ensured like-minded people grouped together and created echo chambers online, reinforcing views and sharing content that agreed with their group biases (Singer & Brooking 2018:123). Within these echo chambers, people spread fake or contentious messages to a supportive audience for attention, which endorsed confirmation biases, strengthened group bonds, and was difficult to halt (Singer & Brooking 2018:125-126; Paul & Matthews 2016:6). In addition to confirmation bias, ‘persuasion bias’ (achieved through repetition and charismatic or authoritative messengers) also explained why individuals' beliefs often evolved to match the views of social groups with which they interacted (DeMarzo et al. 2003:913-914). The combination of internet-accelerated homophily and confirmation biases has fragmented civil society, which Russia and groups like ISIS have expertly exploited to draw support for themselves while exacerbating divisions, and spreading conspiracy theories and hate.

The impacts of social mobilisation in influencing online networks have many implications in effecting behavioural change, both online and offline, which Bond et al. (2012:297-298) argued requires more research to both understand the role of social media in society and to identify which real-world behaviours are amenable to online interventions. DeMarzo et al. (2003:915-917) contend that a person’s beliefs and political orientation could even be predicted based on their social network and certain demographic factors. This assertion also requires further research for democratic nations to understand the dangers of artificial intelligence, metadata collection (e.g. Cambridge Analytica), and to improve civil society’s resilience in the iWar. Cialdini’s research (2003:105) into communications designed to activate social norms to produce societally beneficial conduct, cautioned about the circumstances where such messaging could backfire, and advocated for more research in this area (2003:109). Cialdini (2003) explained that whilst social influence campaigns may be well intentioned, messaging that: ‘many people are doing the wrong thing’ actually provided a powerful normative message that ‘many people are doing something’, therefore justifying people’s decisions to join the bandwagon (Cialdini 2003:105-107).

The other powerful tool in the psychological arsenal of the iWar is influence through emotions, especially anger, which bypasses ‘System 2’ thinking completely (Peters 2012). Emotions arouse audiences, evoke sympathy, provide humour, feed fury, and alter experiences, so unsurprisingly capture the most attention on social media. The stronger the emotions involved, the more likely content would go viral, with anger spreading faster, further and being more influential than any other emotion (Singer & Brooking 2018:161-162). Unlike social contagion, emotional contagion 51 | P a g e 5249168 Cassandra Brooker Masters of Research can spread without direct social interaction and in the absence of non-verbal cues – anger is exciting, addictive and explains the rapid rise of online trolling (Singer & Brooking 2018:162- 165). Hitler recognised the power of stirring emotion in creating persuasive propaganda and appealing to the masses. Being caught up in a mob’s fervour led to changes in mental processing of information, even to the point of believing something different (Wu 2016:113). Hitler realised animated crowds extended the possibilities of influence, the collective excitement opened up their minds to information (Wu 2016:110-115). Table 6 summarises the concept linkages between the influence literature of Part Two and behavioural literature reviewed in Part Three of this chapter.

Part 2: Effective Influence Part 3: Behavioural Theory

MEDIA BIAS, COVERT INFLUENCE • Bias exploits intuitive & automatic ‘System 1’ thinking & AUDIENCE SUSCEPTIBILITY • Certain demographics more impressionable or gullible

EXPOSURE, INUNDATION, • ‘Persuasion Bias’ AMPLIFICATION & REPETITION • ‘Illusory truth effect’ • ‘Exposure effect’ and ‘frequency heuristic’ • Exploits ‘Availability’ heuristic of System 1 • Exploits ‘Association’ heuristic of System 1 • ‘Appeal to Authority’ cognitive bias

DISINFORMATION / FAKE NEWS / • ‘Halo effect’ = resilient first impression FORGERIES / BOT ACCOUNTS • Argument fallacies, e.g. ‘straw-man’ tactics or ‘red herrings’ • Grab the attention of ‘System 1’ cognition = not critically analysed • Occurs outside of conscious awareness, creates accumulated conceptual fluency • Even nonsensical statements will evoke initial belief

ORDERING OF SEARCH RESULTS, • System 1 focusses on higher ranked results NEWS FEED RANKINGS & • Sequence matters, ‘halo effect’ increases weight of first impressions, subsequent info is discarded SEQUENCING • Coherence-seeking ‘System 1 generating intuitive impressions

CONTROLLING THE NARRATIVE • Narratives framed as stories, play on biases, emotions & social norms exploits unconscious, effortless ‘System 1’ thinking • A good narrative that evokes emotion will stimulate action

VISUAL IMAGERY & • System 1 immediately responds to imagery, humour, emotion ENTERTAINMENT • Exploits ‘Representativeness’ bias • ‘Confirmatory Bias’ reinforces discriminatory beliefs

EMOTIONAL CONTAGION • ‘Affect’ heuristic of System 1 – or gut reaction, for rapid decision making. • Anger bypasses ‘System 2’ thinking completely, spreads faster, further, most influential emotion • Anger and outrage are exciting and addictive • Power of emotion influencing ‘System 1’ affects even well-educated, cynical, critical thinkers

SOCIAL CONTAGION • Exploits peer group pressure, social norms, group approval • Appeal to the Masses logic error • ‘Appeal to Authority Bias’ • Power of ‘Homophily’ - endorses ‘Confirmation Biases’ • Exploits ‘Similarity’ heuristic of System 1

PERSONALISATION & TAILORING • Unconscious, effortless ‘System 1’ thinking, • Limited attention spans and computational capacity • System 1 immediately pays more attention to personalised content

ADDICTIVE CONTENT & • Addictive ‘system 1’ behaviours related to emotional & social contagions FEEDBACK MECHANISMS

Table 6: Linkages between the Influence literature of Part 2 and Behavioural literature of Part 3

52 | P a g e 5249168 Cassandra Brooker Masters of Research The power of emotion in influencing ‘System 1’ thinking and subsequent behaviours affects even the most well-educated, cynical, critical thinkers, as it transports them from the role of listener to participant (Patterson et al. 2007:61). This research highlights the importance of developing a good narrative that evokes emotion through storytelling to stimulate action, offers hope, and prevents counter-arguments (Patterson et al. 2007:61-66; Stroh 2015:32; Laity 2015:24-25; Brown 2019). The human dimension plays a significant role in the contemporary iWar (Goble 2002:4), due to the involvement of social systems, therefore, it is imperative Western nations adopts behavioural economics research to confront the complex adaptive system of the human mind in future influencing activities.

How behavioural theories apply to systems analysis Understanding nudge theory and behavioural economics changes mindsets regarding assumptions about people’s behaviour and in analysing what motivates their decisions. This realisation has led to new approaches across various disciplines in understanding the importance of psychological factors and mechanisms that drive the success or failure of programs, policies, or interventions (Datta & Mullainathan 2014:7-10). However, aside from newly introduced initiatives in the ADF such as Information Environment Advanced Analysis training (J. Mark Services Inc. 2018), Potentium wargaming (IWD 2018), and TSA courses; and not withstanding gaps in the literature – democratic nations have been slow to respond to the iWar threat, slow to adopt systems thinking, and have largely ignored research into effective influencing techniques and associated behavioural theories as they apply to the contemporary battlespace. This section reviews the behavioural theory as it applies to complex adaptive social systems and, combined with the research reviewed in Parts One and Two, will inform this thesis’ approach to analysing the influencing effectiveness of two contrasting systems.

Understanding behavioural economics is important for effective analysis of complex adaptive social systems, as human behaviour is often unpredictable and inexplicable. People choose their behaviours based on interpretations of what may happen, as a result of inaccurate mental models of cause and effect, and not all of these interpretations are anchored in reality nor in accordance with their espoused values (Patterson et al. 2007:49,97). Psychological research confirms actual behaviours are inconsistent with standard assumptions surrounding human decision-making due to deficiencies in mental accounting, choice overload and imperfect optimisation (Congdon et al. 2011:18-21; Hiscox 2018). The complexity of analysing the qualitative effectiveness of PSYOPs on the human mind was recognised by Goble (2002:25), who recommended the military adopt a non-traditional systems thinking approach to addressing this issue.

53 | P a g e 5249168 Cassandra Brooker Masters of Research There are six sources of influence (Figure 4), which reflect the separate and highly developed literatures of social psychology and organisation theory, and from which strategies could be derived from a proven repertoire of influence techniques (Patterson et al. 2007:77-78). To appreciate the ability to influence each level of a system, an understanding of the connections between behaviours, motives, abilities and limits of each is required to determine the best approaches, for example, to make the undesirable desirable, harness peer pressure, or create reward structures (Patterson et al. 2007:77-78). Datta and Mullainathan (2014:10) explain that many programs, policies, or interventions fail primarily because of the unpredictable way people behave or make decisions; largely due to the power of social influence and emotional contagion (Patterson et al. 2007:138; Singer & Brooking 2018:131). Therefore, behavioural economics provide useful explanations in both understanding systems that are being influenced and for diagnosing problems within systems. Figure 4: Six Sources of Influence Patterson et al (2007:78) Behavioural economics theories provide insight into the features of a system that are impacted upon by external pressures, such as its interconnections, stocks or flows; as well as identifying the characteristics of a system, such as its fragility or resilience. Understanding which cognitive biases are present in a system allows the effective design of communication and influencing activities to address, exploit, or take advantage of these biases (Datta & Mullainathan 2014:20; Stroh 2015). Whilst Wu (2016:22-23) argued advertising industries are unlike organisms, with no organic limits on their growth, he contradicted this by explaining a classic balancing feedback loop where those same industries took their ‘attention harvesting’ too far, engendered a vehement social reaction, which he termed the ‘disenchantment effect’ and their influence stocks were sunk. Understanding systems processes may have, therefore, helped advertisers during this time and prevented their race to the bottom through quick fixes, sub-optimisation, and drift to low performance (Meadows 2008:187-191; Wu 2016:16-17).

The Canadian SIS (2018:58,101) found that ‘sub-system’ hyper-partisan websites, within the broader ‘disinformation system’, exploited behavioural theory by confirming readership biases, using repetition, encouraging addictive sharing behaviours through viral content and shock value, 54 | P a g e 5249168 Cassandra Brooker Masters of Research and by incorporating audience feedback loops. Whilst these websites remained accessible to anyone, their interactive audience feedback loop ensured their ‘system’ benefitted from ‘filter bubbles’, where like-minded individuals communicating with each other see their beliefs converging faster to common group beliefs, which were difficult for system outsiders to counter or influence (DeMarzo et al. 2003:946; Paul & Matthews 2016:2). Due to these disinformation sub-systems using psychological theories and content bombardment, with humans increasingly relying on ‘System 1’ cognition; the internet has now become “a battlefield with real-world consequences and on which, only losers play fair” (Singer & Brooking 2018:240,261; Paul & Matthews 2016:6).

Systems, and the humans they comprise are influenced by observing the behaviour of others. As Patterson et al. (2007:20,47) explain, if you want to change a system, you have to change how people behave, and to do that, you must first change how they think; which is why understanding behavioural theory is so important. To effectively influence a system, vital behaviours must first be identified that will drive the required changes, and should not be confused with outcomes. Honing the scope of influence activities to target specific behaviours will concentrate efforts and ensure success in solving complex problems, without falling in to the systems trap of ‘quick fixes’ (Patterson et al. 2007:23,27,29,75-76; Meadows 2008:191; Stroh 2015:56).

Understanding behavioural impacts on the interconnections within a system is also important for effective influencing or targeting. The strengths and vulnerabilities of links, relationships and flows of information between nodes ultimately determines the characteristic resilience of the system – and are best identified through cultural analysis, as it relates to the attitudes, thoughts and behaviours of the system (Hall & Citrenbaum 2010:122-123,235-236). Appendix 5 contains a comprehensive table cross-referencing the linkages between the influence and behavioural literature reviewed in Parts Two and Three (as summarised in Table 6) with key systems theories discussed in Part One. This summary includes the twelve archetypes of common system structures (Stroh 2015:67) that produce characteristic patterns of behaviour.

Reviewing the research into behavioural economics provides a deeper understanding of the importance of engaging ‘System 2’ thinking when evaluating information and conducting analysis, thereby improving own-force system resilience and antifragility in the iWar. In realising the extent of unconscious bias and lack of cognition operating within the ‘System 1’ thought process, enables Australian iWar practitioners to imagine the consequences on domestic audiences that fail to use critical thinking (Facione 2015:24-26; Canadian SIS 2018:29,49; Thaler & Sunstein 2009:8). 55 | P a g e 5249168 Cassandra Brooker Masters of Research

Adopting behavioural economics into complex systems analysis improves understanding of target systems and their associated interconnections, as well as assisting in diagnosing own-system problems and enabling the creation of effective influencing strategies in the iWar. As Datta and Mullainathan (2014:12-15,32) explain, understanding behavioural economics is not merely about having better tools of persuasion, rather it provides the potential to change system behaviours without ever changing people’s minds. However, Western democracies must act now in taking a systematic approach to analysing problems and unlocking the potential of behavioural solutions in regaining control and achieving effective influence over their adversaries in the iWar.

Behavioural Summary Part Three of this chapter reviewed the literature surrounding psychological concepts and cognitive traits related to behavioural economics to determine what types of influence activities are most effective on system behaviours and outputs. Section One highlighted the numerous human behavioural traits that have been exploited by propagandists, marketers, advertisers, and campaigners to nudge or persuade audiences more effectively. Many of these techniques, which capitalised on heuristics and bias, sought to exploit target audience ‘System 1’ cognitive processes.

The review of the literature in Section Two sought to evaluate how behavioural literature applied to systems analysis, particularly in the influencing of complex social systems and the human element they comprised. Important gaps were identified, not so much in the academic literature, but in the West’s misapplication of the psychological techniques, behavioural economics, and nudge theory that could improve the effectiveness of influence activities against adversary and neutral systems in the iWar. The literature highlighted that behavioural theory underpins the success or failure of psychological influence operations, as well as provide deeper understanding of human factors supporting effective complex social system analysis.

Conclusion This literature review has critically examined the research of three independent literary disciplines comprising: systems thinking, influence activities, and behavioural economics; as they relate to effective approaches and concepts to be applied in the iWar, to achieve the study aim. It is believed that researching these three disciplines as they relate to each other, within the context of the contemporary iWar, as well as analysing what those findings specifically mean for Australia, is a unique approach. This review has highlighted the interlinking relationships between the

56 | P a g e 5249168 Cassandra Brooker Masters of Research concepts of the three academic fields and identified deficiencies in the literature, thereby establishing a theoretical basis for further academic research into contemporary influencing effectiveness. The chapter has outlined the key gaps in both current systems thinking and the application of behavioural theories and influence literature when applied to the contemporary iWar problem by Western democracies.

Figure 5 demonstrates the important overlaps in theory between the three research disciplines of literature reviewed in this chapter, as they relate to influencing and leveraging behavioural change in a system. The elements related to all three literary disciplines in the centre of Figure 5 comprise the hidden part of the iceberg (Fig 1), which are the fundamental system components that can be leveraged and influenced in the iWar.

Figure 5: Overlapping theory between the three literature disciplines reviewed

The outcome of this literature review was to identify key concepts and highlight gaps in the research, as they relate to the three arguments outlined in the introduction, in order to inform the comparative system analysis methodology of this thesis. As a result, three research questions were confirmed through reviewing the literature through the lens of the initial arguments. 1. How does systems thinking offer an effective approach to understanding complex social systems of our adversaries and provide a methodology for developing effective influence and counter-influence strategies for Western nations in the iWar? 57 | P a g e 5249168 Cassandra Brooker Masters of Research 2. Why is understanding the effectiveness of influencing activities imperative to appreciating how complex social systems adapt their behaviours based on balancing feedback? 3. Why is understanding psychological cognitive theories, with respect to influence effectiveness, central to analysing how external pressures impact on a social system’s behaviours and mental models?

Reviewing the systems thinking theories in Part One underscored the gap between academia embracing systems approaches to research, with that of Western militaries and governments who largely overlooked systems analysis for non-kinetic targeting and influencing activities, in favour of stove-piped, doctrinal processes. The references highlighted that linear analytic approaches were not suited to complex environments and unstructured human problems. When applied to complex problems, current doctrinal analytic approaches tended to over-simplify the system.

In Part Two, gaps were identified in the academic literature surrounding the effectiveness of influence activities in the contemporary, multi-faceted, online iWar; particularly regarding obstacles facing Western democracies and holistic approaches they should employ, instead of relying on risk-averse methodologies.

Important gaps were also identified regarding the application of behavioural theory in Part Three. These gaps were not so much in the academic literature, but in both the grey literature and the West’s ignorance of psychological techniques, behavioural economics, and nudge theories that could improve influence activities in the iWar and provide deeper understanding of human factors underpinning complex social system analysis.

Chapter 3 outlines the systems thinking approach, comparative case study research methodology, and case study selection techniques adopted for this research. The systems thinking literature explored in Part One of this chapter, combined with the effective influencing techniques and behavioural theories covered in Parts Two and Three respectively, provide a basis of understanding of the methodologies and concepts that will be applied critically and holistically to the two case study systems analysed in this thesis. This research aims to determine the associated positive and negative impacts on influence stocks within two comparative complex sociocultural systems, and apply those insights to answering the research questions and identifying the most effective approaches for Western democratic nations to employ in the iWar.

58 | P a g e 5249168 Cassandra Brooker Masters of Research Chapter 3 – Methodology and Case Study Selection

The supreme art of war is to subdue the enemy without fighting. Know your enemy and know yourself and you can fight a hundred battles without disaster.

Sun Tzu 544-496 BCE The Art of War (Giles 2017)

Introduction This chapter describes the research design, data collection, and analytical methodology adopted in this thesis to address the research questions arising from the gaps identified in Chapter 2, which are:  Firstly, how does systems thinking offer an effective approach to understanding the complex social systems of ourselves, our adversaries and target audiences; and provide a methodology for developing effective influence and counter-influence strategies for Western nations in the iWar?  Secondly, why is understanding the effectiveness of influencing activities imperative to appreciating how complex social systems adapt their behaviours based on balancing feedback?  Thirdly, why is understanding psychological cognitive theories, with respect to influence effectiveness, central to analysing how external pressures impact on a social system’s behaviours and mental models?

The research design elements of this thesis comprise an ontology of objectivism by using ‘critical realism’ as a methodological approach, which focuses on configurations, complex systems and case studies that provide narrative data from actual domains and events. As Gerrits and Verweij (2013:171-172) explain, critical realism uses the language of causality to describe the world and can generate provisional explanations of how events follow from previous events, what drives processes, and the mechanisms by which human behaviour transpires. Critical realism therefore, is the best suited epistemology to answer the research questions related to intangible social system’s behaviours and influence effectiveness. Critical realism provides a sound base from which to employ the alternative yet practical epistemological paradigm of ‘pragmatism’ for enquiry, which rejects and replaces the older philosophy of knowledge approach comprising: ‘Objectivism’, ‘Constructionism’ or ‘Subjectivism’ (Morgan 2014:1045-1051). Refer to Table 7 for definitions of the methodological elements of research design utilised in this thesis in hierarchical order.

59 | P a g e 5249168 Cassandra Brooker Masters of Research Pragmatism, as a lens of inquiry, focuses on social research directly connected to actions and consequences, and on examining the difference to outcomes that it makes to behave one way over another. Bohman (2002:499), Sil and Katzenstein (2010:44), and Morgan (2014) showed that pragmatism transforms traditional epistemological, moral, and metaphysical questions into practical problems, and this ‘practical turn’ of epistemology is especially relevant for the social sciences in identifying and solving problems. This meets the intent of this thesis in addressing iWar influence effectiveness problems within complex social systems. Adopting a pragmatic approach was the best way to analyse a “multifaceted, multidimensional process producing differential interconnectedness in different domains” (Bohman 2002:515), which the two case systems subject of this thesis presented.

Hierarchy of Research Design Term Definition Elements

A methodological approach focused on configurations, complex systems and case studies that provide a narrative and causation data from actual domains and events to generate provisional Critical explanations of how events follow from previous events, what drives processes, and the Ontology Realism mechanisms by which human behaviour transpires. Critical realism is a new alternative combining elements from both the traditional opposing ‘Objectivism’ and ‘Constructionism’ ontologies.

An alternative, more practical, epistemological paradigm for enquiry, which rejects and replaces the older philosophy of knowledge approach comprising: ‘Objectivism’, Epistemology Pragmatism ‘Constructionism’ or ‘Subjectivism’. Pragmatism transforms traditional epistemological, moral and metaphysical questions into practical problems for the researcher to identify and address, and complements both Critical Realism and Analytic Eclecticism.

Each system case is a set of elements coherently organised and interconnected in a structure Systems that produces a characteristic set of behaviours, often classified as its “function” or “purpose.” Framework / Mapping of Mapping system structures identifies the source of system behaviour. System behaviour Research Design Case Studies reveals itself as a series of events over time and informs analysis of interconnections, stocks, flows, and leverage points.

Narrative analysis gathers data on diverse events, human actions, and historical happenings and uses narrative analytic procedures to produce explanatory stories. These “stories” are Methodology / Narrative then integrated thematically into a unified temporally organised whole. Lends well to Strategy Analysis qualitative research and in holistically understanding case systems as configurations rather than groups of variables.

Table 7: Definitions of key methodological terms used in this thesis

While this thesis uses pragmatism as an epistemological lens, it will adopt a mixed paradigmatic approach, known as ‘analytic eclecticism’, in synthesising various qualitative research data to holistically understand the respective case configurations and provide practical solutions. As Sil and Katzenstein (2010:40) explain, a single paradigm, such as constructivism, is too one- dimensional for analysing open systems; thus, a holistic, multi-dimensional conception of systems is needed to integrate material, institutional and cultural aspects of social change.

60 | P a g e 5249168 Cassandra Brooker Masters of Research Analytic eclecticism complements a pragmatic epistemology, as it rests on a pragmatic set of assumptions, downplays rigid epistemic commitments, and focuses on the consequences of scholarship for concrete dilemmas (Sil & Katzenstein 2010:2,13). This approach, therefore, directly addresses the gaps in academic literature and knowledge relating to the iWar as identified in Chapter 2, in that despite abundant grey literature, academia has failed to research this problem effectively nor provide practical solutions for iWar strategists. Pragmatism addresses the gap between theoretical academic debate and demands for policy relevance and practicality, it restores the balance between detachment and engagement, and is therefore, strong justification for adopting this epistemological approach in this study.

This thesis addresses the aforementioned research questions by conducting a narrative analysis using a systems’ mapping analytical framework to compare two contrasting case studies as its research methodology. This chapter will build on the justification in Chapter 2 as to why a systems thinking approach has been chosen and how it will be applied as a methodology in comparing the two complex case studies. This chapter will also justify the case study selection, as well as highlighting why a case comparison, using narrative analysis, is the best vehicle for understanding the effectiveness of influence and behavioural models raised by the ‘how’ and ‘why’ research questions (Yin 1994:6); accordingly, this chapter is divided into two parts.

Part One justifies the selection and use of two contrasting systems as case studies to conduct this research into influence effectiveness in the iWar. The two specific case studies comprise: a successful influencing system of the terror organisation ISIS, contrasted with an unsuccessful system case of Hillary Clinton’s US election campaign in 2016. Part Two of this chapter outlines the research design used to conduct a narrative analysis of the system case studies. Part Two describes the methods of data collection and associated narrative analytic procedures used to produce explanatory stories relating to the case systems’ comparison process. Part Two also discusses the internal and external validity of the methodology, before concluding the chapter with a demonstration of how the methodology will be applied.

Part One: Case Study Approach A comparative case study approach, consisting of two contrasting, real-world systems, has been chosen to conduct this research into influence effectiveness; given case study design frameworks’ appropriateness in examining complex social phenomena and developing holistic knowledge of case system structures (Yin 1994:2,6; Flyvbjerg 2006:244). The ‘how’ and ‘why’ research questions are explanatory in this thesis, therefore the use of case studies, and their associated narrative data and contemporary histories, inform the preferred research strategy. This is because 61 | P a g e 5249168 Cassandra Brooker Masters of Research such questions deal with qualitative linkages needing to be traced over time, rather than quantitative frequencies or incidence. As Flyvbjerg (2006:239) contends, “there are more discoveries stemming from the type of intense observation made possible by case studies than from statistics applied to large groups”.

A case study comparative approach is central to human learning, providing context to the problem, greater understanding of the configurations of the system cases, identification of the variables, and whether they are aligned (Wievioka 1992:170; Flyvbjerg 2006:239; Yin 1994). The use of the two case systems to analyse the effectiveness of their respective influence in the iWar also enables a boundary to be drawn around each case. This allows research into the stocks, flows, and interconnections related to the relevant narrative data and research questions – rather than attempting to analyse the entire holistic organisation’s historical activities and outputs, which is beyond the capacity of this thesis.

In accordance with Yin’s advice (1994:45-46), each case was carefully selected to produce contrasting results for predictable reasons (a theoretical replication), and both cases served a specific purpose within the overall scope of inquiry; that is, the stark contrast provided by a successful influencing system compared with an unsuccessful system case study. Researching more than two cases comprising complex social systems in the iWar was also beyond the capacity of this thesis and a single researcher, particularly in developing a certain degree of “intimacy” with each of the cases under consideration (Rihoux & Ragin 2009:23; Flyvbjerg 2006:240). The two contrasting case studies analysed in this thesis are important for examining contemporary events, where the past behaviours contributing to system stocks cannot be manipulated, and to ensure coverage of contextual conditions highly pertinent to this research (Yin 1994:8,13; Polkinghorne 1995:17-18).

There are two main criticisms of case study methodology relevant to this research project. The first criticism is that case studies provide a poor basis for generalisation, and the second is that case analysis is susceptible to bias and the researcher’s influence on conclusions (Yin 1994:9-10). The research design adopted in this thesis attempts to mitigate these criticisms by seeking multiple differing perspectives from anecdotal, historical and narrative qualitative sources to build the most accurate system models possible, and draw out generalisations of why those systems’ configurations are successful or not. The selection of ISIS as a case to be researched in this thesis could be viewed as a biased decision, in that the author’s familiarity with the terrorist organisation is more extensive, compared with her knowledge of other adversary systems, due to recent military experiences. However, researching the unsuccessful case system of US presidential 62 | P a g e 5249168 Cassandra Brooker Masters of Research candidate Hillary Clinton’s 2016 election campaign, rather than another successful analogical adversary case, addresses the question of bias, as this research uncovers findings that are uncomfortable, unpalatable, and revealing of Western democratic weaknesses and vulnerabilities. This thesis also draws on the military intelligence experience of its author in enabling a deeper analytical study of the research problem using narrative data, having a comprehensive understanding of the iWar context and sufficient case-based knowledge; as well as the capacity to remain objective when considering the findings and recommending practical solutions.

Case Study Selection The premise of this thesis is that Western democratic nations are being out-communicated in the contemporary iWar and the best influencers have proven to be their enemies for a multitude of reasons, as detailed in Literature Review and summarised in Tables 2 and 3. Therefore, the selection of the successful case study of ISIS, as an adversary with a resilient closed system, contrasted strikingly with the unsuccessful case of Hillary Clinton’s election campaign, which is part of an open Western democratic system and demonstrated some fragility. The decision to use these two case systems was a deliberate choice for conducting an effective comparative analysis around a single stock of ‘influence’, in order to answer the research questions and provide practical recommendations for enhancing Western democratic participation in the iWar. These two case systems provide the opportunity to ‘close in’ on real life situations and test views directly in relation to phenomena as they unfolded in practice (Flyvbjerg 2006:237).

This section provides the reasons justifying the selection of the two case systems and explains why these cases specifically, are so valuable in exposing the lessons of successful influencing, thereby answering the research questions and providing practical solutions to strategic decision- makers. The three key reasons for selecting these two case studies are:

 Both cases are comparable in size, structure, and purpose.  Both cases are analysed with the same single stock of ‘influence’ and comprise akin sub- systems, meaning that feedback loops, relationships, and information flows are similarly themed for clearer analytical comparison.  Contrasting the resulting vicious or virtuous feedback loops of the two case systems, depending on their respective level of influencing success, highlights the differences in seemingly intangible behaviours, mental models, resilience, and characteristics of each system and contributes to answering the research questions.

63 | P a g e 5249168 Cassandra Brooker Masters of Research Clinton’s election campaign was not only a highly suitable case study for its comparative and contrasting features with ISIS, but also because it answers the ‘how’ and ‘why’ of the research questions as an acceptable methodology (Yin 1994:7). The selection of ISIS is justified in that the group provides a similar system structure, size, and purpose as Clinton’s election campaign for bounding and conducting an appropriate qualitative comparative analysis, as opposed to analysing a large state-based adversary system, such as Russia. The comparison of Clinton’s campaign with Russia’s Internet Research Agency (IRA) was also unfeasible, as the IRA directly targeted Clinton during the election. Additionally, the analysis of such a large, multi-faceted state system in detail is beyond the scope of a single researcher and may also have diplomatic implications.

The stark contrast between ISIS’ system success and the failures of the Clinton campaign is why these two cases were chosen to be compared in this thesis, for best addressing the research questions, clearly defining the outcomes of the system analysis, and highlighting a confronting reality surrounding Western democratic weaknesses in the iWar. As Rihoux and Ragin (2009:23) explain, the primary concern in guiding case selection should be the original research questions. They (2009:24) add that selected cases should share sufficient background characteristics and they recommend including cases with both ‘positive’ and ‘negative’ outcomes. Both of these two contrasting social systems meet these criteria, with their shared focus on increasing influence stocks and persuading populations.

The selection of these two contrasting systems for the case study comparison in this thesis is valuable for defining the outcomes of the research, for logically analysing complex, social systems in a contemporary iWar context, and for revealing the key findings surrounding system effectiveness. Vaughan (1992:177) contends that “case comparison can generate startling contrasts that allow us to discover, reinterpret, and ultimately transform our theoretical constructs”. Vaughan (1992) adds that this method can be particularly advantageous for elaborating theories, models, and concepts focusing on large, complex systems that are difficult to study and may have been previously precluded by complexity. The gap in the literature relating to influence effectiveness of systems operating in the iWar demonstrated that this is a complex area currently under-researched in academia. Wievioka (1992:170) agrees a case provides the opportunity to discover knowledge about how it is both specific to and representative of a larger phenomenon, adding that a comparative analysis may help deconstruct what common sense takes to be unique or unified, or construct the unity of what appears to be broken up into practical categories. This is effectively what the comparative systems analysis in Part Two of this thesis, through decomposition and re-composition, aims to achieve.

64 | P a g e 5249168 Cassandra Brooker Masters of Research Table 8 outlines some significant systems characteristics of the two chosen case studies demonstrating both their similarities and respective suitability for this research methodology.

Systems ISIS Clinton Campaign Characteristics

Establish & hold Muslim Caliphate, influence followers, Win US Election, influence voters/populace, Purpose control terrain & populace, hold power national governance, hold power

Stocks Influence, credibility, power, confidence Influence, credibility, trust, good will, confidence

Media: own media wing (global, local, internet) Sub-systems Media: global, national internet, own PR Political: own shadow governance, local govts Operating as part of Political: Sanders’ campaign, Obama Economic: donors, local business, own shadow branch system. May have administration, intelligence community Military: Al-Qaeda elements, tactical branches, foreign competing Economic: banks, businesses, donors fighters purpose/undermine Supporters: online groups, lobbyists, party Religious: clerics, mosques & communities under ISIS system’s purpose members, volunteers control, Wahhabism

Relationships: political, personal, economic, religious, Relationships: political, personal, economic, societal, membership, global religious, societal, membership, global, feminist Interconnections Communication links: internet, social media, news Communication links: internet, social media, news Between system media, strategy, religious messaging, travel, movements, media, strategy, campaigning activities, travel, elements / nodes and hold over terrain & populace, e-mag policies/messages, debate sub-systems Ideological values: emotional/social contagions, Democratic Values: Ethics, transparency, rules, homophily, biases, protect Muslim rights/way of life laws, ideology, equality

Internet / media commentary: media Internet / media commentary: impacts, influence, filter commentary (skewed), public support (skewed), bubbles, echo chambers, growth, viral spread internet commentary (filter bubbles/echo Information Flows Data analytics and feedback mechanisms: social media, chambers/trolls) Contributing to media consumption, engagement, shares, atmospherics Data analytics and feedback mechanisms: polls Feedback Loops Recruitment/Retention: interest, discussion, numbers, (wrong), engagement, atmospherics strengthened interconnections Voter turnout / levels of support: membership Funding / resource provision: flows, support levels Campaign funding

Virtuous reinforcing cycle: More terrain/populace Vicious reinforcing cycle: Escalation, Accidental Reinforcing Feedback control = more influence/exposure, competitive, Adversaries, Competing Goals, Drift to Low Loops (growth) resilient system, Success to the Successful, sense of Performance, loss of control, counter-influence. purpose & community.

Regain/Rebuild stocks: Successful, proactive Reduction of stock growth: Too violent/horrific, counter-propaganda activities, regain initiative & Balancing Feedback unpalatable violence against Muslims, Coalition counter control of narrative, counter the conspiracies, lies, Loops IO efforts, loss of appeal/credibility/control & authority, trolls, xenophobia, sexism, sense of purpose & drift to low performance community, build credibility

Jihadi ideologues, disenfranchised Muslim population, Democratic ideological supporters, left leaning Sources online fringe elements, unethical/criminal transnational American electorate, Western / Global population, franchises defenders of truth, feminists

Loss of followers, messages ignored/replaced, Loss of voters, supporters and followers, messages Sinks Mistrust/loss credibility, loss power/control, new ignored/replaced, mistrust/loss credibility, biased option/splinter group emerges media

Table 8: Significant systems characteristics of the two case studies

Both cases meet the criteria of being similar systems with sufficient shared background characteristics but also, in having contrasting outcomes due to different system causal impacts, behavioural issues, inherent mental models, information flows, relationships, and structures 65 | P a g e 5249168 Cassandra Brooker Masters of Research (Rihoux & Ragin 2009:24). There has also been a large quantity of qualitative narrative data published about both cases to enable a comprehensive systems analysis of their influence effectiveness.

Lessons can be derived from both system case experiences on why they were successful, or not, which effectively answers the research questions and will provide practical advice to policy makers and iWar strategists in the future. When comparing the two cases from a systems perspective, the overall similarities in their characteristics, combined with the contrasting behaviour of their respective feedback loops, demonstrates why these systems are appropriate for answering the research question regarding the effectiveness of influence activities. All systems have a limit to their growth regardless of whether the growth loop is a virtuous or vicious reinforcing cycle (Meadows 2008:103). Even a successful anti-fragile system, such as ISIS, is susceptible to losing its winning advantage through balancing feedback and associated sinks of its influence stock (refer Table 8). That said, despite losing its physical caliphate, ISIS’ ideological caliphate and affiliated iWar influencing capabilities have not been destroyed. As an anti-fragile, perhaps even anticipatory system (refer Fig 3), ISIS does not need to achieve perfect outcomes, rather it considers any performance as winning over the enemy and maintaining its influential power. Conversely, the Clinton Campaign was a more fragile system than ISIS in gaining and maintaining its influence stocks, and found itself in a vicious reinforcing cycle where behaviours and information flows further undermined its stock, sources, and interconnections.

Flyvbjerg (2006:224) agrees with this premise, explaining that for researchers, the closeness of the case study to real-life situations and the associated details were important for the development of a nuanced view of reality, including the view that human behaviour could not be understood as rational theory expounded; and that cases were important for researchers’ own learning processes. The lack of predictability of human behaviour, which effectively undermines rational theory (Thaler 2015:161) is also why this methodology incorporates behavioural economics theories for better analysing the two case systems, as discussed in Chapter 2. This is also the reason why much of the data used in this thesis is anecdotal, contemporary, informal and non- academic, that is, to ensure the narrative analysis is based on concrete experiences, a close proximity to the studied cases’ reality, and via feedback from those under study. Accordingly, references used include: Hillary Clinton’s autobiography (Clinton 2018), and specific articles relating to Islamic State’s information operations (Ingram 2015) and the group’s own propaganda.

66 | P a g e 5249168 Cassandra Brooker Masters of Research Part Two: Research Methodology The purpose of Part Two of this chapter is to outline the research design, as well as the data collection and analytical methodology used in this thesis. The questions of internal and external validity will also be addressed, before the chapter concludes with a demonstration of how the methodology will be applied in accordance with the flow diagram in Figure 6.

Figure 6: Flow diagram of the Methodological Process

The application of systems analysis in examining the two contrasting case studies is a unique but appropriate methodology for understanding the complexity associated with effectiveness of influence in the iWar and for addressing the research questions outlined in the introduction to this chapter. Braumoeller (2003:209) explains that theories positing complex causation, or multiple causal paths, pervade the study of politics, however, no methodology has been specifically tailored to incorporate the logic of complexity. This is a key reason justifying why a systems thinking approach as a methodology used in this study will enable a deeper understanding of the two cases, the qualitative data relating to each, and their underlying complex characteristics contributing to their relative success.

67 | P a g e 5249168 Cassandra Brooker Masters of Research The methodology employed in this thesis is unique in its conduct of narrative analysis, using the underlying logic of Qualitative Comparative Analysis (QCA) and qualitative sources of evidence, to create systems models to compare and understand complex cases as configurations. This innovative approach is pragmatic in that it overcomes the impossibility of attempting to conduct a structured analysis, employ quantitative measurements or use traditional data collection methods, which are beyond the scope of this thesis. As Polkinghorne (1995:6-7) explained, a narrative analysis studies qualitative data consisting of actions, events, and happenings, and synthesises that data to produce a story, which provides a plotted narrative configuration. This narrative provides meaning and a context for understanding specified outcomes.

Systems Analysis All models are wrong, but some are useful. George Box, statistician

As discussed in Chapter 2, use of systems analysis modelling is an extremely effective methodology for understanding complex socio-cultural systems and their respective structures, configurations, variables, mental models, and behaviour. A system model can identify the emergence of patterns, non-linear relationships, cause and effects, and leverage points for change within a system, based on qualitative data – the findings of which may otherwise be unrealised using traditional linear analytical approaches (refer Table 2). Morris (2001:21) explains that a system dynamics model is an appropriate vehicle for understanding casual mechanisms, and thus the hidden part of the ‘iceberg’ (Fig 1). The systems modelling methodology utilised in Part Two of this thesis is genuinely transdisciplinary and distinguishes between real, actual and empirical events, processes and behaviours, as well as the underlying mechanisms of structure and power (Byrne & Uprichard 2012:2; Gerrits & Verweij 2013:171-172). The added complexity of the research questions in having to measure the intangible effectiveness of influence activities of the two cases, combined with the associated behavioural theories impacting on their system stock flows and feedback loops, meant this adapted narrative approach to systems analysis is unique, specific, and innovative.

Systems thinking offers a more efficient, holistic approach to understanding relationships among organisational components and in comprehensively solving systemic problems, or in this case, answering the research questions relating to influence effectiveness in the iWar (Givens 2012:63- 66). As Byrne and Uprichard (2012:3) contend, most complex social systems do not have states which can be easily understood as linear interactions among simple agents. Rather, these states result from complex relationships between elements both within the system, with external

68 | P a g e 5249168 Cassandra Brooker Masters of Research systems, and from complex assemblages within sub-systems. Therefore, models representing this complexity, attempting to understand trajectories, and establishing causality needed to be reliable and valid at multiple micro and macro levels of observation, as well as considering the relevant context, environmental influences, and history (Byrne & Uprichard 2012:9; Byrne 2005:105; Gerrits & Verweij 2013:169). This is why systems thinking is used to understand complex problems requiring practical solutions, where linear reductionist approaches prove unsuitable (Table 2), and why it is the most appropriate methodological approach for this research.

However, confidence in the model is important (Morris 2001:19) and, whilst difficult to achieve in practice, in order to foster confidence in the two case system models analysed in this thesis, the method used will adhere to Meadows’ (1980:37) three conditions.

 Meadows’ first condition is that “every element and relationship in the model has identifiable real-world meaning and is consistent with whatever measurements or observables are available” (1980:37). This will be achieved by applying qualitative, narrative research data to the values associated with the structures, mental models, and behaviours of the respective case systems. For example: ISIS’ relationship linkages with the local civilian population, and associated feedback loop feeding their influence stocks, was detrimentally impacted upon by extremely violent atrocities they committed against Muslim non-combatants, and thus effectively became a ‘balancing’ feedback loop in the model (refer Table 8). This real-world phenomenon was widely observed and reported at the time by multiple sources. Measuring the level of detriment this had on ISIS’ influence stocks is more problematic and potentially subject to some bias should Qualitative Comparative Analysis (QCA) fuzzy set values be applied to the feedback loop.  Secondly, Meadows (1980:37) advises that “when the model is used to simulate historical periods, every variable exhibits the qualitative, and roughly quantitative, behaviour that was observed in the real system”. As both case systems are bounded by real-world activities and actual processes conducted during a certain period in recent history, and because the qualitative historical data represents reality rather than any forecasts or speculation, this is easily achieved. Neither case model analysed in this thesis will incorporate predictive, theoretical or alternative behaviours that did not actually occur. Again, the issue of bias in measuring the extent of the impacts that certain observed variables had on the two case system behaviours and influence stocks is the real challenge to be considered when applying QCA fuzzy logic.  Meadows’ third condition (1980:37) is that: “when the model is simulated under extreme conditions, the model system’s operation is reasonable (i.e. physical quantities do not become negative or exceed feasible bounds, impossible behaviour modes do not appear). This condition is more difficult to apply to this research as it is replicating actual historical conditions rather than 69 | P a g e 5249168 Cassandra Brooker Masters of Research simulating extreme scenarios that did not occur. However, consideration of this condition is important to ensure the feedback loops and stock flows of the two case models are realistic. Whilst Meadows (2008) explains that “a stock governed by linked reinforcing and balancing loops will grow exponentially if the reinforcing loop dominates the balancing one” (Ibid:45) and conversely, “will die off if the balancing loop dominates the reinforcing one” (Loc. cit.); for both ISIS and Clinton’s campaign, neither systems’ feedback loops were ultimately unbalanced to the point of respectively having infinite growth or dying off – although Clinton came close.

The actual process of conducting the system analysis of the two cases will be in accordance with the methods defined by Meadows (2008) and Stroh (2015), and discussed by Wievioka (1992:171). Specifically, the process first decomposes and identifies all the structural elements, interconnections, sources, sinks, and internal and external impacts on flows, feedback loops and stocks. Before adding value to the systems’ variables and configurations through the application of qualitative narrative data, using the methods discussed in the next section; and then recomposing the models to better understand the causal factors and conduct the comparative analysis. The models will become an extended, more complex version of the ‘one stock Army PSYOPs system’ example contained at Figure 2. In building these system models, the characteristics, behaviours, influences, similarities and differences between the two cases will become apparent and, when considered in the broader social structural context, will contribute to the findings for answering the research questions (Vaughan 1992:178-179).

Methods of data collection and analysis Hall and Citrenbaum (2010) promoted systems thinking approaches for enhancing intelligence analysis processes. They also recommended harnessing the benefits of critical thinking and other complementary analytical techniques, such as link, pattern, cultural, and semiotics analysis, for a more holistic understanding of complex social system problems. This combination of analytical techniques will also be applied to this research to better analyse the narrative data relating to the case systems’ configurations for modelling and their resultant influence effectiveness (refer to Table 9 for definitions).

Therefore, while systems thinking will be utilised as the research design framework to model and compare the two case studies, other analytical techniques, as highlighted by Hall and Citrenbaum (2010), will complement the narrative analysis strategy, to inform the case systems’ configuration, support the decomposition process, identify structures, relationships and archetypes, and to clarify the level of impact certain feedback had on system success. This unique

70 | P a g e 5249168 Cassandra Brooker Masters of Research fused analytical approach to the research problem also applies to the method of data collection and range of sources.

Term Definition

Link Analysis Gaining understanding and insights into behavioural or functional relationships, means of communicating and being connected, and how connections and relationships work between and among people, organisations, internal network nodes, and among networks Pattern Analysis Discerning a consistent series of related actions or events. Patterns involve the composite of traits or features that are characteristic of an individual, a group, or a recognisably consistent series of related acts. Patterns are built upon relationships that drive human behaviour. Cultural Analysis Knowing a particular culture, its people, and their patterns of behaviour deriving from traditional, culturally induced attitudes, behaviours, social norms, and conditions. Semiotics Analysis Discerning meaning, knowledge, or understanding from cultural signs and symbols as reflected in drawings, paintings, photographs, syntax, words, sounds, and body language; plus the mediums upon which they ride and by which messages are delivered.

Table 9: Definitions of analytical techniques (Hall & Citrenbaum 2010:121,139,235,255)

The most effective way to develop accurate and complex systems models of the two cases was to collect, review and synthesise an extensive range of qualitative data from multiple sources, which contained the paradigms of commentators with different agendas, biases, loyalties, opinions, cultures and motivations. Polkinghorne (1995:15) and Sahlstein, Parcell and Baker (2017:1069) explain that narrative data relating to a particular system under study is sought from various sources, with the types depending on the focus of the research. The data must reveal the uniqueness of the individual case or bounded system, and provide an understanding of its idiosyncrasy and particular complexity. The process of narrative analysis then synthesises the data, rather than separating it into constituent parts (Polkinghorne 1995:15-16). This holistic approach to data collection naturally enhances the complete mapping of the case systems and their associated narratives, whilst also assisting in eliminating cognitive biases of the researcher. Collecting and analysing vast amounts of research data, from both traditional and non-traditional sources, also assists in identifying anomalies that may skew the findings, such as bad, deceitful or false data, which is unfortunately prevalent across contemporary media sources in this age of the iWar.

Alongside the risk of obtaining and analysing bad data is the challenge of conducting a systems analysis of a closed, opaque system such as ISIS. Whilst this research might have benefitted from analogous reasoning using a structurally similar, more accessible and transparent system; the fact is, as detailed in Chapter 2; friendly, accessible systems are not as effective as adversary systems in conducting influence in the iWar and are hamstrung by democratic values. Therefore, this thesis addresses the unique challenge of analysing ISIS’ iWar success by using anecdotal qualitative 71 | P a g e 5249168 Cassandra Brooker Masters of Research narrative data and through the application of the persuasion and psychological theories, highlighted in the Literature Review. Narrative analysis of this data will assist in building a system model demonstrating the group’s effective influencing tactics and exploitation of behaviours; thus, exposing the findings related to their success.

The configurative process of conducting a narrative analysis of the disparate qualitative data sources, relating to the system cases, employed a thematic thread of a single ‘influence’ stock, in order to map out activities as parts of an unfolding movement of feedback loops, which culminated in an outcome of success, or not, in influencing effectiveness (Polkinghorne 1995:5). This analytical process of ‘narrative cognition’ complemented and enhanced the system mapping process by providing context and revealing the hidden part of the iceberg (Fig 1) of underlying behaviours, structures and mental models. Polkinghorne (1995:11) explains that human actions are unique and not fully replicable, so narrative reasoning operates by noticing the differences and diversity of human behaviour. It also attends to the temporal context and complex interaction of the elements that make each situation remarkable. Once the activities, events, and happenings are configured within the respective system models, they provide narrative meaning for understanding the whole case and its trajectory, rather than variables in isolation. The result of the narrative analysis is a retrospective explanation, achieved by linking past events to account for the final outcome (Polkinghorne 1995:16).

The narrative analysis undertaken in this thesis differs from the straightforward narrative inquiry employed by qualitative social science researchers, in that the methodology is first and foremost deeply informed by complex systems thinking and a holistic understanding of the case configurations. Secondly, whilst linked to underlying QCA logic in determining influence effectiveness, this methodology did not focus on categorising variables into a predetermined conceptual network to allow for quantitative measurements (Polkinghorne 1995:10); but rather, analyses the interconnections, relationships, behaviours, and system structures producing the outcomes. Polkinghorne (1995:13) contends that much of the literature regarding the conduct of qualitative research focuses on data gathering forms and techniques, rather than procedures for analysing the gathered data; adding that little had been written about the theory of analysis. Therefore, the third pillar of the methodological approach employed in this thesis is the undertaking of critical thinking and other intelligence analysis techniques (Table 9), as recommended by Hall and Citrenbaum (2010), to complement the above-mentioned methods, resulting in a unique, multi-faceted analytical approach.

72 | P a g e 5249168 Cassandra Brooker Masters of Research A qualitative methodology has its limitations and ensuring qualitative research is unbiased can be difficult. The quality, reliability and objectivity of the collected data are important for ensuring balanced and accurate insights are derived from the research. Therefore, this thesis focuses on a wide range of qualitative data collection beyond academic sources, to include: mainstream media or think-tank articles (e.g. Bienvenue et al. 2018; Uren 2017, 2018; Paul & Matthews 2016; The Facebook Dilemma 2018; Wallis 2019); grey literature (e.g. Bosio 2018; Canadian SIS 2018; Mueller 2018; NATO 2016); analytical pieces (e.g. Cave 2018; Mehmetcik & Kursun 2017; Silver 2017; Singer & Brooking 2018; Haywood 2019; Brown 2019; Donovan 2019; Clark 2015; Afzal & Wallis 2016); anecdotal narratives and biographies (e.g. Clapper 2018; Clinton 2018; Watts 2018). This vast and varied collection allowed for an understanding of complex causal relationships to be developed within the two subject systems and also provided a transparent way of scrutinising a wide range of views to better identify any biased, untrue, or skewed reporting – and adjust the system map accordingly. As mentioned in Part One, the use of anecdotal, contemporary, informal, and non-academic data sources enhanced the analytical process and ensured the context and inputs accurately represented the studied cases’ reality.

The method of fusing and synthesising the qualitative data, applying it to each of the case systems’ models to illuminate the interactions between the components, and then supporting the subsequent comparative analysis, is an important step in the process of understanding complex systems. Whilst the collection of a diverse range of extensive data sources provides a richer, more in-depth understanding of the cases, and creates a system story that is effectively a ‘virtual reality’ (Flyvbjerg 2006:241); it also causes case study analyses to contain a density of minutiae. Whilst this density was useful for testing the phenomena relating to the research questions, is also time consuming and laborious to analyse (Flyvbjerg 2006:240). This is a key reason why only two cases were selected for this research. Also, because the selected cases shared a similar purpose and a number of characteristics, this enabled the contrasting differences between the two systems to be more easily identified in the findings. Whilst, this level of data analysis is taxing on the researcher, the benefits of this approach are explained by Flyvbjerg (2006:240) in that “the dense case study is more useful for the practitioner and more interesting for social theory than either factual findings or high-level generalisations of theory. This approach tells the story in its diversity, allowing the story to unfold from the many-sided, complex, and sometimes conflicting stories” (Ibid:240).

One example of how the synthesised data contributes to building a case system model would be the inclusion of ‘Accidental Adversaries’ in the Clinton campaign model, where independently chosen, quick solutions by various systems or sub-systems inadvertently obstructed each other’s 73 | P a g e 5249168 Cassandra Brooker Masters of Research performance in a vicious cycle (Stroh 2015:61-62). This phenomenon was evident during the 2016 election; with the Clinton campaign itself, as well as the media, the Obama administration, and Bernie Sanders’ supporters all having competing goals, messaging, and employing ‘quick fixes’; which effectively undermined all these Democratic systems and diluted their respective influence. The collected anecdotal data supports this analysis, justifies its inclusion into the model, and assists in determining the measurement of detrimental impact that ‘accidental adversaries’ had on Clinton’s campaign (Clinton 2018; Clapper 2018; Silver 2018).

Internal vs External Validity Developing a narrative analysis of the two contrasting case systems, to answer the research questions and determine the findings regarding the effectiveness of influence activities in the iWar, meets the criteria required of internal validity, in that the reported observations correspond with reality. Internal validity is achieved in this thesis through a research structure and methodology that address the research questions, establishes cause-effect configurations and, as discussed, fosters confidence in the models by adhering to Meadows’ (1980:37) three conditions. Morris (2001:19) explains that internal validity raises the question of whether the factors identified are indeed the causes of the described effects. Through the use of a systems analysis approach, combined with the use of real-world case studies and extensive supporting narrative data, it is believed that this research has inferred the correct causation of each case systems’ relative success, or not.

External validity of this qualitative systems research is more difficult to achieve, as this type of case system research is typically unsuited to being generalised and applied to different contexts or circumstances. Additionally, there is nothing externally valid about systems. However, this research can provide lessons and second order principles for Western democratic nations to adopt in similar settings in the iWar context. The aim of this research was to provide deeper insight into how to successfully achieve influence effectiveness in information warfare, through systems analysis and a review of the behavioural and influencing literature. This methodology has enabled the researcher to meet that aim and provide practical solutions, alternative analytical processes and critical lessons to iWar strategists and policy makers, while also contributing to the body of academic literature in this field. As Byrne and Uprichard (2012:18) explain, an emphasis on difference and systematic comparison approaches has considerable policy potential, not least because it enables engagement with changes of kind rather than changes of degree, which is what matters in most policy contexts. However, as discussed, the level to which the findings of the case study analysis are generalizable is difficult to determine, and further work is required to confirm any conclusions relating to why a system is successful, or not. 74 | P a g e 5249168 Cassandra Brooker Masters of Research

Practical Demonstration of Methodology In accordance with the methodological flow diagram contained in Figure 6, this section will demonstrate how those steps translate into conducting the narrative analysis and creating the systems model, using an expanded adaptation of Stroh’s (2015:217) system model demonstrating ‘The Vicious Cycles of Climate Change’ (Figure 7).

As described, a large range of diverse qualitative source data would be collected and reviewed to enable the emplotment of a meaningful narrative based on the theme being researched. In this example, sources would include a wide range of scientific, political, economic, and environmental data relating to climate change. Because this demonstration expands upon Stroh’s model focused just on vicious reinforcing cycles; non-traditional, emerging, and contemporary sources are important inclusions for determining appropriate and relevant balancing feedback loops for example: CO2 sequestration solutions recommended in the recent ‘2040’ documentary (Gameau 2019).

The collected data is then fused and synthesized into different strands supporting the overall thematic narrative and analysed to determine relevance to the system model’s components, configuration and flows. In this example, the narrative analysis is focused on identifying meaningful elements, patterns and linkages proven to contribute to either increasing or reducing the climate system’s stock of greenhouse gases in the atmosphere, and discarding irrelevant or non-credible sources.

The next step is to consolidate the analytical findings and create a systems model to holistically represent the thematic narrative and outcomes being researched. Figure 7 provides an example of the relevant narrative analysis findings being mapped into a systems model. This model could be developed further by introducing more complexity, i.e. contributing internal or external sub-systems, or quantifying feedback loop contributions - dependant on research boundaries.

The final step is to conduct a comparative analysis with the second case system and determine key findings relating to how the respective systems’ configurations, flows, behaviours, mental models, and structures resulted in the observed outcomes and impacted on stock levels. These findings contribute to the thesis’ analytical discussion and potentially provide solutions to the research questions. In this example, a comparative analysis could be conducted against the same system under

75 | P a g e 5249168 Cassandra Brooker Masters of Research temporally different boundaries, for example, greenhouse gases in the atmosphere prior to the industrial revolution; or against a micro-level system that has reversed vicious reinforcing cycles, such as: the repair of the Ozone layer through banning Chlorofluorocarbons (CFCs).

Figure 7: System Model demonstrating the methodology

Conclusion This chapter has described the elements of research design and outlined the methodological approach that will be used in this thesis. It also justified the selection of the two contrasting case study systems to be modelled and analysed, and the validity of this research. An ontology comprising critical realism and a pragmatic epistemological approach, enables the effective research strategy of employing systems’ thinking approaches, supported by narrative analysis, to model the two case studies and conduct a comparative analysis to answer the research questions.

This research aims to determine the associated positive and negative impacts on influence stocks within complex sociocultural systems, and apply those insights to answering the gap of which approaches are most effective for Western democratic nations to use in the iWar. The comparative case study research methodology and case study selections were justified in Part One of the 76 | P a g e 5249168 Cassandra Brooker Masters of Research chapter as the most suitable and apt for answering the research questions and highlighting the outcomes, through the stark contrasts that will become apparent between a successful adversary system and an unsuccessful Western democratic system attempting to win influence.

Part Two of this chapter reinforced the key arguments contained in the Literature Review, by justifying a holistic, narrative, systems analysis approach to modelling the two contrasting cases as a highly effective and rather unique methodology, and also described how confidence in the models was to be fostered. Part Two outlined key aspects of narrative data collection and analytical techniques that will be applied as part of the system modelling process. The chapter concluded with a discussion on internal and external validity and a demonstration of how the methodology will be applied to mapping and analysing the case systems in Part Two of this thesis.

Despite the potential for a density of minutiae associated with the case research becoming laborious and time consuming to analyse, the application of systems thinking approaches, via narrative inquiry, in examining the two contrasting case studies is a unique but suitable methodology for understanding the complexity associated with effectiveness of influence in the iWar, and for addressing the research questions. The actual comparative analysis of the two case systems’ influencing effectiveness is conducted and the associated findings are summarised across the next three chapters, which forms Part Two of this thesis.

77 | P a g e 5249168 Cassandra Brooker Masters of Research Thesis Part Two – Comparative Analysis of System Effectiveness

The behaviour of a system cannot be known just by knowing the elements of which the system is made.

Donella H. Meadows Thinking in Systems: A Primer (2008)

Introduction Part Two of this thesis, comprising the next three chapters, will demonstrate the process of conducting the narrative analysis of the two cases’ respective qualitative data to consequently map each system’s model surrounding their influence stocks and present the associated findings. It will do so in accordance with the methodology and approaches defined by Meadows (2008) and Stroh (2015), and discussed by Wievioka (1992:171). Noting Meadows’ epigraph, the process firstly takes a holistic look at the type of system being analysed, its centre of gravity, and associated resilience. It then decomposes and identifies all the structural elements, interconnections, sources, sinks, and the internal and external impacts on the flows, feedback loops, and influence stocks of each case (see Table 1 for systems definitions).

Analytical value is added to each system’s variables and configurations through the application of qualitative narrative data, using the methods discussed in Chapter 3. This enables the recomposition of the emplotted events and elements into models, providing detailed understanding of the case systems’ behaviours, structures and mental models resulting from its feedback loops. Figure 8 below provides an example of an ideal system flow, with one balancing loop and one reinforcing loop maintaining a steady stock level. However, as Meadows (2008:34) explains, in real systems, feedback loops rarely come singly, they are interlinked, often in fantastically complex patterns. A single stock is likely to have several reinforcing and balancing loops of differing strengths pulling it in several directions. Therefore, the models constructed in Chapters 4 and 5, with their multiple intertwined feedback loops impacting on influence stock levels, will become an extended, more complex version of the ‘Army PSYOPs system’ example contained at Figure 2 and the ‘Greenhouse Gases’ model at Figure 7.

Meadows (2008:45) explains that a stock governed by linked reinforcing and balancing loops would grow exponentially if the reinforcing loop dominated the balancing one. Conversely, it will die off if the balancing loop dominated the reinforcing one. This feedback loop theory provides the foremost justification for the selection of the two case systems analysed in this thesis – with ISIS in Chapter 4 demonstrating dominant reinforcing feedback loops contributing to its exponential growth of influence; whilst Hillary Clinton’s election campaign in Chapter 5 demonstrates the opposite effects of a downwards spiral of balancing feedback and associated

78 | P a g e 5249168 Cassandra Brooker Masters of Research archetypes undermining her influence stocks. The rationale for selecting the contrasting case studies was detailed in Chapter 3. In building the two case studies’ respective system models in these chapters; the characteristics, behaviours, influences, similarities and differences between them will become apparent (Vaughan 1992:178-179). When this comparative systems analysis is considered in the broader social structural context, in later chapters, the findings contribute to answering the research questions.

Part Two of this thesis concludes with Chapter 6 providing the findings of a comparative analysis of the two case systems’ models, to better understand the configurational, behavioural and causal factors contributing to their respective levels of influence. These findings provide a comprehensive base for discussion in Chapter 7 into the gaps relating to effective influence and counter-influence strategies, how complex social systems adapt their behaviours based on feedback loops, and why understanding psychological cognitive theories is central to analysing and leveraging social system behaviours and mental models.

Figure 8: Example of balancing & reinforcing loops maintaining stock level. (Source: Systems and Us website, 2014)

79 | P a g e 5249168 Cassandra Brooker Masters of Research Chapter 4 – System Analysis of Influence Effectiveness: ISIS Fighting battles is not about territory, it's about people's attitudes and perceptions - the battlefield is in there.

Lt. Gen. Sir Mike Jackson (2007) British Army

Introduction The selection of ISIS as the case system example used to understand why the group has been so effective in influencing target audiences was not only due to their unparalleled success and demonstrated adaptation, self-organisation and resilience as a highly-functional social system; but also to address the myopia of Western nations in crediting and learning from their adversaries, and equally, the failure to address and adapt their own systems’ deficiencies to anticipate adversary actions in the iWar (Brown 2019; IWD 2018:7; Clark 2015; Babbage 2019; Chua 2018).

Mehmetcik and Kursun (2017:67-68) explain that ISIS, directly or indirectly, affected the redistribution of power in the entire Middle East region and had a far greater influence than any other terror network in history. Singer and Brooking (2018:154) state that the ISIS legacy will live on long after the group has lost all its physical territory, because it was one of the first conflict actors to fuse warfare with the foundations of attention in the social media age. ISIS mastered the key influencing elements of narrative, emotion, authenticity, community and inundation (Singer & Brooking 2018; Afzal & Wallis 2016:67-88; Murray & Blannin 2017). Therefore, this systems analysis of ISIS as an effective IO case study will provide a holistic understanding of the components, structures, feedback loops, and behaviours contributing to an anti-fragile, anticipatory adversary system in the iWar; which was difficult to target and arguably, its ‘virtual tribe’ remains undefeated in the cognitive battlespace (Dhanaraj 2018:3).

ISIS: Antifragile, anticipatory system Resilience is a measure of a system’s ability to survive and persist in a variable environment, and arises from a rich structure of feedback loops working in different ways to restore a system even after a large perturbation (Meadows 2008:76). In an anti-fragile system, elements possess characteristics, such as rapid learning, adaptation, experimentation with different responses, and exploitation of accumulated knowledge to ensure successful strategies or to avoid unsuccessful ones (Albino et al. 2016:13-14). Chapter 2 details how ISIS promoted and exploited its anti- fragility to formulate a strategy of dynamic, anticipatory response to stressors. This is an intrinsic capability of complex systems and the ideal (Albino et al. 2016:12-13; Lieber & Reiley 2016:50). As a self-organising system, ISIS also demonstrated the ability to evolve, using selection and 80 | P a g e 5249168 Cassandra Brooker Masters of Research replication, to reinforce success, eliminate weakness, and enhance robustness (Meadows 2008:79). This evolutionary adaptation also shifted ISIS’ intentions, in response to balancing feedback, as over the years ISIS cared less about borders or territorial continuity and more about ideological and procedural continuity in its territoriality (Mehmetcik & Kursun 2017:60).

Understanding ISIS’ system response characterisation provides insight into its hidden behaviours and structures (see Fig 1). This exposes leverage points and archetypes that may be exploited. For example, Albino et al. (2016) state that an anti-fragile system “must be opposed by means of other dimensions of activity” (Ibid:12). This means dislocating a military anti-fragile system through its political, social, or economic fragility, and which history reveals was not done well against ISIS. Albino et al. (2016:12) explain this effect further by citing contemporary examples of the insurrection of the Taliban and Al-Qaeda, which both gained members and support as a result of the conditions of conflict. In these cases, as with ISIS, the actions of the Coalition (seen as the occupier) fuelled the will, unity, and organisation of these terrorist systems, strengthened their narrative; whilst also providing them learning opportunities and accumulated knowledge about how to successfully respond to a stress and adapt accordingly (Albino et al. 2016:14; Watts 2018:22; Clark 2015).

In order to deconstruct ISIS’ complex system into its key elements, a holistic analysis of its centre of gravity (COG) and supporting critical capabilities (CCs) was an important step for bounding the research methodology to specifically focus on the system’s influence stocks. The military definition of COG is: “a characteristic, capability or locality from which a military force, nation or alliance derives its freedom of action, strength, or will to fight” (Australian Army 2002). Adapting this definition to relate to its influencing activities, the ISIS COG was assessed to be: ‘the ability to effectively influence target audiences to maintain power and control’, with its critical capabilities being: ideology, means, control, media/IO capabilities, and resources (Fig 9).

This COG assessment corresponds with Mehmetcik and Kursun’s (2017:66) explanation of influence comprising two dimensions and the conditions for each:  The first dimension addresses sustainability, i.e. does ISIS have deterrent-resistant motivations, such as an ideology, which are flexible and adaptable enough to cope with rapidly changing environments and events?  The second dimension refers to impact, i.e. can ISIS force a state response that acknowledges their autonomy? Can they force major changes in states’ policies, practices, and understanding? Success is being able to force significant changes in government policies

81 | P a g e 5249168 Cassandra Brooker Masters of Research and actions, using violence and non-violence, which provides them recognition and legitimacy.

As the CCs are broken down further into critical requirements (CRs) and, which if targeted also double as critical vulnerabilities (CVs), then the key elements providing the scaffold to ISIS’ system model become apparent, as well as the consequences of targeting those elements and linkages. Figure 9 provides the COG analysis for ISIS (see Appendix One for definitions).

Figure 9: ISIS’ Influence Centre of Gravity Analysis

System Elements Building upon the holistic understanding provided by the system characterisation, the COG analysis, and the qualitative narrative data analysis; the first step of the system mapping process, or ‘decomposition’, can commence with confidence that the complexity and key elements will be appropriately captured. The following key system elements were identified during the decomposition process and mapped accordingly (Fig 10).

82 | P a g e 5249168 Cassandra Brooker Masters of Research Stock and Purpose: The stock being researched to determine system effectiveness is influence and, as a result, both case system models will be single stock models bounded by configurations and activities related to the stock and purpose. The purpose of ISIS as a complex social system was to establish control and hold power by declaring a Muslim Caliphate, influencing followers, maintaining legitimacy and credibility, and controlling the terrain, the populace and its ideological narrative.

Source: The source of ISIS influence stocks initially comprised marginalised, disempowered Sunnis, previous Al Qaeda supporters and splinter groups, anti-Western Muslim populations around the world, fundamentalist Wahhabi clerics, and Jihadi ideologues (Clapper 2018:174; Ehlers 2017:7; Mehmetcik & Kursun 2017:58, Ward 2018:4,66); all of whom were open to ISIS’ narrative and ideology, pledged deep allegiance, and actively contributed to the viral spread of their messages and media content (Dhanaraj 2018:2). As Hall and Citrenbaum (2010:257) explain, ISIS’ influence stemmed from the shared Sunni culture shrouding them, as they were embedded within a large population with common cultural orientations, characteristics and grievances, and were also able to exploit secure social media forums as echo chambers to personalise their messaging and reinforce their extreme ideology.

As ISIS’ power, control and reach expanded physically, territorially and online, so did their influence, which in turn expanded their source base to include fringe elements, disenfranchised youth in non-Muslim societies, and Islamic terrorists emboldened by the Arab Spring. Combining their skilled media propaganda efforts with successful military actions, an international scare campaign comprising filmed atrocities, and the seizing of territory to declare a physical Caliphate; provided overwhelming media and political attention and further reinforced their source. This growth in turn strengthened the group’s sub-systems, through interlinked reinforcing feedback loops, to rapidly increase influence stocks. As Mehmetcik and Kursun (2017:55,58) explain, ISIS’ ideology and tactics were designed to take advantage of Sunni grievances as a source of funding, manpower, and legitimacy; and enabled the group to follow a territorially-expansive agenda.

Sink: Despite the virtuous cycle of reinforcing feedback exponentially growing ISIS’ influence, balancing feedback loops existed to provide checks on unlimited growth and contribute to the sink. The balancing feedback’s resultant sinks on influence included a scale of disengagement by target audiences starting from disinterest and a loss of trust, rising up to open hostility, disgust, and anger. ISIS’ influence sinks resulted from losses of credibility, legitimacy, and appeal; which arose from: a loss of power and control over target audiences, atrocities committed against Muslims, their extreme application of Wahhabism, and physical losses of resources, deaths of 83 | P a g e 5249168 Cassandra Brooker Masters of Research charismatic leaders, fighters, and media specialists and, significantly, Caliphate territorial losses. Additionally, Mehmetcik and Kursun (2017:58) explain that because ISIS relied heavily on ideological support of disenfranchised Sunnis at the local level to recruit, fund, and organise its quasi-state formation; then over time, a fundamental slowdown in their virtuous cycle occurred because ISIS encountered non-Sunni towns, whose populations staunchly resisted them.

Internal Sub-systems Terrorist groups require secrecy to survive, which places constraints on how the group communicates and conducts operations. Due to having to balance efficiency with a need for security, many terrorist groups adopt a decentralised ‘hub-and-spoke’ or ‘market’ hierarchical structure, where the nodes have individual autonomy to plan and carry-out attacks and makes them difficult to target (Kilberg 2012:811-813). However, as Neriah (2014) explains, ISIS is much more than a terrorist organisation; it is a terrorist state comprising almost all required governing elements and a bureaucratic hierarchy.

From a systems perspective, Meadows (2008:83) explains that hierarchies give a system stability and resilience, reduces the amount of information any part of the system has to monitor, and also that relationships within each sub-system are denser and stronger than relationships between sub- systems. By employing a centrally-commanded and controlled bureaucratic hierarchy over its sub-systems, ISIS allowed its sub-systems to regulate and maintain themselves, while still serving the needs of the larger system. This resulted in a stable, resilient and efficient structure (Meadows 2008:82), which would not have been possible in a decentralised hierarchy.

ISIS evolved from a dispersed, extremist, marginal faction employing an ‘all channel’ type hierarchical network, as used by Al Qaeda. The group exploited the information revolution (Kilberg 2012:814-815); to become the strongest, most capable militia in the region, with a bureaucratic ‘pyramid of power’ providing centralised leadership over shadow government departments, councils, internal sub-systems and provincial ‘wilayats’ (Neriah 2014). The rise of ISIS’ power, control and influence was enabled by the seizure of territory, including the disenfranchised human terrain, infrastructure, services, judiciary and governance mechanisms contained within, which provided increased legitimacy. As Kilberg (2012:815,822) explains, the more popular support for dissent that exists, the more likely terrorist groups would adopt centralised structures to enable the organisation of numerous members – particularly in poor countries, where the capacity for conducting counter-terrorism operations is low, thus providing safe haven for overt terrorist activity and ideological propagation. Kilberg (2012:816-817) adds that terrorist groups planning on initially attacking hard targets, religiously motivated groups with 84 | P a g e 5249168 Cassandra Brooker Masters of Research built-in pools of potential support, and groups with narrow, defined goals were more likely to adopt centralised hierarchical structures in anticipation of achieving their goal. These factors allowed ISIS to operate overtly, in a relative safe haven, and adopt an elaborate, centrally- controlled ‘bureaucratic’ structure, which is the most hierarchical of all the structures (Kilberg 2012:812-813).

Rather than decompose and analyse specific geographical, operational, or governance-specific sub-systems within ISIS’ organisational structure, the sub-systems of interest to this system model of influence have been classified into five broad disciplines of: Political/Leadership, Military, Media, Religious, and Economics to better understand their interlinking relationships and the feedback loops generated by each.

Sub-system: Political Wing / Leadership Having charismatic, authoritative, likeable leaders, some of whom were minor celebrities, exploits the ‘appeal to authority’ bias. The aura projected by the leadership also engendered loyalty, enhanced an image of authenticity, and maintained the sense of community and social contagions that ISIS promoted. Having a centralised command structure and strong interconnected relationships with the four other sub-systems ensured ISIS’ leadership could effectively maintain their image, control the narrative, project power, and exercise the same political authority as other state governments. ISIS’ organisational leadership was not viewed as elitist, like the aloof leaders of Al Qaeda, and it used this perception for engaging with, influencing and appealing to disaffected Muslim audiences both locally and globally. Mehmetcik and Kursun (2017:53) explain that ISIS were able to amass new-found revenue, support, institutions, and legitimacy achieved through civil war to form a Caliphate. ISIS created competing forms of authority, governance, and community services; which were further enabled by its bureaucratic hierarchical structure. Ruthven (2007:54) adds that modern fundamentalisms, such as ISIS, provide sources of authority and psychological reassurance in a global environment where actual power is diffused and impersonal.

Sub-system: Military ISIS’ influence stocks were significantly increased by the battlefield successes, seizure of caliphate territory, and frontline actions conducted by its military fighters as ‘propaganda by deed’, all of which were captured in high quality, full motion video, manipulated, and disseminated globally. As Murray and Blannin (2017) explain, violence itself is a form of communication, which is known in terrorism parlance as ‘propaganda by the deed’. ISIS’ military wing was presented as a highly capable, adaptable, and ruthless military force (Ingram 2015:739) 85 | P a g e 5249168 Cassandra Brooker Masters of Research to leverage funding, increase recruitment, inspire support and evoke reactions. Again, the other four sub-systems’ interlinking relationships with the military sub-system provided considerable strength to the resilience of the whole system and its reinforcing feedback loops. For example: the media sub-system provided IO content as a force-multiplying PSYOPS tool to enhance its scare campaign and thereby frighten Iraqi troops into fleeing their defensive positions, whilst simultaneously strengthening the legitimacy of their rhetoric and enhancing recruitment through the conduct of actual deeds. With the military, media, and leadership sub-systems working in concert to reinforce their influence stock, Ingram (2015:740) explains ISIS was able to portray the strategic and historical context behind its operations, weave those narratives into a theme of a struggle between good and evil, and out-communicate its enemies in the iWar with its own version of the “truth”, backed by deeds, across multiple channels.

Sub-system: Media ISIS’ media sub-system, comprising all its critical capabilities, requirements (Fig 9), architecture, specialists, resources, and tactics; was the dominant sub-system contributing to the whole system’s influence stock and virtuous cycle of reinforcing feedback. Accordingly, the media element had the strongest relationships and interconnections with all other system elements. The other sub-systems were configured to align with the media campaign to protect the centre of gravity, control the narrative, and maintain the virtual Caliphate – anything else was secondary to the system’s purpose. There are three levels to ISIS’s media architecture: propaganda and content designed for transnational audiences, high profile announcements targeting regional audiences, and communiqués focused on localised issues and events (Ingram 2015:734-5). This three-tiered approach contributed to ISIS’ media proficiency, exposure, reach, and legitimacy, and reflected the strategic, operational, and tactical level hierarchical delineation of the whole system. ISIS’ media sub-system excelled at psychologically exploiting the online activities of its virtual tribe to ensure rapid, high-volume dissemination of its appealing, multilingual content and the domination of its targeted messaging in a multi-pronged approach; to create a sense of community, manipulate heuristics, and effectively cement its branding as a social movement (Dhanaraj 2018:1-2; Afzal & Wallis 2016:67-88; Watts 2018:45-46; Clark 2015; Murray & Blannin 2017).

Sub-system: Religious ISIS’ fundamentalist Wahhabi ideology (Ward 2018) was a critical capability underpinning the centre of gravity, providing legitimacy to the narrative, and credibility to its media and leadership sub-systems. Having authentic religious clerics who provided simple, resonant and appealing Wahhabi edicts and justifications for ISIS’ actions, strengthened the reinforcing feedback loops and shaped the whole system’s ‘hidden part of the iceberg’ (Fig 1). The religious sub-system also

86 | P a g e 5249168 Cassandra Brooker Masters of Research benefited from the threat provided by the military sub-system and scare campaign of the media sub-system in ensuring compliancy. Working in concert with the other sub-systems and exploiting psychological and behavioural economics theories (Table 6), such as: tailoring messages, appeal to authority biases, indoctrination practices, and social and emotional contagions, ensured continued inflows of influence. This cognitive exploitation also provided a unified purpose and high resilience, recruitment and retention rates within the system. Ward (2018:10,14) explains that for their guiding principles, ISIS’ leaders were clear about their almost exclusive commitment to the Wahhabi movement; which was inspired, funded, and imported from Saudi Arabia to exploit state failings in northern Iraq and eastern Syria.

Sub-system: Economic Underpinning all the sub-systems, and ISIS’ system as a whole, was the economic sub-system providing the resources and incentives to enable all influencing activities, shadow governance, battlefield success, and associated reinforcing feedback flows. The economic sub-system also relied on the activities of the other sub-systems to reinforce its own virtuous cycle of fundraising and spending to secure further influence, for instance, payments to widows and martyrs’ families, providing services and infrastructure to win hearts and minds of the local populace, or purchasing communications and media platforms and services to dominate the information domain. A major source in establishing ISIS’ economic sub-system was the inflow of Saudi funding and foreign fighters to support the spread of Islamic extremism in the Wahhabi tradition (Ward 2018:4-5). ISIS’ economic wing also benefited from other regional revenue sources; the theft of oil, antiques and agricultural products; and heavy taxation on populations in held territory.

External Systems Local Target Audience Initially local, civilian populations living in seized Caliphate territory, marginalised Sunnis, and former Ba’athists across Syria and Iraq were targeted to expand ISIS’ source support and recruitment. As ISIS’ power and influence grew with its territorial expansion, so too did the size and diversity of the local civilian population it sought to control. However, the further ISIS’ Caliphate expanded, the more its influence inflows slowed, as regional-level balancing feedback was introduced from non-Sunni populations, moderate Islam, and local political, clerical, and military counter-influence (Mehmetcik & Kursun 2017:58).

Global Target Audience Through its virtual Caliphate, ISIS’ global target audience initially comprised those foreign terrorists, ideologues and disenfranchised Muslims around the world, as listed under the ‘Source’ 87 | P a g e 5249168 Cassandra Brooker Masters of Research element. However, due to its extremely effective multi-pronged, multi-platform and multilingual influencing activities, ISIS rapidly gained a massive global audience that experienced viral self- perpetuating growth, predominantly through ‘super spreaders’ on social media, in a reinforcing feedback loop that was never really counter-balanced effectively by other external systems’ influence in the iWar (Watts 2018:30-32,45-46; Singer & Brooking 2018:151; Murray & Blannin 2017). Rather ISIS’ influence over global audiences was mitigated by the direct targeting of personnel, communications systems, resources, and commercial networks.

Global Mainstream Media (MSM) As an external system, global media can both be exploited to provide legitimacy to ISIS’ content and reinforce their influence stocks, or can provide balancing feedback through transparent reporting mechanisms to expose ISIS’ hypocrisy and undermine their credibility. However, as discussed in Chapter 2, traditional mainstream media are hamstrung by tone, fairness, allegiance to facts, and context over conclusions. This results in MSM often being out-published and out- manoeuvred by ISIS’ media sub-system, using tactics of constant inundation in multiple languages, creation of salacious, addictive clickbait, and the rapid production of high quality, credible content.

Coalition Governments / Military Like the global media, Coalition counter-influencing actions can backfire in the iWar and actually reinforce ISIS’ influence stocks, particularly in the event of kinetic targeting causing civilian casualties (known as CivCas) and damaging local infrastructure, or by overtly favouring Shia concerns. ISIS recognised this threat to Coalition legitimacy and conducted activities to bait Western forces, increase CivCas numbers and worsen damage caused by Coalition targeting (Ingram 2015:744; Clark 2015). Also, whilst counter-IO activities could undermine ISIS’ influence and credibility with target audiences, often Coalition efforts were reactive, delayed and hampered by ethics, operational security, democratic values, and rules of engagement (Table 4). Coalition governments and their militaries also played a role in strengthening the systems of regional governments and local militaries fighting ISIS within their borders to create domestic- level balancing feedback.

Regional Governments – Iraq / Syria Like the conundrum faced by the Coalition, state governments and their military forces could also strengthen ISIS’ influence; both passively by creating a vacuum of governance, investment, and services for minority Sunni communities, or actively through authoritarianist actions and committing atrocities against civilian populations, which undermined their own legitimacy and 88 | P a g e 5249168 Cassandra Brooker Masters of Research narrative. However, if states were able to win the trust and support of the population through ethical deeds, such as the provision of aid by the military, avoiding civilian casualties, nation- building activities, and providing transparent, honest, community engagement; this dislocated ISIS’ influence and created strong balancing feedback loops, as evidenced in Marawi (Knight & Theodorakis 2019). As Mehmetcik and Kursun (2017:66) explain, the most important factor contributing to ISIS’ resistance was its ideological roots and the social/political context that created them. As long as those issues remained unaddressed, ISIS would not be politically or ideologically defeated. Mehmetcik and Kursun (2017:66) add that while regional and international actors were mainly focused on the military battle, there was an urgent need to address the aftermath of ISIS, as well as the structural failures and societal grievances that allowed ISIS’ rise in the first place.

Competing systems Other external systems relating to ISIS’ system influence model, which initially supported ISIS but later developed ‘competing goals’ or become ‘accidental adversaries’ (resulting in balancing feedback), included: al-Qaeda, non-Wahhabi Muslim clerics, splinter groups, and elements of the online community. However, in terms of competitive exclusion, ISIS largely won against al- Qaeda and other competitors in the ‘Success-to-the-Successful’ archetype, which saw the losers gradually forced out (Meadows 2008:127). Arguably, Saudi Arabia also became an accidental adversary when its duplicity in supporting ISIS while co-operating with the Coalition was exposed (Ward 2018:67-69). However, a more in-depth analysis of Saudi Arabia’s opaque, secretive, nation-state system is beyond the scope of this model.

Linkages / Relationships In Figure 10, the interlinked relationships are depicted by arrows, which are coloured red if they contribute to a reinforcing feedback loop or blue for balancing feedback. The specific feedback loops will be covered in more detail in the following sections, however, in decomposing the system into key elements, it is important to detail some of the fundamental relationships depicted in the model. These relationships can be broadly split into two categories: ideological links and communications links.

Ideology and Narrative In building up its complex adaptive system of influence, ISIS recognised that the relationships and linkages between its disparate elements, sub-systems and supporters must be strong, resilient, and unified in purpose under an authentic, resonant ideology. In this they were assisted by Sunni Islam generally, which is described as a triumphalist faith, ‘programmed for victory’ (Ruthven 89 | P a g e 5249168 Cassandra Brooker Masters of Research 2007:26-27), and Wahhabism specifically, which is extremist, fundamentalist and exclusionist (Ward 2018:17). Clark (2015) explains that ISIS’ ubiquitous narrative of: “Islam is under attack” was a driving force in uniting and motivating Muslim followers, which was reinforced anytime the Coalition carried out kinetic targeting. This narrative also created a sense of purpose around the cause and a sense of community around the shared struggle. ISIS’ ideology and associated narrative were critical capabilities (Fig 9) for system success, which were synchronised with deeds for maximum effectiveness, and designed to provide supporters with a competitive system of meaning and to strengthen the ‘lens’ fundamentally shaping audience perceptions (Ingram 2015:730; Afzal & Wallis 2016:75).

ISIS identified the need to not only fill a politico-military void but also a vacuum of values and meaning, which they addressed by influencing audiences and recruits towards its vision (Ingram 2015:735), to varying degrees of success. For example, as Mehmetcik and Kursun (2017:65) explain, foreign fighters joining ISIS were generally motivated by its propaganda of building Islamic states, while local forces joined because it was the least bad option of rational choice. Additionally, Gharajedaghi (2011:66) explains that ideologies have proven to be major obstructions to the viability of a social system, where underlying assumptions were not to be questioned by true believers.

Ingram (2015:730) states that ISIS’ narrative and ideological messaging was particularly potent, because it fused pragmatic and perceptual factors in its communiqués to resonate with diverse global and local audiences. ISIS’ messaging was simple, resonated, always aligned with its strategic narrative, and exploited the rapid, evocative power of imagery suited to the modern communications environment (Singer & Brooking 2018:158-179). They also employed rational- choice and identity-choice appeals, which further strengthened ‘in-group’ bonds. ISIS’ media sub- system disseminated messages designed to build an affinity within its system and sub-systems through appeals to a shared identity, an image of the group as champion and protector; whilst framing its enemies as evil ‘Others’ (Ingram 2015:733; Afzal & Wallis 2016:73; Ruthven 2007:102).

Communications Links ISIS’ shrewd exploitation of the information environment via multi-pronged communications links, to disseminate its narrative and imagery depicting ‘propaganda of the deed’; strengthened its system interconnections by unambiguously reinforcing perceptions and polarising support of friends and foes alike, while capturing global media attention (Ingram 2015:746; Taylor 2007:203; Singer & Brooking 2018:154; Afzal & Wallis 2016:75; Watts 2018:18). ISIS’ 90 | P a g e 5249168 Cassandra Brooker Masters of Research influence system comprises an extensive network of physical, virtual, human, machine, and multi- media communications links – producing content, disseminating messages, providing feedback, and reinforcing growth through influence. While many communications links within the ISIS system model may be categorised as ‘hard system’ linkages such as hardware, software, internet connections, broadcasting networks, applications and services; it is the ‘soft system’ definition of communications links, for example, relationships and human interconnections, that are of most importance to ISIS’ influence.

Communications links, combined with the successful propaganda techniques and associated psychological theories discussed in Chapter 2, were exploited by ISIS to create an expansive interconnected virtual tribe. This community communicated through ideological echo chambers, content-manipulated filter bubbles, and encrypted, personalised chat rooms, to create deeply- bonded, loyal relationships and contributed to the rapid virtuous reinforcing cycle of growth in influence and power. Dhanaraj (2018:4) explains the power of these communications links, where individuals who are receptive to certain content seek out one another and undergo processes of ‘social bonding’, they exchange grievances, and expose one another to similar ideological material. Dhanaraj (2018:4,6) also describes the ‘virtual caliphate’, where exchanges via communication links create a semblance of geographical space, transcending traditional boundaries, and enabling social bonding and a unity of identity; which strengthens the virtual tribe, motivates members, and provides access to leaders. Mehmetcik and Kursun (2017:54) agree that as result of its vast and diverse transnational terror network and communication links, ISIS became the strongest, best-resourced, and most ideologically potent “terrorist quasi-state” of recent times.

The other ‘soft system’ communications links central to ISIS’ influence is the engagement and bonds formed and reinforced from its multi-axis, multi-platform, multi-level, multi-lingual media strategy. Information operations were a central pillar of ISIS’ campaign, which were choreographed to sow terror, disunion, and defection amongst its enemies, create shock value, and gain the attention of global audiences (Ingram 2015:734; Singer & Brooking 2018:4-5,152); whilst conversely strengthening its own internal relationships, support, loyalty, and branding. ISIS understood the power of cyberspace for projecting power, weaponising viral content, and distributing messages by a global network of ‘super spreaders’ beyond any one state’s control (Singer & Brooking 2018:151).

Using hard and soft system communications links, ISIS was the first terrorist group to hold both physical and digital territory. They effectively controlled the narrative and perceptions of reality, 91 | P a g e 5249168 Cassandra Brooker Masters of Research that is, as long as most observers believed that ISIS was winning, it was winning – and their influence stocks continually increased as a result (Singer & Brooking 2018:152-153). Patrikarakos (2017:246) contends that ISIS’ physical and virtual battlefields had merged and therefore, its virtual caliphate would live on. ISIS’ communication and ideological links connect all the decomposed system elements, depicted at Figure 10, and contribute to the reinforcing and balancing feedback loops created by the resultant interconnected flows between internal and external sub-systems and elements.

92 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 10: ISIS Baseline System Model – key elements and concepts

93 | P a g e 5249168 Cassandra Brooker Masters of Research Reinforcing feedback loops As demonstrated through the decomposition process and modelling of the key elements (Fig 10), and despite being bounded by a single stock and specific time period; ISIS’ system is extremely complex with multiple components, interconnections, and behaviours contributing to the seemingly infinite array of configurations that could be modelled. Therefore, in recomposing the system and analysing the evident feedback loops in the next two sections, this research will step back to broadly understand the cycles holistically, therefore avoiding the time consuming and laborious analysis of this case study’s density of minutiae, as cautioned in Chapter 3 (Flyvbjerg 2006:240). Table 10 summarises the four reinforcing feedback loops discussed in this section.

Reinforcing Feedback Key Description Loop

Sense of Influence stocks are reinforced through the increased participation and support of R1 members identifying with in-Group / Us vs Them narratives, social contagions, Community unified tribal dynamics, and heuristic biases manipulated by ISIS.

Sense of Influence over target audiences grows, motivation of members is reinforced, and R2 ideological narratives spread when there is belief in a justified cause, the purpose Purpose is perceived as authentic/legitimate, and pride is derived from participation.

Success-to- Influence grows with power and authority resulting from winning actions, which in R3 turn provides reinforcing feedback and momentum for the adaptive, anticipatory, the- Successful escalation of continued success in a virtuous cycle. Influence stocks are reinforced through the effective exploitation of IO tactics, communications links, and behavioural economics as an “Information Jihad” – Exposure R4 designed to virally inundate and control audience perceptions to build legitimacy, reinforce the narrative, and dominate the IE.

Table 10: ISIS Reinforcing Feedback Loops

Sense of Community (R1) The first reinforcing feedback loop (R1) provides influence inflows generated by a sense of community, achieved through increased participation and support of members and audiences. R1’s ‘aspirational tribalism’ is derived from leveraging identity-choice appeals, and exploiting social contagions, ‘confirmation biases’, ‘homophily’ and ‘similarity’ heuristics (Table 6). As Dhanaraj (2018:1,3) explains, in an increasingly globalised world, resulting homogenised identities in cyberspace now threatened and diluted cultural distinctiveness, which causes feelings of marginalisation or discrimination within minority communities. This phenomenon leads to vulnerable segments of society, particularly disenfranchised Sunnis, seeking a sense of belonging and solidarity; and, when combined with ISIS propaganda tailored to trigger feelings of

94 | P a g e 5249168 Cassandra Brooker Masters of Research disassociation and alienation, increased the group’s attractiveness, influence, and ability to radicalise.

ISIS combined numerous ‘soft power’ tactics focused on charity, unity, and winning hearts and minds (Ward 2018:118; Clark 2015; Patrikarakos 2017:217) with an ‘Us Versus Them’ narrative of the ‘benevolent in-group’ confronting the ‘crisis-generating others’, which resonated particularly strongly in the socio-historical context of intergroup conflict in Iraq and Syria (Dhanaraj 2018:3; Ingram 2015:730,741). ISIS offered a common identity and unifying ideology that represented shared interests, values, and goals among its members. ISIS also dominated local information environments, which were unsophisticated and barely penetrated by native media, let alone by external broadcasters (Lieber & Reiley 2016:48). When combined with resonant frames of familiar language, culture, and an authenticity achieved by relatable online videos, this contributed to the sustainability of this reinforcing feedback loop (Mehmetcik & Kursun 2017:62; Singer & Brooking 2018:159,167,170; Ciovacco et al. 2011:44; Patrikarakos 2017:215).

In addition to messaging around perceptual factors, such as: ‘in-group’ identity-choice appeals and crisis-solution constructs designed to shape audience understanding of the conflict; ISIS simultaneously appealed to pragmatic factors, such as: the security, stability and livelihood of its members, using rational choice appeals, and framed itself as the champion of Sunni Muslims, fighting their shared enemies (Ingram 2015:730).

This two-pronged approach was not limited to local or regional Caliphate audiences but extended globally to its online community. Cyberspace enabled extremists to band together, allowed their views to reinfiltrate mainstream discourse, and empowered them to “be themselves” (Singer & Brooking 2018:170,172). The ‘virtual tribe’ and its shared hatred of perceived enemies propagated by ISIS, provided companionship, comradery, embraced social outcasts, and reinforced confirmation biases through echo chambers and filter bubbles, in turn increasing recruitment, support, influential reach and motivating violence. Figure 11 depicts ISIS’ Sense of Community reinforcing feedback loop (R1). The causes are listed for all internal and external sub-systems and elements contributing to ISIS’ in-group narrative and therefore, the inflows of influence via R1.

95 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 11: ISIS Reinforcing Feedback Loop (R1): Sense of Community

96 | P a g e 5249168 Cassandra Brooker Masters of Research Sense of Purpose / Belief in Cause (R2) The second reinforcing feedback loop (R2) contributes to influence stocks from the ‘sense of purpose’ and ‘belief in a justified cause’ feedback, as garnered by ISIS, which strengthens drive, passion, and motivation, persuades target audiences of its legitimacy, and spreads its ideological narrative through the feeling of pride it invokes. This sense of purpose, nurtured by deeds fused with an effective IO campaign that mastered the key elements of narrative, emotion, authenticity, community, and inundation (R4); would ensure ISIS’ legacy and influence would live on long after the group had lost all physical territory (Singer & Brooking 2018:154; Patrikarakos 2017:246).

ISIS provided a competitive system of meaning and purpose by targeting audiences with emotive imagery, powerful symbolism, a strong narrative, and appeals to both pragmatic and perceptual factors. This legitimised their actions and resonated with a broad spectrum of supporters – especially marginalised populations faced with poor living conditions, oppression, unemployment and instability (Ingram 2015:736; Clapper 2018:157-158). The engendered ‘belief in the cause’ contributing to this reinforcing feedback was primarily enabled by ISIS’ mastery of modern technologies, such as open social media platforms and closed messaging applications. ISIS radicalised individuals and created an impenetrable virtual tribe with a self-organising resilience, derived from a united purpose and unwavering belief system (Dhanaraj 2018:1,3,5; Mehmetcik & Kursun 2017:58; Singer & Brooking 2018:154).

The simultaneous appeals to pragmatic and perceptual factors apply for both the R1 and R2 feedback loops in ISIS’ system of influence. Ingram (2015:736) explains this approach created mutually reinforcing narrative cycles, whereby the veracity of ISIS’ system of meaning was evidenced in the efficacy of its politico-military apparatus (and vice versa), as demonstrated by the strong interlinking relationships between its internal sub-systems (Fig 10). Ingram (2015:736) adds this fusion of factors (aligning rational and identity-choice decision-making) provides understanding into why ISIS messaging rapidly radicalised its supporters. The more that rational- choice decisions were processed from the perspective of ISIS’ system of meaning, the greater the potential for its messages to resonate, influence, and act as a driver of mobilisation (Ingram 2015:736); and thereby also contributed to the R3 Success-to-the-Successful reinforcing feedback loop. Figure 12 depicts ISIS’ Sense of Purpose reinforcing feedback loop (R2) and the causes of contributing inflows for key internal and external sub-systems.

97 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 12: ISIS Reinforcing Feedback Loop (R2): Sense of Purpose

98 | P a g e 5249168 Cassandra Brooker Masters of Research Success-to-the-Successful (R3) In the R3 ‘Success-to-the-Successful’ reinforcing feedback loop, ISIS’ influence grows with the power and authority resulting from winning actions, which in turn provides reinforcing feedback and momentum for the adaptive, anticipatory, escalation of continued success in a virtuous cycle. Essentially, the more terrain and populace ISIS controls under the caliphate, the more influence they have. This cycle enables them to build their brand, which attracts more support, yields demonstrable results, and further strengthens the brand, their resilience, and associated influence. Success-to-the-Successful reinforcing feedback was not built on individual factors but rather in how all the elements interacted to reinforce one another over time (Stroh 2015:48).

The feedback provided by this virtuous growth cycle, enhanced ISIS’ system resilience by becoming adaptive and then, anticipatory. ISIS’ improbable momentum continued: out- communicating adversaries in the iWar, recruiting over 30,000 fighters from nearly a hundred countries, and exporting its message internationally and in the mainstream (Singer & Brooking 2018:9). Singer and Brooking (2018:9) described ISIS as being “like a demonic McDonalds” in rapidly opening new franchises and spreading a global contagion of fear, which in turn acted as a force multiplier strengthening its internal sub-systems. The anticipatory characteristic of ISIS’ system resulting from this R3 loop, also produced a higher level of autonomy, which lead to better representation, and in turn, higher levels of influence. This provided ISIS the means to compete even more effectively at the strategic level and continue to dominate the iWar (Mehmetcik & Kursun 2017:56; Meadows 2018:127).

Further strengthening ISIS’ R3 virtuous cycle were the actions and inactions of external systems, often unwittingly, as depicted in Figure 13. ISIS was able to exploit the Coalition’s quest for legitimacy, and attempts to minimise collateral damage and civilian casualties by manipulating the narrative to its own version of the truth, creating emotional contagions, and baiting its enemies into reacting with misguided IO responses (Albino et al. 2016:4; Ingram 2015:744). ISIS positioned itself as an ethical actor, leveraged cultural fissures and entrenched grievances, exploited popular resistance, and accused security forces of hypocrisy and use of indiscriminate force – which was reinforced with fake content, staged events, and examples of US military and security forces human rights abuses, such as: Abu Ghraib and Guantanamo (Knight & Theodorakis 2019). ISIS also exploited the vacuum of governance, services, investment, and infrastructure in Sunni Communities caused by systemic Iraqi Government failures, as well as the disruption of social stability and related adverse socio-economic conditions resulting from state military force (Albino et al. 2016:9; Ingram 2015:735). This shadow governance provided system learning and therefore, ISIS adapted from coercion-centric control to portraying its governance 99 | P a g e 5249168 Cassandra Brooker Masters of Research apparatus as multi-dimensional, sophisticated, bureaucratised, and well-resourced; thereby further reinforcing its success and associated influence (Ingram 2015:736,739).

Whilst there were balancing feedback loops in place to prevent the exponential escalation (and subsequent implosion) of the R3 reinforcing feedback loop, they initially comprised weaker, disparate efforts, conducted by external systems in isolation, with competing priorities (Meadows 2018:125-126). The next section details how these balancing feedback loops were continually improved and how those systems evolved to eventually slow ISIS’ R3 reinforcing feedback and contribute to its influence sinks. Figure 13 depicts ISIS’ Success-to-the-Successful reinforcing feedback loop (R3).

100 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 13: ISIS Reinforcing Feedback Loop (R3): Success-to-the-Successful

101 | P a g e 5249168 Cassandra Brooker Masters of Research Exposure (R4) The final reinforcing feedback loop (R4) relates to exposure, where ISIS’ influence stocks were reinforced through the effective exploitation of IO tactics, communications links, and behavioural economics conducted as an “Information Jihad” (Singer & Brooking 2018:154). Exposure tactics, such as inundation, amplification, and repetition were designed to virally inundate and control audience perceptions to build legitimacy, reinforce the narrative, and dominate the information environment (IE). The exposure techniques employed by ISIS manipulated cognitive biases and heuristics, such as: persuasion bias, appeal to authority, illusory truth effect, as well as the frequency, association, and availability heuristics (Table 6).

ISIS’ effective use of exposure to reinforce its influence stocks was assisted through its projected image of success (resulting from the R3 feedback loop and high-profile ‘propaganda by deed’ activities). It also employed a global network of recruiters and super spreaders, a highly professional media wing, a steady torrent of viral online content, advertising tactics, and an unassailable brand that made terrorism “sexy” (Dhanaraj 2018:2-3; Singer & Brooking 2018:65,154,167; Ruthven 2007:2; Afzal & Wallis 2016:83-83; Patrikarakos 2017:233; Knight & Theodorakis 2019).

ISIS used exposure to reinforce its influence, strengthen its internal sub-systems, and to run its military offensive like a viral marketing campaign. As Singer and Brooking (2018:6-8) explain, before invading northern Iraq, ISIS employed a similar tactic as the German blitzkrieg at the Maginot Line in World War II, which used radios to sow confusion, doubt and fear. However, ISIS pioneered a different sort of blitzkrieg, using the internet as a weapon (Patrikarakos 2017:251; Afzal & Wallis 2016:75; Singer & Brooking 2018:151). Even though ISIS had not yet arrived, fear was already ruling the military and security ranks with thousands of Iraqi soldiers and police streaming out of Mosul, leaving weapons and vehicles behind. In addition, nearly half a million civilians fled. When ISIS arrived, they were astounded at their good fortune and learnt important lessons for strengthening system resilience. ISIS’ use of exposure and the associated feedback it received, to enhance its scare campaign and grow its influence stock, was further reinforced by videos and images that moved faster than the truth and were simultaneously horrifying and intoxicating to global audiences (Singer & Brooking 2018:8; Watts 2018:28-29,45- 46). Their rapid multi-pronged, multi-platform, multi-lingual production and distribution of content won them a victory that probably should not have been possible. Figure 14 depicts the R4 Exposure reinforcing feedback loop, including the external causes contributing to ISIS’ success.

102 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 14: ISIS Reinforcing Feedback Loop (R4): Exposure

103 | P a g e 5249168 Cassandra Brooker Masters of Research Balancing feedback loops Meadows (2018:125-126) explained that if reinforcing feedback loops were allowed to exponentially escalate, with no effective balancing feedback, then this would result in an eventual collapse of the system. Whilst ISIS’ system initially comprised a virtuous cycle of strong reinforcing feedback loops creating rapid growth in influence stock, eventually the balancing feedback caught up, due to a combination of internal and external factors. The three key balancing feedback loops identified in this case system analysis, and discussed in this section, are summarised in Table 11.

Balancing Key Description Feedback Loop ISIS’ influence was diminished through the development of effective, diverse, Counter- counter-influence activities designed to shift ISIS’ iWar system from anticipatory B1 Influence to reactive, dislocate its power, and break its Success-to-the-Successful cycle (R3). Loss of ISIS’ influence was undermined by a loss of control & authority resulting from a Authority & B2 number of internal and external system impacts and delays in the associated Control feedback.

ISIS’ reinforcing feedback loops deteriorated and sunk influence stocks due to an Drift to Low internal drift to low performance, which resulted from balancing feedback B3 Performance including: lost leadership & resources, degraded communications, competing systems’ goals & policy resistance, and a decline in system resilience.

Table 11: ISIS Balancing Feedback Loops

Counter-Influence (B1) The B1 Counter-Influence balancing feedback loop strengthened over time as external systems learnt, adapted, became more anticipatory themselves, and therefore more effective in countering ISIS’ virtuous cycle of influence. Meadows (2008:129-130) explains that success-to-the- successful loops (R3) could be controlled by putting into place balancing feedback that prevented competitors from taking over entirely. This “levelling of the playing field” was achieved through a diversification of equalising mechanisms employed indirectly by various external systems to dislocate ISIS’ anti-fragile influencing power through other more fragile political, social, or economic dimensions (Albino et al. 2016:12). Counter-influence activities were designed to shift ISIS’ anti-fragile iWar system from its characteristic anticipatory resilience and pressure it to become more fragile and reactive; whilst concurrently enabling State and Coalition IO and targeting cycles to overtake ISIS’ own planning and decision cycles.

Counter-influence activities contributing to the effectiveness of the B1 loop (and consequently the B2 loop) included the use of soft power tactics, indirect effects, community engagement, and 104 | P a g e 5249168 Cassandra Brooker Masters of Research authentic, ethical IO messaging backed up with deeds; to develop trust, credibility and improved relationships with local populations, as well as providing access to local information networks relevant to undermining ISIS’ influence (Albino et al. 2016:10; Lieber & Reiley 2016:47). Counter IO and non-kinetic targeting of communications links were designed to expose ISIS’ hypocrisies, undermine its influence over target audiences, and break its Success-to-the- Successful cycle. ISIS’ ‘propaganda by deed’ tactics were countered with own-force deeds to reinforce Coalition messaging of being the ethical, truthful, and transparent actor in the ‘battle of the narratives’ with ISIS.

The efforts involved in making B1 an effective balancing loop were not without difficulties. Where soft power and nation-building activities were instrumental in gaining the trust and support of the population and undermining ISIS’ control and influence in Marawi in the Philippines in 2017 (Knight & Theodorakis 2019); such success was not automatically translatable to Iraq and Syria. With Iraq’s historical context of the State actively marginalising Sunni populations, and with the effects of airstrikes against civilian communities in Syria, along with the difficulties of conducting urban operations in both countries, all combined to provide further reinforcing feedback to ISIS’ narrative and influence stocks. The most successful tactics contributing to B1 were counter-IO strategies based on ‘reverse-engineering’ the core principles underpinning ISIS’ strategic logic, these included: linking ISIS to perceptions of crisis, denigrating its system of control, and exposing ruptures between its narrative and actions (Ingram 2015:746).

Counter-influence strategies contributing to B1 also had to recognise the challenges of competing against ISIS’ extremely effective, sustained, ideological, multi-dimensional iWar campaign, from which its entire system derived its stocks, support, legitimacy, and centre of gravity (Ingram 2015:743-744; Lieber & Reiley 2016:48). Accordingly, ISIS was not bound by convention or ethics, and was a master of manipulation and exploiting influence effectiveness, behavioural economics, and hard and soft communications links. Additionally, as Ingram (2015:746) cautioned, government counter-narrative strategies had to avoid confronting ISIS ideologically. For instance, even the most jurisprudentially sound counter-ISIS case commands zero credibility, undermines ‘moderate’ ideologues, and reinforces ISIS’ central narrative if it is delivered by non- Muslim actors.

Luckily, in September 2016, another external system addressed this ideological gap in counter- influence and contributed significantly to B1’s growth. As Ward (2018:102-103) explained: “globally renowned Sunni leaders spoke out in unison against Wahhabism during an international conference in Grozny. All one hundred clerics unanimously took a stand against the Wahhabi 105 | P a g e 5249168 Cassandra Brooker Masters of Research terrorists who condemn and murder non-believers. This was the first time Sunni Islamic scholars had clearly rejected Wahhabism as part of the larger Sunni family” (Ibid:102-103).

The counter-influence activities contributing to B1 growth, combined with improved counter- terrorism approaches, and increased political rights and civil liberties for the populace, resulted in also strengthening the B2 and B3 balancing feedback loops. All these factors eventually succeeded in slowing ISIS’ influence stock growth, upsetting its interlinked relationships between internal sub-systems, and reshaping ISIS’ whole system towards a more decentralised organisation (Kilberg 2012:825). The B1 loop is depicted at Figure 15 and comprises the key counter-influence contributions of various systems and elements.

106 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 15: ISIS Balancing Feedback Loop (B1): Counter-Influence

107 | P a g e 5249168 Cassandra Brooker Masters of Research Loss of authority / control (B2) ISIS’ influence was further undermined by the B2 balancing feedback loop comprising a loss of control and authority across both its physical and virtual caliphates. This loss of control, which is a critical capability supporting ISIS’ COG (Fig 9), resulted from a number of internal and external system impacts combined with delays in the associated feedback surrounding the detrimental effects on their influence stock (Meadows 2008:117-121). Along with the effective counter- influence balancing feedback of B1, the physical losses of critical requirements such as: personnel, resources, leaders, communications links, territory, and accesses to target audiences resulted in sinks of influence.

ISIS’ use of indiscriminate violence and atrocities committed against Muslims, particularly by heavy-handed foreign fighters with contrary motivations, further undermined their legitimacy and credibility. This was a lesson, previously provided by Al Qaeda, that ISIS seemingly forgot as they grew more desperate on the physical battlespace (Mehmetcik & Kursun 2017:65; Ciovacco et al. 2011:40-42; Albino et al. 2016:4). Additionally, ISIS’ over-reach of the Caliphate saw a dilution of influence and control, as they expanded into non-Sunni territory and encountered communities empowered to oppose them (Mehmetcik & Kursun 2017:58).

Further contributing to B2 feedback effectiveness was the combined effect of separate activities conducted by external systems, which included: the online community, technology companies, mainstream media, and transnational law enforcement. As recommended by Dhanaraj (2018:6) and Ciovacco, Clark and Van de Velde (2011:48), these individual systems successfully regulated, restricted, and prevented flows of ISIS content, undermined their legitimacy and authority using transparency and humour, and denied services, network accesses, and online safe havens. This resulted in an accidentally co-ordinated, multi-pronged front, which created compound dilemmas for the ISIS system to respond to effectively, when it was already degraded.

ISIS’ loss of authority, resulting from this B2 feedback, also applied internally and therefore, is linked to the B3 balancing loop: ‘Drift to Low Performance’. As ISIS lost caliphate territory, communications links, and key personnel; its image, branding and appeal were undermined, many ISIS supporters became disillusioned, and internal sub-systems and elements developed competing goals and alternative motivations (Dhanaraj 2018:3). Additionally, as ISIS lost its territory and ‘geographic centre’, it relied increasingly on erratic electronic communications across vast distances, which limited its ability to maintain a centralised hierarchical bureaucracy and control its dispersed units and sub-systems (Kilberg 2012:824). The B2 loop is depicted in Figure 16 below.

108 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 16: ISIS Balancing Feedback Loop (B2): Loss of Control

109 | P a g e 5249168 Cassandra Brooker Masters of Research Drift to Low Performance (B3) The performance of ISIS’ reinforcing feedback loops deteriorated and increasingly sunk their influence stocks due to an internal ‘drift to low performance’ archetype, resulting from factors covered by the B3 balancing feedback loop, such as: battlefield and leadership losses, degraded communications and media standards, eroding goals and policy resistance, and a rise in competing systems. ISIS’ reinforcing feedback loops that should have kept the system state at acceptable levels, were overwhelmed by the increasing strength of the balancing feedback loops countering its influence. Following the aforementioned system impacts generated by the B1 and B2 balancing feedback loops, the cohesion and unity bonding ISIS’ internal sub-systems started to corrode, leading to various actors pulling the influence stock toward various competing goals, resulting in policy resistance and a decline in system resilience (Meadows 2008:116). This drift to low performance was a gradual process, where ISIS as a system became more reactive, employing quick fixes, tending to believe bad news over good, dwelling on failures, and developing a perception that the system was in a worse state than it actually was (Meadows 2008:122-123).

Where previously ISIS were extremely successful, resilient and difficult to counter, the drift to low performance balancing feedback eventually resulted in internal system elements lowering their expectations, efforts, and performance standards (Meadows 2008:123). This drift, when combined with other losses and the effects of delayed feedback on the system, was difficult for ISIS to prevent or reverse.

The loss of authority and subsequent undermining of ISIS’ centralised hierarchy, created by the B2 loop, enabled the rise of internal system elements developing competing goals, and supportive external systems becoming accidental adversaries. Certain actors’ heavy-handedness in the application of Sharia law eroded ISIS’ support among local populations, which was central for its state-building efforts. Foreign fighters coming from Western countries were particularly violent and had their own external motivations, not aligned with ISIS’ ideology (Mehmetcik & Kursun 2017:65). As Dhanaraj (2018:5) explains, where disparate actors and sub-groups broke away from the organisation and conducted random, unjustified attacks, such leaderless resistance ultimately became a message-less resistance; which further balanced the R2 reinforcing feedback garnered from ISIS’ ideological legitimacy and sense of purpose. Such acts of message-less resistance undermine legitimacy, fail to provide meaning, and do not inspire others to act; which further reinforces the system’s drift to low performance. The B3 loop is depicted in Figure 17.

110 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 17: ISIS Balancing Feedback Loop (B3): Drift to Low Performance

111 | P a g e 5249168 Cassandra Brooker Masters of Research This is not to say ISIS did not adapt to this balancing feedback. ISIS was still a highly resilient system and effectively maintained its influence stocks among its sources and virtual tribe. Accordingly, ISIS evolved into a more decentralised, partially leaderless resistance movement in the face of territorial, leadership, and iWar losses – effectively withdrawing back to its terrorist organisational roots to regroup, recoup, and recover. ISIS shifted its strategy away from the physical caliphate in Iraq and Syria, towards leading the global Salafi-jihadist movement. As such, ISIS recognised the importance of declaring a global caliphate and securing territory in different parts of the world to maintain its influence (Dhanaraj 2018:6).

Delayed Feedback As depicted in the balancing feedback loop diagrams, ISIS suffered from delays in feedback when its influence stock started to decline and communications links were degraded. These delays undermined ISIS’ ability to react effectively and therefore impacted on the whole system’s anticipatory resilience. As Meadows (2008:23) explains, a stock usually changes slowly, because flows take time to flow, and can cause delays, lags, buffers, or sources of momentum in a system. The time lags imposed by slowly changing stocks can provide room for a system to manoeuvre or revise its approaches (Meadows 2008:23-24). However, in ISIS’ case this delayed feedback caused problems in its system behaviours, responses and decision-making. If a decision is made based on delayed information, the resulting action will be misguided, off-target, may amplify variations, and can create instability through oscillations (Meadows 2008:104-105). The loss of control (B2) and broken communications links were key contributors to the delays in feedback received by ISIS as they were defeated in Iraq and Syria.

Conclusion Figure 18 concludes this chapter by depicting the holistic system diagram for ISIS, as an effective iWar case study. The model includes all of the elements, and the reinforcing and balancing feedback loops identified from the narrative analysis conducted in this chapter.

Four key themes become apparent as contributing to ISIS system’s effectiveness at maintaining their influence stocks as follows:  Their system characteristics of being a relatively closed system with a high level of resilience.  The efficiency of system flows both via feedback loops and through the strength of hard and soft interconnections between key elements.

112 | P a g e 5249168 Cassandra Brooker Masters of Research  Having and maintaining control, which is a critical capability and a vital reinforcing element to increasing influence stocks.  Having a centrality of focus, where influence activities are the main priority and purpose driving all system outputs, and which contributed to strengthening all four reinforcing feedback loops.

The same systems analysis process will be applied to an ineffective iWar case study in Chapter 5, of Hillary Clinton’s 2016 Election Campaign. Chapter 6 concludes Part Two of this thesis with a comparative analysis of the findings arising from the modelling of both case systems.

113 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 18: Holistic ISIS System Diagram

114 | P a g e 5249168 Cassandra Brooker Masters of Research Chapter 5 – System Analysis of Influence Ineffectiveness: Clinton Campaign

Fact checkers couldn’t win the culture war in the US election.

Mark Laity NATO Strat-Com

Introduction In contrast to ISIS’ influence system, where virtuous reinforcing feedback enabled success-to- the-successful; the Clinton campaign of the 2016 US Presidential Election suffered a vicious reinforcing cycle resulting from feedback associated with internal and external system events and activities. This impacted on its influence stock in a downward spiral. There is a psychological and political tendency to blame external influences for the cause of system problems (Meadows 2008:4) and Clinton had just cause in doing so; given the escalating counter-influence reinforcing feedback pitted against her campaign. However, competing goals of sub-systems, the rise of accidental adversaries, and a loss of both control and system resilience resulted in a ‘Drift to Low Performance’ archetype and contributed to this vicious cycle (Meadows 2008:15,122). The modelling of Clinton’s campaign highlights the intrinsic system problems, undesirable perceptions, and behavioural characteristics, which, when combined with weak balancing feedback loops, undermined the effectiveness of the campaign.

Importantly however, the activities of other external systems generated significant impacts on Clinton’s system behaviour due to the inherent nature of an open, public election campaign. For example, when comparing Clinton’s campaign with ISIS’ closed, secretive, opaque system; external systems had a limited effect on the terrorist group’s dynamics and influence stocks. In fact, as Benkler, Farris and Roberts (2018:45-71) explain, tens of thousands of entities and a diverse range of sources form the complex ecosystem of America’s political media, all of which contributed feedback into Clinton’s system.

To understand the US media ecosystem, Benkler et al. (2018:59-61,75) created a network map and identified two distinct, structurally different sub-systems. The right-wing media sub-system was dominated by densely interconnected and insular partisan outlets; whilst the moderate mainstream media (MSM) sub-system spanned the rest of the centre-right to left spectrum, and adhered to journalistic standards, allegiance to facts, and context over conclusions. The asymmetric split in the US national media ecosystem means politicians and media outlets face starkly different incentives. Benkler et al. (2018:80) explained right-wing propagandists and the Trump campaign benefited and were rewarded for spreading lies and disinformation, whereas

115 | P a g e 5249168 Cassandra Brooker Masters of Research Clinton was held accountable by the centre-left media that diligently conducted fact-checking and reinforced truth-seeking behaviour.

Clinton Campaign: Robust system suffering some fragility As determined in Chapter 2 (see Fig 3), Clinton’s campaign system was assessed to be an open system with a robust level of resilience that suffered increased fragility throughout the bounded time period of the 2016 Election, due to an asymmetry of balancing and reinforcing feedback. Whilst, like ISIS, the campaign was a complex social system, it suffered from a lack of diversity, making it more vulnerable to external shocks and unable to effectively balance the escalating vicious reinforcing feedback cycle (Meadows 2008:84). Additionally, systems that are constant over time can lack resilience, which is difficult to identify without a holistic system overview and due to the gradual nature of a drift to low performance. The Clinton campaign applied the same approaches to achieving the same outputs as Democrat systems had done during previous election campaigns. Unfortunately, as Meadows (2008:77) explains, having static stability, business-as- usual productivity, or conducting familiar, repetitive activities can sacrifice system resilience. Meadows (2008:78) adds that “a loss of resilience can come as a surprise, because the system is usually paying much more attention to its play than to its playing space. One day it does something it has done a hundred times before and crashes” (Ibid:78).

Meadows (2008:30) contends that the presence of a feedback mechanism does not necessarily mean the mechanism works well. The balancing feedback loops in Clinton’s campaign were not strong enough to increase influence stocks to desired levels. Additionally, like ISIS, Clinton’s system suffered from delayed feedback regarding its stock levels – where the interconnections’ flows failed due to information arriving too late, at the wrong place, or it was unclear, incomplete or hard to interpret. Therefore, reactions or decisions triggered by this feedback were often misguided, delayed, or ineffective, which further weakened system resilience. The delayed feedback dynamics of Clinton’s campaign are analysed at the end of this chapter.

In order to deconstruct Clinton’s campaign, as a complex system, a holistic assessment of the centre of gravity (COG) and supporting critical capabilities (CCs) was conducted (Fig 19). This analysis informed the identification of key elements and the associated feedback loops, whilst bounding the model to a single influence stock. The campaign’s COG was assessed to be: ‘the ability to effectively influence target audiences to secure the election win’, with its critical capabilities being: ideology, means, control, media capabilities, and resources. The CCs are divided further into critical requirements (CRs), which if targeted become critical vulnerabilities (CVs). The consequences of such targeting are also included in Figure 19. 116 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 19: Clinton Campaign Centre of Gravity

System Elements Building upon the holistic understanding provided by the system characterisation, COG construct, and qualitative narrative data analysis, the key elements of Clinton’s campaign system were identified and considered in detail. The system elements, sub-systems and external systems derived from the de-composition process, and their interconnections, are described below and mapped in a baseline system model at Figure 20.

Stock and Purpose: As per the ISIS system model, the single stock being researched in the Clinton campaign model is influence. The purpose of the campaign, as a complex social system, was to win the 2016 US election by effectively influencing target audiences to vote for Clinton and the Democratic Party. Priorities associated with maintaining influence stocks and achieving campaign aims included: building credibility, trust, and confidence in the system through a strong ideological narrative. Activities contributing to these aims included: controlling the associated messaging, campaign and policy debates, maintaining the “experienced”, “front runner” reputation of Clinton, and building on goodwill generated by the Obama Administration.

117 | P a g e 5249168 Cassandra Brooker Masters of Research Source: Clinton’s campaign influence source primarily comprised the Democrats’ ideological support base, specifically current supporters of the Obama Administration and previous supporters of Bill Clinton. However, sources also included: centre-left media outlets, the left- leaning, socially conscious American electorate; minority communities; feminists, and other voters wanting to see the election of the first female US President. There was also a groundswell of grassroots supporters who valued Clinton’s experience as Secretary of State and her pragmatic, socially responsible campaign policies.

Sink: As a result of the vicious cycle of reinforcing feedback, the sinks to campaign influence stocks included: the loss of swing voters, presidential contender Bernie Sanders’ supporters, centralist media, and online followers (Silver 2018; Reynolds 2018; Clinton 2018:277,399-400). Additional sinks comprised a loss of reputation and credibility, reduced messaging effectiveness, and campaign communications being ignored, misdirected or dislocated by events. These sinks further increased mistrust across the electorate and contributed to a loss of control, lowered system resilience, and undermined belief in the cause (all CCs), which when combined, contributed to the system’s drift to low performance.

Internal Sub-systems Four internal sub-systems were identified as integral to contributing to the system’s influence stocks and maintaining the balancing feedback loops.

Sub-system: Campaign Leadership Clinton herself, as leader of the campaign, was the primary influential contributor to system stocks. Her relevant experience as Secretary of State, combined with her husband’s legacy, her social policies appealing to minorities and women, and the fact that she was the antithesis of Trump, set her up as the legitimate front runner (Clapper 2018:328-329). The leadership and media sub-systems worked together to exploit the paradigms surrounding Clinton’s proven reputation in building balancing feedback relating to her credibility, her proposals to address inequality, a united sense of purpose, and a sense of community with a social conscience. Additionally, many staffers involved in the leadership team were experienced professionals, adding to the campaign’s legitimacy, authority and influence. The campaign leadership sub- system represented the critical capabilities of ideology and control in protecting the system COG.

Sub-system: Campaign Media Team As in ISIS’ system, the media sub-system represented a critical capability and was well-resourced (Silver 2017). It worked effectively to adopt successful influencing tactics used during former 118 | P a g e 5249168 Cassandra Brooker Masters of Research President Barack Obama’s campaign and apply them to Clinton’s campaign media strategy. The team collected metadata and conducted advanced data analytics to model the electorate, target voters, test messages, conduct grassroots organising, and win delegates (Clinton 2018:64-65). It also undertook polls, opinion research, and focus groups to create personalised messaging and targeted advertising. The media sub-system built an effective narrative surrounding balancing feedback associated with integrity, a sense of purpose, belief in the cause, and an inclusive community, to further boost system influence and strengthen linkages between sub-systems.

However, as demonstrated by subsequent feedback, the media team’s tailored approaches paled in comparison to the aggressive, industrial-scale, data-driven, behavioural exploitation operation conducted by the Trump campaign, with the assistance of Cambridge Analytica (Singer & Brooking 2018:177-179; Watts 2018:221-222; Benkler et al. 2018:19,215; Clapper 2018:333, 353). Also, due to the asymmetric media ecosystem, Clinton’s media team was hamstrung, by ethics and transparency requirements, in the influence tactics they could effectively employ. These obstacles correspond to those encountered by Western democracies fighting ISIS in the iWar (Table 4). Additionally, whilst the media team was heavily focused on public relations (PR) to maintain Clinton’s image and credibility, it forgot key influencing lessons from the Obama Campaign. The Democratic Party (DNC) alienated Obama’s organic, grassroots support base, disempowered them, and dislocated their self-organising capabilities (Benkler et al. 2018:344). The failings of this corrosive approach towards grassroots supporters were neither recognised nor rectified for Clinton’s campaign.

Sub-system: Campaign Membership The DNC members, volunteers, and supporters (described under source elements) represented the operational and tactical level sub-system that was a driving force in organising and conducting influence activities – both physically at rallies, visits, and speeches, and virtually through online action groups, advertising, and lobbying support. This sub-system contributed to the CCs of ‘means’ and ‘control’ for the whole system, and relied on the linkages and support of the other three sub-systems to operate effectively and achieve its outputs for increasing influence. However, the insidious impacts of reinforcing feedback, which included multi-pronged attacks against members, networks, Clinton’s credibility, and the truth; slowly eroded the membership sub-system’s resilience (Singer & Brooking 2018:16,142-143,175; Clinton 2018:229,345; Benkler et al. 2018:200,270).

119 | P a g e 5249168 Cassandra Brooker Masters of Research Sub-system: Campaign Resourcing The resourcing sub-system represented the CCs of ‘resources’ and ‘means’ in protecting the campaign’s COG, whilst also supporting and enabling the other three sub-systems’ influencing activities. Clinton’s campaign budget, and resulting expenditure of USD1.2 billion, was double that of the Trump campaign, and had 3320 more paid staffers than Trump (Silver 2018). The resourcing sub-system benefited from established Democratic revenue streams, traditional donor sources, and the goodwill generated by the Obama Administration, Clinton’s reputation, and their respective policy platforms. However, as with the other internal sub-systems, delayed feedback arising from negative reinforcing loops, resultant poor decision-making surrounding influence activities, and misguided expenditure; meant this sub-system’s resilience was also impacted upon negatively, resulting in a drift to low performance.

External Systems Target Audience The target audience of voters, particularly swing voters, and moderate mainstream media comprised the main external system that the Clinton campaign sought to influence. The campaign also relied upon this external system to bolster its balancing feedback loops and assist internal sub-systems in countering the vicious cycle of reinforcing feedback. However, the impacts of the counter-influence reinforcing feedback increasingly normalised and regularised political meaning derived from right-wing propaganda and disinformation, thereby undermining Clinton’s influence stock. If the opinions of citizens are poorly formed, weak, or subject to manipulation informed by social and cognitive psychology (as discussed in Chapter 2), “then the idea of deliberative democracy by an informed citizenry exercising self-governance is a utopia” (Benkler et al. 2018:26-27).

Trump Campaign The external system of the Trump campaign included an internal sub-system comprising an online troll army that evolved in concert with Russian iWar activities and Trump’s domination of the spotlight (Singer & Brooking 2018:175-176; Watts 2018:143). Whilst this troll army was not directly linked to Trump’s media team, it was later backed by his campaign strategy, and contributed to his effective massive-scale, social media influencing activities. Trump’s campaign used tactics that included: personalised micro-targeting, marketing techniques, sentiment manipulation, and inundation (Wardle 2019; Singer & Brooking 2018:175-176). Watts (2018:19) compared the Trump campaign’s influencing tactics in overtaking the Republican Party, as eerily similar to how ISIS overtook al-Qaeda on social media. The Trump campaign was aided by other external systems, including: Russia’s Internet Research Agency (IRA), right-wing media, 120 | P a g e 5249168 Cassandra Brooker Masters of Research WikiLeaks, and Cambridge Analytica. As a combined force, these disparate systems were able to dominate the information environment (IE) and heavily contributed to the vicious reinforcing feedback impacting negatively on Clinton’s influence stock.

Due to the support of these external systems, Trump was not held to the same standards of accountability and transparency as Clinton (Benkler et al. 2018:80-81; Clinton 2018:265). Freedom from democratic norms and political ethics enabled his campaign to be fuelled by audacity, outrage, and divisiveness, as well as facilitating the propagation of conspiracy theories, fear mongering, ‘flame wars’, character assassinations, and lies (Benkler et al. 2018:18; Clinton 2018:76). Like ISIS and Russia’s IRA, the Trump campaign was not hamstrung by moral conscience (Table 4) and employed many of the same iWar tactics as America’s adversaries. Trump’s counter-influencing tactics and reinforcing feedback were so successful that James Clapper, the former US Director of National Intelligence, asked whether Trump meant it literally when he said: “I could stand in the middle of Fifth Avenue and shoot somebody and I wouldn’t lose voters” (Clapper 2018:353). The vast extent of Trump’s campaign tactics is summarised in Table 12 alongside the related effective influencing and behavioural theories that were discussed in Chapter 2 (Table 6).

121 | P a g e 5249168 Cassandra Brooker Masters of Research Trump Campaign iWar tactics Effective Influence Behavioural theory

• Used Twitter to exert power & create feedback loop with media • MEDIA BIAS • Exploits ‘System1’ thinking • Influence flowed in both directions via unusual multi-media relationship • COVERT • Exploits impressionable/ gullible • Trump in mutually reinforcing dynamic with Breitbart & RW media - “Fox News effect” INFLUENCE • Exploits exposure bias • Trump’s tweets became legitimate news - across entire media ecosystem • AUDIENCE • Trump was centre of media attention = influenced the media agenda SUSCEPTIBILITY

Troll Army: never stopped, endlessly available, always producing: • ‘Persuasion Bias’ • Frantic mania - set a tempo no traditional campaign could match • ‘Illusory truth effect’ • Helped steer online trends to promote Trump & shape the election • EXPOSURE • ‘Exposure effect’ & ‘frequency heuristic’ • Ensured impactful attacks continued to fester & never left public attention • INUNDATION • Exploits ‘Availability’ heuristic Inundation: Trump had the most social media followers & vastly greater media coverage • AMPLIFICATION • Exploits ‘Association’ heuristic • Deployed network to scale, sent most messages, on the most platforms, to the most people • REPETITION • ‘Appeal to Authority’ cognitive bias • Trump’s Twitter loudspeaker drove the national conversation at pace & volume where • NEWS FEED • System 1 focus on higher ranked results journalists and opponents couldn’t keep up - contributing to information disorder RANKINGS • Sequence matters / ‘halo • Understood wealth of information = poverty of attention, generated 4-5 storylines a day • SEQUENCING • ‘System 1 intuitive impressions • Campaign focused on social media, tailoring, sentiment manipulation & machine learning. • Rankings related to shares/likes/clicks Largest digital effort ever in US political history

• Troll Army provided best of both worlds: Trump could deny or incorporate into campaign • ‘Halo effect’ = resilient first impression • Troll Army created controversies & conspiracy theories to waste opponents’ resources • DISINFORMATION • Argument fallacies • Bots & sock puppets amplified messages, expanded support base • MALINFORMATION • Grab the attention of ‘System 1’ cognition • Promoted topics/themes of audience preferences (filter bubbles) • FAKE NEWS • Exploits lack of critical thinking • Use of Framing: Immigration, fear of Muslims, Islamic terrorism, Clinton taking jobs / guns • FORGERIES • Nonsensical statements evoke initial belief • Manipulated truth (Clinton’s emails), enabled character assassination & conspiracies • BOT / TROLL • Links to social/emotional contagions • Aligned with RW Media to reinforce confirmation biases, propaganda feedback loop ACCOUNTS • Reinforces flawed beliefs, confirmation bias • Disorientation - bizarre stories circulated widely in RW media re: Clinton • Addictive content / political clickbait

• Diversion: of attention by introducing new themes of greater impact or interest • Narratives framed as stories - play on • The more a narrative grew in popularity, the more Trump amplified it = more support biases, emotions and social norms CONTROLLING THE • Skewed versions of events perfectly suited Trump’s political narrative • Exploit unconscious, effortless ‘System 1’ NARRATIVE • Benefited from manipulating the facts (intelligence allowed to become politicised) • Narratives evoking emotion stimulate • Shifted the narrative - which the public and media willingly followed action

• Understanding and mastery of self-made marketing via reality television • System 1 immediately responds to • Reality TV formula: keep people watching with an endless soap opera, sustained drama, imagery, humour, emotion constant conflict, heroes, villains, simple storylines VISUAL IMAGERY & • Exploits ‘Representativeness’ bias • Entertained & exploited audience biases, weaponised memes ENTERTAINMENT • ‘Confirmatory Bias’ of discriminatory beliefs • Expert at clickbait populist narratives, repeated regardless of facts or made sense • Entertaining content more memorable • Outrageous & disrespectful statements, personal insults, trafficked in conspiracy theories • Exploits limited attention spans

• Generated Fear & Anger, re: Immigration, Muslims, gun control, “crooked Hillary”, threats • ‘Affect’ heuristic of System 1 • ‘Flame wars’ succeeded at drawing attention • Anger bypasses ‘System 2’ thinking EMOTIONAL • Emotive content stoked cycle of attention & outrage that kept Trump in the spotlight • Anger is most influential emotion CONTAGION • Riled up support base re: Clinton shredding constitution, confiscating guns • Anger / Outrage are exciting & addictive • Fear lowers people’s ability to distinguish fact from fiction, lies easier to sell • Emotion affects educated, critical thinkers

• Appealed to similarity & homophily biases, justified racism, xenophobia - difficult to counter • Exploits peer group pressure, social norms, • Immigration topic - created emotional/ social contagions group approval • Reinforced nationalism / ‘in-group’ narrative by cognitive appeals • ‘In group’ vs ‘others’ dynamics • Tailored messages targeted low socio-economic areas, suffering job losses SOCIAL CONTAGION • Appeal to the Masses logic error • Trump’s persona on ‘The Apprentice’ appealed to disadvantaged rural populace • ‘Appeal to Authority Bias’ • Didn’t use big words, used soaring rhetoric, voiced their concerns • Power of ‘Homophily’ ‘Confirmation Biases’ • Flouted convention & upset progressive politicians & liberal media Exploits ‘Similarity’ heuristic of System 1

• Low-income households heavily targeted with ads focused on immigration & racial conflict • Unconscious, effortless ‘System 1’ thinking • Cambridge Analytica = tailored messaging, population modelling, psychometric analysis • Limited attention spans PERSONALISATION & • Research identified political dispositions, psychological profiles, and creation of “perfect” • Limited computational capacity TAILORING messages for engaging voters • System 1 pays more attention to • Multi-platform inundation linked to demographic data personalised content

• Clickbait, fake news, conspiracies – widespread, echo chambers, filter bubbles • Addictive ‘system 1’ behaviours ADDICTIVE CONTENT • Use of reality TV formula, entertainment, contagions • Emotional / Social contagions & FEEDBACK • Group reward: “Build the Wall”= hot topic, got people excited, a rallying call • Rewards from peers, monetary gain, MECHANISMS confirmation biases, in-group

Table 12: Trump’s iWar Tactics

Sources: Benkler et al. (2018:18,19,36,78,105,108,114,274); DellaVigna & Kaplan (2007); Singer & Brooking (2018:173-179); Watts (2018:134,221-222); Clapper (2018:181,328-329); Clinton (2018:187,225); Donovan (2019)

122 | P a g e 5249168 Cassandra Brooker Masters of Research Right-Wing Media The right-wing (RW) media ecosystem, comprising all forms of media, enabled both the Trump campaign and the IRA’s iWar tactics during the 2016 election campaign. The Trump campaign’s extensive creation, replication and propagation of fake news, conspiracies, and counter-influence content, which often contained an element of truth for improved effectiveness, continued to the point where stories appeared natural, normal, and credible and were therefore picked up by mainstream media outlets (Watts 2018:91-93; Benkler et al. 2018:73-74; Patrikarakos 2017:15; Clinton 2018:361; Clapper 2018:395) It was not one message, media outlet, or social media troll that mattered, but rather the sum of echoed content creating an environment of seemingly independent sources repeating the same message. As discussed in Chapter 2, inundation and repetition of disinformation leads to familiarity, belief and iWar success.

The radicalisation of RW media content was driven by a group of extreme sites including: Breitbart, Infowars, Truthfeed, Zero Hedge, and the Gateway Pundit, none of which adhered to the norms or processes of professional journalistic objectivity (Benkler et al. 2018:14; Tucker 2016; Canadian SIS 2018:47-49). The RW media also employed iWar tactics in its production and dissemination of content, including: emotional appeals, evocative political clickbait, information saturation, and reinforcing biases and prejudices through the exploitation of closed loop discourse such as: filter bubbles, algorithms, echo chambers, and propaganda feedback loops (Benkler et al. 2018:78; Donovan 2019; Canadian SIS 2018:17; Watts 2018:216; DellaVigna & Kaplan 2007:1228).

Building on the “Fox News effect” discussed in Chapter 2, which increased Republican votes in the 2000 US election; Fox News was exclusively biased in its coverage towards the Trump campaign, whilst reinforcing the RW media’s counter-influence feedback against Clinton (Benkler et al. 2018:14). In 2017, Fox News effectively become the propaganda arm of the White House in all but name (DellaVigna & Kaplan 2007:1228; Epstein & Robertson 2015:1; Benkler et al. 2018:14; Clinton 2018:9). Additionally, RW media and politicians conditioned their supporters to distrust mainstream media (MSM), whilst pushing conspiracy theories. Fox News, in particular, used its prominent platform to demonise and delegitimise MSM that tried to adhere to traditional standards of objectivity and accuracy (Clinton 2018:366). The result of this multi- platform, multi-axis RW media barrage was what Benkler et al. (2018:37) called an ‘anomic disorientation’, where audiences could no longer tell truth from fiction, and where there is no truth, the most entertaining conspiracy theory often wins (Ibid:37).

123 | P a g e 5249168 Cassandra Brooker Masters of Research While the RW media may have been initially defined as a sub-system nested within the national media ecosystem, throughout the election period it demonstrated increased independence, isolation, developed competing goals, and became an accidental adversary of MSM. This division between the RW and the remainder of the political media ecosystem widened throughout the election campaign. While the moderate national media ecosystem remained stable, adhered to professional journalistic norms, and was resilient to external escalation; the RW media was increasingly caught up in a reinforcing feedback loop creating a system shift towards more extreme versions of itself and perpetuating the information disorder (Benkler et al. 2018:73-74).

National Media Ecosystem: Centre / Left The pattern of increasing hyper-partisan radicalisation demonstrated by the RW media was not mirrored on the left-wing, largely because extreme left sites do not enjoy the same visibility or prominence as RW sites. This is due to having to serve a more critical audience, and the integration of left-wing media into moderate, traditional, mainstream media outlets, which are committed to journalistic truth-seeking norms (Benkler et al. 2018:14-15).

Whilst media outlets seek to attract large audiences and ensure profits through influence (which can lead to competing priorities), for the majority of the election campaign, the national media ecosystem remained true to its journalistic principles. Unfortunately, as the RW media and counter-influence tactics of the Trump campaign, the IRA, and WikiLeaks gained traction, MSM sought to win back some of this influence by letting standards slip, becoming caught up in publishing salacious stories and supposed ‘scoops’, in its own system’s drift to low performance (Meadows 2008:122; Benkler et al. 2018:40; Canadian SIS 2018:47-49; NATO 2016:4). As Benkler et al. (2018:189) explain, only 30 per cent of the American population inhabits the insular, RW media ecosystem, indicating that it alone could not have secured the result of the 2016 election. The MSM’s “scoop culture” and attempts to appear balanced in reporting made it particularly susceptible to being manipulated into spreading RW propaganda. As a result, Trump received more coverage of his core substantive issues (immigration, jobs, and trade) than Clinton, whose coverage was dominated by scandals and her campaign being associated with the terms: ‘emails’, ‘lie’, ‘scandal’, and ‘foundation’ (Benkler et al. 2018:17,40,189,356-358; Clinton 2018:99,222; Silver 2018).

Clinton explained (2018:222-223) that discussion of public policy only accounted for 10 per cent of all campaign news coverage during the election, the other 90 per cent comprised obsessive coverage of controversies. In 2008, the major networks’ nightly newscasts spent a total of 220 minutes on policy, however in 2016 just 32 minutes was spent on policy and 100 minutes spent 124 | P a g e 5249168 Cassandra Brooker Masters of Research covering Clinton’s emails (Clinton 2018:222-223). MSM also refused to report on Clinton’s detailed policy plans and criticised her for being too prepared for debates whilst, conversely, never challenging Trump on his lack of policy, deceitful promises, nor his political ignorance (Clinton 2018:232). Most of the press were busy chasing ratings and scandals, and were lured into over- reporting on the IRA’s document dumps of hacked emails (Benkler et al. 2018:358). This shifted system behaviours, with increased irresponsibility and a degradation of professional standards undermining the MSM system’s purpose and contributing to the rise of competing goals of sub- systems within the wider national media ecosystem.

Concurrently, the spread of ‘misinformation’ by the MSM ecosystem also increased due to pressures associated with the 24-hour news cycle and the race for breaking news. Journalists were making mistakes in haste and failing to fact check; while the internet enabled numerous untrained, amateur reporters to rapidly and widely communicate inaccurate stories, which were never redacted or corrected (Benkler et al. 2018:37; Clinton 2018:99; Canadian SIS 2018:10; Wardle 2019). Contributing to this spread of political disinformation, misinformation, and tabloid clickbait was the public’s quest for media coverage and political messaging that informed them as best as possible, whilst not causing too much cognitive discomfort (Benkler et al. 2018:76; Hochschild & Einstein 2014:471). Clinton (2018:99) explained that political press coverage increasingly became more superficial, no longer covering important election issues, largely because the way news is consumed has changed – now, getting clicks is all important and encourages sensationalism.

The social media component of the national media ecosystem further contributed to reinforcing feedback impacting on Clinton’s system influence. Watts (2018:247) explains social media makes threats feel urgent in an unprecedented way, and instils a desperate need to access information through “breaking news”, which heightens fear and weakens people’s filter for falsehoods. Wardle (2019) agrees that internet platforms encourage rapid sharing, identity performative behaviours, and emotional contagions – because slowing down to fact-check content before sharing is far less compelling than reinforcing one’s political opinions to an audience and gaining instant feedback. The Facebook newsfeed algorithm and online echo chambers further reinforced patterns of sharing and social contagions within the insularity of tightly clustered user communities. All these dynamics enabled Trump to conduct an online iWar campaign, gave voice to millions in Americans who had not been heard before in Washington, as well as enabling Russia’s propaganda strategy by providing legitimacy and reach (Watts 2018:216; Benkler et al. 2018:342; Canadian SIS 2018:9-10; Clinton 2018:359; Singer & Brooking 2018:139; Epstein & Robertson 2015:2; The Facebook Dilemma 2018). The contemporary phenomenon of ever- 125 | P a g e 5249168 Cassandra Brooker Masters of Research present smart phones and mobile social media access also meant any gaffes made on the campaign trail were documented and uploaded instantly, propelling words or deeds beyond the control of any politician or journalist (Singer & Brooking 2018:56).

Competing systems / Democratic Sub-systems The three competing external systems that impacted on Clinton’s system influence stocks and became accidental adversaries were Bernie Sanders’ campaign, the Obama administration, and the intelligence community (IC). The Obama administration faced predicaments of whether to openly deal with the Russian interference and defend Clinton during the Benghazi investigation, whilst not appearing biased towards or viewed as assisting the Democrats’ campaign. Partisan politics played a key role in preventing a timely response by the Obama administration to IRA iWar activities. (Clinton 2018:356; Uren 2017). Subsequently, the decision by the IC to delay exposing the extent of Russia’s massive cyber and propaganda efforts to undermine American democracy further strengthened the counter-influence reinforcing feedback undermining Clinton’s influence stocks, and exposed the IC’s failures in assessing the extent of the threat against domestic audiences (Clapper 2018:2-3).

In addition to this passive inaction by the IC, which contributed to the erosion of Clinton’s influence by other external systems, were the active attacks on the system’s resilience by the FBI, specifically the Director, James Comey. Not only did the Republican-strong FBI actively leak information to damage Clinton’s campaign, but Comey’s letter to Congress two weeks before polling day, regarding the Clinton email server investigation, effectively cost Clinton the election (Clinton 2018:317; Silver 2018; Benkler et al. 2018:181). Combined with the counter-influencing effects of other external systems, even when Comey cleared Clinton of wrongdoing, it made no difference to those in the US electorate who were susceptible to the Russian campaign’s psychological exploitation tactics (Clapper 2018:355).

As a Democratic sub-system, Sanders’ campaign disrupted the holistic DNC system and contributed to its drift to low performance, while also reinforcing the counter-influence feedback against Clinton’s system by becoming an accidental adversary. Sanders’ retaliatory attacks on Clinton, designed to undermine her control, reputation, and credibility, were based on delayed feedback, competing goals, and a downward spiral of his own influence. Clinton (2018:229) explained that because Sanders could not make arguments on policy areas they agreed on, he resorted to innuendo, questioning her character, attacking her supporters. This made it difficult for Clinton to unify progressives and paved the way for a new theme in IRA messaging and

126 | P a g e 5249168 Cassandra Brooker Masters of Research Trump’s ‘crooked Hillary’ campaign (Watts 2018:149,161-162; Silver 2018; Reynolds 2018; Benkler et al. 2018:255,270).

Russian Internet Research Agency While analysts were divided on the level of impact Russia’s IRA had on the election outcome, there was consensus that US democracy, values and institutions were attacked by a concerted, sustained, complex Russian iWar strategy, which was assisted by Trump’s campaign influencing tactics, and reinforced by the RW media (Clapper 2018:4; Clinton 2018:325). Singer and Brooking (2018:16) argue the Russian offensive dwarfed all previous efforts in terms of staffing, online accounts, automation, and infiltrating every aspect of US political dialogue – effectively launching the most politically consequential attack in history. Like ISIS, the IRA’s influence system is relatively closed and anticipatory in its resilience (Fig 3). The IRA’s original goal was to undermine democracy and initially targeted both sides with its multi-axis, multi-platform propaganda and information manipulation. However, due to the asymmetric political media ecosystem, it was easier to target Clinton’s campaign and was effectively enabled by the RW media’s propaganda feedback loop (Benkler et al. 2018:99).

Russia’s iWar success can be credited partly to the seamless integration of its intelligence and security services, as well as the effective application of influencing and psychological behavioural theories discussed in Chapter 2 (Watts 2018:94; Babbage 2019; Donovan 2019). Their general iWar tactics included: fomenting socio-economic divides, mixing true and false messages, seeding new conspiracies, inflaming tensions, infiltrating RW audiences, manipulating filter bubbles and algorithms, fanning anti-government sentiment, and most significantly, hacking email servers for compromising material or ‘kompromat’ (Watts 2018:91,93; Clapper 2018:303; Singer & Brooking 2018:16; Benkler et al. 2018:99; Reynolds 2018; Wardle 2019; Canadian SIS 2018:46).

Whilst attacking Clinton’s campaign specifically, Russian active measures blended three layers of propaganda messaging (white, grey, black) to influence different target democratic audiences with aggressive anti-Clinton rhetoric, contradictory theories to obscure reality, and discrediting known facts (Watts 2018:132,143; Clapper 2018:268,395). Their multi-faceted campaign created lies to help Trump and hurt Clinton, which were promoted through social media and state- sponsored channels to the point that traditional US media unwittingly spread Russian propaganda (Clapper 2018:395). The IRA’s unprecedented intensity and inundation of social media content promoting Trump resulted in computational distortions, making Trump’s appeal and support base appear greater than it was (Watts 2018:143; Paul & Matthews 2016:1). When Trump won the 127 | P a g e 5249168 Cassandra Brooker Masters of Research election, the Russians were shocked, they had succeeded beyond their wildest imagination and were completely unprepared for it (Clapper 2018:3). Clinton (2018:326) contends the Russian attack succeeded because democratic immunity had slowly eroded, and many Americans had lost faith in the institutions previous generations relied on for objective information, including: government, academia and the press; therefore, leaving them vulnerable to a sophisticated misinformation campaign. It was certainly a combination of these external systems’ activities that provided the counter-influence reinforcing feedback contributing to Clinton’s vicious downwards cycle.

Linkages / Relationships In Figure 20, the interlinked relationships between the system elements are depicted by arrows, coloured red if they contribute to a reinforcing feedback loop or blue for balancing feedback. The specific feedback loops will be detailed in the following sections, however, in de-composing the system into key elements, it is important to detail some of the fundamental relationships depicted in the model. These relationships can be broadly divided into two categories: ideological and communications.

Ideology and Narrative Much of the Clinton campaign’s system influence stocks were derived from the ideology and campaign narratives that contributed to the balancing feedback loops. Providing a ‘sense of purpose’ to campaign supporters through messaging that championed Democratic values, social ethics, and political morality, worked in conjunction with Clinton’s proven experience, existing support base and transparent approach, which in turn strengthened the system’s credibility, legitimacy and influence.

A ‘sense of community’, created by campaign policies and pledges addressing equality, diversity and inclusivity, was strengthened by the relationships formed through engaging minority audiences, women, and centralist-left moderates, and then reinforced through deeds that supported the messaging and countered the divisive, nationalist messaging of the Republicans. The interlinkages between internal sub-systems, external systems and target audiences were strengthened through an ideology and narrative that exploited social and emotional contagions, and which heightened along partisan lines throughout the election campaign. Unfortunately, the resilience of these links was eroded by asymmetric reinforcing feedback that undermined the system’s control of the narrative and influence stocks.

128 | P a g e 5249168 Cassandra Brooker Masters of Research Communications Links Both hard and soft communications links enabled the interlinkages created through ideology and narrative. Soft interlinkages included political, personal, economic, religious, societal, DNC membership, global and feminist relationships between the campaign system and its audiences. Strengthening these links was achieved through open, effective community engagement, visits, listening, conducting focus groups, winning debates, and through the dissemination of policies, messages, and ideas – and then receiving and acting on the associated feedback.

Like ISIS, hard communication links were a critical capability providing the means to influence audiences and provide data analytics for more tailored approaches. Hard links comprised the internet, social media, the news media ecosystem, as well as the campaign media strategy, visits and engagement activities. Information flows via these links enabled the system’s influence stocks to increase as they contributed to the balancing feedback. Unfortunately, these information flows also contributed to the negative reinforcing feedback loops by enabling biased media commentary, inundation of counter-influence content, skewed inaccurate feedback, and the strengthening of filter bubbles, echo chambers and trolling. The baseline system model for Clinton’s campaign depicting all the key elements and interconnections is at Figure 20.

129 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 20: Clinton Baseline System Model – key elements and concepts

130 | P a g e 5249168 Cassandra Brooker Masters of Research Reinforcing feedback loops As demonstrated through the decomposition process and modelling of the key elements (Fig 20), and like the ISIS model, Clinton’s campaign system, with a single stock, is also extremely complex with multiple components, interconnections, and behaviours contributing to a countless array of configurations that could be modelled. Therefore, in recomposing the system and analysing the evident feedback loops in the next two sections, this research has again broadly represented the feedback cycles holistically, as per Chapter 4, resulting in the identification of three reinforcing and three balancing loops. Table 13 summarises the reinforcing feedback loops discussed in this section.

Reinforcing Feedback Key Description Loop

Influence was diminished through the conduct of effective, diverse, multi-axis, counter-influence activities designed to undermine Clinton system resilience & ideology (a CC). Influence stocks were sunk by target Counter- R1 audiences being manipulated and subsequently identifying with in-group narratives, social/emotional Influence contagions, unifying nationalist dynamics, and heuristic biases exploited by Trump, the IRA & RW Media. Influence stocks further weakened by the effective manipulation of communication links, info flows & behavioural economics in multi-platform attacks by external systems.

Influence diminished concurrently with reduced authority and a loss of control resulting from misguided Loss of reactions to R1 feedback and the targeting of Clinton’s credibility (B2). The loss of control (a CC) contributed to a downward spiral of influence, and which provided momentum to negative reinforcing Authority & R2 feedback and escalated the vicious cycle. The loss of control and authority resulted in reputational damage, Control undermined Clinton’s legitimacy, and reinforced false perceptions; which eroded balancing feedback relating to credibility and sense of community (B2 & B3 loops).

Clinton’s balancing feedback deteriorated further and sunk influence stocks due to an internal drift to low performance, which resulted from a loss of system resilience caused by the R1 & R2 feedback loops. This drift to low performance resulted from vicious reinforcing feedback including: lost support & resources, Drift to Low R3 ineffective communications, competing systems’ goals, sub-system policy resistance, accidental Performance adversaries, and a decline in system resilience. Clinton’s influence was undermined by a loss of credibility and control resulting from a number of internal and external system impacts and delays in the associated feedback. The R3 loop impacted the ‘sense of purpose’ balancing feedback (B1), due to a lowering of morale, undermining of the ideology (a CC), and increased disillusionment & division.

Table 13: Clinton Reinforcing Feedback Loops

Counter-Influence (R1) As described in Table 13, the counter-influence reinforcing feedback was relentless, aggressive, multi-pronged, and created a vicious escalating cycle, which eroded Clinton’s balancing feedback, as well as feeding the other reinforcing feedback loops. The intent driving the inundation of counter-influence feedback was the manipulation and control of target audience perceptions, which undermined Clinton’s legitimacy, reinforced opposing narratives and dominated the election IE. The R1 loop was further enabled by the ‘propaganda feedback loop’, described by Benkler et al. (2018:33) as a dynamic where: “media outlets, political elites, and

131 | P a g e 5249168 Cassandra Brooker Masters of Research public figures form and break connections based on statements that progressively lower the costs of telling lies consistent with a shared political narrative, and increase the costs of resisting that narrative in the name of truth” (Benkler et al. 2018:33).

The R1 loop comprised various forms of effective influencing tactics and the exploitation of behavioural and cognitive biases by isolated external systems, as described in Chapter 2, in order to undermine Clinton’s influence stock. As the campaign progressed however, individual systems’ counter-influencing efforts became increasingly complementary and subsequently reinforced each other’s outputs. For example: the IRA took elements of a story developed by Fox News, created a new layer of fabrication, repeated variations on the story, which was then further warped and rapidly disseminated by right-wing agitators on multiple platforms. RW news outlets subsequently reported these lies as truths, they were picked up by MSM not wanting to miss the scoop, and amplified by Trump (Table 12) – effectively reaching millions of people (Benkler et al. 2018:96-97). This propaganda feedback loop demonstrates the power of inundation, repetition, emotional/social contagions, and personality bias confirmations; as well as demonstrating behaviours of people preferring to access entertaining content, not requiring ‘System 2’ critical thinking. Audiences encountered multiple versions of the same story, propagated over months, through their favoured media sources; to the point that both recall and credibility were enhanced, fact-checkers were overwhelmed, and a ‘majority illusion’ was created (Benkler et al. 2018:96- 97,280-281; Watts 2018:90).

The combined counter-influence activities contributing to R1 resulted in American audiences relying on biases, believing disinformation, doubting facts, turning on each other, and becoming overwhelmed by the tsunami of conflicting information (Watts 2018:16,19; Clapper 2018:333,355). The fusion of separate counter-influence campaigns against Clinton was further demonstrated by the identical themes underlying conspiracy theories espoused by external systems seemingly operating in parallel. Both Russia and the Trump campaign promoted themes that Clinton was corrupt, was physically and mentally unwell, and had ties to Islamic extremism – which specifically undermined all three of Clinton’s balancing feedback loops and reinforced the R2 and R3 loops (Clapper 2018:349; Silver 2018; Watts 2018:161-162,176).

The extent of online counter-influence feedback impacting on Clinton’s system was enabled by hard and soft communications links, and described by Singer and Brooking (2018:130,142-143) as an ecosystem so vast, it dwarfed all previous hoaxes and disinformation operations in election history, and demonstrated an unrivalled level of algorithmic manipulation and virulent political attack. The effectiveness of this online feedback contributing to the R1 loop was further reinforced 132 | P a g e 5249168 Cassandra Brooker Masters of Research by the “most damaging technique” of hacking, manipulating, and leaking DNC and Clinton emails, in cleverly timed releases, in order to undermine credibility, damage reputations, dominate headlines, and attack the legitimacy of the election process (Canadian SIS 2018:46; Wardle 2019; Clinton 2008:236). Clinton (2018:349) described the relentless WikiLeaks email dumps as being like “Chinese water torture”, where no single day was bad, but the effect added up over time, contributed to an acceleration of fake news, and they “could never get past it”. This meant that Clinton’s balancing feedback was not strong enough to counter the vicious escalating R1 loop (Figure 21). Combined with counter-influence strategies persuading audiences to vote against their economic interests by appealing to cultural issues, such as race, “gays, guns, and God”, as Clinton put it (2018:274), this led to a loss of authority and control and the second reinforcing feedback loop.

133 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 21: Clinton Reinforcing Feedback Loop (R1): Counter-Influence

134 | P a g e 5249168 Cassandra Brooker Masters of Research Loss of Control and Authority (R2) As the R1 loop escalated and undermined Clinton’s influence stock, misguided system reactions by the Clinton campaign combined with those iWar effects to trigger momentum in the ‘Loss of Control’ negative reinforcing feedback loop (R2). The loss of control and authority, which were critical capabilities to the system’s COG (Fig 19), resulted in reputational damage, undermined Clinton’s legitimacy, and reinforced false perceptions; which eroded balancing feedback relating to credibility and sense of community (B2 & B3 loops).

Both internal and external systems contributed to the reinforcing feedback of the R2 loop, which, like R1, also escalated in a downward spiral over time – albeit not as rapidly. The R2 feedback contributing to the loss of control was achieved by external system activities undermining perceptions of Clinton, exacerbating social divisions, controlling the narrative, and eroding communications links caused by the asymmetric media ecosystem. These factors combined with increased partisan political discourse, a resultant loss of trust and loyalty among supporters, a disunity among internal sub-systems, and damage to ideological legitimacy and associated relationships.

The more hits the R2 loop succeeded in landing against Clinton’s reputation and political bipartisanship more generally, the more damage was done to overall system resilience. This occurred in Clinton’s case by undermining sub-system unity, influence stocks, and balancing feedback loops; and to democracy as a whole, through the weakening of institutions, and creation of disparate values and societal fissures – the relentless momentum of which, was very difficult to counter or reverse (Rosenberger 2019; Uren 2018).

The most aggressive, effective contribution to the R2 loop was the multi-axis attacks on Clinton’s credibility and reputation using psychological manipulation and influencing techniques discussed in Chapter 2. Such tactics included: associating Trump with Jesus and Clinton with Satan to target conservative audiences, manipulating grievances and emotional contagions of Sanders’ supporters and black communities, and exploiting fears of gun control, job losses, immigration, corruption, the Washington ‘swamp’, and a loathing of Muslims to target veterans, racists, misogynists, pro-gun groups, and low socio-economic audiences to vote against Clinton (Reynolds 2018; Benkler et al. 2018:132,270; Clinton 2018:50,115,277,362; Silver 2018; Watts 2018:161-162; Clapper 2018:288). Effective content was focused on polarising both Clinton and various hot-button policies by fear-mongering using targeted messaging, oversimplified narratives, conspiratorial explanation, and demonising others (Wardle 2019). Quoting lines from the leaked emails, out of context, reinforced this paranoid logic and seemingly verified false 135 | P a g e 5249168 Cassandra Brooker Masters of Research conspiracy stories attacking Clinton’s credibility and integrity (Benkler et al. 2018:137). Additionally, the inundation of salacious content attacking Clinton’s reputation skewed the media coverage and further reinforced false perceptions, as residual associations remained in voters’ minds, thereby overshadowing any truthful counter-messaging or policy discussion attempted by Clinton (Benkler et al. 2018:190-191; Clinton 2018:45,406).

The R2 reinforcing feedback contributing to Clinton’s loss of control and authority was facilitated by the same communications links enabling the escalation of R1. Benkler et al. (2018:201) explain stories attacking Clinton were repeated and linked through an interconnected network of sites, those associations were then reinforced through inundation and disseminated through multiple, diverse sources, thereby increasing the stories’ credibility. This further demonstrated a loss of control by Clinton’s system over those hard and soft communication links, as well as over its messaging, ideology, media strategy and narrative, which were all critical for maintaining influence stocks and protecting the COG.

From an internal system perspective, actions and reactions by the Clinton campaign also contributed to the R2 loop. The campaign media and leadership sub-systems were not as shrewd as their adversaries in controlling the narrative, having anticipatory responses, or in targeting influence. Where the Democrats provided factual, analytical explanations surrounding its campaign plan and policy positions as was required for System 2 thinking and received well in previous elections; the Republicans shifted the focus, manipulating System 1 cognitive behaviours to stir up anger, resentment, despair, and to divide the electorate. The DNC’s carefully considered, detailed messaging was ineffective, boring and unsatisfying to voters with existing cognitive biases and high emotions (Clinton 2018:275). The politics of cultural identity and resentment overwhelmed any evidence, reason and personal experience presented to the voting public and control over the messaging was lost to the Republicans (Clinton 2018:50,272).

As a result of these contrasting approaches, the Clinton campaign was viewed as “overly- educated”, liberally elite, wealthy, un-relatable, and aloof, leading to a loss of control over its image portrayal and unable to effectively counter the pessimism and social contagions espoused by Trump and the IRA (Clinton 2018:277-278; Clapper 2018:288,321; Silver 2018). Additionally, the internal sub-systems of Clinton’s campaign failed to realise the extent of rage and resentment building across America resulting from R1 and R2 feedback, therefore undermining the system’s resilience in its ability to adapt or respond appropriately. They did not assess the attacks on Clinton as credible or worth responding to, given Clinton’s reputation, record of independence, and experience in politics. They also underestimated the levels of sexism and misogyny across 136 | P a g e 5249168 Cassandra Brooker Masters of Research the electorate further undermining Clinton’s authority (Clinton 2018:10,45,114-115,311). This inaction, combined with delayed feedback from a lack of situational awareness and inaccurate polling, further reinforced the loss of control (R2) and system influence stocks (Silver 2017; Clinton 2018:399-401).

The leadership sub-system inadvertently contributed to the R2 loop considerably. Not only because Clinton herself was as unpopular as Trump and the electorate was potentially suffering “Clinton fatigue”, but also because their messaging lacked the authenticity of Trump’s unpredictable, hyperbole (Clinton 2018:301). Trump’s captivating Twitter feed, where people felt he was talking to them, was in stark contrast to Clinton’s, whose tweets were sometimes crafted by a team of 11 staffers (Clinton 2018:301; Silver 2018; Singer & Brooking 2018:168- 169).

Notwithstanding the specific error of the “deplorables” comment, as the R1 and R2 loops escalated, the leadership sub-system increasingly employed misguided reactions. Fact-checking was powerless in countering the rapid inundation of disinformation, denials and explanations strengthened scandals and allowed them to remain front page news and foremost in voters’ minds, and accidental adversaries, such as Sanders and Comey, reinforced adversary attacks against Clinton’s image (Clinton 2018:301; Silver 2018). As a result, the Clinton team was unable to regain the initiative and strengthen balancing feedback, thereby losing control of the narrative, the presidential candidate’s reputation, and the loyalty of her supporters; which in turn, further undermined the system’s influence stocks and the leadership’s interconnections with the other three internal sub-systems (Clinton 2018:399-400). The R2 reinforcing loop is depicted in Figure 22.

137 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 22: Clinton Reinforcing Feedback Loop (R2): Loss of Control

138 | P a g e 5249168 Cassandra Brooker Masters of Research Drift to Low Performance (R3) The effects of R1 and R2 feedback on Clinton’s campaign system as a whole, as well as on the behaviours and linkages of the individual sub-systems, resulted in a drift to low performance archetype, resulting in a downward spiral feeding a third reinforcing loop (R3). Like ISIS, this descending performance drift was a gradual process, where exponential growth of trivial problems worsened over time, and this slow growth failed to raise alarms until the system experienced the effects in real time (Stroh 2015:48; Meadows 2008:123; Clinton 2018:363). But unlike ISIS, Clinton’s campaign was not as resilient in its reactions and ability to counter the negative feedback, therefore its balancing feedback, influence stocks, and overall performance deteriorated more rapidly.

Clinton’s system influence had already deteriorated from counter-influence attacks (R1) and a loss of community cohesion (B3), which caused a loss of credibility (B2) and control (R2). The R3 loop added to the vicious cycle of reinforcing feedback by predominantly impacting on the ‘sense of purpose’ balancing loop (B1), due to a lowering of morale, undermining of the ideology (a CC), increased disillusionment, and internal division. This drift to low performance effected all four sub-systems as a result of lost support, donations and resources; ineffective communications; competing systems’ goals; sub-system policy resistance; accidental adversaries; and a decline in system resilience.

Where Clinton’s leadership sub-system contributed to the R2 reinforcing feedback, so too did its behaviours and actions reinforce the R3 drift to low performance. The hierarchy was too large, unwieldy, and slow to react. Conducting the campaign in the same old, traditional, conventional manner, without regard for modern communications technology and not anticipating iWar attacks, resulted in a malfunctioning sub-system failing to meet its goals. Clinton was increasingly dislocated from the messaging, it was no longer her voice or vision being disseminated, which undermined authenticity and the CC of ideology. The whole system itself became increasingly isolated from the ground truth. It lacked awareness of the actual situation across the electorate as it relied on skewed polling, delayed feedback, and was unaware of counter-influence success being achieved by external systems. The poor performance and sub-optimisation resulting from this bounded rationality, further weakened the campaign’s balancing feedback, led to misguided decision-making, competing goals and quick fixes, all of which created unwanted behaviours and drove results down further (Meadows 2008:15,84-85,106,122).

Meadows’ (2008:184) description of modern culture eroding the “goal of morality” – by magnifying bad behaviours and affirming them as typical, whilst ignoring examples of “human 139 | P a g e 5249168 Cassandra Brooker Masters of Research goodness”, therefore lowering expectations and narrowing the gap between desired and actual behaviours – perfectly illustrates the process contributing to Clinton system’s drift to low performance. Meadows (2008:184) explains this erosion leads to cynical public discourse, amoral or immoral public leaders not being held to account, idealism is ridiculed, and statements of moral belief are suspected. All of which were experienced during the 2016 election as morality, transparency, and ethical behaviours were undermined, thereby contributing to the R3 loop and undermining system influence (Clinton 2018:263,265,272; Benkler et al. 2018:81,97-98).

‘Accidental Adversaries’ is where independently chosen, misguided solutions by various systems or sub-systems inadvertently obstruct each other’s performance in a vicious cycle (Stroh 2015:61- 62). This phenomenon was evident during the 2016 Election and contributed to the R3 loop. The Clinton campaign, the media, the Obama administration, and Sanders’ supporters were all trying to optimise their own performance, independent of the other systems, therefore resulting in competing goals; contradictory, undisciplined messaging; and quick fixes, which ultimately undermined them all (Clinton 2018:66,74; Stroh 2015:31).

The R3 loop was further reinforced by system overload resulting from high aspirations for Clinton as the inevitable front runner, incorrect resource allocation, escalating goals and expectations, and difficulties in measuring progress, which all impacted on system resilience and sub-system performance (Stroh 2015:147; Silver 2017). Specifically, the drift to low performance resulted in Clinton’s approval numbers dropping, increased distrust, messages being blocked or overwhelmed, greater criticism, analytical blind spots, and more mistakes (Clinton 2018:72- 73,292; Silver 2017). This downward spiral of Clinton’s campaign system could be defined as a ‘sticky trajectory’, in that it failed to achieve its goals despite having more resources, funds, experience, and credibility than Trump (Byrne & Uprichard 2012:11). The R3 loop comprising all of these factors is depicted in Figure 23.

The balancing feedback loops that should have kept the system state at an acceptable level were overwhelmed by the downwards cycle created by the reinforcing feedback loops. When perceived system performance slips, the desired state is allowed to decline, and its associated goals are eroded (Meadows 2008:122). The next section describes the balancing feedback loops contributing to Clinton’s influence stocks at their height, before the reinforcing feedback took effect.

140 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 23: Clinton Reinforcing Feedback Loop (R3): Drift to Low Performance

141 | P a g e 5249168 Cassandra Brooker Masters of Research Balancing feedback loops As described in Chapter 4, a system will eventually collapse should the vicious downward spiral resulting from reinforcing feedback loops be allowed to escalate unabated, without any effective balancing feedback to slow the flow (Meadows 2018:125-126; Stroh 2015:50). Whilst Clinton’s campaign was increasingly dominated by a vicious cycle of strong reinforcing feedback loops creating a rapid downturn in influence stock, balancing feedback ensured the system survived and maintained base-level sources and influence stock. The three key balancing feedback loops identified in this case system analysis, and discussed in this section, are summarised in Table 14.

Balancing Feedback Key Description Loop Influence over target audiences grows, motivation of members is reinforced, and ideological narratives spread when there is justified belief in the cause. Achieved when the purpose is Sense of perceived as authentic, legitimate, and just; and when pride is derived from participation. B1 Clinton was marketed as an ethical, democratic champion – fighting for the rights of Purpose minorities, women, and socio-economic equality. B1 also benefited from Obama’s legacy, strong social policies, and resonant, authentic messaging - a stark contrast to hateful, xenophobic Republican messages. Exploited Clinton’s reputation to reinforce credibility and increase influence. Key aspects contributing to B2 were: Clinton’s political experience, transparency, ethical policies, authentic Credibility B2 messaging, knowledgeable / factual claims, willing to admit mistakes, proven performance, and residual political currency from Bill Clinton’s legacy and her time as secretary-of-state. Influence gained through a sense of community. Clinton portrayed as a mentor/leader for Sense of women, fighter for minority rights, champion of diversity and democratic/left social policies. B3 Presented as the polar opposite of racist, sexist RW Media & Trump. System source included Community diverse support base and strong relationships. Policies addressed wide range of social solutions, campaign knew, researched and understood target audiences and issues.

Table 14: Clinton Campaign Balancing Feedback Loops

Sense of Purpose (B1) Like ISIS’ reinforcing feedback loop that provided a ‘sense of purpose’, Clinton’s B1 loop also provided a competitive system of meaning by targeting audiences with resonant messaging, a strong narrative, and authentic appeals to both pragmatic and perceptual factors, which contrasted starkly with the pessimistic Republican ideology. The B1 loop not only enhanced influence and countered negative reinforcing feedback for Clinton’s system, but also projected balancing feedback against other external systems, such as the Trump campaign and RW media. As described in Table 14, a justified belief in the cause, garnered from the sense of purpose and optimism, reinforced the source support and influence amongst the Democrats’ own ‘tribe’ and resonated with other target audiences still undecided on how to vote.

142 | P a g e 5249168 Cassandra Brooker Masters of Research

Clinton’s campaign marketed itself as an ethical, righteous, democratic champion – fighting for the rights of minorities, women, diverse social groups, and socio-economic equality. The B1 loop also benefited from espousing policies that would continue Obama’s legacy, and protect democratic values and the attractiveness of America on the world stage, which had proven to be the most successful narrative in previous elections (Rosenberger 2019). The media and leadership sub-systems were also able to reinforce the sense of purpose balancing feedback through the exploitation of centralist and left wing cognitive biases and social contagions of fear, disgust and disbelief regarding Trump’s abrasive personality, outrageous rule-breaking tactics, and extreme, polarising RW messaging (Clapper 2018:328-329; Wu 2016:16)

Clinton explained (2018:63) that unlike the Trump campaign, she wanted her policies to be bold, innovative, industrious and responsive to people’s real-life needs, and that her associated messaging have substance and be based on community feedback. This connection of ideology with sub-system actions in the B1 loop was the most effective method of influencing, i.e. propaganda by deed; it motivated people to co-operate as they could see the big picture and their role in it.

The B1 loop allowed both audiences and the system’s elements to appreciate how their individual success depended on the success of all stakeholders and their effective collaboration in overcoming the polarisation of US politics (Stroh 2015:33). Figure 24 shows the B1 loop and system contributions to enhancing the sense of purpose and subsequently the influence stocks.

143 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 24: Clinton Balancing Feedback Loop (B1): Sense of Purpose 144 | P a g e 5249168 Cassandra Brooker Masters of Research Credibility (B2) The importance of having and maintaining system credibility in ensuring legitimacy, control, authenticity, and effective influence was discussed in Chapter 2. The B2 loop reinforcing Clinton’s campaign credibility was fed by: her established reputation and political experience, the media sub-system’s authentic messaging and reputation management, the leadership sub-system’s researched, ethical, targeted policies based on lessons learnt and community engagement; the membership sub-system’s belief, support and dissemination of the ideology at the grassroots level as a social contagion; and the resource sub-system’s funding of targeted activities that backed up the narrative with deeds.

External systems also contributed to the B2 loop with the positive legacies of Obama and Bill Clinton’s administrations, the supportive unbiased media coverage from the central-left media ecosystem, and from the initial lack of credibility surrounding Trump as a candidate and the Republicans’ recent lack of political experience.

Clinton’s sub-systems applied lessons learnt from the Obama campaign regarding advanced data analytics to further strengthen the B2 loop and enhance influence stocks (Clinton 2018:64-65). Specifically, the sub-systems collaborated in developing holistic system B2 feedback in their determination to obtain the best data, deploy more field organisers, create the biggest fund-raising network, and establish the deepest political relationships – with no in-fighting or rivalries (Clinton 2018:70). Clinton individually contributed to her credibility by seizing the initiative through early policy development and inspired messaging in order to mobilise support and inform target audiences on her vision, what they could expect, and how her election success would affect their lives (Clinton 2018:224).

The B2 loop also complemented the other balancing feedback loops in strengthening both internal and external system relationships. This was achieved by reinforcing the ‘in-group’ sense of community, and the righteous, sense of purpose of the campaign; and through building credibility via authentic influencing activities.

The B2 loop remained strong throughout the election campaign despite the escalation of reinforcing feedback working against it. Benkler et al. (2018:257) partially explain this phenomenon, citing the example of why the DNC leaks failed to splinter the Democratic Party as intended. They contend the leaks failed because the national media ecosystem did not follow the same practices available for propagandists to harness when their content was aimed at the right, that is, “they did not devolve into polarising hatred-inducing stories” (Benkler et al. 2018:257). 145 | P a g e 5249168 Cassandra Brooker Masters of Research The B2 loop also increased system resiliency due to a large majority of voters being turned off by fake news, lies, hate speech, propaganda, RW conspiracy theories, and abuse of internet commons. Additionally, these same moderate voters were critical thinkers, more highly-educated, and received their news from a diverse range of sources – external to the RW media echo chambers.

That is not to say the R1 and R2 loops did not have an effect on the B2 loop. Clinton’s campaign largely ignored fake conspiracy scandals attacking her credibility or reputation, so as not to give the stories validity in the MSM. However, this presented a paradoxical conundrum for the system. On one hand, their silence or omissions were viewed as suspicious, aloof, and dislocated their control of the narrative; whilst on the other, when they did try to directly refute the lies, clarify the facts, or own up to mistakes; it often backfired and reinforced the vicious cycle (Clinton 2018:46). This was starkly demonstrated when Bill Clinton announced the Clinton Foundation would stop taking foreign donations if Hillary Clinton was elected president. This resulted in even more scrutiny and coverage by a broad range of global media, outside of the RW sphere (Benkler et al. 2018:205). Additionally, the external system of MSM failed in its journalistic objectivity and lost its balanced perspective towards the end of the campaign, with the associated coverage impacting on Clinton’s B2 loop and ultimately affecting the outcome of the election (Clinton 2018:319,321; Benkler et al. 2018:74). The B2 ‘credibility’ balancing feedback loop is depicted at Figure 25.

146 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 25: Clinton Balancing Feedback Loop (B2): Credibility

147 | P a g e 5249168 Cassandra Brooker Masters of Research Sense of Community (B3) Complementing the B1 and B2 feedback loops was an inherent ‘sense of community’, carefully managed by the system, and uniquely achievable by Clinton playing on her strengths – as a female Democratic candidate campaigning on a platform of social justice, minority rights and championing equality and diversity. As discussed, Clinton’s supporters were also more highly educated and engaged with her narrative, meaning her social policies and inclusive narratives interested and resonated with audiences, while reinforcing the sense of community and an ‘us versus them’ confirmation bias against the ‘hateful, xenophobic, nationalist, and misogynist’ right wing.

Whilst the DNC made mistakes with Obama’s campaign and undermined grassroots initiatives (Benkler et al. 2018:344), Clinton’s system approach to engaging with and receiving feedback from communities, organisations, minorities, and social groups to adjust policy platforms and fund activities, reinforced the B3 loop. These activities combined with an exploitation of emotional and social contagions related to the increasing polarisation of US politics, resulting in increased buy-in, participation and investment of target audiences along partisan lines, and created a sense of belonging to the righteous, inclusive ‘tribe’. Benkler et al. (2018:306) explain that social identity is linked to party affiliation and when people choose sides, they become emotionally invested in promoting their side and opposing the other side. These group bonds determine opinions on issues, the interpretation of political events, and exacerbates polarisation, biases and anger.

Whilst this phenomenon was wholly apparent on the RW of politics with Republican supporters reacting with increasing levels of anger and outrage, Clinton’s B3 feedback and interconnections also benefited from this partisanship. Just as ISIS did in creating their sense of community, Clinton’s system created messaging around both perceptual and pragmatic factors. Perceptual aspects included the aforementioned ‘in-group’ identity-choice appeals and crisis-solution constructs designed to shape audience understanding of potential election outcomes; while appeals to pragmatic factors included: the security and stability of the nation and the socio- economic quality of life of the voters. Rational choice appeals were also employed using themes regarding Trump’s lack of credibility, hypocrisy, and evil intent, while framing Clinton as the champion of minorities, women and equality. The B3 balancing feedback loop is depicted at Figure 26.

148 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 26: Clinton Balancing Feedback Loop (B3): Sense of Community

149 | P a g e 5249168 Cassandra Brooker Masters of Research Delayed Feedback As depicted in the reinforcing feedback loop diagrams, the Clinton campaign suffered from delays in feedback when its influence stock started to decline, which undermined its ability to react effectively. This delayed feedback also reduced Clinton’s system resilience at a much faster rate when compared to ISIS, due to the nature of the campaign being more open and fragile on the starting scale of system characterisation (Fig 3). As Meadows (2008:39) explains, the information delivered by a feedback loop can only affect future behaviour; it cannot correct behaviour that drove the current feedback in real time. This means there will always be delays in responding by decision-makers. If a system is not anticipatory in its resilience and lacks situational awareness, as Clinton’s system experienced due to extensive reinforcing feedback clouding reality, then further reactive, misguided decisions and actions are implemented, further contributing to the downwards spiral and loss of influence.

A prime example of delayed feedback was that nearly every private and public poll over the two years of the election campaign showed Clinton to be way ahead, which led the system to continue behaving as usual, employing the same mental models, unable to break through the noise or realise the reality until it was too late, and when the gradual drift to low performance had gained momentum (Clinton 2018:399-400; Meadows 2008:122-123).

The loss of control (a CC), particularly of the narrative, also contributed to delays in receiving associated feedback about the degradation of influence stocks. As Singer and Brooking (2018:159) explain, the big losers in the ‘control of the narrative’ battle are those systems that are too big, slow or hesitant to weave such stories, adding that “these are not the kinds of battles that a plodding, uninventive bureaucracy can win” (Ibid:159). This further demonstrates the lack of anticipatory resilience of Clinton’s system in not foreshadowing the potential effects of the iWar campaign waged against it, as well as the game-changing nature of social media and communications technology (Benkler et al. 2018:273).

To counter the effects of delayed feedback, an analysis of historical trends is recommended, rather than focusing on current events or fluctuating flows, in order to properly understand system dynamics and reasons for behaviours (Meadows 2008:90). As discussed, the DNC ignored important lessons from past elections and Clinton’s system lacked situational awareness and resilience to anticipate and correct these behaviours. Clinton herself admitted (2018:76): “I didn’t realise how quickly the ground was shifting under our feet. I was running a traditional presidential campaign, playing by the rules, with carefully thought out policies and painstakingly built coalitions. Trump was running a reality TV show that expertly and relentlessly stoked Americans’ 150 | P a g e 5249168 Cassandra Brooker Masters of Research anger and resentment” (Ibid:76). Additionally, the corrective processes employed by Clinton’s system in response to the delayed feedback failed due to an employment of quick fixes, a failure to appreciate the time required to effect change, and a weakening of the balancing feedback loops once the R3 loop gained momentum (Stroh 2015:51).

Conclusion Figure 27 concludes this chapter by depicting the holistic system diagram for Clinton, an ineffective iWar case study, which includes all of the elements, reinforcing and balancing feedback loops identified in Chapter 5. As with ISIS in Chapter 4, four key themes become apparent as contributing to Clinton’s campaign system’s ineffectiveness at maintaining its influence stocks as follows:

1. The system characteristics of Clinton’s campaign being an open system with lower levels of resilience and therefore, a higher propensity towards increasing fragility and a lack of responsiveness (Fig 3). 2. The inefficiency of system flows both via feedback loops and through hard and soft interconnections between key elements, which led to delays in feedback, misguided responses, a lack of situational awareness and an asymmetric increase of reinforcing feedback. 3. Increasingly losing control, specifically of the narrative, the campaign’s credibility and Clinton’s image. Control is both a critical capability and a vital balancing element for increasing influence stocks (Fig 19). 4. Lacking a centrality of focus, where unlike ISIS, Clinton’s campaign influence activities were not the main priority driving all system outputs. The stove-piped, slow, traditional campaign media strategies contrasted sharply with the agile, adaptive iWar tactics of Clinton’s adversaries. This lack of focus on the system’s stock and purpose contributed to the asymmetric escalation of all three reinforcing feedback loops.

Part Two of this thesis concludes with a comparative analysis of the findings arising from the modelling of both ISIS and Clinton’s case systems in the next chapter. Chapter 6 will also present the findings from the narrative analysis of the two systems as it relates to influence effectiveness.

151 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 27: Holistic Clinton Campaign System Diagram

152 | P a g e 5249168 Cassandra Brooker Masters of Research Chapter 6 – Comparative Analytical Findings

There are but two powers in the world, the sword and the mind. In the long run, the sword will always be beaten by the mind.

Napoleon I, Conversation 1808

Introduction Part Two of this thesis concludes with the findings of a comparative analysis of the two case systems’ models analysed in Chapters 4 and 5, to better understand the configurational, behavioural and causal factors contributing to their respective levels of influence. This chapter comprises two parts, the first analyses the findings from comparing the two systems, which separate into four key factors contributing to rises and declines in influence stock. The second part follows with a discussion of the findings specifically related to influence effectiveness. These findings provide key themes and a comprehensive framework for discussion in Chapter 7. This discussion will cover the research gaps relating to Australia’s influence effectiveness and counter- influence strategies, how complex social systems adapt their behaviours based on feedback loops, and why understanding psychological cognitive theory is central to analysing and leveraging social system behaviours and mental models.

Part One: Comparative Systems Analysis Findings Chapter 3 provided justification of the two case studies selected, which was due to the number of similarities between these complex social systems and because both provided the ability to isolate their influence stock as a bounded ‘systems thinking’ analytical device. The analysis and modelling of the two systems in Chapters 4 and 5 respectively, determined the shared characteristics of both systems focused on maintaining their influence stock, such as: their purpose, centre of gravity (COG), balancing and reinforcing feedback, key elements, sub-systems and interconnections formed through hard and soft communication links. These parallels provide a sound base to then contrast the effectiveness of each system in maintaining its influence stock, thus enabling a straightforward comparative analysis to determine the findings.

The conclusions of the two case analyses in Chapters 4 and 5 identified the same four themes contributing to both systems’ level of effectiveness in maintaining their influence stocks:  System Characteristics – open or closed, level of resilience versus fragility;  System Information Flows – feedback loops, interconnections, delayed feedback;  Control – as a critical capability and balancing element; and  Centrality of Focus – regarding the level of importance placed on influence activities. 153 | P a g e 5249168 Cassandra Brooker Masters of Research These four themes will inform the comparative analysis between the two systems. The findings of which provides the framework for further discussion in Chapter 7 and in answering the research questions.

System Characteristics The system characteristics were the first key theme, to become apparent from the modelling, as contributing to both systems’ effectiveness at maintaining their influence stocks (as depicted in Figure 3). Comparatively, ISIS was a relatively closed system with a high level of anticipatory resilience and self-organisation; while Clinton’s campaign was an open system with lower levels of resilience and therefore, a higher propensity towards increasing fragility and reduced responsiveness.

These starting points in terms of system characteristics determined the future trends in escalating feedback that each case experienced throughout the period, in either a vicious or virtuous cycle, as well as relating to the associated behaviours they exhibited in response. The anticipatory characteristic of ISIS’ system resulting from its R3 success to the successful loop produced higher levels of autonomy, representation, and influence; which provided ISIS the means to compete effectively at the strategic level and dominate in the iWar. Whereas Clinton’s open, public election campaign was a less diverse and less robust system, making it more vulnerable to external shocks, counter-influence attacks, and suffering a more extreme loss of control.

ISIS was a diverse, highly-functional, social system that demonstrated adaptation, was self- organised, and more robust and resilient to stressors. However, like Clinton’s campaign, ISIS did eventually suffer a drift to low performance – as balancing feedback predictably countered the escalating reinforcing loops. However, because Clinton’s system started with a lower level of resilience, its drift to low performance was more pronounced, albeit still gradual. Additionally, external systems and other stressors enjoyed greater access, leverage, and their iWar attacks had a more significant effect against Clinton’s open system. Clinton’s campaign was less anticipatory in its reactions and in its ability to counter the negative feedback; therefore, its balancing feedback, influence stocks, and overall performance deteriorated more rapidly than observed with ISIS.

Both case systems experienced an asymmetry of balancing and reinforcing feedback, with their respective system characteristics significantly contributing to either the vicious or virtuous escalation of those cycles, which in turn further effected the levels of overall system resilience.

154 | P a g e 5249168 Cassandra Brooker Masters of Research Another characteristic impacting on resilience levels was the difference between remaining constant over time, conducting business-as-usual productivity and lacking diversity like Clinton’s campaign – or being adaptable, shifting focus and evolving in order to maintain optimal system performance, as ISIS demonstrated in counterbalancing negative feedback.

As Meadows (2008:159) explains, the ability to self-organise is the strongest form of system resilience, and ISIS was able to evolve its system structure and adapt behaviours to survive attacks against it. Whereas Clinton’s campaign lacked variability, diversity, innovation and experimentation, particularly when compared with ISIS or even Clinton’s adversaries’ tactics in the iWar. This stagnation by Clinton’s campaign ultimately blocked system learning and impacted on information flows and overall resilience. Allowing a system to experiment, adapt, evolve and self-organise means relinquishing some control, which despite the benefits to system resilience in the long-run, was not even considered in Clinton’s conventional campaign strategy (Meadows 2008:160). Such risk-averse, traditional, linear approaches lead systems, such as democratic Western nations, to become stunted, static and unable to anticipate or respond appropriately to unconventional adversary actions in the iWar. As Lieber and Reiley (2016:50) explain, there is no realistic way to traditionally hunt an asymmetric adversary that is constantly changing. ISIS relied on cutting edge technology, exploited inventive processes, and correctly expected that its adversaries, being traditional in structure, would be slow in adapting to the threat.

The effectiveness of these factors in the iWar will be explored in part two of this chapter. Also contributing to enhancing system resilience is the efficient flow of information within and between key elements to enable appropriate, timely responses, superior decision-making, and heightened situational awareness.

System Flows A system with high levels of resilience and an ability to self-evolve also enjoys effective flows of information in the form of feedback loops via interconnections formed by hard and soft communications links. As the system modelling demonstrated, the efficacy of ISIS’ system flows both via its feedback loops and through the strength of its interconnections contributed to the group’s effectiveness in maintaining influence stocks and their virtuous ‘success to the successful’ R3 reinforcing cycle (Fig 13). Whereas, the inefficiency of system flows for the Clinton campaign led to delays in feedback, misguided responses, a lack of situational awareness, and an asymmetric increase of negative reinforcing feedback. These factors combined to create the vicious reinforcing cycle, which the balancing feedback loops were too weak to counter, thereby impacting on the campaign’s influence stock in a downward spiral. 155 | P a g e 5249168 Cassandra Brooker Masters of Research

When comparing the two case systems and analysing information flows contributing to their influence stock, the similarities in the types of feedback loops affecting both systems are starkly apparent, with the major differences being the speed of flow and the contrasting effects of those flows in either a vicious or virtuous cycle. For example, both systems required feedback flows relating to a ‘sense of community’ (Figs 11 & 26) and a ‘sense of purpose’ (Figs 12 & 24) to enhance their influence stock, whilst a ‘loss of control’ (Figs 16 & 22) and ‘counter-influence’ activities (Figs 15 & 21) by their respective adversaries undermined this positive feedback.

Changes in stocks sets the pace of a system’s dynamics and the speed of evolution, meaning the speed of the associated information flows about the stock determines the appropriateness of reactions, counterbalancing actions, and overall system resilience. While both systems suffered delays in feedback, Clinton’s campaign was less anticipatory in its responses, which further slowed the pace of the positive feedback flows contributing to stock levels, therefore creating further delayed feedback relating to those often misguided responses. Additionally, changes to interconnections may dramatically alter the system or system’s behaviour (Meadows 2008:16). This was observed where the negative reinforcing feedback attacking the Clinton campaign’s relationships and soft communications links undermined the flows of positive balancing feedback and source support.

The ‘sense of purpose’ feedback loop in both case systems was integral to maintaining influence stocks, as it provided the paradigm from which each system derives its source support, goals, structure, and parameters. As Meadows (2008:163) explains, a system’s paradigm provides a shared social agreement about the nature of reality, and produces system goals, information flows, feedbacks, and stocks. Meadows (2008:164) adds that those who have managed to intervene in systems at the paradigm level have hit a leverage point that totally transforms that system. Leverage points associated with the paradigm also directly targets both case systems’ identified critical capability of ‘ideology’. This leverage point for both case systems was the ‘counter- influence’ feedback loop attacking their paradigms related to the sense of purpose and belief in the cause, which in turn contributed to other negative feedback loops impacting on their control, credibility, and performance.

Analysing the system flows for both cases also provided insight into their performance over time relating to growth, decline, oscillation, or evolution. Long-term behaviour provides clues to the underlying system structure, which is key to understanding what is happening with its interlocking stocks, flows, and feedback loops (Meadows 2008:88-89). De-composing both systems into key 156 | P a g e 5249168 Cassandra Brooker Masters of Research elements and then re-composing them around the main feedback loop flows provided a holistic understanding of the patterns, relationships, behaviours and leverage points contributing to their respective stock changes, which are often not obvious or intuitive (Meadows 2008:146).

Manipulation of information flows can provide important leverage points for intervening in systems, and missing or inefficient information flows are the most common causes of system malfunction (Meadows 2008:157). Delayed feedback, a lack of situational awareness, and misunderstanding actions and reactions’ impacts on feedback loops and stocks, were all factors that undermined both case systems’ performance to varying degrees. As Meadows (2008:157) explains, there is a systematic tendency by human beings to avoid accountability for their own decisions, which is why there are so many missing feedback loops and a regular lack of holistic understanding of one’s own system. In the two case examples, this was demonstrated through the blind spots of both systems surrounding the impact of the leverage being applied against their system paradigms and influence stocks by negative feedback loops flowing more rapidly than positive feedback. Instead of taking a holistic systematic review of the problems arising from negative feedback, both systems blamed external systems and unwittingly contributed to these problems through their respective ‘drifts to low performance’ (Figs 17 & 23), rather than optimising the whole, improving relationships, resilience, and sub-system collaboration (Stroh 2015:15,31, 33). In turn, this led to a loss of control, which was a critical capability and is the next theme to be analysed as it relates to the two cases.

Control Control was identified as a critical capability for both cases in protecting the COG and as a vital reinforcing element to increasing system influence stocks. Specifically, a control of the narrative, messaging, media quality, the systems’ credibility, image and legitimacy, and of sub-system behaviours were the most important elements requiring control by both systems for effective influencing. Whilst, ISIS’ system initially proved strong in having and maintaining control over its sub-systems and iWar activities, it was a ‘loss of control and authority’ feedback loop that contributed to a deterioration of influence and a drift to low performance for both systems (Figs 16 & 22).

Notwithstanding, this control only extends to the associated feedback loops, behaviours, and information flows within the system, because, as Meadows (2008:169-179) contends, systems cannot be controlled, rather they can only be designed and redesigned. This is where ISIS proved more effective than the Clinton campaign in maintaining this control. Due to a higher level of anticipatory resilience and a more dynamic hierarchical structure than Clinton’s campaign, ISIS 157 | P a g e 5249168 Cassandra Brooker Masters of Research seemingly understood it could not impose its will on the system. ISIS observed its own system behaviours and reactions to events, considered mental models, and prioritised information flows to ensure high levels of situational awareness to evolve, enhance decision-making, dominate the IE, and achieve successful influence (Meadows 2008:169-179). It was only after receiving delayed feedback relating to a decline in influence stocks, resulting from counterbalancing feedback, that ISIS became more reactive and less considered in its responses. This shift towards fragility resulted in a loss of control and authority and a drift to low performance.

The Clinton campaign mirrored this behavioural archetype in response to negative and delayed feedback, however its base-level responsiveness was less anticipatory than ISIS, so it experienced a loss of control much more rapidly. As demonstrated by the R2 reinforcing feedback loop (Fig 22), both internal and external systems contributed to the negative reinforcing feedback undermining Clinton’s control, and were facilitated by the same hard and soft communications links enabling the escalation of counter-influence attacks (Fig 21). As Meadows (2008:173) explains, information is power, and the media, politicians, organisations, and advertisers who regulate much of the public flow of information have far more power than most people realise. However, Clinton campaign’s degraded information flows and delays in receiving feedback, as discussed in the previous section, undermined its power and authority in the information environment (IE). Unlike ISIS, the Clinton campaign failed to pay attention to important qualitative information to improve system responses and maintain control, choosing instead to focus on quantifiable polling data and statistics, which are poor indicators of system health. Meadows (2008:169-179) contends the idea that something able to be measured is more important than unmeasurable qualitative data is flawed, adding that pretending something does not exist, because it is difficult to quantify, leads to faulty models.

Just as both cases could not control their respective overarching system, only the feedback, behaviours and flows within their systems; neither could they control the actions of external systems, such as the media or their adversaries, only their responses to those other systems’ outputs. In responding poorly to the negative feedback generated by external systems, both case systems undermined their own positive feedback loops, influence stocks, and system resilience. This resulted in both case systems suffering a loss of control and a subsequent ‘drift to low performance’ archetype, where internal sub-systems developed competing goals, the system purpose and outputs were degraded, and allies became accidental adversaries. This drift to low performance was further enabled by eroded communications links, increased risk aversion, and shifts within both systems to more rigid, less responsive, and increasingly isolated hierarchies.

158 | P a g e 5249168 Cassandra Brooker Masters of Research Centrality of Focus The fourth theme extrapolated from the comparative analysis was the importance of having a ‘centrality of focus’ around building and maintaining influence. Whilst the ability to effectively influence target audiences was a COG for both systems, there was a manifest contrast with the level of importance placed on influence activities by each. Like Russia’s Internet Research Agency (IRA) and other adversaries of Western democracies, ISIS had a centrality of focus, where influence activities were the main priority and purpose driving all system outputs, at all levels, receiving majority resource allocation, and which ultimately contributed to strengthening all four of its positive reinforcing feedback loops and the escalating virtuous cycle.

Conversely, like Western democratic nations and their militaries, the Clinton campaign lacked that centrality of focus. The campaign’s influencing strategy was not the main priority driving all system outputs, and other stocks not as critical to the COG were competing for priority and diluting resource allocation. The stove-piped, slow, traditional campaign media strategies and a focus on quantitative data contrasted sharply with the agile, qualitative iWar tactics of Clinton’s adversaries. This lack of focus on the system’s stock and purpose contributed to the asymmetric escalation of all three negative reinforcing feedback loops and the associated vicious cycle.

This contrast in focus underpins much of the discussion in Chapter 7 surrounding practical approaches to answering the research questions, addressing blind spots in the iWar, and improving Australia’s own influence effectiveness. Centrality of focus on influence is important for future warfare; and the West’s adversaries, such as ISIS and the IRA, recognise this and place importance on it accordingly. Meanwhile, Western democratic nations, and systems such as Clinton’s campaign, continue to apply linear, traditional, doctrinal approaches to iWar targeting (Table 2), which underestimate adversary capabilities, lack innovation, are risk averse, and fail to account for holistic system-based strategies. If influence is not central to all system aims, it will not be seen, it will not be effective, and the system ultimately will not succeed in iWar.

Many pundits believe ISIS was defeated through kinetic targeting on the physical battlespace, whereas in truth, their virtual caliphate remains undefeated in the information domain (Singer & Brooking 2018:154,235; Dhanaraj 2018:1-3; Patrikarakos 2017:246). Whilst ISIS suffered a decline in influence stocks as a result of balancing feedback, the system itself did not collapse – rather it contracted, shifted purpose, and is likely adapting lessons learnt and rebuilding its resilience before potentially rapidly escalating again in the future. This self-evolution is only possible due to the system’s centrality of focus on influencing target audiences through dynamic

159 | P a g e 5249168 Cassandra Brooker Masters of Research iWar tactics, technological innovation, exploiting cognitive behaviours, and by understanding adversary systems and leverage points.

Despite doctrine stating: “in war, the fundamental goal is to make the adversary comply with your will” (ADF 2012); Western militaries, in contrast to their enemies, decentralise influencing activities and segregate the associated analysis and targeting disciplines into niche, professional stove-pipes. Western militaries and government organisations rarely internally analyse their own systems holistically, fail to employ a fused approach to the iWar, and fail to harness the benefits of behavioural economics, as an integrated centralised focus of the whole system.

Egocentric tendencies are common among senior leaders in both military culture and politics, which presents a significant barrier to adopting critical thinking and innovative iWar solutions. ‘Egocentric memory’, where people tend to forget information that does not support their line of thinking, is related to these tendencies, and also applies to failings of the Clinton campaign. Additionally, ‘egocentric righteousness’, a tendency to feel superior based on a belief that one has it all figured out, and ‘egocentric blindness’, which is a natural tendency not to notice facts or evidence that contradict our beliefs; also contribute to undermining own-system situational awareness and detrimental mental models (J. Mark Services Inc. 2018; Elder & Paul 2002). One of the approaches Facione (2015:11) recommends for becoming a critical thinker is being honest in facing one’s own biases, prejudices, stereotypes, and egocentric tendencies. The system analysis in Chapter 5 of Clinton’s campaign demonstrated the egocentric tendencies of its leadership and media sub-systems, which enabled her adversaries to successfully exploit a number of behavioural economics tactics as described in Chapter 2. Such blind spots to psychological manipulation also apply to other Western democratic organisations, which are further bound by various legal, moral and ethical constraints, risk aversion, a lack of creative thinking, and are easily targeted in the iWar as open, transparent systems.

Such constraints do not apply to the closed, secretive systems of democratic state adversaries. Closed states, such as Russia and China, demonstrate this centrality of focus on influence dominance from the political-strategic level right down to the tactical grassroots (Gleeson 2019). The blurring of lines between military and civilian target audiences and a shift in the social- cultural context of rules, norms and accepted behaviours has also changed how military activities are conducted. These factors now affect freedom of action, impede the design and execution of operations, and sees military forces having to justify the legitimacy of their actions; whereas adversaries are unfettered by restrictive policies or the need for truth (US DOD 2018:6-8). Furthermore, despite vast changes in global geopolitical landscapes that continue to strengthen 160 | P a g e 5249168 Cassandra Brooker Masters of Research the significance of understanding cultures and society, resource allocation for militaries to conduct any social science or cognitive behavioural research remains a low priority (Buckle 2018:6).

Part Two: Findings Related to Influence Effectiveness The second part of this chapter provides a summary of findings related to influence effectiveness derived from the research and analysis presented in the previous chapters in order to inform the discussion in Chapter 7. Chapter 2 revealed there are key themes and linkages between effective influencing activities and behavioural theories that enable the successful manipulation and exploitation of target audiences (as summarised in Table 6). These tactics and psychological hacks were proven to be understood, fused, and applied by adversaries of Western democratic nations in the iWar. The comparative analysis demonstrated that a ‘centrality of focus’ by systems operating in the iWar provides the foundation environment, focused mental models, and reinforcing conditions and resources to ensure their success in influence effectiveness. Further, these systems enjoying influence success in the iWar are not hamstrung by unwieldy, bureaucratic, traditional, linear thinking nor by the ethics and values held by democratic societies (Table 4); and in Russia’s case, they got away with it, suffering no discouragement or consequences.

The key findings arising from the research are categorised under the five most effective influencing concepts, described by Singer and Brooking (2018:21), for winning the iWar:  Narrative  Emotion  Authenticity  Community  Inundation

Whilst some influencing tactics, such as ‘nudge theory’ and social contagions, manipulate multiple cognitive behaviours that overlap across all five categories, the findings will be presented according to the category they best fit. This provides a more holistic discussion around influence effectiveness, rather than focussing on specific iWar tactics.

Narrative Controlling the narrative is an imperative for achieving success in the iWar. A narrative must be simple, must resonate, and must have a certain novelty for target audiences (Singer & Brooking

161 | P a g e 5249168 Cassandra Brooker Masters of Research 2018:158-179). As demonstrated by the case systems, the narrative is the underpinning device driving the critical capability of ideology, and providing the ‘sense of purpose’ feedback. Control of the narrative is therefore critical for protecting an influencing system’s COG. Bosio (2018) explains that the West’s enemies formulate narratives that resonate far greater than democratic ideals, and create a sense of belonging, which ties into the key influencing concepts of ‘community’ and ‘emotion’. The importance of a resonant narrative was clearly evident in the differing levels of influencing success achieved by ISIS compared with the Clinton campaign. Adversary systems are adept at disseminating their narratives using simple language, cultural memes, evocative imagery, and humour to ensure the novelty and memorability of the messaging. As Stroh (2015:30) explains, stories shape our identity, communicate our values in memorable form, and inspire others to act. Therefore, narratives not only bind system elements to their purpose but are also a key source of conflict.

Through its recent iWar battles with Russia in the Baltic States and Ukraine, NATO realises the importance of having a strong narrative to counter adversary messaging, and now has doctrine to ensure its messaging achieves the desired effect. Key principles guiding this narrative development include: ensuring it captures NATO’s values, is empowered down to all levels of the organisation, and that the message is credible, aligned, understood, integrated and resourced (NATO 2016; Laity 2015:24).

For Western democratic nations, conceptual agreement is needed for a unity of effort across government, military, academia, and society to develop and disseminate an agreed strategic narrative to wrest back control in the iWar. Watts (2018:190) contends that “America sucks at information warfare”, largely because the US government lacks consensus about what should be done, how, and by whom – that is, the message, the messenger and the method. Whereas America’s adversaries all know exactly what they seek to accomplish on the social media battlefield, are better at communicating it, and their success is further strengthened by their ‘centrality of focus’. At a micro-level these failings were demonstrated by Clinton’s conventional campaign strategy, which lacked a strong resonant narrative and associated messaging to counter adroit adversary attacks.

Western democracies struggle to commit to having a strong narrative largely due to risk-averse, stove-piped bureaucracies that dilute, micro-manage, or overly sanitise iWar messaging. This interference dulls innovation, slows responsiveness, and fails to resonate with target audiences (Watts 2018:192). Additionally, Western media contributes to these messaging failures. Inexperienced journalists lacking critical thinking skills often misunderstand, believe and spread 162 | P a g e 5249168 Cassandra Brooker Masters of Research enemy propaganda, as well as invariably criticising any government messaging effort, thereby undermining democratic states’ iWar campaigns. These weaknesses were replicated at the micro- level by Clinton’s campaign leadership and media sub-systems. The US DOD (2018:3) advises that the Joint Force must better understand how relevant actors “assign meaning to information”, in order to understand how to better leverage information in the iWar battlespace. This was demonstrated in Chapter 4, where understanding the meaning behind ISIS’ narrative and ideological messaging provided insight into their Sense of Community (R1) and Sense of Purpose (R2) reinforcing feedback loops (Figs 11 & 12). However, influencing activities complementing the narrative must also incorporate the four other key concepts of emotion, authenticity, community and inundation.

Emotion As revealed through the review of the literature and demonstrated by the IRA, ISIS’ reinforcing feedback loops (Figs 11 & 12), and the Trump campaign (Table 12), the primary element influencing behaviour and underpinning cognitive manipulation is emotion. Emotion also contributes to the success of the other concepts of narrative, authenticity and community, through social and emotional contagions, and in exploiting biases and heuristics. Tucker (2016) explains that when there is a strong emotional response to messaging or content, this correlates to changes in real world behaviours.

The exploitation of emotions and ‘System 1’ thinking (Table 5) is the reason why polarising subjects, conspiracy theories, humorous or evocative clickbait, and sensational fake news stories are so compelling and become viral contagions. In addition, Wardle (2019) expounds on what she says social scientists and propagandists have long known: “that humans are wired to respond to emotional triggers and share misinformation if it reinforces existing beliefs and prejudices”. This presents a significant challenge for Western governments trying to understand their audiences and develop emotionally compelling content. Additionally, most people are loathe to share their personal information with the state as willingly as they do with marketers, social media platforms, and websites (Tucker 2016). The system analysis of the Clinton campaign highlighted the impacts of counter-influence activities, which exploited emotional contagions and heuristic biases of target audiences, in undermining her influence stocks (Fig 21).

As the literature review revealed, whilst evoking any emotion proves to be an advantageous influencing device, anger and outrage are the most effective emotional contagions. Adversaries intent on deepening existing societal fissures, such as ISIS or the IRA, design content to anger or excite audiences and incite them into becoming messengers. This leads to users spreading the 163 | P a g e 5249168 Cassandra Brooker Masters of Research emotive content virally throughout their networks, while also adding their own social capital to reinforce and give credibility to the original message (Wardle 2019). This iWar tactic proved extremely effective and efficient for Russia’s interference in the 2016 US election, due to the contribution of social media enabling the profiling and targeting of users with emotive, addictive content that reinforced their biases, fears, desires, and social anxieties (Jakubowski 2019). This was also demonstrated acutely by Trump’s ‘in-group’, nationalist rhetoric, which stirred outrage, generated fear, and reinforced racist stereotypes about migrants and Muslims (Table 12). The combination of these multi-pronged, multi-platform, highly emotive, viral attacks resulted in extensive counter-influence feedback and a loss of control and authority by the Clinton campaign (Figs 21 & 22).

Notwithstanding Western government messaging and content surrounding national security and counter-terrorism; unfortunately evoking anger, outrage, and fear in an attempt to influence domestic audiences is counter to democratic social norms and values, and would undermine the credibility of the strategic narrative. Therefore, such tactics are largely the reserve of the West’s adversaries, RW media outlets, and political systems that hold little regard for such mores. The importance of maintaining legitimacy and having authentic messaging were critical requirements for both case systems, with Clinton’s credibility contributing to a significant balancing feedback loop (B2) underpinning the campaign’s influence effectiveness (Fig 25).

Authenticity The comparative case system analysis starkly demonstrated the importance of being authentic, having credibility, and maintaining legitimacy for ensuring strong balancing feedback and abundant influence stocks. Having an authentic narrative and credible ideology were critical capabilities for protecting the COG. Conversely, this analysis also revealed the powerful impacts that counter-influence activities had when attacking the two systems’ authenticity through negative reinforcing feedback. Without authenticity, trust cannot be established and for the case systems, the ‘sense of purpose’, ‘credibility’, ‘success to the successful’, and ‘sense of community’ feedback loops would have been severely degraded or even ceased to exist.

Being authentic and building credibility is a significant problem for Coalition nations operating in the Middle East. As Watts (2018:192) explains, America’s counterpropaganda, against ISIS for example, suffers from a ‘messenger problem’, for example, their spokespeople fail to elicit sympathy or fear among target audiences. Watts (2018:192) adds that America’s use of moderate Muslim clerics as messengers against terrorism undermined their credibility with diaspora audiences. These failings highlight the importance of conducting in-depth target audience 164 | P a g e 5249168 Cassandra Brooker Masters of Research analyses, preferably using systems thinking approaches, to gather insights for better influencing. Blackmore (2003:22) contends that while an intimate knowledge of culture, languages, economics and politics is vital for effective influence operations, it’s extremely difficult to achieve; and this is where ISIS had the advantage over the Coalition. Understanding target audiences, local issues, and linguistic nuances not only reinforces authenticity but enables the development of dynamic messaging strategies, rapid adaptation, and anticipatory responses in the iWar.

The ability to embed into the local population and understand cultural nuances, local social issues and community grievances was the reason ISIS was initially so effective at influencing both their physical and virtual caliphate audiences (Figs 11 & 12). Conversely, ISIS’ barbaric responses to local culture destroyed their legitimacy in the Muslim world and ultimately undermined their authenticity, thus sowing the seeds of their own destruction (Siyech 2016:25-26). Similarly, as mentioned, continuous attacks on Clinton’s credibility and associated positive (B2) feedback loop, contributed to the vicious cycle undermining her influence and thereby the entire campaign’s COG (Fig 19).

A key iWar tactic that contributes to authenticity and develops trust is the tailoring of messages to target different audience interests, sharing relatable human interest stories, and personalising content for individuals. Both ISIS and the Trump campaign used tailored content, relatable stories, psychometric analysis, and personalised advertising to great effect, in order to enhance their influence, manipulate sentiment, and ensure brand engagement (Singer & Brooking 2018:65,158- 179). Audience engagement tactics contribute to creating a sense of belonging, encouraging partisanship and heightening in-group versus out-group tensions, and were used by both case systems in the development of their respective ‘sense of community’ feedback loops (Figs 11 & 26).

Community Nudge theory and social contagions are much more effective in influencing behaviours when they exploit real or perceived community expectations and anxieties surrounding in-group conformity. Community linkages, social relationships, trust, loyalty, and accesses to target audiences were all critical requirements for the two case systems in maintaining their ‘sense of community’ feedback loops and overall influence stocks. iWar tactics that exploit emotions and display authenticity also reinforces the sense of community among target audiences.

165 | P a g e 5249168 Cassandra Brooker Masters of Research In developing an ‘in-group’ community feel for achieving effective influence, systems should employ cognitive exploitation devices such as: using targeted, personalised messaging; reinforcing polarising ‘them vs us’ narratives and in-group ideals; manipulating biases, heuristics, fears, and cultural guilt; and exploiting social pressures to belong to a tribe, conform to group norms, and be accepted by the group (Table 6, Fig 4). The Trump campaign demonstrated a canny use of such tactics (Table 12), which contrasted starkly with the Clinton campaign’s safe, traditional, influencing strategy; and contributed to negative reinforcing feedback that undermined Clinton’s own ‘sense of community’ feedback loop (Fig 26). Extremist narratives that tap into shared grievances or cultural identity, such as ISIS’ messaging that “Islam is under attack”, resonate far greater than democratic ideals in Coalition-deployed locations, because they provide a sense of belonging and unity (Bosio 2018; Clark 2015). Just as understanding target audience culture, language, and symbolism is important for ensuring authenticity, Bosio (2018) and Blackmore (2003:21) agree that understanding societal frames and integrating human factors are fundamental to conducting any complex social system analysis for informing information operations or in formulating themes and messages.

Developing this sense of community also now applies to social media and its associated ‘virtual tribes’ for enhancing online influence. Watts (2018:194-195) explains that successful social media influencers must be human to be social, and that Western governments’ official social media messengers are neutered in this sense due to a lack of agility, a failure to integrate content across multiple media platforms, and due to power struggles between agencies over control of the messaging. Compared with their adversaries, Western official messaging comes across as cold and clinical, and fails to build that sense of community or human connection. In contrast, ISIS provided a networked, digital, community space where all the accumulated propaganda resided, where fighters and fans could mingle, where it could track and manipulate global opinion, and where they still held power and influence even after losing physical terrain (Singer & Brooking 2018:152; Dhanaraj 2018:1). As a result, ISIS’ ‘sense of community’ (R1) reinforcing feedback loop (Fig 11) was extremely strong, contributed information flows into other reinforcing loops, which contributed to their virtuous cycle and success to the successful archetype. The echo chambers and filter bubbles exploited by the Trump campaign, right-wing media and Russia’s IRA also demonstrated this powerful partisan “virtual tribe” effect, which was not recognised nor replicated by the Clinton campaign (Fig 20).

Understanding the psyche of a community and having a thorough knowledge of cultural and social issues contributing to its in-group dynamics enables effective influencing of that target audience (Blackmore 2003:21). Ethnic identity plays a particularly large role in in-group versus out-group 166 | P a g e 5249168 Cassandra Brooker Masters of Research dynamics. Benkler et al. (2018:309) contend this explains the extraordinary passion and enthusiasm of Trump political rallies, where nationalist in-group dynamics combined with anger and outrage to motivate supporters and played an instrumental role in political mobilisation. Russia’s troll army also leveraged social fissures, community grievances and group outrage to manipulate political discourse, effectively attack the Clinton campaign, and undermine US democratic institutions more broadly. These in-group factors also contributed to Russia’s information operations’ success in the Baltic States and Ukraine. As Jakubowski (2019) explains, the target nations’ historical relationships with Russia, percentage of ethnic Russians in the population, ethnic homogeneity, racial conflict, migration, and Russia’s control of those states’ media and internet services all contributed to developing a sense of community and belonging to the Russian motherland for the ex-patriot diaspora.

Further contributing to Russia’s iWar success was its ‘centrality of focus’ and the networked nature of social media enabling content saturation across multiple types of media outlets. Russia was the first entity to incorporate the entire social media ecosystem into its information operations. The associated inundation overwhelmed Western democratic systems’ ability to respond or counter this “firehose of falsehood” (Jakubowski 2019; Paul & Matthews 2016:1).

Inundation After combining the four abovementioned concepts into an iWar influencing strategy, the effectiveness of that strategy is further enhanced through inundation. Saturating target audiences with an authentic narrative and emotive ideology, using a repetitive, multi-pronged, multi- layered, multi-platform media approach proved highly successful for attaining exposure and influence for the IRA, ISIS, and the Trump campaign.

Disinformation will often be built around an element of truth to enhance believability and make it more difficult to expose (Wardle 2019). However, it is then the volume of content, the repetition leading to familiarity, and the long-term effect of the disinformation campaign’s multi-media inundation that forms the basis of its success. Wardle (2019) explains that human brains are wired to rely on heuristics to help us judge credibility. As a result, repetition and familiarity are the most effective mechanisms for ingraining misleading narratives, even when audiences receive contextual information explaining why a narrative is false. It becomes a reinforcing cycle as audiences with declining attention spans, develop familiarity with the messages, which leads to easier acceptance of the message, and is then spread further via their social networks by amplifiers and multipliers to dominate the IE as contagions, overwhelming and undermining adversary

167 | P a g e 5249168 Cassandra Brooker Masters of Research narratives, content, and reach. As Singer and Brooking (2018:240) explain, social media is now a battlefield with real-world consequences and on which only losers play fair.

Homophily, filter bubbles and echo chambers, and associated viral contagions created by an inundation of iWar content now shape reality. This was demonstrated by the ineffectiveness of Clinton’s campaign to appeal to ‘System 1’ thinking and grab the attention of audiences. Unlike the Trump campaign, which employed constant exposure, inundation, amplification and repetition methods (Tables 6 & 12) to dominate the information environment with pithy, superficial, and emotive messaging; Clinton’s campaign focused on detailed policy announcements, substantive discussions, deliberate messaging and researched, factual statements. As a result, Clinton’s media strategy haemorrhaged audience attention, bored journalists, appeared overly stage-managed, and lacked the agility to counter opposition attention-grabbing content or respond to headline grabbing controversies (Benkler et al. 2018:191; Clinton 2018:63,222-225,232,236,272).

Further contributing to the effectiveness of inundation are darker forces associated with automation, the rise of bots, sock-puppets, micro-targeting using artificial intelligence, and troll farms operating numerous social media accounts. These tactics have fuelled information campaigns designed to manipulate public opinion on a massive scale, and overwhelmed fact- checking processes and technology companies’ content moderation practices (Wardle 2019; Singer & Brooking 2018:235). ‘The Facebook Dilemma’ documentary (2018) highlighted the results of such inundation tactics that saw the rise of digital nation states conducting massive propaganda production on an enormous scale, and effectively weaponised social media. Russia’s IRA, ISIS, and America’s RW media ecosystem have all conducted these techniques to shape the narrative and dominate the information environment. Wardle (2019) describes the failures of technology companies in taking action against these exploitative inundation tactics as having resulted in a “polluted information ecosystem” and “intentional chaos”. Wardle (2019) adds that most of this content is not designed to persuade people but rather to cause confusion, overwhelm, and undermine trust in democratic institutions.

Inundation, saturation, mass exposure and Russia’s “firehose of falsehood” are difficult challenges to counter – as moderators, journalists and fact-checkers are unable to keep up with the immense flows of disinformation. Additionally, this constant bombardment overwhelms people’s ‘System 2’ thinking and their ability to critically analyse large volumes of content. As Paul and Matthews (2016:9) explain, it takes less time to make up facts than it does to verify them

168 | P a g e 5249168 Cassandra Brooker Masters of Research and psychological evidence shows that retractions and refutations are seldom effective, especially after significant time has passed.

However, Benkler et al. (2018:236) contend that by overstating the prevalence and effect of Russian attacks against democracy further reinforces their success, just as terrorism succeeds most when it evokes an overreaction or emotive uncalculated response. This was the reason ISIS baited Coalition forces into reacting, in order to reinforce its narrative with target audiences and therefore, increase its influence (Ingram 2015:744; Clark 2015). The research suggests, therefore, the most effective way to counter these iWar tactics is to reciprocate in kind. But Western democracies must also ensure that transparency, credibility, and legitimacy are maintained to ensure both an effective counter-influencing strategy and the ongoing support, trust and belief by target audiences.

Conclusion This chapter concludes Part Two of the thesis, which comprised the modelling of the two case systems, a comparative analysis of the two systems, and a summary of the findings resulting from both the comparative analysis and the research surrounding key concepts contributing to influence effectiveness.

The four key themes contributing to both case systems’ levels of influence arose from the findings of the system modelling conducted in Chapters 4 and 5, and were analysed in the first part of this chapter. Those themes were: 1. System Characteristics – open or closed, level of resilience versus fragility; 2. System Information Flows – feedback loops, interconnections, delayed feedback; 3. Control – as a critical capability and balancing element; and 4. Centrality of Focus – regarding the level of importance placed on influence activities.

The five key concepts contributing to influence effectiveness were discussed in the second part of this chapter and comprised: narrative, emotion, authenticity, community, and inundation.

These themes will provide a framework for further discussion in Part 3 of the thesis, specifically in Chapter 7. The discussion and conclusion chapters will address the research gaps relating to: Australia’s influence effectiveness and counter-influence strategies, how complex social systems adapt their behaviours based on feedback loops, and why understanding psychological cognitive theory is central to analysing and leveraging social system behaviours and mental models.

169 | P a g e 5249168 Cassandra Brooker Masters of Research Chapter 7 will structure the discussion in three parts:  Part One: How much do we know? This part summarises the research and findings derived from the literature review and case systems analysis.  Part Two: How has the research answered the gaps? This part is further segregated into three research areas of: Theoretical, Methodological (systems), and Practical solutions.  Part Three: What does this mean for Australia? This part also includes discussion surrounding the associated future challenges in applying the findings.

170 | P a g e 5249168 Cassandra Brooker Masters of Research Thesis Part Three – Analytical Conclusions America’s problem in counterinfluence is that we don’t know what to say, because we don’t know what we believe in.

Clint Watts Messing with the Enemy (2018:210)

Introduction The premise of this research is that Australia is already losing in the iWar and Western democratic nations are being out-communicated by their adversaries. A key gap is that psychological operations (PSYOPs) and information warfare are spoken about extensively in the grey literature, however those discussions are largely omitted from the corpus of academic knowledge. As a result of rapid worldwide changes, such as: globalisation, urbanisation, population growth, technological advances, communications links, and network flows; power has essentially devolved to lesser nations and non-state groups. The contemporary battlespace has evolved into a ‘grey zone’ existing between peace and war and the line is increasingly blurred between military and civilian, allies and adversaries, in the iWar.

As the findings of this research exposed, a new form of hybrid warfare, which combines political, economic, diplomatic, media, civilian and military measures, presents new challenges, emerging threats to civilisation and considerable uncertainty to democratic nations and their military forces. The information domain of warfare favours the West’s adversaries – who are unbounded by democratic values, ethics, journalistic standards, or any commitment to the truth (Table 4). Linear thinking, conventional, doctrinal influencing tactics, and slow, stove-piped, bureaucratic approaches enacted by the West are stymied in the iWar, resulting in the dislocation of state power and military might (Table 2).

Through controlling the narrative, exploiting cognitive behaviours, and employing asymmetric influencing tactics, the West’s adversaries threaten democratic values and institutions, create fear, panic, uncertainty, and emotional contagions. This insidious threat weakens the will of populations; erodes trust in governments; and paralyses systems. The iWar is cheaper to fight than conventional war for asymmetric threat groups and its impacts are immediate.

The iWar is political warfare, where the ability to influence and motivate target audiences is essential to success. The unpredictability and complexity of modern communications platforms, particularly social media, diminishes situational awareness and the ability to control the message or predict events and behaviours. As Watts (2018:229) explains, shifts in political support and the international order stemming from social media preference bubbles were swift and volatile, with 171 | P a g e 5249168 Cassandra Brooker Masters of Research bad actors worldwide adapting this technique from the Russians. This is why it is important for a nation to holistically understand its own system, and address any fragility, before attempting to deploy into the iWar battlespace.

Whilst there is no one solution to countering the modern iWar threat, this research has shown that a proactive, anticipatory, cooperative, non-conventional approach, which leverages adversary system weaknesses and biases, whilst building own system resilience and stocks, is the most effective method. Complacency is the biggest threat. Part Three of this thesis, comprising Chapters 7 and 8, discusses and concludes the results of this research as they relate to the gaps in knowledge, the three key arguments, and the research aim, before making key recommendations for addressing Australia’s iWar problems.

Chapter 7 discusses the implications of the findings of the literature review in Part One, and the results of the case system analyses in Part Two of this thesis, to achieve the aim of investigating solutions for improving Australia’s influence effectiveness in the iWar. Chapter 8 closes Part Three in providing the conclusions identified in response to the aim and the research questions, which arose from the discussion of the research findings in Chapter 7. These conclusions distil the key concepts arising from the research and analytical findings and inform the associated recommendations for improving Australia’s influencing effectiveness as a complex social system operating in the iWar. Chapter 8 concludes with a synopsis regarding the significance of this research and accounts for its contribution to Australia’s future iWar capability development.

172 | P a g e 5249168 Cassandra Brooker Masters of Research Chapter 7 – Discussion It’s not whose Army wins but whose story wins.

Joseph Nye Soft Power (2005)

Introduction The discussion of this research will contribute to addressing gaps in the current literature where systems analysis approaches to the iWar are misunderstood and underutilised by the ADF, where behavioural economics theory can be better exploited in the iWar, and in providing an analytical argument for what makes certain influence activities effective in the contemporary, complex, globalised information environment. The discussion contained in Chapter 7 is structured as follows:

 Part One: How much do we know? This part segregates and summarises the research and findings, derived from the literature review in Part One and case systems analysis in Part Two of the thesis, into the three research disciplines of: influence, cognitive behaviours and systems thinking.

 Part Two: How has the research answered the gaps? This part is further segregated to provide solutions into three research areas of: Theoretical (highlighting the strengths and weaknesses), Methodological (systems thinking), and Practical (military and civilian applications).

 Part Three: What does this mean for Australia? This part also includes discussion surrounding the associated challenges with applying the findings to future iWar approaches and future research opportunities.

Part One: How much do we know? The literature review demonstrated that governments, corporations, analysts, strategists, and academics have all been investigating how to better influence target audiences and exploit cognitive human behaviour for many years. There is a long and colourful history associated with the evolution of media and advertising, whilst propaganda and controlling the narrative are established influencing activities as old as warfare itself (Giles 2017). The collective interest in these fields increased following the results of Russian interference in the 2016 US election and Brexit. Systems thinking theory has also developed over a number of years and the associated

173 | P a g e 5249168 Cassandra Brooker Masters of Research literature continues to evolve to better apply the principles to improving our own complex social systems in the modern interconnected world.

However, researching all three disciplines as they relate to each other, the West’s adversaries, and the contemporary iWar; as well as analysing what those findings specifically mean for Australia, had, until now, not been attempted. Because influencing activities, cognitive behaviours, and systems thinking have all been impacted upon by the interconnected global information environment of cyberspace, research into the results of such modern effects is required to address the gaps in the accepted knowledge. As Jensen and Sear (2019) explain, cyberspace, especially social media, has enabled the strategic use of information to pursue a competitive advantage in amplifying, influencing, and degrading the capacity of adversaries to make decisions, even at a societal level. They add that human consumption of cyberspace content – with all our human cognitive biases – combines with a tribal, adversarial democratic system, which has blurred the line between politics and war, creating a new type of information warfare.

Figure 5 demonstrated the important overlaps in theory between the three research disciplines of literature reviewed in Chapter 2, as they relate to influencing and leveraging behavioural change in a system. The elements that related to all three literary disciplines comprised the hidden part of the iceberg (Fig 1), which are the fundamental system components that can be leveraged and influenced in the iWar. The Venn diagram (Fig 5) was updated following the conduct of the case systems’ modelling in Part Two of this thesis, where application of the theory arising from the literature review informed the methodology and analysis of the two case systems. Therefore, Figure 28 summarises the key concepts and overlapping findings of this research, and the following paragraphs summarise how much we now know across these three disciplines as a result of this combined research approach.

174 | P a g e 5249168 Cassandra Brooker Masters of Research Figure 28: Links and findings of the three research disciplines

Influence Lessons The review of literature in Chapter 2 showed there were ten key thematic areas identified that proved most effective in the conduct of influencing activities. These are summarised in Table 3 and include concepts such as: emotional and social contagions, exposure, audience susceptibility, sequencing, biases, entertainment, personalisation and addictive feedback. A number of these thematic tactics also overlapped with the findings related to the research into cognitive behaviours and nudge theory (Table 6). In Chapter 6, following the systems analysis of two cases’ levels of effective influencing, those ten concepts were distilled further into a framework of five important themes required for optimal system feedback flows and overall iWar success. These five themes were: narrative, emotion, authenticity, community, and inundation.

As ISIS and other adversaries of the West in the iWar have demonstrated, controlling the narrative using resonant language and an authentic ideology is key to attaining informational power. They have demonstrated further strength in their strategic communications by employing soft power tactics in the community and conducting ‘propaganda by deed’, designed to persuade target audiences through economic, security or cultural influences, which exploit emotional and social

175 | P a g e 5249168 Cassandra Brooker Masters of Research contagions and links into the cognitive behavioural lessons. Further contributing to influence effectiveness is inundation. The advent of unprecedented volumes and speeds of information flows, combined with less time and manpower available to manage and analyse this data, impacts on Western systems’ strategic decision-making, system responsiveness, and overwhelms communications links (IWD 2018:7-8). These factors contribute to a further loss of situational awareness and cedes the advantage in the iWar.

The hard and soft communications links provided the ways and means of producing, disseminating and reinforcing the effective tactics of influence identified in the research. In the two case system analyses, contained in Chapters 4 and 5, the interlinkages and relationships relating to ideology, narrative, and communications links were defined as key system elements due to their importance in maintaining influence stocks, flows and function. Controlling the communication environment and its flows is now just as important as controlling the narrative and associated content. As Watts (2018:219) explains, “social media users now reside in virtual social media nations, where they identify allegiances through the use of virtual cues that allow connections with other like-minded users, reinforce beliefs, shape collective values, and fight digital adversaries” (Ibid:219). Furthermore, fundamentalists have benefited from the communications revolution, with target audiences being more accessible and susceptible to manipulation than previously achieved using mainstream media.

Ruthven (2007:131-133) describes that now fundamentalist groups, such as ISIS, are able to exploit the para-personal, electronically-amplified relationships between charismatic leaders and their audiences. This new information age capability contributed significantly to ISIS reinforcing feedback loops relating to a ‘sense of community’ and ‘sense of purpose’ (Figs 11 & 12). However, the internet also creates negative feedback by undermining information monopolies on which authoritarian and fundamentalist movements depend. This is the reason for the unprecedented increase in the scale and volume of enemy iWar content in recent years. It is designed to inundate target audiences, obscure the facts, and dominate the IE.

The West’s adversaries have demonstrated a deep understanding of effective influencing tactics and how best to exploit hard and soft communications links. Both Russia’s IRA and ISIS exacerbated social divisions, exploited psychological behaviours, and created echo chambers and filter bubbles in multiple languages, on multiple platforms to amplify, inundate and dominate the information environment and its associated interconnections. The effectiveness of adversary iWar tactics has been further enabled by a relative freedom of action available to them in the information domain, whereas Western democratic nations face numerous obstacles and ethical 176 | P a g e 5249168 Cassandra Brooker Masters of Research limitations to their influencing strategies (Table 4). Ehlers (2017:3-4) explains, their willingness to use new iWar capabilities in ways unhindered by Western ethical or moral considerations, such as unethically exploiting the rules of engagement used in Western targeting processes, only adds to their comparative advantage.

Additionally, linear thinking and conventional, risk-averse approaches employed in the asymmetric, rapidly evolving iWar battlespace further undermines the influencing effectiveness of democratic nations and their militaries (Table 2). This lack of creative thinking and freedom of action also impacted on Clinton’s campaign influence effectiveness and system resilience, as it relied on rigid, traditional political strategies (Fig 23). As Singer and Brooking (2018:159) outline: “the big losers in the control of the narrative battle are those people or institutions that are too big, slow or hesitant to weave such stories. These are not the kinds of battles that a plodding, uninventive bureaucracy can win” (Ibid:159). Furthermore, non-state actors, hackers and propagandists now occupy a level playing field against powerful nation states, which has completely altered the threat environment. Individuals can now compete for strategic-level power and control, influence audiences, or create behavioural changes, which could potentially perpetuate a terrorist attack, sway votes, or even ignite a civil war (Watts 2018:19).

What do these influence lessons mean for Australia? Successful influencing tactics utilised by Russia in the iWar have already started to shape the geo- political situation in Europe, America and the Middle East. The IRA’s counter-influence attacks on Clinton’s campaign were a micro-level demonstration of their wider strategy (Fig 21). Russia sees the West’s migration into preference bubbles as an opportunity to use democratic freedoms as a wedge to subvert reality and further exacerbate social divisions into irreconcilable chasms (Watts 2018:227). Australia is not immune nor isolated from hostile influences arising from this rapidly changing, contested, strategic environment. Australia is being out-communicated by its adversaries as they present a consistent narrative, alter the rules to their advantage, and manoeuvre to embrace conflict below the threshold of conventional warfare (Bienvenue et al. 2018b).

Australia’s complacency and inaction in the iWar can be likened to the ‘frog in the pot’ analogy, where as a system it is drifting to low performance and any future reactions may be ineffective, misguided and essentially dislocated (Meadows 2008:122; Clinton 2018:363). The Information Warfare Division (IWD 2018:7) in Canberra has warned against the emerging threats from both state and non-state actors exploiting this new ‘Information Age’. IWD explains that while Australia has not been directly threatened since World War II, dark forces operating in cyberspace changes this paradigm – enabling adversaries to operate behind Australia’s borders and 177 | P a g e 5249168 Cassandra Brooker Masters of Research potentially attack government, industry, and the public directly. Ehlers (2017:3-4) agrees that the dramatic and continuing diffusion of technology to both major powers and smaller, non-state actors presents a significant challenge for democratic nations in the IE. Adversary abilities to control messaging in order to pre-empt Western policymakers and iWar operators, as well as their sophisticated manipulation of strategic communications for influencing public opinion (as demonstrated by ISIS, Fig 13), has upended traditional forms of conflict, thereby impacting on national security and undermining conventional means of achieving successful resolution.

The information age is transforming how humanity is informed, determines judgements, and ultimately makes decisions (IWD 2018:8). Human cognitive behaviours and the psychological warfare tactics associated with the iWar are complex matters that can only be countered through a scrutiny that develops a nuanced appreciation of their sophistication. This comprehension complements the five influencing themes of: narrative, emotion, authenticity, community, and inundation.

Cognitive Behaviours Lessons Understanding and exploiting human cognitive behaviours is central to conducting successful influencing activities and leveraging target systems (Table 6). As the review of the literature revealed, cognitive behaviours that can be exploited in the iWar are separated into either ‘System 1’ or ‘System 2’ thinking heuristics (Table 5). The effective manipulation of these heuristics has been demonstrated in practical form by Russia’s IRA, ISIS, and the Trump Campaign in recent years (Fig 10, Table 12), and their success underlines the importance of developing critical thinking skills in Western democratic audiences.

The psychological manipulation of target audiences, use of social engineering, and heuristic exploitation; sees social media becoming “anti-social media”, and enables adversaries to reinforce their influence and effect behavioural change (Watts 2018:18). Combined with a revolution in communications technology, traditional media broadcasters are losing control over their target audiences. As Ruthven (2007:131) explains, in regions where levels of literacy are low, audio- visual technologies have enabled charismatic religious figures to acquire massive followings. YouTube provided an extremely powerful platform for ISIS to reach target audiences using carefully crafted videos to inspire, recruit, fundraise, establish physical and virtual caliphates, and essentially control the narrative through the manipulation of heuristics (Fig 10).

The human cognitive behaviours exploited by Western adversaries in the iWar reinforce the five influencing themes of: narrative, emotion, authenticity, community, and inundation. As 178 | P a g e 5249168 Cassandra Brooker Masters of Research demonstrated in the case system analyses of Chapters 4 and 5, playing on System 1 and 2 heuristics to develop a ‘sense of purpose’ and an ‘in-group’ sense of community were very powerful tactics in strengthening positive feedback loops and building influence stocks (Figs 11, 12, 24 & 26). The US Joint Concept for Operations in the Information Environment (US DOD 2018:4) explains that informational aspects derived from sensory inputs that are physical in nature are generally more impactful. This is why videos or activities that demonstrated ‘propaganda by deed’ proved so successful for both systems and reinforced their authenticity and credibility. The manipulation of social and emotional contagions and the creation of addictive feedback mechanisms further reinforces influencing success, as target audiences decide and react using System 1 thought processes.

The second cognitive behaviours lesson relates to employing critical thinking over linear thought processes (Table 2). The rapid advent of communications technology and globally connected social media communities have heralded a new information age, where target audiences are becoming increasingly “dumb” and more easily manipulated. As Watts (2018:216) explains, nowadays people are “driven by ideology, desire, ambition, fear, and hatred – or what might collectively be referred to as preferences” (Ibid:216). These preferences have led to insulated filter bubbles, echo chambers, algorithm-selected newsfeed content, and an overall loss of critical-thinking skills across the online population, which are easily exploited by adversary systems (Tables 6 & 12). This use of the automatic, quick, effortless System 1 thinking, relies on heuristics and unconscious processes of perception and memory. Biases cannot be turned off, audiences become more gullible to disinformation, and increasingly fail to exercise conscious doubt. This lack of critical thinking is further exacerbated by shortened attention spans and increased time spent on social media. Addiction to devices and social media feedback mechanisms essentially rewires users’ brains, which impacts on cognition, and causes strange behaviours outside traditional social norms or interpersonal relationships (Watts 2018:249).

The manipulation of these human behaviours, cognitive processes and mental models, has been termed “cognitive warfare” and takes the ‘weaponisation of information’ a step further into the realm of “neuro-weapons”. When cognitive warfare is coordinated and directed at open, liberal, democratic societies, it has realised significantly positive results for adversary systems (Bienvenue et al. 2018a). This scrambling of Western democratic societal orientation, and the undermining of situational awareness in the iWar, contributes to a vicious reinforcing feedback loop of misguided reactions and increased system fragility, as demonstrated by the counter- influence feedback undermining the Clinton campaign and contributing to its drift to low performance (Figs 21 & 23). As Bienvenue et al. (2018a) explain, adversary actors have 179 | P a g e 5249168 Cassandra Brooker Masters of Research strategised to avoid a confrontation with the US and allied forces at their strongest point – namely, high intensity conventional warfare – instead pursuing gains in the information domain by denying, disrupting, and countering the narratives that underpin Western legitimacy and by diminishing their influencing power. As demonstrated by the research and case analyses in this thesis, the most effective way to counter such asymmetric iWar attacks is by holistically understanding one’s own complex social system and identifying leverage points, behaviours, structures, and mental models to be adjusted for increased system resilience.

Complex Social Systems Lessons Because complex social systems are inherently human, the lessons of the influencing and cognitive behaviours research are very clearly associated with the conduct of system analysis. The lessons derived from conducting systems analyses of the two case studies revealed the utility of this methodology in holistically understanding the structures, mental models, and behaviours of both own systems models and those of our adversaries for maximum effectiveness in the iWar. Furthermore, the review of systems literature demonstrated both the utility of applying systems thinking approaches over linear thinking in the iWar, as well as justifying the methodology employed in this thesis (Table 2).

In distilling the findings identified from the case system analyses, four key themes emerged that provided an analytical framework for determining the level of effectiveness of influencing activities. The first theme was identifying the indicative characteristics of systems operating in the iWar. In that, the level of resilience and whether a system was open or closed determined the levels of system effectiveness, self-awareness, and anticipatory responses to vicious or virtuous reinforcing feedback (Figure 3).

Linked closely to the characteristics were the associated system flows of information and feedback. The interconnecting relationships and interpersonal networks of complex social systems provide important leverage points, dissemination tools, reinforcing enablers, and nudges to change behaviour and opinions in the iWar. These linkages and flows also reinforce ‘in-group’ sense of community feedback loops and preference-orientated social contagions and biases in the two case systems (Figures 11 & 26). The efficient flows of such influencing information and positive feedback, using both hard and soft communications links, also reinforces system success and thereby increases the overall resilience of the system. As the case studies demonstrated, it is just as important to understand one’s own system characteristics, structures and flows as it is to analyse adversary systems in the iWar.

180 | P a g e 5249168 Cassandra Brooker Masters of Research The other two key themes for system success in the iWar were the maintenance of control and having a centrality of focus. This research indicates that adversary systems have seemingly grasped and prioritised both concepts accordingly. Having control was identified as a critical capability for both case systems in maintaining their COG (Figures 9 & 19). An increased loss of control observed in both case systems – over their narratives, image, credibility, content and situational awareness – contributed to their decline in influence and falling victim to certain system archetypes, such as a drift to low performance (Figures 16, 17, 22 & 23). Having a centrality of focus ensured the critical capabilities of means, resources and ideology for ISIS were appropriately, prioritised, supported and reinforced. This ensured all the individual system elements worked together as a united whole towards achieving the common purpose and maintaining influence stocks in a ‘success to the successful’ reinforcing loop.

The benefits of employing a systems approach in the iWar for holistically understanding oneself and an adversary system will be discussed in greater detail in the methodology section in the next part of this chapter. Part Two of this chapter describes how the research across the three interlinked disciplines has answered the gaps and provides solutions for Theoretical, Methodological (systems), and Practical iWar problems.

Part Two: How has the research answered the gaps? The research undertaken in this thesis aimed to address the gaps in knowledge and investigate solutions for improving Australia’s influence effectiveness in the iWar. The conduct of this research was to provide findings that confirmed the three initial propositions. 1. Systems thinking offers an effective approach to understanding complex adaptive social systems of ourselves, our adversaries, and neutral audiences, and provides a methodology for developing effective influence and counter-influence strategies for Western militaries and societies. 2. Understanding the effectiveness of influencing activities is imperative to appreciating how a complex social system adapts its behaviour based on balancing feedback. 3. Understanding psychological cognitive theories, with respect to influence effectiveness, is central to analysing how external pressures impact on a social system’s behaviours and mental models.

Accordingly, Part Two will aim to address the theoretical, methodological and practical gaps in knowledge by applying solutions, derived from the findings, to answer the research questions and improve democratic nations’ effectiveness in influencing.

181 | P a g e 5249168 Cassandra Brooker Masters of Research Theoretical The theoretical findings arising from this research link to the lessons highlighted in Part One of this chapter and inform the methodological and practical approaches recommended Part Two, which were also derived from the research. To demonstrate their utility, the theoretical research findings are presented in terms of strengths and weaknesses, i.e. the theory contributing either to influencing effectiveness or, conversely, understanding why systems have failed to compete effectively in the iWar.

Strengths for success The findings, which ultimately determine success in the iWar, were distilled down into a framework of key concepts, arising from the three research disciplines and as described in Part One of this chapter (Figure 28). These included the five key elements for ensuring influencing effectiveness, the four system characteristics contributing to iWar success, and the importance of critical thinking and understanding behavioural science theories and applying ‘cognitive warfare tactics’ to influencing strategies.

As the research showed, successful systems operating in the iWar understand the game-changing nature of modern hard communications technology for dominating the IE, manipulating heuristics, inundating media feeds, and creating online communities, echo chambers and filter bubbles (Tables 6 & 12). This research also identified the importance of using soft communication links to create a shared identity and sense of belonging among target audiences around common principles, ideology or events (Figs 11 & 26). Soft communications links were also instrumental in reinforcing the narrative (a critical capability) and developing a ‘sense of purpose’ in both case systems (Figs 12 & 24). Exploiting relationships, through emotional and social contagions, is an extremely powerful influencing tactic, not only for shifting beliefs and attitudes but also for motivating individuals to act (Jensen & Sear 2019).

To paraphrase Sun Tzu (Giles 2017): to succeed in the iWar is to win the war of the wits. The IE presents a multi-domain, multi-level, multi-spectrum media battlespace where long-term mental combat is fought for strategic influencing outcomes. This research has shown that to compete effectively, systems must be anticipatory, self-aware, resilient, creative, and willing to take risks. Successful influencing and manipulation of target audiences involves more than disseminating disinformation, propaganda, or stage-managed, risk averse content. Success in the iWar relies upon the fusion of the theoretical lessons provided by the three research disciplines in this thesis and the findings of the case systems’ holistic analysis. iWar strategies must engage audiences, dominate the IE, amplify the messaging, deflect attention from adversaries, undermine the 182 | P a g e 5249168 Cassandra Brooker Masters of Research capacity of adversaries to respond coherently, and maintain control – so as to steer, persuade, direct or activate target audiences in accordance with your will.

Weaknesses leading to ineffectiveness Through the analysis of narrative data pertaining to predominantly Western democratic systems and the system analysis of Clinton’s election campaign, this research identified weaknesses and common mistakes that undermine influence effectiveness. Notwithstanding the fact that Western nations, their militaries, and media outlets are hamstrung by democratic values, journalistic standards, adherence to the truth, and ethical targeting (Table 4); they have failed manipulate these obstacles to their advantage, failed to develop a centrality of focus, and therefore failed to keep pace with their iWar adversaries.

Western nations are cognisant of the ‘holy trinity’ of strategic communication in grey zone conflicts, that is: narrative, framing and ideology – however, often their well-crafted messaging is dislocated as they face ambiguity, opacity, uncertainty, and misinformation in the iWar. Their outputs, resilience, and system COG are further undermined when combined with linear thinking approaches to information operations (Table 2). This research shows that bureaucratic behaviours, such as risk aversion, competing priorities, misunderstanding the IO strategy, fixed mental models, and micromanaging the creative process; undermines successful influencing campaigns and has led to the demise of successful Western iWar systems (Sampson 2019; Watts 2018:204; Patrikarakos 2017:244).

This research shows that while Western militaries have recognised the shift of conventional warfare into the asymmetric information domain, there remains a weakness by democratic states to address the gaps in their strategic failure to dominate the IE and exploit hard and soft communications links. Technological advances mean the military information environment is now global, complex, dispersed, no longer isolated into military and civilian silos. The diffusion of power over information provides adversaries alternative ways to offset diminishing physical overmatch (J. Mark Services Inc. 2018; US DOD 2018:5). This enables enemy organisations, such as ISIS and Russia’s IRA, to affect strategic decision-making, undermine the legitimacy of Coalition forces, and has caught the military unprepared, lacking strategic guidance and a strong narrative, and unable to respond to high volumes of multi-channel, multi-source propaganda.

This research has also shown that unlike adversary systems that have control, resilience, and a centrality of focus; a lack of own system understanding and integrated approaches in the iWar 183 | P a g e 5249168 Cassandra Brooker Masters of Research are key weaknesses of Western democratic nations. The US DOD (2018:7) recognises that joint forces lack emphasis, policy, resources, training, and education to harness the full power of information. Specifically, the US DOD (2018:7-8) lists the following issues as undermining cohesive iWar strategies: an inadequate understanding of the civilian IE, a lack of effective inter- agency coordination, ineffective application of IO capabilities, doctrinal ambiguity, poor measures of effectiveness (MoE), a limited ability to recognise and understand narratives, insufficient authorities to execute, and a reluctance to acknowledge that physical capabilities create informational effects.

This research reviewed literature where authors encourage military influencers to apply critical thinking, different cultural lenses, and unbiased paradigms to their target system analysis and creation of PSYOPs content; in order to better understand adversary systems, the significance of relationships, the meaning of events, and to anticipate future behaviours (Lieber & Reiley 2016:47,51; Hall & Citrenbaum 2010:122-123; Blackmore 2003:22; Goble 2002:12,41; Buckle 2018:13,16,20; Brown 2019; Clark 2015). Unfortunately, non-linear approaches to IO are still relatively embryonic, practised by a minority of niche specialists; while own-system analysis is superficial, and planning, targeting and intelligence processes are still dominated by a conventional, kinetic mindset.

The lack of priority placed on iWar capabilities, creative solutions and behavioural economics, combined with failures in conducting holistic target systems analysis, are key weaknesses of Western democratic nations that were identified through this research. ISIS has created virtual safe havens in the psychological and sociological aspects of the human domain; which can only be countered through comprehensive engagement in the narrative space, and therefore, will require a major paradigm shift by Western military leaders, strategists, and policy-makers in the iWar.

The combination of these weaknesses of Western militaries and government institutions in the iWar, contrasts starkly with their adversaries who are bolder, accept more risk, are more anticipatory, and therefore can adapt to a rapidly changing IE and create political, social and military advantages exceeding their traditional combat power. The identification of the theoretical concepts arising from the findings of this research, contributes partially to answering the gaps and confirming the three initial propositions. However, this research shows that only in applying a systems thinking methodology and implementing the practical solutions derived from this research may Western military forces, and their governments, hope to catch up and become more effective contenders in the iWar. 184 | P a g e 5249168 Cassandra Brooker Masters of Research

Methodological – Systems Thinking This research has shown that systems thinking approaches provide a powerful analytical tool for holistically understanding ourselves and our adversaries, as complex social systems operating in the iWar. The methodological conduct of qualitatively modelling and comparing two contrasting case systems and the resulting analysis of those findings also determined the four key requirements contributing to system success in the iWar, being: the system characteristics, information flows, maintaining control, and having a centrality of focus. The systems thinking approach demonstrated in Chapters 4 and 5, enabled the holistic analysis of two contrasting systems’ key elements, structures, behaviours, linkages and flows contributing to their influence effectiveness. This research shows that a systems thinking methodology enables practitioners to locate responsibility and leverage points within a system to enhance the analysis of behaviours towards outside influences and in the design of triggering events to alter system behaviours (Meadows 2008:179).

Some paradigm shifts have recently occurred to improve influencing activities, information operations, and cognitive manipulation in the iWar. Despite the weaknesses identified in the theoretical findings of this research, Western militaries are now also increasingly embracing the benefits of a systems thinking methodology over traditional linear doctrinal approaches in intelligence analysis, targeting strategies, and operational planning. As Byrne and Uprichard (2012:5) explain, holistically analysing systems is fundamental to understanding complex causality and generating useful knowledge. As this research has shown, systems thinking helps to differentiate causal paths toward particular outcomes; as well as providing options for intervention that might produce desired changes in that system. Understanding adversary systems and identifying leverage points enhances the effectiveness of iWar strategies. Additionally, applying the same analysis and leverage to own joint force systems can increase awareness, rewire cause-effect relationships, shift mental models, address negative archetypes, improve resilience, and reinforce the purpose (Stroh 2015:148,155).

Adopting systems thinking approaches to iWar strategies enables the analysis of the interactions and interconnections between informational, physical, and cognitive domains. Understanding these interconnections provides valuable insight into target audience worldviews that frame perceptions, attitudes, and mental models and drive behaviours (US DOD 2018:4). The case system modelling and analysis conducted in Part Two of this thesis, demonstrated that the level of resilience or fragility of an iWar system can be identified, along with leverage points in their information flows, which provide a greater array of targeting options to the joint force (Albino et 185 | P a g e 5249168 Cassandra Brooker Masters of Research al. 2016:17). For example, the leveraging of counter-influence balancing feedback in ISIS’ system (Fig 15), weakened its anticipatory responses, impacted upon relationships, and undermined trust and credibility, which, in turn, increased the balancing feedback contributing to the system’s ‘loss of control’ and ‘drift to low performance’ (Figs 16 & 17).

This research has demonstrated the importance of having a ‘centrality of focus’, which is a key characteristic of successful iWar adversary systems. However, this lesson not only applies to the resourcing, structure, planning, and conduct of influence activities but also for driving systems thinking approaches to targeting adversaries and audiences in the IE. The West’s adversaries in the iWar have demonstrated this centrality of focus in systems thinking; both informally, as in ISIS’ and the IRA’s influence-focused system structures, behaviours and information flows, and formally, as demonstrated by China, where system-of-systems thinking pervades every aspect of conducting modern warfare as a unified whole (Engstrom 2018:113,119-120).

Conversely, this research has shown that whilst systems thinking can improve attacks on enemy capabilities, leverage their vulnerabilities and enhance iWar targeting; it also provides heightened understanding of own-force system responses and how negative feedback impacts on resilience. For example: during conflict, military force systems can suffer increased fragility if they rely on controlling the initiative or specific capabilities, positions, or materiel; or if they suffer from limited logistics, popular disaffection, or psychological vulnerability due to defeatism (Albino et al. 2016:17). This phenomenon was also demonstrated by the Clinton campaign’s misguided responses to delayed negative feedback (Figs 22 & 23). Conducting a holistic own systems analysis during the planning phases provides a mechanism for improving own-force resilience, cohesion, and responsiveness, while shortening the time and cost of learning, and improving chances of success (Goble 2002:5). Therefore, a holistic systems-based approach that considers the totality of functions performed by the own-force system is the only methodology that can enable reform and increase resilience, without creating unintended consequences or have negative cascading effects upon other friendly force systems.

This research has aimed to address the gaps surrounding Australia’s influence effectiveness in the iWar and specifically to demonstrate that systems thinking offers an effective methodological approach to understanding complex adaptive social systems of ourselves, our adversaries, and neutral target audiences. As Engstrom (2018:iii) asserts, warfare is no longer centred on the annihilation of enemy forces, rather, it is won by those who can disrupt, paralyse, or destroy the operational capability of an adversary system. As the benefits and lessons of systems thinking approaches in the iWar are increasingly understood, military doctrine is being updated 186 | P a g e 5249168 Cassandra Brooker Masters of Research accordingly. For example, the Target Systems Analysis (TSA) tradecraft handbook used by the ADF espouses utilising systems approaches to build a more holistic understanding of target systems. The handbook also explains that a systems approach is particularly valuable for understanding and contextualising complex systems and ‘wicked’ problems across the three domains of the IE. However, whilst systems thinking is being increasingly embraced in doctrine, the practical application of systems theory in the non-kinetic iWar battlespace and towards own- force influencing systems’ analysis is still largely misunderstood, misapplied or mismanaged at the operational and strategic planning levels in Australia. This research aims to provide recommendations and practical solutions for addressing this gap in Australia’s iWar capabilities.

Practical Practical solutions for improving Australia’s influence effectiveness in the iWar were derived from the theories and methodologies provided from the three research disciplines covered in Part One of this thesis, and the case study system modelling and analysis conducted in Part Two, and applied here in both military and civilian contexts. Traditionally, in Western democratic nations, the conduct of information operations and PSYOPs have been the domain of the military on operational battlefields. However, the lines between civilian and military, political and commercial, state and non-state have blurred in the iWar, everyone is a combatant whether they are aware of it or not. Implementing effective defensive iWar solutions now requires the involvement, education and engagement of the entire nation’s population. Whilst concurrently, the effectiveness of offensive iWar activities requires the military and government to better understand this asymmetric, non-conventional, complex IE and evolving social threats contained within.

Military applications Whilst the ADF is leading the way in its development of iWar capabilities, non-kinetic targeting, and implementation of offensive and defensive influencing strategies; this problem is now bigger than just what the military can affect at the operational level. A holistic national, strategic, political approach is required, involving the engagement and input of commercial entities, economic organisations, technology companies, law enforcement, aid agencies, and academia working alongside military and government systems, to truly address this new threat to national security and democratic society. Ehlers (2017:5) agrees that the implementation of diverse, information sharing IE planning groups, comprising a range of civilian and government organisations and individuals, is fundamental for resolving “wicked” problems, understanding the human terrain of the IE, and achieving iWar success. Ehlers (2017:4) further advocates for a

187 | P a g e 5249168 Cassandra Brooker Masters of Research whole-of-government (WoG) approach and close cooperation with Western allies, explaining the Department of Defence cannot and must not try to act alone in the iWar.

However, as the ADF and other government agencies will always lead the iWar effort, this section provides the practical solutions derived from the research that could be applied to immediately improve influence effectiveness. Most importantly, the macro-level holistic approach recommended above must first be applied to remove the disparate, stove-piped iWar elements contained in the national security ecosystem. Notwithstanding other government agencies’ iWar activities, military influence operations are divided into different strata, partitioned confusingly between: public affairs, information operations, PSYOPs, joint effects, CIMIC, human intelligence, offensive cyber, and non-kinetic targeting spread across three services. As Watts (2018:202-203) explains, during the Iraq war, the efforts of disparate elements were inconsequential, limited in effectiveness, regularly overlapped, and often proved contradictory. The consequences of isolating iWar capabilities into ‘cylinders of excellence’ can combine with other legal, communications, and technological issues to seriously undermine the critical capabilities of ‘authenticity’ and ‘controlling the narrative’ of an influencing system in the iWar (Figs 9 & 19). Therefore, information warfare and ‘soft power’ strategies must be included in initial preparatory and planning phases of operations, with collaboration at all levels, to ensure the lesson of having a ‘centrality of focus’ is applied by all stakeholders.

Western military targeting doctrine and operational planning is still primarily conventional, kinetically focused and risk averse, whereas adversaries in the iWar have re-orientated their concept of warfare to be all encompassing and unrestricted in their approaches across social, economic, cultural, and political lines. This research identified a paradigmatic shift is required by Western militaries towards adopting more creative approaches, alternative problem-solving methodologies, allowing freedom of critical thought, and removing restrictions on iWar practitioners’ unconventional ideas. As Chua (2018) explains, unlike businesses and non-state actors, militaries have been slow to adapt to the information age and exploit opportunities inherent across global networks. Chua (2018) adds that while some natural conservatism is understandable, an inability to adapt to the ever-changing IE condemns government organisations to irrelevance.

This research has demonstrated that holistic system analyses must be conducted, from an influencing perspective rather than a kinetic targeting perspective, of target audiences and adversaries in order to deliver behavioural change, identify leverage points, and inform the conduct of informational, attitudinal, and behavioural influencing activities. By analysing target 188 | P a g e 5249168 Cassandra Brooker Masters of Research systems holistically, their mental models, behaviours, and structures are better understood – which enhances non-kinetic targeting, counter-influence, appropriate allocation of resources, and strengthens own-system credibility when exposing adversary methods, deception and iWar tactics. Furthermore, understanding socio-cultural dynamics of target audiences and defining influencing factors by exploiting cognitive behavioural theory, particularly in the consideration of ‘System 1’ heuristics, is critical for effective influencing against the minds of decision makers and in creating desired effects.

Western militaries also face the challenge of trying to attain ‘information superiority’, by operating under the same paradigmatic planning cycles used to achieve ‘air superiority’, that is by having better capabilities and faster decision-making processes. However, applying this conventional approach to dominating the IE is problematic in that, with modern, rapidly evolving technology, media, and tools being available to anyone leads to a lack of control over other stakeholder systems and their associated hard and soft communications links – thereby dislocating any tangible technical or doctrinal advantage. This research has shown that to even attempt to control aspects of the IE and attain the level of information superiority enjoyed by their adversaries, Western democratic nations must understand own-force systems and discard linear thought patterns. The behaviours, mental models, reactions, and structures of own-force systems can be affected, leveraged, aligned and improved to enhance resilience, robustness and effectiveness in the iWar, despite any unpredictable actions of adversaries, negative feedback, or asymmetric impacts presented by the global IE.

In the iWar, the ADF must be innovative, creative, and address the problem using a two-pronged approach, i.e. to enhance the effectiveness of their influencing strategy (publicly) and to optimise their own influencing ecosystem (privately). As Stroh (2015:91) explains: “if you really want to understand something, try to change it” (Ibid:91), adding that the link between insight and effective change is critical. Solutions that are implemented without a deep appreciation for how a system operates and why, usually result in no observable difference or often make matters worse. By addressing the challenges of combining different perspectives, diverse interests, and disparate stakeholders to establish common ground, embrace diversity, uncover the ‘hidden part of the iceberg’, and improve the ability of the system to work together with a ‘centrality of focus’ is the single most practical solution to fixing systemic failures and refining Australia’s iWar approaches (Figures 1 & 28). Contributing to the effectiveness of new paradigms and approaches in solving iWar problems are the insights and understanding that can be generated by creating systems maps, which provide a visual form of story-telling around own-force and adversary systems (Stroh 2015:123). 189 | P a g e 5249168 Cassandra Brooker Masters of Research

However, information warfare strategies used in isolation will never compel a complete capitulation by the enemy. Coalition forces still require kinetic warfare capabilities across all domains, resources to conduct soft power activities (propaganda by deed), and the threat of economic or political leverage to back up any IO campaign and effectively impose their will on the enemy. Therefore, as modern militaries respond to growing threats, it is important to remember that war in the information age is not just warfare in the information domain (Chua 2018). However, alongside tough conventional measures, Western nations must improve civilian integration for informing asymmetric iWar tactics, for example: incorporating lessons from advertising agencies, psychologists, cyber experts, and social scientists. They must also conduct innovative, realistic training relating to: the weaponisation of information, creative messaging, manipulating behavioural economics, and exploiting social media. The line between civilian and military has blurred in the global IE, this has been recognised and exploited by adversarial elements, and now some democratic systems, such as the Israeli Defence Force, are recognising this new dynamic as an opportunity to regain the initiative in the iWar.

The ADF must leverage from the lessons learnt by other nations, as well as exploiting the power of networks, to develop capabilities in a more cost-efficient and operationally-effective way, particularly in the cyber domain, where the military cannot compete with the private sector in paying for talent (Chua 2018). The government and military must lift the shroud of secrecy surrounding iWar threats to the nation and engage the civilian populace with factual information and transparency to enhance resilience, as successfully demonstrated in Israel and the Baltic states.

The research has highlighted how Western democratic nations are falling behind in the iWar, how they are being out-communicated by their adversaries, as well as highlighting the multitude of interrelated challenges and spectrum of threats their militaries face in the contemporary information battlespace, with its contested norms and persistent disorder (IWD 2018:4; US DOD 2018:9). The IE is both a threat and force multiplier to Western democratic nations and their militaries. Traditional approaches, linear thinking, complacency, risk aversion and unwieldy bureaucracies are proving increasingly redundant in dealing with contemporary iWar challenges. There is a need for Western forces to blend emerging technological capabilities with socio- cultural analysis, in a sustained approach, to inform the integration of physical and informational activities and enable freedom of action in countering adversary influence (US DOD 2018:10). As the leading organisation in the iWar for the national strategy, the ADF must treat the IE like it would any key terrain, shape conditions by manipulating and leveraging the inherent 190 | P a g e 5249168 Cassandra Brooker Masters of Research informational aspects of the iWar battlespace, and ensure the intended narrative is controlled and able to mitigate any unintended interpretations of military activities (US DOD 2018:14-16). As General Mattis, NATO’s Supreme Allied Commander for Transformation, stated in 2009: “capturing perceptions is the new ‘high ground’ in today’s conflicts, as the moral is to the materiel as three is to one” (US DOD 2018:16).

Just as the national security apparatus operating in the iWar needs to develop innovative, critical thinking skills and non-linear, creative problem-solving strategies, so too does the wider civilian population. An informed, engaged citizenry will build socio-cultural resilience and strengthen national audiences against iWar attacks on democratic institutions, sovereign freedoms and the truth itself.

Civilian applications Much of the iWar literature, particularly articles analysing Russian interference in Brexit and the 2016 US Election, is dominated by commentators espousing the importance of an informed citizenry as the most effective defensive counter-influencing solution. This research, therefore, has identified practical solutions for improving Australia’s iWar effectiveness that can be applied to the civilian populace, specifically the strategies required to empower and inform the populace, encourage critical thinking, and build grass-roots resilience.

Hochschild and Einstein (2014:467) explain that political thinkers have always insisted on the importance of a knowledgeable public to ensure a successful democracy. However, despite such calls for an educated population, the general public frequently falls considerably short of reformers’ democratic ideal of critical thinkers. With the advent of modern communications technology and ever decreasing attention spans, critical thinking skills have atrophied across vast swathes of the population. Furthermore, in contemporary democratic political discourse, prominent public figures often find it in their political or personal interest to encourage the spread of misinformation and political ignorance, reinforced by filter bubbles, echo chambers, social contagions, and addictive feedback mechanisms – as demonstrated by Trump (Table 12) and the RW media during the 2016 US Election (Hochschild & Einstein 2014:468,472-474).

These iWar tactics encouraging misinformed political activity, disinformation, and manipulation of ‘System 1’ heuristics leads to increasingly engaged but ignorant audiences, who blur reality through an inundation of skewed content. This ultimately fuels vicious feedback cycles, reinforces partisanship, undermines democratic institutions, and dislocates the positive influence of ethical, truth-seeking actors, whilst simultaneously strengthening the influence stocks of adversaries. 191 | P a g e 5249168 Cassandra Brooker Masters of Research

It is very difficult to shift the paradigms of a misinformed person into the realm of informed political engagement, and democratic political actors seldom try (Hochschild & Einstein 2014:472). It requires a great deal of cognitive effort for people to question their biases, loyalties and experience; to address their social and emotional driven beliefs; research the facts, and then to publicly engage with others outside of their partisan in-group. As the case system analysis showed, Hillary Clinton tried and failed to engage audiences and was unable to encourage them to think critically about the long-term future. Based as it was on a platform of factual statements, detailed policy announcements and pragmatism, Clinton’s campaign failed to resonate with target audiences or achieve any significant media coverage. Clinton was unable to break through the ‘System 1’ barriers of gender bias, personality politics, emotional contagions or vilification levelled against her (Clinton 2018:222-223).

The importance of education and engagement with the population cannot be overstated for improving media literacy, critical thinking and state-wide resilience in the iWar. Singer and Brooking (2018:264) explain, information literacy is no longer merely an education issue but a national security imperative. The challenge in forcing targeted populations to dedicate cognitive effort and ‘System 2’ logical thought processes when engaging media messaging is not insignificant. However, by inculcating certain truths, demonstrating an objective reality, explaining how technology exacerbates the problem of disinformation, and exposing subversive iWar elements and their tactics; current and future generations of democratic audiences can be rendered more resistant (Canadian SIS 2018:29,87). Also, overcoming inherent biases using tools, such as the ‘Ladder of Inference’, helps people distinguish how their assumptions and conclusions were formed based on the reality or data they were presented (Stroh 2015:87).

In addition to this research showing that raising public awareness of misinformation and manipulation efforts, enhancing media literacy and critical thinking skills, and exposing enemy influencing activities and threats all contribute to improving national iWar resilience; other practical solutions in the civilian context also emerged. A number of recommendations for countering adverse iWar tactics and content arose from the literature, which apply to three functional areas of: the media, the government, and the technology companies (Canadian SIS 2018:47-49,66; Waldman 2019; Paul & Matthews 2016:10; NATO 2016:4-5; Watts 2018:253- 254; Dhanaraj 2018:6).  Media: Regulate and uphold journalistic standards, improve source appraisal and fact- checking, create media rating systems; and register and monitor media outlets, while blocking or sanctioning those found to be hostile. 192 | P a g e 5249168 Cassandra Brooker Masters of Research  Government institutions: Identify and unmask sources of disinformation, expose the threat and educate the citizenry, engage civilian specialists in WoG iWar planning, provide online media and social media analysis reports freely to the public, provide fact-checking applications and tools, develop unifying national narratives, learn from other countries’ experiences.  Tech companies: Implement measures to improve online monitoring and regulation, improve social media standards, provide controls and proactive responses to protect democracy, sovereign rights, to uphold laws, and counter the exploitation of their platforms for hostile purposes.

This research, particularly the Clinton campaign case study, has demonstrated the importance of Western democratic nations empowering their citizens and institutions to uncover the truth behind iWar content and in order to enhance the community’s ability to verify information. As the Canadian SIS (2018:29) reports, enemies, both foreign and domestic, using disinformation to undermine democracy and the rule of law must be confronted and exposed; adding: there are no front lines, there is no neutrality, the iWar is total. Had these practical measures been implemented during the bounded time periods of the two case studies; their system models, specifically the strength of their reinforcing feedback loops, may have been significantly altered (Figs 18 & 27).

Taking a systems thinking approach to countering the effects of adverse propaganda on the target audience, rather than the content, is also a powerful practical solution that enables the prioritisation of resources, hones targeted counter-measures and improves responsiveness (Paul & Matthews 2016:10). To better protect sovereign audiences, enemy intentions to force changes in attitudes or behaviours must be identified and exposed to effectively counter those tactics from a systems feedback perspective. This can be achieved by using leverage points provided by the structures, mental models, and relationships of both own and adversary systems. As discussed in the methodological section, holistically understanding own system elements, structures, linkages and behaviours is imperative for improving resilience, effectiveness, anticipatory responsiveness and success in the iWar. For example, if the Clinton campaign had conducted a holistic system analysis of both its target audience and its own system, it would have improved its situational awareness and understanding of the leverage points, and the effects of the counter-influence feedback would have been more obvious, thereby enabling more timely and effective responses.

Taking systems thinking approaches a step further to create civilian defence force sub-systems for deploying in the iWar battlespace, in support of the national effort, has already been successfully introduced in Estonia and Lithuania. In Estonia, the Cyber Defense League is a 193 | P a g e 5249168 Cassandra Brooker Masters of Research voluntary network of information technology professionals who can be mobilised to defend against cyberattacks (Chua 2018). Building upon the development of critical thinking skills, media literacy and influencing awareness among the general populace, is empowering them to become active contributors in defending the nation in the iWar. Watts (2018:183) even recommends arming the public with information and awareness and encouraging a grassroots online counter-insurgency to expose and challenge adversarial iWar tactics. More formally, civilian or reservist cyber warriors could be organised into units to compete against and overtake, diminish, or prevent adversary misinformation efforts, seize back the initiative and dominate the IE (Paul & Matthews 2016:10). These iWar defenders could hone their skills in dealing with iWar threats in their civilian lives without bureaucratic or hierarchical obstacles, while familiarising themselves with military networks, capabilities and tactics during combined training and strategic planning activities. As such, these cyber warriors, straddling both military and civilian domains in the asymmetric iWar battlespace, are likely to perform much more effectively than active duty personnel (Chua 2018). However, such a creative, non-conventional and revolutionary approach to enhancing national security may face resistance in the near term, as the iWar threat is not as menacing nor as urgent for Australia, as it is to the Baltic States.

Part Three: What does this mean for Australia? The lessons and solutions arising from the research findings, detailed in the first two parts of this chapter, have been delivered through the lens of providing recommendations for improving Australia’s influencing effectiveness in the iWar. Therefore, Part Three of this chapter will distil the results of the research into a discussion of what this actually means for Australia, specifically with respect to the challenges it faces in the future.

Whilst cognitive warfare is not a new concept, and influencing the will and morale of target audiences is a long held military and strategic tactic; the globalised, hyper-connected, accessible, media saturated IE presents a significant contemporary threat to democratic societies like Australia. Until now, Australia was a key beneficiary of the rules-based global order and geographically protected from conventional threats (Bienvenue et al. 2018). These contemporary threats against democracy, sovereignty, and truth, enabled by modern communications technology, and which blur the lines between East and West, civilian and military, and between innocent civilians and government agent (Clapper 2018:81), means Australia needs to urgently engage, compete, and assert its dominance in the IE.

The research and case system analyses have shown that Australia must adopt a holistic, fused approach to developing its iWar strategy and create a resonant, credible, strategic narrative 194 | P a g e 5249168 Cassandra Brooker Masters of Research that reinforces democratic values and truth, rather than being hamstrung by them. Coherent communication of Australia’s strategic objectives is critical to build trust, inform global and regional audiences, and reinforce Australia’s reputation as a strong and capable ally. Where the internet has connected the globe, it has also distorted the delineation between domestic and international affairs, hierarchical structures have been replaced by networks, and the power and influence of nation states have been diffused down to individuals and non-state actors (Chua 2018). Complacency, risk aversion, linear thinking, maintaining national security stovepipes, or only investing in technical or kinetic solutions, which can be easily dislocated; are no longer viable options for winning in the iWar (Maher 2018; IWD 2018:21).

As revealed by the system analyses in this thesis, to effectively influence and dominate in the iWar, Australia needs to identify the virtuous reinforcing feedback that needs to be maintained and cultivated for system success, as well as those critical capabilities vital to protecting Australia’s iWar centre of gravity, increasing resilience, and countering negative external impacts. Australian sub-systems operating in the IE, must analyse and improve own system behaviours, relationships, and resilience, be appropriately resourced, have control and a centrality of focus; as well as understanding the asymmetric, non-conventional systems and objectives of iWar adversaries.

Australia’s iWar sub-systems, both military and civilian, must be empowered, informed and enabled through effective hard and soft communications links to ensure rapid receipt of feedback, enhanced situational awareness, and anticipatory responses. Australia needs to capitalise on its greatest strengths: openness and information, and provide education that enhances critical thinking skills, truth-seeking activities, cognitive scepticism, and open-mindedness in order to counter enemy influence (Clapper 2018:50; Facione 2015:25-26; Wardle 2019).

Adapting to this new dynamic IE and applying the research lessons and findings to Australia’s future iWar approaches is not without significant challenges and obstacles. Priority has not been afforded to non-kinetic, non-cyber, non-technical information warfare capabilities; specifically regarding how information gets distorted, magnified and amplified; and how content engages, shapes, distracts, misleads, and potentially influences those that consume it and pass it on (Cave 2018). There has also been a complacency surrounding Australia’s iWar capability development due to a reliance on more powerful and proficient allies to provide solutions. The pace of change in the IE is so rapid and disruptive that action and reaction cycles of conventional military thinking struggle to comprehend emerging trends and threats, whilst at the political level, a paradigm shift

195 | P a g e 5249168 Cassandra Brooker Masters of Research is required from demanding quantifiable, kinetic battlefield outcomes to qualitative cognitive strategic successes in changing target audience behaviour.

The increasing trend towards wars fought in the digital realm, which are psychologically coercive in nature, technologically adroit, and increasingly complex, means Western democracies are already unwittingly embroiled in the iWar. In order to regain the initiative, Australia must address the significant challenges, overcome the obstacles, improve situational awareness, and counter unpredictable threats; while simultaneously maintaining democratic values, remaining transparent in messaging, playing catch up, and rapidly learning lessons arising from balancing feedback. This research has provided some theoretical, methodological, and practical solutions for addressing this gap in Australia’s influencing effectiveness, however future research is required to further develop enhanced solutions to Australia’s unique future challenges.

Challenges The challenges facing Australia in addressing its influencing effectiveness in the iWar can be classified into the following categories, which also correlate to three of the critical capabilities shared by both case systems: Resources, Control, and Ideology (Figures 9 & 19).

Resources Deploying iWar tactics in cyberspace is a relatively inexpensive endeavour, as proven by the disproportionate power imbalance enjoyed by individuals and non-state actors conducting attacks against conventionally powerful nation states, using low-cost equipment and cheap networks from undetectable locations. However, improving iWar capabilities in centralised, stove-piped, risk averse, hierarchical bureaucratic systems, such as the ADF, will prove resource intensive and face resistance to the associated cultural change. As the findings of this research uncovered, having a ‘centrality of focus’ is a key characteristic of successful systems operating in the iWar. Flowing on from this honed focus are the benefits of: having a shared purpose, clear objectives, effective communication links, creative freedom, and the appropriate allocation and use of resources and capabilities to achieve those goals. Therefore, a clear paradigm shift is required in Australia, and specifically by the ADF, in adopting unconventional approaches to the iWar, delineating responsibility, and resourcing those strategies accordingly.

A holistic WoG approach to developing Australia’s iWar strategy is the most effective way of bridging the gaps in capability caused by the blurring of boundaries between political, military, civilian, and commercial information domains. As Chua (2018) contends, the military should leverage the power of networks with media professionals and technology companies in 196 | P a g e 5249168 Cassandra Brooker Masters of Research developing effective influencing activities and supporting cyber technologies such as artificial intelligence and big data analytics. The ADF should also experiment with methods to integrate emerging unconventional approaches with existing military capabilities to support the tactical fight in the information domain. Working with civilian and commercial media and advertising companies opens up the global realm of the IE to government and provides exponential access, means and reach to better compete against adversaries. However, there is no requirement for vast expansion or investment into another new branch of iWar capability, the Government just needs to re-role, re-allocate, and reprioritise existing manpower, units, capabilities and technologies to address the challenges associated with reasserting dominance in the iWar.

A continuing focus on conventional, kinetic targeting and linear, doctrinal approaches to imposing Australia’s political will, means that any allocated resources will effectively be wasted; as traditional military capabilities are being increasingly dislocated by contemporary, agile systems dominating the contested global IE. Therefore, any subsequent degradation in conventional military effects and situational awareness will see increased emphasis placed on the need for offensive and defensive, technical and non-technical iWar solutions; as well as increased demand for the rapid development of new doctrine, tactics, techniques, and procedures for the employment of these strategies (Chua 2018; Bray 2016). Additionally, consideration is required surrounding the allocation and prioritisation of resources to support more sophisticated iWar intelligence collection and analytical capabilities, which are focused on cognitive warfare, behavioural sciences, leveraging systems, and measuring influence effectiveness. The priority of resourcing in the iWar should be allocated to capabilities conducting the functions of: find, listen, respond, influence.

Any half-hearted, ill-prepared, risk-averse efforts lacking a ‘centrality of focus’ and prioritisation of resources, to compete in this new congested iWar; will see the continuation of adversary dominance in communication manipulation, information superiority, and the undermining of social cohesion, political institutions, and democratic structures. This research has shown the competition for influence in the global IE is becoming increasingly contested, congested, and concentrated, and the ability of Australia’s adversaries in accessing iWar resources represents a significant future challenge. Any commercial, private or non-state actor can now purchase a disinformation campaign and associated resources, such as trolls, bots, or metadata, from iWar contractors (Singer 2019).

Additionally, influence operations designed to swing public opinion require relatively few resources to achieve significant effect – by taking advantage of freedom of speech and using the 197 | P a g e 5249168 Cassandra Brooker Masters of Research media to amplify their message; it is, therefore, much easier to carry out such attacks than to prevent them (Uren 2017). Cave (2018) explains that complacency surrounding this new threat environment and external interference into political systems and social structures is a dangerous oversight by Australia, and questions whether Australia is actually prepared to face a deliberate, strategic, multi-platform iWar attack. As Government, and the ADF, have traditionally had a risk- averse approach to online engagement and have focused on cybersecurity rather than actual content, effects, or information integrity; vast resources have been allocated to protecting secrets and developing cyber technologies rather than iWar solutions (Cave 2018; Chua 2018).

The final resource challenge faced by Australia is finding the political will to implement the practical solution of harnessing the civilian population as a grassroots sub-system for deploying alongside official systems operating in the iWar battlespace. Whilst improving education, information, critical thinking and exposing adversarial iWar tactics, are approaches recommended by numerous commentators for developing a well-informed, engaged and resilient populace; the actual public exposure of national influence activities faces the difficulties of information classifications to protect sources and methods (Uren 2017). Taking this information sharing, transparency, and trust to the next level by employing the general public as active players in the national iWar strategy, may be beyond comprehension for Australia to achieve in the near term and therefore lacks any urgency of implementation – because, unlike the Baltics and Israel, Australia is not facing imminent attack. Conversely, academic environments that pride themselves on honesty, reflexivity, and ethical research standards may be unable to ethically reconcile with supporting national iWar strategies that weaponises information, breaches privacy, or utilises social science research for targeting purposes (Buckle 2018:25). To address these challenges, Australia must implement the recommended practical solutions described in Part Two, starting with a paradigm shift across the national security ecosystem.

This strategic shift in focus and the associated reallocation and re-prioritisation of resources to support influencing effectiveness in the iWar, in a national combined effort, would clearly be met with resistance, obstructionism and criticism. Additionally, Australia’s adversaries would seize upon, exploit and amplify the associated polarisation and arguments surrounding any new iWar approaches, non-kinetic prioritisation, and associated changes to resource allocation and defence spending (Chua:2018). This is why the challenges in control and ideology must also be addressed in conjunction with any resource allocation reorientated towards an iWar centrality of focus.

198 | P a g e 5249168 Cassandra Brooker Masters of Research Control Control, as a critical capability to achieving effective influence in the iWar, relates to internally controlling system behaviours, processes, responses, and outputs; rather than attempting to control the activities of other stakeholder systems operating in the IE. The case studies demonstrate (Figures 9, 18, 19, 27), that losing control over internal sub-systems and system iWar elements such as: the narrative, messaging, credibility, image, security, knowledge, and responses can escalate negative reinforcing feedback, which ultimately undermines influence stocks and contributes to a drift to low performance archetype. Holistically, understanding and addressing own system problems arising from the hidden part of the iceberg (Figure 1), not only ensures control is maintained but improves system self-awareness, responsiveness and resilience (Stroh 2015:138).

Whilst the ADF conducts impressive target system analysis against its enemies and tangible threats that can be effected kinetically, there remains a glaring gap in the self-analysis of own force systems operating in the unconventional, non-kinetic iWar space. This oversight results in a loss of control in the coherence of Australia’s joint force influencing plan, undermines the effectiveness of the resulting disparate, clumsy and risk-averse information operations, and ultimately weakens the ADF’s centre of gravity in the iWar. At the strategic level, government conducts even less self-examination of its effectiveness as a system influencer in the global IE. The associated failures in: developing a resonant strategic narrative, providing iWar leadership and cohesion, fusing influencing planning strategies at a WoG level to overcome stovepipes, and delegating responsibilities for defensive and offensive IO; results in a loss of control over Australia’s narrative, credibility and domination of the IE. The weaknesses in Australia’s control mechanisms at the strategic and operational levels were highlighted in Part Two of this chapter, along with the provision of solutions to both address these gaps and the greater challenge of losing control in the iWar and an increase in the associated negative reinforcing feedback.

After relinquishing the misapprehension of any control of the global IE that Australia may be labouring under, to then actually re-gain internal control through an own system analysis and associated behavioural change, leads logically to the application of systems thinking approaches for strategic iWar planning, to further reinforce Australia’s control and power feedback loops. Stroh (2015:168,195) advises to apply systems thinking for prospective strategic planning and performance evaluation, in order to take advantage of the insights that own system reflection provides, and to ensure more anticipatory responses, continuous learning, goal achievement, and improved situational awareness, all of which contribute to greater control in the unpredictable iWar. As recommended in Part Two, by taking a systems approach to future iWar strategising 199 | P a g e 5249168 Cassandra Brooker Masters of Research also enables more efficient and effective collective planning, delegation, and resource allocation; the active engagement of all sub-systems and stakeholders; and ensures appropriate responses are formulated to negative feedback and attacks. To assert control in the iWar, Australia must implement a coordinated, whole-of-government, strategic communication plan, grounded in systems thinking, that ensures the nuanced distinctions between using information to inform, educate, persuade, or influence, depending on the audience, are identified accordingly, resourced appropriately, and targeted effectively (Lieber & Reiley 2016:52-53).

Addressing the aforementioned self-awareness gap by conducting an own-force system analysis and implementing the practical solutions discussed in Part Two, would lead to improvements in Australian systems’ resilience and control over own-system responses in the iWar, The cringing reluctance by Western democratic nations to expose and publicly address iWar threats and activities being conducted by other nation states, dilutes the success of counter-influence efforts, fails to hold those states to account, undermines any education measures designed to inform the population, and contributes to Australia’s loss of control (Singer 2019). In choosing not to offend China or Russia, and in hiding or overclassifying information regarding iWar threats, Australia effectively hands all power and control to its adversaries and reinforces their influence stocks. Singer (2019) recommends democratic nations unify globally in sharing threat information and devising multi-lateral strategies to expose and respond to iWar attacks. Australia should work with other Western nations to implement transparent, overt, cooperative international approaches, based on facts and evidence, to raise democratic system resilience, improve threat awareness across democratic societies, and regain control over own system iWar activities and responses. Such approaches would concurrently undermine the credibility and integrity of adversaries in the IE. Improving control over own system responses in the iWar, has the added benefit of increasing understanding of adversary aims, trends, leverage points and activities; thereby enabling enhanced detection, analysis, anticipation, and counter-measures.

Such control measures must be devolved across all of Australia’s systems and sub-systems operating in the IE, it is no longer the sole responsibility of the government or military to fight independently in the iWar. Social media companies are no longer just tech companies, they must holistically analyse their own systems through this new threat lens and develop proactive solutions to regulate weaponised content in the media ecosystem and counter hostile influence, including implementing the practical solutions discussed in Part Two. Likewise, consumers must be encouraged to improve their collective awareness and critical thinking skills in their consumption of content, in exposing and reporting hostile elements, and pressuring media and tech companies to act. As Wardle (2019) explains, understanding how we are all subject to iWar campaigns—and 200 | P a g e 5249168 Cassandra Brooker Masters of Research might unwittingly participate in them—is a crucial first step to fighting back against those who seek to up-end the target society’s sense of shared reality.

Ideology In order to address future iWar challenges, ensure the success of a coordinated, whole-of- government, strategic communication plan, and implement the associated resource allocation and budget spending to support that strategy; Australia must develop a strong, resonant, and credible narrative. This strategic narrative must unify the nation ideologically, build trust – both internally in the community and externally with neighbouring states and allies – and effectively enhance state-system resilience in the iWar. Creating an inclusive national ideology that generates trust, will require increased transparency on matters of national security and a clearly articulated strategic narrative. Australia must overcome cultural cringe or ‘tall poppy’ criticisms and formulate a compelling narrative at the political level, which engages audiences by telling a positive story incorporating democratic values, national cultural pride, mateship, and strength in unity and diversity. Any strategic-level negative, divisive messaging or cheap political point scoring in the iWar creates socio-cultural fissures, reinforces negative feedback cycles, and hands power and influencing success to Australia’s enemies.

As demonstrated by the findings of this research, influencing success in the iWar is reliant on having a strong narrative, employing emotional devices and contagions, ensuring authenticity is prevalent in messaging, engaging target audiences by creating a sense of community, and reinforcing all these tactics using inundation. War is an extension of politics, therefore the government needs to take a leadership role and set the narrative at the political level, champion Australia’s ideology, encourage cultural change and a paradigm shift, and unify isolated iWar efforts into an overarching WoG framework combining capabilities from various information domains. As Bienvenue et al. (2018b) explain, the core purpose of cognitive warfare is to erode trust and undermine social cohesion therefore, adversaries specifically target the credibility of narratives which underpin open, rules-based, liberal democracies.

However, there are significant challenges with implementing a national ideology and associated narrative within democratic systems. Meadows (2008:169) highlighted some of these challenges through her rhetorical questions of: “how is it that when exposed to the same information, different people absorb different messages and draw different conclusions?”; “how can you key a feedback loop to qualities you can’t measure, such as values, rather than to quantities you can?”; and “what are values and where do they come from? Are they universal, or culturally determined? What causes a person or a society to give up on attaining ‘real values’ and settle for cheap 201 | P a g e 5249168 Cassandra Brooker Masters of Research substitutes?” (Ibid:169). These questions expose the significant cognitive effort required by decision-makers developing Australia’s iWar strategy. They face uncertainty exposed by the current gaps in own system understanding, being out-communicated ideologically by adversaries, lacking a centrality of focus, and due to the unpredictability of the IE and its associated feedback loops. Additionally, iWar threats are continuous, persistent, harder to identify, harder to stop, and part of everyday life for increasingly connected societies, such as Australia (Bienvenue et al. 2018b).

In creating messaging and a narrative espousing Australia’s ideology, decision-makers must also be cognisant of the potential for IO tactics to backfire. Whilst a lack of effective deterrents or repercussions and a blurring of the line between war and peace undermines Australia’s ability to counter iWar attacks on society, business and politics; fear-mongering will only amplify anxieties, fuel conspiracies, and degrade trust in quality-information sources and institutions of democracy (Wardle 2019; Clinton 2018:371). Additionally, whilst offensive state-level iWar messaging needs to be faster, louder and dominate the IE to counter adversary content, caution must be taken not to accidentally offend the target audience (Lieber & Reiley 2016:48). As demonstrated by this research, increased socio-cultural division, partisanship, communications technological advances, and a degradation of journalistic standards all contribute to Australia’s vulnerability in the IE – with the increased susceptibility of the population believing hostile iWar content over friendly strategic messaging (Benkler et al. 2018:268; Chua 2018). Wardle (2019) explains, there are no permanent solutions to weaponised narratives. In order to address this ideological challenge, Australia must adapt to this new normal and create new habits and strong strategic communication strategies based on positive reinforcing feedback, in order to build resiliency against a disordered information environment.

This research has highlighted the future challenges Australia faces in the iWar, specifically in relation to resourcing, maintaining control, and strengthening strategic narratives based on a resonant ideology. This research has also contributed a number of theoretical, methodological and practical solutions for addressing the gaps in Australia’s influencing effectiveness, both immediately and in facing future challenges. However, further research across all three research disciplines is required to provide: a deeper understanding of shifting complex dynamic of the contemporary IE, a holistic appreciation of Australia’s iWar system and the systems of adversaries and target audiences, to identify further influencing strategies tailored to exploit Australia’s strengths whilst exposing enemy vulnerabilities, and to measure the effectiveness of those implemented solutions.

202 | P a g e 5249168 Cassandra Brooker Masters of Research Future Research Just as Australia must reorientate, re-think, adapt to the ‘new normal’ of the contemporary IE, and act according to the recommendations arising from the findings of this research; academia and iWar practitioners must continue exploring this emerging research discipline. There are two key areas of future research into influencing effectiveness in the iWar that were beyond the scope of this thesis, comprising: the conduct of Qualitative Comparative Analysis (QCA) on the case systems, and determining appropriate measures of effectiveness (MoE) for IO, which is a clear gap highlighted across the iWar literature.

Whilst the underlying philosophy of QCA, and its associated fuzzy set logic, were considered in the methodology in Chapter 3 to determine key configurations and qualities for inclusion during the narrative analysis; the actual implementation of QCA to measure research outcomes was beyond the scope of this thesis and the capabilities of a solitary researcher. The application of QCA would provide an interesting pathway for future iWar research, by expanding upon this methodology and converting qualitative data to quantitatively measure influence effectiveness stock levels of complex social systems.

Additionally, much of the grey literature espoused the need for clear and answerable measures of effectiveness for influencing activities in the iWar. A military or national-level influencing strategy is irrelevant if the enemy thinks something else is occurring, it is ignored, or target audiences are completely unaware of the shaping activity being conducted. These failings were demonstrated starkly by the disinterested reception or outright disregard of Hillary Clinton’s election campaign’s messaging strategy and policy announcements. Additionally, improving MoE contributes valuable feedback to enhancing own system resilience and outputs, as well as providing situational awareness and holistic understanding of iWar impacts on adversary systems. As Bray (2016) explains, information warfare is an enabler to traditional warfare domains but its contribution is hard to measure, and can vary substantially between conflicts and even within a single conflict.

Qualitative Comparative Analysis Gerrits and Verweij (2013:166) contend that many methods were used in research on complexity and one of these was qualitative comparative analysis (QCA). QCA techniques allow the systematic comparison of cases, and its strength is the ability to examine complex causal configurations, where particular combinations of factors are needed to explain a case. Consequently, QCA offers a bridge between more traditional quantitative analysis and case studies; and its underlying logic is strongly linked to narrative analysis (Rihoux & Ragin 2009:5). 203 | P a g e 5249168 Cassandra Brooker Masters of Research Therefore, the application of QCA and fuzzy set logic would be the most appropriate approach for measuring the findings and validating the methodology of this case-based comparison research; however, as discussed, the actual conduct of QCA in quantifying measures of effectiveness is beyond the scope of this thesis.

If a variation of QCA was to be conducted against any related research extending this methodology in future, applying fuzzy set logic membership scores to the levels of influence effectiveness of the various feedback loops in each system case would be the recommended approach. As Rihoux and Ragin (2009:12) explain: “it is perfectly possible to work with ‘subjective’ or ‘qualitative’ data using QCA...the only practical requirement is to be able to transform the data into categories or numbers” (Ibid:12).

QCA using fuzzy set logic does not view individual variables in isolation nor use a variable-based analysis to understand what has occurred, rather it supports the logic of conducting case-based comparisons, limited to a relatively small number of observations, and therefore, QCA’s underlying philosophy validates the configuration-focused methodology employed in this thesis. Gerrits and Verweij (2013:167) confirm that QCA is particularly suitable for analysing the complexity of reality presented by systemic case-based methods, because it is able to account for the contingency of social phenomena and allows for an exploration of multiple causal configurations. Byrne (2005:101) agrees that a comparative complexity method is needed for comparing cases within the social sciences and fuzzy set QCA certainly provides that method.

Measures of Effectiveness Joint IO doctrine (US DoD 2014) highlights the importance of measures of performance (MoP) and MoE in helping accomplish the assessment process by qualifying or quantifying the intangible attributes of the IE. Additionally, and specific to this research, MoEs provide valuable feedback mechanisms to measure system performance, influence stocks and resilience, and could also inform fuzzy set logic values for future QCA applications.

The US DoD (2014) states that MoEs should be observable, to aid with collection; quantifiable, to increase objectivity; precise, to ensure accuracy; and correlated with operational progress, for timeliness. However, useful tangible indicators are difficult to identify; resource intensive in the collection, management and analysis of the data; and problematic to apply to global, civilian, or commercial contexts. Similarly, UK Joint IO doctrine outlines four stages to the MoE process, which are also extremely demanding of man-power, collection assets, intelligence resources and time. These four stages comprise: identifying the immediate impact of an IO activity, tasking 204 | P a g e 5249168 Cassandra Brooker Masters of Research intelligence assets to detect MoE indicators to confirm long-term responses, measuring the cumulative effect over time of an IO plan, and inform future IO planning (Blackmore 2003:24). This doctrine lacks specific procedures to support this process, ignores the issue of extensive resources involved, fails to address the significant challenges this poses to IO staff, given the high competency and experience levels necessary to achieve this. This doctrine is completely inappropriate for the contemporary, non-conventional, global IE and impossible to apply at the strategic WoG level.

Lieber and Reiley (2016:50-51) recommend social science theory–based MoE designed to measure attitudes and opinions, using qualitative interviews, focus groups and quantitative (survey) instruments. Again, these methods are time consuming, represent a small proportion of a target audience, are highly resource intensive, and by their own admission, the resultant attitudes and opinions identified can predict but never cause behaviour. Fawcett (2018:4) also highlights the difficulties of assessing the success of effects-based operations on enemy systems without effective intelligence, surveillance or reconnaissance (ISR) efforts and where holistic multi-level understanding and a centrality of focus is lacking.

Western democratic nations do not have the same freedoms of their adversaries in conducting extensive, multi-platform, multi-media, multi-lingual, influencing tactics without any care or concern regarding the success of those tactics. Hostile systems, such as Russia’s IRA, inundate the IE with all manner of iWar content and merely hope that some of it has an effect in creating disorder and division. Unlike Western nations, they are unconcerned about credibility, ethics, or privacy, and are unhindered by media standards, global rules-based order, and democratic values. In line with this contrasting dynamic, attempts to measure social media effectiveness continues to prove an elusive task for Western democratic nations, leading to an overreliance on quantitative trend data that can be manipulated by hostile systems (Lieber & Reiley 2016:49).

Therefore, Western military IO doctrine, approaches and associated MoE must be reinvented under a modern paradigmatic lens to practically address the inherent problems presented by the new complex, globalised, interconnected iWar battlespace. The military no longer holds the monopoly on the conduct of IO and cannot continue to operate in isolation nor apply conventional MoE to unconventional operations. Accordingly, there is a major gap in academic research on developing holistic, strategic, geo-political MoE that contribute to improved performance of democratic national iWar plans, the appropriate allocation of responsibilities and resources, and in improving situational awareness around system effectiveness.

205 | P a g e 5249168 Cassandra Brooker Masters of Research Future research into the utility and formulation of new, appropriate MoE for Australia in the iWar, would benefit from employing a systems thinking approach. By logically assessing iWar adversaries, as complex social systems, to identify weaknesses, leverage points, centres of gravity, and critical capabilities, would provide tangible indicators of MoE, and garner improved understanding of their targetable system vulnerabilities (Fawcett 2018:6). Conversely, future research may determine that the monitoring of MoEs in the rapidly evolving iWar context is unfeasible and a waste of time, effort and resources – as adversaries continue to out-communicate and react faster than democratic nations in the global IE.

Conclusion This chapter has discussed how the findings of the research contribute to addressing gaps in the current literature where systems analysis approaches to the iWar are misunderstood and underutilised by the ADF, and provided an analytical argument for what makes certain influence activities effective in the contemporary, complex, globalised iWar environment. Researching all three disciplines as they relate to each other, our adversaries, and the contemporary iWar; as well as analysing what those findings specifically mean for Australia, had not been attempted until now.

Part One of this chapter discussed how much is known across the three research disciplines as derived from the findings of the literature review and shown through the case systems’ analyses. The specific lessons arising from the three research disciplines of: influence, cognitive behaviours and systems thinking were identified and considered alongside the findings of the contrasting case systems, before being distilled into key discussed in relation to Australia’s specific iWar context.

Because the disciplines of influencing activities, cognitive behaviours, and systems thinking have all been impacted upon by the interconnected global IE of cyberspace, research into the effects of this contemporary dynamic, and the associated impacts on the two case systems, was required to address the gaps in the accepted knowledge. Therefore, Part Two provided solutions for addressing the gaps identified by the research and was segregated into: Theoretical (highlighting the strengths and weaknesses relating to influence effectiveness), Methodological (systems thinking), and Practical (military and civilian applications).

Part Three discussed what the findings of this research means for Australia by highlighting the associated future iWar challenges facing the nation relating to resources, control and ideology, and summarised the solutions recommended in Part Two into iWar approaches Australia must consider in addressing those challenges. The chapter concluded with a discussion around future 206 | P a g e 5249168 Cassandra Brooker Masters of Research research opportunities that are beyond the scope of this thesis. The two key areas recommended for future research were the conduct of QCA to measure the causal configurations of the contrasting case systems, and determining appropriate measures of effectiveness for monitoring Australia’s influence effectiveness in the contemporary iWar.

207 | P a g e 5249168 Cassandra Brooker Masters of Research Chapter 8 – Conclusion No matter how enmeshed a commander becomes in the elaboration of his own thoughts, it is sometimes necessary to take the enemy into account.

Winston Churchill

Introduction The rise of influencing activities as the new ‘sharp power’ in information warfare (iWar) and a blurring of lines between political, civilian, and military information environments (IE) means Western democracies and their citizens are unwittingly embroiled in contemporary war in the information domain. The West has been complacent, reactive, and risk averse in dealing with this new frontier of hybrid political warfare and, as a result, democratic societies are being manipulated, exploited, and thoroughly out-communicated by their adversaries. In Australia, through complacency or wilful ignorance, the ADF has been largely left to deal with this problem in isolation. Its stove-piped, doctrinal, linear approaches are unsuited to addressing the unstructured, evolving, complex social problems of the iWar. Additionally, as identified from a review of the literature, there are key gaps relating to the three research disciplines covered by this thesis in systems thinking, influencing activities and behavioural science.

Therefore, the aim of this research was to investigate solutions for improving Australia’s influence effectiveness in the iWar. This thesis comprised three parts and covered three different research disciplines, comprising systems thinking, influence effectiveness, and behavioural science to achieve the study aim. The fusion of these three research disciplines was a unique approach which enabled a deeper understanding of cognitive warfare, enemy systems, effective influencing tactics, and adversary aims of achieving maximum psychological effect through exploitation of heuristics, mental models, and cognitive biases in the iWar.

This chapter completes Part Three of this thesis by providing the conclusions identified in response to the aim, which arose from the discussion of the research findings in Chapter 7. These conclusions demonstrate how the research and analytical findings fulfilled the aim and informed the associated recommendations for improving Australia’s influencing effectiveness as a complex social system operating in the iWar. This chapter concludes with a synopsis regarding the significance of this research and accounts for its contribution to Australia’s future iWar capability development.

208 | P a g e 5249168 Cassandra Brooker Masters of Research Conclusions The three arguments driving the research contained within this thesis highlighted the gaps in the iWar literature, shaped the discussion pertaining to the analytical findings, and contributed to the formulation of the conclusions to address the aim. The three propositions were: 1. Systems thinking offers an effective approach to understanding complex adaptive social systems of both ourselves and our adversaries and provides a sound methodology for developing effective influence and counter-influence strategies for Western militaries and societies. 2. Understanding the effectiveness of influencing activities is imperative to appreciating how a complex social system adapts its behaviour based on balancing feedback. 3. Understanding psychological cognitive theories, with respect to influence effectiveness, is central to analysing how external pressures impact on a social system’s behaviours, reactions, and mental models.

Chapter 7 discussed the findings arising from the three research disciplines and the case study comparison, and distilled them into a framework of key concepts that ultimately determined success in the iWar (Figure 28). These included the five key elements for ensuring influencing effectiveness, the four system characteristics contributing to iWar success, and the importance of understanding behavioural science and applying ‘cognitive warfare’ to influencing strategies. The significance of the theoretical, methodological and practical findings discussed in Chapter 7 led to a number of key conclusions being drawn across the three research disciplines.

Influence Effectiveness Conclusions Understanding the effectiveness of influencing activities is imperative to appreciating how a complex social system adapts its behaviour based on balancing feedback. Therefore, the first conclusion supporting the research aim is that Western democratic states, such as Australia, must adapt contemporary non-kinetic iWar approaches to incorporate the theory and master the practicalities relating to the five key themes of narrative, emotion, authenticity, community, and inundation. The literature review identified ten key themes that proved most effective in the conduct of influencing activities (Table 3), which were further reinforced during the systems comparative analysis, and overlapped with the findings related to the research into cognitive behaviours and nudge theory. Those ten concepts were distilled further into a framework of these five important themes required for optimal system feedback flows and overall iWar success, as demonstrated by the analysis of the two case systems.

209 | P a g e 5249168 Cassandra Brooker Masters of Research The second conclusion is that the control and exploitation of hard and soft communications links should not be underestimated as a critical enabler in the iWar. These links provide the ways and means of producing, disseminating and reinforcing the feedback associated with effective influencing tactics identified in the research. In the two case system analyses, the interlinkages and relationships relating to ideology, narrative, community, and communication were defined as key system elements underpinning the CoG, due to their importance in maintaining influence stocks, flows and function.

The third conclusion relating to influence effectiveness relates to the importance of having freedom of creative action in the iWar. Systems able to swiftly respond to unprecedented volumes of content, exploit emerging technologies on multiple platforms, and employ flexibility and inventiveness in their tactics enjoy greater influencing success, positive reinforcing feedback, and anticipatory levels of resilience. The research showed that linear thinking and conventional, risk-averse, doctrinal approaches employed in the asymmetric, rapidly evolving iWar battlespace undermines the influencing effectiveness of Western nations and their militaries, and who are also bound by ethics, democratic values, bureaucratic hierarchies, and privacy, security and targeting legalities (Table 4). This lack of freedom available to democratic states’ iWar sub-systems also restricts their strategy options, reach and access, as well as overwhelming the sub-optimal allocation of manpower and resources, leading to a downward spiral of ineffectiveness. Additionally, a level of complacency and inaction in the iWar, combined with stove-piped efforts by disparate agencies, leads to a ‘drift to low performance’ archetype, with any future reactions being ineffective, misguided and essentially dislocated.

Behavioural Economics Conclusions The manipulation of human behaviours, cognitive processes and mental models, has been termed “cognitive warfare” and takes the weaponisation of information a step further into the realm of “neuro-weapons” (Bienvenue et al. 2018a). Therefore, understanding psychological cognitive theories and exploiting human behaviour, is central to analysing how external pressures impact on a complex social system’s behaviours and mental models, and also for achieving influence effectiveness in the iWar. The importance of cognitive theory in the iWar was demonstrated by correlating linkages between the concepts extracted from the influence literature and the behavioural literature (Table 6), and which were distilled further into two key behavioural conclusions during the case system analyses, to ultimately answer the aim of the research.

The first conclusion relating to behavioural economics is realising the importance of identifying and manipulating System 1 and System 2 heuristics for maximum influence effectiveness in the 210 | P a g e 5249168 Cassandra Brooker Masters of Research iWar (Table 5). The psychological manipulation of target audiences, use of social engineering, and heuristic exploitation has enabled adversaries to reinforce their influence, dominate the IE, and effect behavioural change within democratic society. Exploiting System 1 and 2 heuristics, such as biases, personalisation, a ‘sense of purpose’, ‘in-group’ dynamics, and environmental cues proved very powerful tactics in strengthening positive feedback loops and building influence stocks in the two case systems.

The second conclusion, which is not unique to this research, is the importance of employing critical thinking over linear thought processes in the iWar. As adversaries of the West increasingly exploit biases in societies that cannot be turned off, and as audiences become increasingly gullible to disinformation, fail to exercise conscious doubt, spend excessive time on social media, and suffer from ever-shortening attention spans – this lack of critical thought has undermined democratic state structures, strengths, systems and overall effectiveness in the iWar. Employing cognitive shortcuts, laziness in analysis, emotive-based responses, and adopting traditional linear thought processes are not just symptoms of the general population but also demonstrated weaknesses of Western governments and militaries in the iWar (Table 2).

This failure by democratic nations to apply critical thought processes to rapidly evolving, unconventional, intangible threats allows enemies to predict, manipulate, exploit and anticipate influencing sub-systems’ activities and reactions in the IE. Western adversaries know and understand the importance of manipulating cognitive biases and heuristics in their influencing activities for attaining power, control and dominance over target audiences. As this research has demonstrated, the importance of education, engagement, raising awareness, transparency and a willingness to take risks and employ non-conventional, Whole of Government (WoG) creative solutions to iWar approaches is an urgent necessity to improve critical thinking, overcome system fragility and ultimately regain the initiative in the IE.

Systems Thinking Conclusions The results of this research confirmed the assertion that systems thinking offers an effective approach to understanding the complex adaptive social systems of our own-forces, our adversaries, and neutral target audiences; as well as providing a powerful methodology for developing effective influence and counter-influencing strategies. In distilling the findings identified from the case system analyses, four key themes emerged that provided an analytical framework for determining the level of system effectiveness in influencing activities. The first conclusion therefore, is that any iWar systems thinking methodology must analyse the four key themes at a minimum, which comprise: the indicative system characteristics such as resilience 211 | P a g e 5249168 Cassandra Brooker Masters of Research levels and openness, the associated flows of information via interconnected hard and soft communications links, the maintenance of control by system elements, and the system’s centrality of focus.

The second systems thinking conclusion is that in order for systems operating in the iWar to succeed in influencing effectiveness, they must first conduct an own-force system analysis and optimise performance surrounding the findings resulting from consideration of the four key themes. As the case studies demonstrated, it is just as important to understand one’s own system characteristics, structures and flows as it is to analyse adversary systems in the iWar. Even in military planning doctrine, the own-force analysis step is cursory and merely consists of analysing the mission, tasks, and listing available platforms and capabilities. Even if current doctrinal tools used for analysing the enemy (i.e. non-systems) were applied to own-force considerations, this would be a marked improvement on current practice.

Conducting a more critical analysis of friendly force capabilities may also shift the paradigms of operational staff who demonstrate extreme distaste for any intelligence analyst who conducts red team attacks on blue force vulnerabilities during wargaming, which had been overlooked during the planning and decision-making phases and ignored during ‘course of action’ development. Furthermore, this own-force systems analysis must be undertaken by all WoG sub-systems operating in the iWar, for day-to-day, business as usual activities, and not just in reaction to an attack or in support of a military operation. This identification of reinforcing and balancing feedback loops, system archetypes, and levels of fragility would enable self-correcting feedback, positive leverage, improved collaboration, and optimisation solutions to be implemented and tested before engaging with the enemy in the IE.

The third conclusion is to then build on current intelligence and targeting analytical approaches and apply the same level of systems analysis to complex adversary systems to ensure holistic understanding of the mental models, structures, behaviours, and interconnections contributing to the hidden part of the iceberg (Fig 1). A non-doctrinal systems thinking approach to analysing enemy systems, as a form of critical thinking and which includes complementary analytical techniques (Table 9), would enable the identification of leverage points, weaknesses, and critical vulnerabilities to be targeted. Additionally, this holistic analysis would provide insight into levels of enemy system effectiveness, degree of situational awareness, and how anticipatory their responses will be to vicious or virtuous reinforcing feedback. This research indicates that adversary systems have seemingly grasped and prioritised the concepts of control and a centrality of focus for protecting their COG and achieving iWar success. Whereas Western democratic 212 | P a g e 5249168 Cassandra Brooker Masters of Research nations have overlooked own-force system optimisation and apply superficial analytical target systems approaches, which are focused on kinetic effects and fail to identify appropriate leverage points, specific IO targets and the consequences of their actions.

Recommendations The following dot points are a summary of recommendations for Australia to improve its influencing effectiveness in the iWar, which were derived from the literature review, the case systems’ research findings, and the associated discussion.  Australia must develop a strong, credible, resonant narrative as a matter of priority. It must allocate resources and shift traditional national security paradigms towards having a centrality of focus on iWar capabilities across multiple agencies and stakeholders. Australia’s influencing strategy needs to be anticipatory, engage audiences, dominate the IE, amplify the messaging, deflect attention from adversaries, and undermine the capacity of adversaries to respond coherently. Incorporating the five key themes of influence effectiveness and the cognitive warfare insights identified in this research will ensure control is maintained and resilience is increased so as to steer, persuade, direct, or activate target audiences in accordance with Australia’s will.  Australia has to re-think its communications strategy, its associated risk-averse obstructionism, and its siloed cyber-warfare approaches to better exploit both hard and soft communication links. Successful systems operating in the iWar understand the game- changing nature of modern communications technology for dominating the IE, dissemination of content, manipulating heuristics, inundating media feeds, and creating online communities, echo chambers and filter bubbles (Tables 3, 6 & 12). Australia’s iWar strategy must regain control of interconnecting communications links to engage and educate audiences, create a shared identity and sense of community around common principles, to increase trust and transparency, shift beliefs and attitudes, shape behaviours, and motivate individuals to take positive action. Australia’s iWar sub-systems, both military and civilian, must be empowered, informed and enabled through effective hard and soft communications links to ensure rapid receipt of feedback, enhanced situational awareness, and anticipatory responses. A reassertion of communication dominance in the IE, as a holistic, open, shared national approach, will ultimately undermine the influencing power currently exerted by adversary systems against democratic society.  Australia needs to exploit current obstacles to effective influencing, such as: democratic values, journalistic standards, adherence to the truth, ethical targeting, and privacy and security concerns (Table 4); and turn them to its advantage in the iWar, in order to undermine adversary credibility and legitimacy. Turning the narrative back on our enemies 213 | P a g e 5249168 Cassandra Brooker Masters of Research and manipulating these impediments to Australia’s advantage, reinforces the strategic narrative, provides a centrality of focus, builds trust based on values and honesty, and outmanoeuvres Australia’s iWar adversaries. Taking control of strategic communication and basing it on a foundation of democratic values and transparency holds Australia’s enemies to account in the court of public opinion, exposes their weaknesses and hypocrisy, and contributes to negative balancing feedback undermining their iWar systems.  Australia must widely adopt systems thinking approaches to both own-force and adversary systems operating in the iWar, as well as for target audiences. Holistically analysing systems is fundamental to understanding complex causality, feedback flows, hidden behaviours, leverage points, and generating useful knowledge for understanding the contemporary, non-linear IE and enhancing the effectiveness of iWar strategies within it. Only in applying a systems thinking methodology and implementing practical solutions derived from the three research disciplines covered by this thesis can Western military forces hope to catch up and become more effective contenders in the iWar.  Australian government and military influencers should be encouraged to apply critical thinking, different socio-cultural lenses, and unbiased paradigms to their target system analysis and creation of PSYOPs content, in order to better understand adversary systems, the significance of relationships, the meaning of events, to seek hidden values, and anticipate what might happen and the reasons why. They should be encouraged to research, take risks, seek expert advice from unconventional sources, explore social sciences, and provide creative solutions to support IO planning. There should be increased co-operation and information sharing with both traditional and non-traditional allies, organisations, agencies, and civilian entities for maximum understanding and holistic integrated approaches in the iWar. Soft power strategies must be included in initial preparatory and planning phases of operations, at all levels, and not conducted in isolation by IO specialists in ‘cylinders of excellence’. Current doctrine and training should also be amended accordingly as new approaches and ways of thinking are tested and are proven successful.  Australia may consider developing its own civilian cyber-warrior force as a sub-system contributing to national efforts, similar in structure to those in other democratic nations facing imminent iWar threats. Implementing effective defensive iWar solutions now requires the involvement, education and engagement of the entire nation’s population. The government and military must lift the shroud of secrecy surrounding iWar threats to the nation and engage the civilian populace with factual information and transparency to enhance resilience. A holistic national, strategic, political approach is required, involving the engagement and input of commercial entities, economic organisations, technology companies, law enforcement, aid agencies, and academia working alongside military and 214 | P a g e 5249168 Cassandra Brooker Masters of Research government systems, to leverage lessons learnt and to truly address this new threat to national security and democratic society. An informed, engaged citizenry will build socio- cultural resilience and strengthen national audiences against iWar attacks on democratic institutions, sovereign freedoms and the truth itself.  Practical solutions for Australia to regain control over the media ecosystem include: regulating journalistic standards, improving source appraisal and fact checking, exposing enemy influencing activities, raising awareness of misinformation and manipulation efforts, enhancing media literacy and critical thinking across the population, creating media rating systems, registering and monitoring media outlets, blocking or sanctioning hostile media, and holding tech companies to account for the regulation and monitoring of content on their platforms.  Further research is required across all three academic disciplines covered by this thesis, as well as into Qualitative Comparative Analysis (QCA) of systems operating in the iWar and regarding Measures of Effectiveness (MoE) in order to enhance understanding, refine approaches, improve effectiveness, situational awareness and resilience, and for monitoring feedback relating to Australia’s future iWar strategy.

The Contribution of this Research The significance of this research is its unique contribution to addressing gaps in the current literature and the lack of innovative practices where systems approaches for effective iWar influencing are misunderstood and underutilised by the ADF, and Australia’s national security organisations more broadly. This research provides understanding into the effectiveness of influencing activities used throughout history and how these lessons apply to the contemporary global battlespace of the internet commons. By fusing three disparate research disciplines, this study also contributes to the development of original theories surrounding what makes certain influence activities effective and has contributed sound conclusions in support of the aim.

Researching all three disciplines as they relate to each other, our adversaries, and contemporary iWar audiences; as well as analysing what those findings specifically mean for Australia, has never before been attempted. Because influencing activities, cognitive behaviours, and systems thinking have all been impacted upon by the interconnected global information environment of cyberspace, research into the results of such modern effects is required to address the gaps in the accepted knowledge. This research offers recommendations for practical solutions, in both the military and civilian contexts, to address Australia’s weaknesses in dominating the contemporary, complex, cognitive iWar environment.

215 | P a g e 5249168 Cassandra Brooker Masters of Research The theoretical portion of this research, covering the three academic disciplines of systems thinking, influencing effectiveness and behavioural science, has contributed to enhanced understanding of important concepts, themes, machinations and methodologies that lead to iWar success. Systems thinking was fused with the behavioural and influencing theories to reveal the power of this research methodology in holistically revealing previously hidden system structures, leverage points, cognitive behaviours, mental models and archetypes contributing to the stocks and flows of effective influence in complex social systems. The theoretical understanding and the methodological case systems analysis achieved by this research answered a gap in knowledge set out in the aim, justified the three propositions, and provided practical solutions for application to Australia’s iWar problem.

To date, a key gap in the literature exists regarding the use of systems analysis for non-kinetic information warfare and combining that knowledge with influencing theory and behavioural science. The two contrasting case studies analysed in this thesis were highly suitable for examining the dynamic, non-linear IE and related contemporary iWar events, where the past behaviours contributing to system stocks could not be manipulated, and to ensure coverage of contextual conditions highly pertinent to this research. This research highlights the importance of improving control over own system responses in the iWar, and the benefits of having a holistic understanding of adversary system aims, trends, leverage points and activities; which thereby enables enhanced detection, analysis, anticipation, and counter-measures.

This research has also contributed to the realisation that responsibility must be devolved across all of Australia’s systems and sub-systems operating in the IE, it is no longer the sole charge of the government or military to fight independently in the iWar on behalf of the nation. This research has underlined the need for Australia to develop a resonant strategic narrative and powerful communication links, to adopt unconventional and creative approaches to the iWar, delineate responsibilities, and resource iWar strategies accordingly. The strategic paradigm shift, a new centrality in focus, and the associated reallocation and re-prioritisation of resources to support a national combined iWar effort, as recommended by this research, must be implemented in conjunction with increased co-operation, engagement, information sharing, and transparency across democratic society.

This research highlights the need for Australia to reorientate, re-think, adapt to the ‘new normal’ of the contemporary IE, and act according to the recommendations arising from the findings. Academia, government and iWar practitioners must continue exploring this emerging fused research discipline and rapidly incorporate those lessons into offensive and defensive iWar 216 | P a g e 5249168 Cassandra Brooker Masters of Research doctrine, education, messaging, and planning processes; in order to address the clear gaps in the literature and ensure own-force system optimisation. This research has highlighted the dangers of a continued focus on conventional, kinetic targeting and linear, doctrinal approaches to imposing Australia’s political will in the contested, complex global IE. The future will see an increased emphasis placed on the need for offensive and defensive, technical and non-technical iWar solutions; as well as increased demand for the rapid development of new research, doctrine, tactics, techniques, and procedures for the effective employment of these strategies.

Conclusion This chapter concludes Part Three of the thesis with a summary of the key conclusions arising from the findings and recommendations for Australia to enhance its influencing effectiveness in the iWar, in accordance with the research aim. There is a requirement for new approaches in researching, fusing, and applying theories to analytical methodologies, such as systems mapping, for discovering practical solutions to enhance Australia’s iWar strategy, as well as contributing to the body of literature surrounding contemporary influencing activities.

The research in this thesis realised the aim in demonstrating that Australia can achieve effective influence over population, military, and government systems in the iWar from non-technical information operations. This was achieved by fusing the three research disciplines of systems thinking, influencing activities, and behavioural science to highlight the gaps in the literature (Chapter 2) and inform the research methodology (Chapter 3).

Part Two of this thesis compared and contrasted two complex case studies using a systems modelling approach to demonstrate why Western democracies have proven ineffective in the iWar (Chapters 4, 5 & 6). The two complex social case studies selected for this research comprised a successful influencing system of ISIS, and an unsuccessful influencing system of Hillary Clinton’s 2016 US presidential election campaign. A pragmatic narrative analysis of the two cases’ respective qualitative data was undertaken to consequently map each system’s model surrounding their influence stocks and present the associated findings. The utility of adopting a systems thinking analytical approach was validated in the understanding that the findings provided into both the modern, complex social systems of key adversaries, as well as of own- force iWar systems. The systems analysis findings determined the reasons why the case studies had been effective, or not, in raising their influence stocks, dominating the IE, and manipulating cognitive behaviours; whilst also highlighting the configurational, behavioural, and causal factors contributing to each system’s respective levels of influence.

217 | P a g e 5249168 Cassandra Brooker Masters of Research Part Three of this thesis, comprising Chapters 7 and 8, discusses and concludes the results of the research as they relate to the gaps in knowledge, the three key arguments, and the research aim, before making key recommendations for addressing Australia’s iWar problems. Chapter 7 discusses the implications of the findings of the literature review in Part One, and the results of the case system analyses in Part Two of this thesis, to achieve the aim of investigating solutions for improving Australia’s influence effectiveness in the iWar. The discussion in Chapter 7 revolves around responding to questions surrounding how much is known, how the research answered the gaps, and what this research means for Australia. This chapter concludes the thesis with a summary of the key conclusions arising from the three research disciplines, it provides recommendations for Australia, and highlights the contribution of this research to Australia’s iWar capabilities, using a systems thinking methodology, as well as to the body of literature surrounding this study.

218 | P a g e 5249168 Cassandra Brooker Masters of Research References

Afzal, W. & Wallis, J. (2016), ‘Informational properties of extremist content: An evidence-based approach’, Australian Army Journal, Vol. 13, No. 2, Spring 2016, pp. 67-90. Retrieved 09 January 2020 from: https://search.informit.com.au/documentSummary;dn=673263594698137;res=IELAPA

Albino, D., Friedman, K., Bar-Yam, Y. & Glenney IV, W. (2016), ‘Military Strategy in a Complex World’, New England Complex Systems Institute Report 2016-02-02, Cornell University, 18 February 2016, pp. 1- 22. Retrieved 15 February 2019 from: https://arxiv.org/abs/1602.05670

Aquino, M.A. (1987), ‘Psychological Operations: The Ethical Dimension’, Industrial College of the Armed Forces, National Defense University, Fort McNair, Washington D.C, USA, pp. 1-26.

Ashenden, D. & Lawrence, D. (2013), ‘Can we sell security like soap? A new approach to behaviour change’, NSPW 2013: Proceedings of the 2013 New Security Paradigms Workshop, ACM, New York, USA, pp. 87-94. Retrieved 09 January 2020 from: https://dl-acm-org.wwwproxy1.library.unsw.edu.au/doi/pdf/10.1145/2535813.2535823?download=true

Australian Army (2002), Land Warfare Doctrine 1 (LWD-1): The Fundamentals of Land Warfare, Doctrine Wing, Land Warfare and Development Centre, Puckapunyal VIC, Australia.

Australian Army (2006), Grade 3 Operations Battle Book, Officer Training Wing, Land Warfare Centre, Canungra, QLD, Australia, released 27 June 2006.

Australian Army (2018), ‘Land Warfare Procedures–General (LWP-G) 2-1-1: Psychological Operations (Interim)’, Defence Intelligence Training Centre, Canungra, QLD, Australia, dated 4 June 2018.

Australian Defence Force, (2012), ‘The Nature of War’, Australian Defence Doctrine Publication (ADDP- D), Chapter 2, ADF Joint Warfare Centre, RAAF Williamtown, NSW, Australia.

Australian Defence Force, (2018), ‘ADDP 3.14—Targeting’, Australian Defence Doctrine Publication (ADDP), Edition 3, Operations Series, 12 July 2018.

Babbage, R. (2019), ‘Hastie’s right: it’s time to protect ourselves’, The Strategist, Australian Strategic Policy Institute (ASPI), 16 Aug 2019. Retrieved 10 January 2020 from: https://www.aspistrategist.org.au/hasties-right-its-time-to-protect-ourselves/

Benkler, Y., Farris, R. & Roberts, H. (2018), ‘Network Propaganda: Manipulation, Disinformation, and Radicalization in American Politics’, Oxford University Press, New York, USA. Retrieved 27 August 2019 from: http://www.oapen.org/search?identifier=1001606

Berger, J. & Fitzsimons, G. (2008), ‘Dogs on the Street, Pumas on Your Feet: How Cues in the Environment Influence Product Evaluation and Choice’, Journal of Marketing Research, Vol. 45, No. 1 (February 2008), Sage Publications Inc. pp. 1-14. Retrieved 13 February 2019 from: https://www.jstor.org/stable/30162517

Bienvenue, E., Rogers, Z. & Troath, S. (2018a), ‘Cognitive Warfare: The Fight We’ve Got’, The Cove website, posted 19 September 2018. Retrieved 13 February 2019 from: https://www.cove.org.au/adaptation/article-cognitive-warfare-the-fight-weve-got/

Bienvenue, E., Rogers, Z. & Troath, S. (2018b), ‘Trust as a Strategic Resource for the Defence of Australia’, The Cove website, posted 29 October 2018. Retrieved 12 February 2019 from: https://www.cove.org.au/war-room/article-trust-as-a-strategic-resource-for-the-defence-of-australia

Birney, A., Drobac, P. & Nicholls, A. (2019), ‘Is It Possible to Measure Systems Change?’, Skoll World Forum, Skoll Foundation. Retrieved 27 May 2019 from: http://skoll.org/video/is-it-possible-to-measure- systems-change/

219 | P a g e 5249168 Cassandra Brooker Masters of Research Blackmore, T.R. (2003), ‘Assess the importance of Measures of Effectiveness (MOE) for the further development of UK information operations’, British Army, Defence Studies, Vol 3:3, pp. 9-37. Retrieved 12 February 2019 from: https://doi.org/10.1080/14702430308405075

Blannin P. (2018), ‘Meet the Neighbours: ISIS on Israel’s Doorstep’, Small Wars Journal, published by Small Wars Foundation, published 01 July 2018. Retrieved 09 January 2020 from: https://smallwarsjournal.com/jrnl/art/meet-neighbours-isis-israels-doorstep

Bohman, J. (2002), ‘How to Make a Social Science Practical: Pragmatism, Critical Social Science and Multiperspectival Theory’, Millennium, Vol 31, No.3, pp. 499–524. Retrieved 29 May 2019 from: https://doi.org/10.1177/03058298020310030701

Bond, R.M., Fariss, C.J., Jones, J.J., Kramer, A.D.I, Marlow, C., Settle, J.E & Fowler, J.H. (2012), ‘A 61- million-person experiment in social influence and political mobilization’, Research Letter, Nature, Vol 489, 13 September 2012, Macmillan Publishers Limited, pp. 295-298. Retrieved 13 February 2019 from: https://www-nature-com.wwwproxy1.library.unsw.edu.au/articles/nature11421

Bosio, N. (2018), ‘IO Theory: Creating Resonance’, Cognitive Edge webpage, Land Power Forum, Australian Army Research Centre website, published 06 April 2018. Retrieved 06 March 2019 from: https://www.army.gov.au/our-future/blog/cognitive-edge/io-theory-creating-resonance

Braumoeller, B.F. (2003), ‘Causal Complexity and the Study of Politics’, Political Analysis, Vol. 11, No. 3 (Summer 2003), pp. 209-233, Cambridge University Press on behalf of the Society for Political Methodology. Retrieved 29 May 2019 from: https://www.jstor.org/stable/25791730

Bray, W.R. (2016), ‘Intelligence Is Not Warfare!’, U.S. Naval Institute, Proceedings Magazine, December 2016, Vol. 142 / 12 / 1,366. Retrieved 24 February 2018 from: https://www.usni.org/magazines/proceedings/2016-12-0

Brooking, E. & Singer, P.W. (2016), ‘War Goes Viral: How social media is being weaponized across the world’, The Atlantic, November 2016 issue, published 15 October 2016, pp. 1-19. Retrieved 12 March 2019 from: https://www.theatlantic.com/magazine/archive/2016/11/war-goes-viral/501125/

Brown, Z.T. (2019), ‘Myth Versus Lethality: Losing the Plot in the Information War’, The Strategy Bridge, published 02 December 2019. Retrieved 16 December 2019 from: https://thestrategybridge.org/the- bridge/2019/12/2/myth-versus-lethality-losing-the-plot-in-the-information-war

Buckle, K. (2018), ‘A Framework for Effective Relations: Social Scientists and the Military in operations other than war’, The University of Adelaide, SAGE International Australia, July 2018, pp.1-37. Retrieved 8 September 2018 from: https://www.sageinternational.org.au/general-discussion/a-framework-for-effective-relations-social- scientists-and-the-military-in-operations-other-than-war/

Byrne, D. (2005), ‘Complexity, Configurations and Cases’, Theory, Culture & Society, Vol. 22(5), October 2005, pp. 95-111. Retrieved 11 April 2019 from: https://journals-sagepub-com.wwwproxy1.library.unsw.edu.au/doi/pdf/10.1177/0263276405057194

Byrne, D. & Uprichard, E. (2012), ‘Useful Complex Causality’, The Oxford Handbook of Philosophy of Social Science, August 2012, Harold Kincaid (ed.), Oxford University Press, pp. 1-23. Retrieved 11 April 2019 from: http://www.oxfordhandbooks.com.wwwproxy1.library.unsw.edu.au/view/10.1093/oxfordhb/9780195392 753.001.0001/oxfordhb-9780195392753-e-6?print=pdf

Canadian Security Intelligence Service (2018), ‘Who Said What? The Security Challenges of Modern Disinformation’, Academic Outreach Program, February 2018, World Watch: Expert Notes, Canada, pp. 1-124. Retrieved 24 January 2019 from: https://www.canada.ca/en/security-intelligence- service/corporate/publications/who-said-what-the-security-challenges-of-modern-disinformation.html

220 | P a g e 5249168 Cassandra Brooker Masters of Research Capps, J. (2019), ‘The Pragmatic Theory of Truth’, The Stanford Encyclopedia of Philosophy, Summer 2019 Edition, Edward N. Zalta (ed.), first published 21 May 2019. Retrieved 03 June 2019 from: https://plato.stanford.edu/archives/sum2019/entries/truth-pragmatic

Cave, D. (2018), ‘Content and ideas + cyber—the sleeper in our national security defences’, The Strategist, Australian Strategic Policy Institute (ASPI) website, posted 17 April 2018. Retrieved 13 February 2019 from: https://www.aspistrategist.org.au/content-ideas-cyber-sleeper-national-security-defences/

Chua, E. (2018), ‘Fighting and Winning in the Information Age’, 2018 Writing Contest, The Strategy Bridge website, dated 26 September 2018. Retrieved 14 November 2019 from: https://thestrategybridge.org/the-bridge/2018/9/26/fighting-and-winning-in-the-information-age

Cialdini, R.B. (2003), ‘Crafting Normative Messages to Protect the Environment’, Current Directions in Psychological Science, Vol. 12, Number 4, August 2003, pp. 105-109. Retrieved 12 March 2019 from: https://journals.sagepub.com/doi/abs/10.1111/1467-8721.01242?journalCode=cdpa

Ciovacco, C., Clark, H.G. & Van de Velde, J. (2011), ‘Ending al-Qaeda’, The American Interest, Vol. 6, Issue 6, Jul/Aug 2011, Washington, pp. 40-49. Retrieved 09 January 2020 from: https://search-proquest- com.wwwproxy1.library.unsw.edu.au/docview/895048611?rfr_id=info%3Axri%2Fsid%3Aprimo

Clapper, J.R. with Brown, T. (2018), ‘Facts and Fears: Hard Truths from a life in Intelligence’, Penguin Random House LLC, New York, USA.

Clark, H.G. (2015), ‘Go Local’, The American Interest, Vol. 10, Issue 6, Jul/Aug 2015, Published 10 June 2015, Washington, USA, pp. 1-12. Retrieved 09 January 2020 from: https://search-proquest- com.wwwproxy1.library.unsw.edu.au/docview/1689848760?rfr_id=info%3Axri%2Fsid%3Aprimo

Clinton, H.R. (2018), ‘What Happened’, Simon & Schuster UK Ltd, London, UK.

Congdon, W.J., Kling, J.R. & Mullainathan, S. (2011), ‘Policy and Choice: Public Finance through the lens of Behavioural Economics’, Chapter 2: “Psychology and Economics”, The Brooking Institute, Washington DC, USA, pp. 17-39. Retrieved 12 March 2019 from: https://www.brookings.edu/wp-content/uploads/2016/07/policyandchoice_book.pdf

Datta, S. & Mullainathan, S. (2014), ‘Behavioural Design: a New Approach to Development Policy’, Review of Income and Wealth, Series 60, Number 1, March 2014, pp. 7-35. Retrieved 12 March 2019 from: https://onlinelibrary.wiley.com/doi/epdf/10.1111/roiw.12093

DellaVigna, S., & Kaplan, E. (2007), ‘The Fox News Effect: Media Bias and Voting’, The Quarterly Journal of Economics, Vol. 122(3), pp. 1187-1234. Retrieved 13 February 2019 from: http://www.jstor.org/stable/25098871

DeMarzo, P., Vayanos, D. & Zwiebel, J. (2003), ‘Persuasion Bias, Social Influence, and Unidimensional Opinions’, The Quarterly Journal of Economics, Vol. 118(3), pp. 909-968. Retrieved 13 February 2019 from: http://www.jstor.org/stable/25053927

Dhanaraj J. (2018), ‘Evolution of the Islamic State after its Territorial Defeat’, Counter Terrorist Trends and Analyses, Vol. 10, No. 5 (May 2018), Published by: International Centre for Political Violence and Terrorism Research, pp. 1-7. Retrieved 08 August 2019 from: https://www.jstor.org/stable/10.2307/26435152

Donovan, J. (2019), ‘How memes got weaponized: A short history’, MIT Technology Review, Humans and Technology, published 24 October 2019. Retrieved 16 December 2019 from: https://www.technologyreview.com/s/614572/political-war-memes-disinformation/

Ecker, U.K.H., O’Reilly, Z., Reid, J.S. & Chang, E.P. (2019), ‘The effectiveness of short-format refutational fact-checks’, British Journal of Psychology (2019), John Wiley & Sons Ltd, pp. 1-19. Retrieved 25 September 2019 from: https://doi.org/10.1111/bjop.12383

221 | P a g e 5249168 Cassandra Brooker Masters of Research Ehlers, R (2017), ‘Making Old Things New Again: Strategy, the Information Environment, and National Security’, 3 January 2017, IEAA Course Reading, pp. 1-8.

Elder L. & Paul, R. (2002), ‘Taking Charge of the Human Mind’, Thinker's Guide Series, Foundation for Critical Thinking, California, USA. Retrieved extract 18 January 2020 from: https://www.criticalthinking.org/resources/articles/natural-egocentric.shtml

Engstrom, J. (2018), ‘Systems Confrontation and System Destruction Warfare - How the Chinese People’s Liberation Army Seeks to Wage Modern Warfare’, Rand Corporation, Santa Monica, California, US, pp. 1-143. Retrieved 18 August 2019 from: https://www.rand.org/content/dam/rand/pubs/research_reports/RR1700/RR1708/RAND_RR1708.pdf

Epstein, R. & Robertson, R.E. (2015), ‘The search engine manipulation effect (SEME) and its possible impact on the outcomes of elections’, posted 18 August 2015, Proceedings of the National Academy of Sciences (PNAS) USA website, pp. 1 – 10. Retrieved 13 February 2019 from: https://www.pnas.org/content/112/33/E4512

Facione, P.A. (2015), ‘Critical Thinking: What It Is and Why It Counts’, Measured Reasons LLC, Hermosa Beach, CA, USA, distributed by Insight Assessment, pp. 1-31. Retrieved 8 March 2019 from: https://www.insightassessment.com/Resources/Importance-of-Critical-Thinking/Critical-Thinking-What- It-Is-and-Why-It-Counts

Fawcett, J.R. (2018), ‘Paradigm for Integrated Warfighting: Kinetic and Non-kinetic Solutions’, Information Superiority / Information Operations, US Air Force Command and Control Training and Innovation Group (AFC2TIG), Hurlburt Field, FL, pp. 1-6. Retrieved 28 January 2019 from: http://www.dodccrp.org/events/7th_ICCRTS/Tracks/pdf/117.PDF

Flyvbjerg, B. (2006), ‘Five Misunderstandings about Case-Study Research,’ Qualitative Inquiry, Vol. 12, no. 2, April 2006, pp. 219-245. Retrieved 28 January 2019 from: http://qix.sagepub.com/content/12/2/219.abstract

Freedman, L. (2006), ‘The Transformation of Strategic Affairs’, Routledge for the International Institute for Strategic Studies, Abingdon, New York, USA.

Game, E.T., Meijaard E., Sheil D. & McDonald-Madden, E. (2013), ‘Conservation in a Wicked Complex World; Challenges and Solutions’, Conservation Letters, May/June 2014, Vol. 7(3), Wiley Periodicals Inc., pp. 271–277. Retrieved 06 May 2019 from: https://onlinelibrary.wiley.com/doi/full/10.1111/conl.12050

Gameau, D. (2019), ‘2040: Join the Regeneration’, documentary, released 29 May 2019, Madman Entertainment, Australia.

Gerber, A.S., Karlan, D. & Bergan, D. (2009), ‘Does the Media Matter? A Field Experiment Measuring the Effect of Newspapers on Voting Behavior and Political Opinions’, American Economic Journal: Applied Economics, April 2009, Vol. 1:2, pp. 35–52. Retrieved 13 February 2019 from: http://www.aeaweb.org/articles.php?doi=10.1257/app.1.2.35

Gerrits, L. & Verweij, S. (2013), ‘Critical Realism as a Meta-Framework for Understanding the Relationships between Complexity and Qualitative Comparative Analysis’, Journal of Critical Realism, Vol. 12(2), pp. 166–182. Retrieved 03 June 2019 from: https://doi.org/10.1179/rea.12.2.p663527490513071

Gharajedaghi, J. (2011), ‘Sociocultural System, Chapter 3 in: ‘Systems Thinking: Managing Chaos and Complexity - A Platform for Designing Business Architecture’, third edition, Elsevier Inc, Burlington MA, USA, pp. 59-66.

Giles, L. (2017), ‘The Art of War – Sun Tzu’, Arcturus Publishing Ltd, London, UK.

222 | P a g e 5249168 Cassandra Brooker Masters of Research Givens, A. (2012), ‘A Systems-Based Approach to Intelligence Reform’, Journal of Strategic Security, No. 1, Volume 5, Spring 2012, Article 9, Henley-Putnam University Press, pp. 63-84. Retrieved 26 May 2019 from: https://scholarcommons.usf.edu/cgi/viewcontent.cgi?referer=https://www.google.com/&httpsredir=1&arti cle=1135&context=jss

Gleeson, E. (2019), ‘Internet Sovereignty as a Gateway to Global Norms: The Battle over Global Internet Governance’, Strife Blog, 18 July 2019, Strife Journal, Department of War Studies, King’s College London. Retrieved 27 August 2019 from: http://www.strifeblog.org/2019/07/18/internet-sovereignty-as-a- gateway-to-global-norms-the-battle-over-global-internet-governance/

Goble, J.J. (2002), ‘Combat Assessment of Non-Lethal Fires: the Applicability of Complex Modeling to Measure the Effectiveness of Information Operations’, A Monograph, School of Advanced Military Studies, United States Army Command and General Staff College, Fort Leavenworth, Kansas, pp. 1-56. Retrieved 12 February 2019 from: https://apps.dtic.mil/dtic/tr/fulltext/u2/a402626.pdf

Guevin, P. R. (2004), ‘Psychological Operations’, Air & Space Power Journal, Summer 2004, Vol. 18, Issue 2, p. 30. Retrieved 15 March 2019 from: https://search-proquest- com.wwwproxy1.library.unsw.edu.au/docview/217765638?rfr_id=info%3Axri%2Fsid%3Aprimo

Hagmann, D., Ho, E.H. & Loewenstein, G. (2019), ‘Nudging out support for a carbon tax’, Nature Climate Change, Vol. 9, published 13 May 2019, pp. 484–489. Retrieved 29 May 2019 from: https://www-nature- com.wwwproxy1.library.unsw.edu.au/articles/s41558-019-0474-0

Hall, W.M. & Citrenbaum, G. (2010), ‘Intelligence Analysis: How to think in complex environments’, an Association of the United States Army book, Praeger Security International, Santa Barbara, California, USA.

Haywood, L. (2019), ‘Future Joint Force Information Warfare Operations’, Part 1 and Part 2, Cognitive Edge Web page, Land Power Forum, Australian Army Research Centre Newsletter 01/19, published 28 February 2019. Part 1 retrieved 06 March 2019 from: https://www.army.gov.au/our-future/blog/cognitive- edge/future-joint-force-information-warfare-operations-part-1-major-lee Part 2 retrieved 06 March 2019 from: https://www.army.gov.au/our-future/blog/cognitive-edge/future- land-force-information-warfare-operations-part-2-major-lee

Hellman, M. & Wagnsson, C. (2017), ‘How can European states respond to Russian information warfare? An analytical framework’, European Security, Vol. 26, No. 2, Routledge, pp. 153-170. Retrieved 12 February 2019 from: https://doi.org/10.1080/09662839.2017.1294162

Hiscox, M, (2018), ‘Applying Behavioural Insights to Improve Policy Outcomes’, Nudging for Good – Executive Education Seminar, 19 – 20 November 2018, Criterion Conferences, Canberra, Australia.

Hochschild, J. & Einstein, K.L. (2014), ‘It Isn’t What We Don’t Know that Gives Us Trouble, It’s What We Know that Ain’t So’: Misinformation and Democratic Politics’, British Journal of Political Science, Volume 45, Issue 3, July 2015, Cambridge University Press, published online 22 December 2014, pp. 467-475. Retrieved 14 November 2019 from: https://doi- org.wwwproxy1.library.unsw.edu.au/10.1017/S000712341400043X

Information Warfare Division (2018), ‘Information Warfare Strategy 2018 – 2028’, Joint Capabilities Group, Australian Defence Force, pp. 1-33, distributed 15 January 2018.

Ingram, H.J. (2015), ‘The strategic logic of Islamic State information operations’, Australian Journal of International Affairs, Vol. 69:6, pp. 729-752, published 25 August 2015. Retrieved 12 February 2019 from: https://doi.org/10.1080/10357718.2015.1059799

Jackson, M. (2007), ‘Soldier: the autobiography of General Sir Mike Jackson’, Bantam, London, UK.

223 | P a g e 5249168 Cassandra Brooker Masters of Research Jakubowski, G. (2019), ‘What’s Not to Like? Social Media as Information Operations Force Multiplier’, Joint Force Quarterly, No. 94, 24 July 2019, National Defense University Press. Retrieved 27 August 2019 from: https://ndupress.ndu.edu/Media/News/News-Article-View/Article/1914738/whats-not-to-like- social-media-as-information-operations-force-multiplier/

Jensen, M. & Sear, T. (2019), ‘Hashtag war: Russian trolls and the project to undermine Australian democracy’, Griffith Review, Edition 64: ‘The New Disruptors’, April 2019, Brisbane, Australia. Retrieved 10 May 2019 from: https://griffithreview.com/articles/hashtag-war-russian-trolls-undermine-australian- democracy/

J. Mark Services Inc. (2018), ‘Information Environment Advanced Analysis (IEAA) Course’, for US Department of Defense, 26 November – 7 December 2018, Ottawa, Canada.

Kahneman, D. (2011), ‘Thinking, Fast and Slow’, Farrar, Straus and Giroux, MacMillan Publishing, New York, USA, Chapter 7, pp. 79-88.

Kilberg, J. (2012), ‘A Basic Model Explaining Terrorist Group Organizational Structure’, Studies in Conflict & Terrorism, Vol. 35:11, Taylor & Francis Group, LLC, pp. 810-830. Retrieved 12 August 2019 from: https://doi.org/10.1080/1057610X.2012.720240

Kim, D.H. (1999), ‘Introduction to Systems Thinking’, Innovation in Management series, Pegasus Communication Inc, pp. 1-21. Retrieved 01 April 2019 from: https://thesystemsthinker.com/wp- content/uploads/2016/03/Introduction-to-Systems-Thinking-IMS013Epk.pdf

Knight, C. & Theodorakis K. (2019), ‘The Marawi Crisis – urban conflict and information operations’, Special Report, The Strategist, July 2019, Australian Strategic Policy Institute, Barton, Australia.

Laity, M. (2000), ‘The media, the military and policy‐making – who's calling the shots?’, The RUSI Journal, Vol. 145, Issue 6, pp. 15-20. Retrieved 09 January 2020 from: https://www.tandfonline.com/doi/abs/10.1080/03071840008446583?journalCode=rusi20

Laity, M. (2013), ‘Perception Becomes Reality. The Afghan Strategic Communication Challenge’, Politique étrangère, winter issue, No.4, 2013/2014, pp. 161-174. Retrieved 09 January 2020 from: https://www.cairn-int.info/article-E_PE_134_0161--perception-becomes-reality.htm

Laity, M. (2015), ‘NATO and the Power of Narrative’, Information at War: From China’s Three Warfares to NATO’s Narratives, Beyond Propaganda series, September 2015, Legatum Institute, pp. 22-28. Retrieved 09 January 2020 from: http://www.lse.ac.uk/iga/assets/documents/arena/archives/information- at-war-from-china-s-three-warfares-to-nato-s-narratives-pdf.pdf#page=24

Lawson, D., Stevenson, K., Peterson, N., Carrier, S., Strnad, R. & Seekamp, E. (2019), ‘Children can foster climate change concern among their parents, Nature Climate Change, No. 9, 06 May 2019, pp. 458–462. Retrieved 30 August 2019 from: https://www.nature.com/articles/s41558-019-0463-3

Lieber, P.S. & Reiley, P.J. (2016), ‘Countering ISIS’s Social Media Influence’, Special Operations Journal, Vol. 2:1, pp. 47-57. Retrieved 13 February 2019 from: https://www-tandfonline- com.wwwproxy1.library.unsw.edu.au/doi/pdf/10.1080/23296151.2016.1165580?needAccess=true

Lieberson, S. (1992), ‘Small N’s and big conclusions: an examination of the reasoning in comparative studies based on a small number of cases’, Chapter 4 in: Ragin, C.C. & Becker, S.B (1992), ‘What is a Case? Exploring the Foundations of Social Inquiry’, Cambridge University Press, New York, USA.

Maher, A. (2018), ‘Proxy Wars and the Demise of Conventional Warfighting’, Real Clear Defense, published 15 October 2018. Retrieved 25 January 2019 from: https://www.realcleardefense.com/articles/2018/10/15/proxy_wars_and_the_demise_of_conventional_wa rfighting_113892.html

224 | P a g e 5249168 Cassandra Brooker Masters of Research Maltby, S. & Thornham, H. (2012), ‘The disembodiment of persuasive military discourse’, Journal of War & Culture Studies, Vol. 5:1, pp. 33-46. Retrieved 15 March 2019 from: https://www-tandfonline- com.wwwproxy1.library.unsw.edu.au/doi/pdf/10.1386/jwcs.5.1.33_1?needAccess=true

Meadows, D.H. (1980), ‘The Unavoidable a Priori’, Elements of the System Dynamics Method, ed. Randers, J., Productivity Press, Cambridge, MA, USA, pp. 23-57.

Meadows, D.H. (2008), ‘Thinking in Systems: A Primer’, ed. Wright, D. Sustainability Institute, Earthscan, Dunstan House, London, UK.

Mehmetcik, H. & Kursun, A.M. (2017), ‘Making Sense of the Territorial Aspirations of ISIS: Autonomy, Representation, and Influence’, Part 1: Actorness, Violent Non-State Actors and the Syrian Civil War: The ISIS and YPG Cases, pp. 53-72. Retrieved 13 February 2019 from: https://ebookcentral.proquest.com/lib/unsw/detail.action?docID=5178327

Mersol, J. (2019), ‘Leaders Agree that Cybersecurity is About People’, Government CIO media, USA. Retrieved 16 September 2019 from: https://www.governmentciomedia.com/leaders-agree-cybersecurity-about-people

Mishler, E.G. (1995), ‘Models of Narrative Analysis: A Typology’, Journal of Narrative and Life History, Vol. 5 (2), pp. 87-123, Retrieved 12 July 2019 from: http://h10.relais-host.com/posttowebnun/DownloadDocument?4464658.pdf

Morgan, D.L. (2014), ‘Pragmatism as a Paradigm for Social Research’, Qualitative Inquiry 2014, Vol. 20(8), pp. 1045–1053. Retrieved 04 June 2019 from: https://journals-sagepub-com.wwwproxy1.library.unsw.edu.au/doi/pdf/10.1177/1077800413513733

Morris, D.R. (2001), ‘Causal Inference and System Dynamics in Social Science Research: A Commentary with Example’, The 19th International Conference of The System Dynamics Society, System Dynamics website, Atlanta, Georgia, USA, pp. 1-25. Retrieved 03 June 2019 from: https://www.systemdynamics.org/assets/conferences/2001/papers/Morris_1.pdf

Mueller, R.S. (2018), ‘Internet Research Agency LLC Indictment’, Case 1:18-cr-00032-DLF, United States District Court for District of Colombia, Filed: 16 February 2018, pp. 1 – 37. Retrieved 13 February 2019 from: https://www.justice.gov/file/1035477/download

Murray, S. & Blannin P. (2017), ‘Diplomacy and the War on Terror’, Small Wars Journal, published by Small Wars Foundation, published 18 September 2017. Retrieved 09 January 2020 from: https://smallwarsjournal.com/jrnl/art/diplomacy-and-the-war-on-terror

NATO Strategic Communication Centre of Excellence (2016), ‘Internet trolling as a tool of hybrid warfare: the case of Latvia’, Published 25 January 2016, pp. 1-5. Retrieved 27 February 2019 from: http://www.stratcomcoe.org/internet-trollinghybrid-warfare-tool-case-latvia

Neriah, J. (2014), ‘The Structure of the Islamic State (ISIS)’, Jerusalem Centre for Public Affairs, posted 08 September 2014. Retrieved 12 August 2019 from: http://jcpa.org/structure-of-the-islamic-state/

Nimmo, B. (2015), ‘The Case for Information Defence: A Pre-Emptive Strategy for Dealing with the New Disinformation Wars’, Information at War: From China’s Three Warfares to NATO’s Narratives, Beyond Propaganda series, September 2015, Legatum Institute, pp. 29-36. Retrieved 09 January 2020 from: http://www.lse.ac.uk/iga/assets/documents/arena/archives/information-at-war-from-china-s-three- warfares-to-nato-s-narratives-pdf.pdf#page=24

Nye, J. (2005), ‘Soft Power: The Means to Success in World Politics’, Ingram Publisher Services, La Vergne, TN, USA.

Patrikarakos, D. (2017), ‘War in 140 characters: how social media is reshaping conflict in the twenty-first century’, first edition, Basic Books, New York, USA.

225 | P a g e 5249168 Cassandra Brooker Masters of Research

Patterson, K., Grenny, J., Maxfield, D., McMillan, R. & Switzler, A. (2007), ‘Influencer: The Power to Change Anything’, McGraw-Hill, New York, USA. pp. 5, 45-51, 57-63. Retrieved 09 March 2019 from: https://epdf.tips/influencer-the-power-to-change-anything.html

Paul, C. & Matthews, M. (2016), ‘The Russian “Firehose of Falsehood” Propaganda Model: Why It Might Work and Options to Counter It’, Published Research, Perspectives webpage, RAND Corporation, Santa Monica, CA, pp. 1-16. Retrieved 28 January 2019 from: https://www.rand.org/pubs/perspectives/PE198.html

Peters, S. (2012), ‘The Chimp Paradox – The Science of Mind Management for Success in Business and in Life’, Vermilion of Ebury Publishing, Penguin Random House, London, UK.

Pirsig, R.M. (2005), ‘Zen and the art of motorcycle maintenance: an inquiry into values’, 1st Harper Perennial Modern Classics ed. HarperCollins Publishers, New York, USA.

Polkinghorne, D.E. (1995), ‘Narrative Configuration in Qualitative Analysis’, International Journal of Qualitative Studies in Education, Vol. 8, No. 1, pp. 5-23, Retrieved 09 July 2019 from: https://doi-org.wwwproxy1.library.unsw.edu.au/10.1080/0951839950080103

Ragin, C.C. & Becker, S.B. (1992), ‘What is a Case? Exploring the Foundations of Social Inquiry’, Cambridge University Press, New York, USA.

Read, E. (2019), ‘The Meme Justifies the Means: Political Warfare & Group-Based Mass Manipulation on Social Media’, ANU Research, Australian Army Research Centre Seminar Series, 02 May 2019, Russell Offices. Retrieved 06 May 2019 from: https://www.army.gov.au/our-future/australian-army-research- centre-aarc/australian-army-research-centre-seminar-series/the-meme-justifies-the-means-ms-emily-read

Reynolds, E. (2018), ‘Russian trolls’ sinister interference in Trump’s favour exposed’, Finance page, News.com.au website, 19 December 2018. Retrieved 25 January 2019 from: https://www.news.com.au/finance/work/leaders/russian-trolls-vast-interference-in-trumps-favour- exposed/news-story/f32db6a4ab2146512f33796d3b19f705

Rihoux, B. & Ragin, C.C. (2009), ‘Applied Social Research Methods: Configurational comparative methods: Qualitative comparative analysis (QCA) and related techniques’, (Vols. 1-51), SAGE Publications Inc., Thousand Oaks CA, USA. Retrieved 30 May 2019 from: http://methods.sagepub.com.wwwproxy1.library.unsw.edu.au/book/configurational-comparative-methods

Rosenberger, L. (2019), ‘US foreign policy in the 2020 presidential race, and where to after Trump?’, Presentation at the National Security College, 27 March 2019, Crawford School of Public Policy, ANU College of the Asia & the Pacific, Canberra, Australia. Retrieved 28 March 2019 from: https://nsc.crawford.anu.edu.au/news-events/podcasts/video

Ruthven, M. (2007), ‘Fundamentalism: A very short introduction’, Oxford University Press, Oxford, UK.

Sahlstein Parcell, E. & Baker, B.M.A. (2017), ‘Narrative Analysis’, The SAGE Encyclopaedia of Communication Research Methods, M. Allen (ed.), SAGE Publications Inc., Thousand Oaks CA, USA, pp. 1069-1072. Retrieved 09 July 2019 from: http://dx.doi.org/10.4135/9781483381411.n368

Sampson T. (2019), ‘Gruen’ episode, Australian Broadcasting Corporation, aired 30 October 2019, ABC, Sydney, Australia.

Schleifer, R. (2006) ‘Psychological Operations: A New Variation on an Age Old Art: Hezbollah versus Israel’, Studies in Conflict & Terrorism, Vol. 29, Issue 1, pp. 1-19, Retrieved 15 March 2019 from: https://www-tandfonline-com.wwwproxy1.library.unsw.edu.au/doi/abs/10.1080/10576100500351185

226 | P a g e 5249168 Cassandra Brooker Masters of Research Schneider, D.A. (2002), ‘Measuring the Intangible: How Can An Operational Commander Know If His Information-Based Operations Are Effective?’, 13 May 2002, Joint Military Operations Department, Naval War College, Newport, USA, pp. 1-23. Retrieved 12 February 2019 from: https://apps.dtic.mil/dtic/tr/fulltext/u2/a406201.pdf

Sil, R. & Katzenstein, P.J. (2010), ‘Beyond paradigms: Analytic Eclecticism in the Study of World Politics’, Palgrave Macmillan, MacMillan Publishers Ltd, Basingstoke, England.

Silver, N. (2017), ‘Donald Trump Had a Superior Electoral College Strategy: How Hillary Clinton and the media missed the boat’, posted 6 February 2017, FiveThirtyEight website. Retrieved 13 February 2019 from: https://fivethirtyeight.com/features/donald-trump-had-a-superior-electoral-college-strategy/

Silver, N. (2018), ‘How Much Did Russian Interference Affect the 2016 Election?’, posted 16 February 2018, FiveThirtyEight website. Retrieved 13 February 2019 from: https://fivethirtyeight.com/features/how-much-did-russian-interference-affect-the-2016-election/

Singer, P.W. (2019), ‘The rise of information warfare: in-conversation with Peter W. Singer’, Australian Strategic Policy Institute (ASPI) event, 29 October 2019, ASPI Auditorium, Barton ACT, Australia.

Singer, P.W. & Brooking, E.T. (2018), ‘LikeWar: The Weaponization of Social Media’, Houghton Mifflin Harcourt Publishing Company, 3 Park Avenue, New York, USA.

Siyech, M.S. (2016), ‘Islamic State: Sowing the seeds of its own destruction’, Counter Terrorist Trends and Analyses, Vol. 8, No. 11, November 2016, JSTOR, pp. 24–28. Retrieved 26 November 2019 from: www.jstor.org/stable/26351470

Spaulding, S., Nair, D. & Nelson, A. (2019), ‘Russia’s Attacks on Democratic Justice Systems’, Center for Strategic & International Studies website. Retrieved 08 July 2019 from: https://www.csis.org/features/russias-attacks-democratic-justice-systems

Stanley, J. (2018), ‘How Fascism Works – The Politics of Us and Them’ Penguin Random House, New York, USA. Retrieved associated SysQ model of Stanley’s theories on 17 May 2019 from: https://exchange.iseesystems.com/public/pontifex/fascism/index.html#page2

Stroh, D.P. (2015), ‘Systems Thinking for Social Change’, Chelsea Green Publishing, 85 North Main Street, Suite 120, White River Junction VT, USA.

Systems and Us website (2014), ‘Reinforcing and Balancing Loops and Delays’ jpeg, San Francisco, USA. Retrieved 09 October 2019 from: https://systemsandus.files.wordpress.com/2012/07/reinforcing-and- balancing-loops-and-delays.jpg?w=439&h=191

Taylor, P.M. (2007), ‘Munitions of the mind': A brief history of military psychological operations’, Place Branding and Public Diplomacy, July 2007, Vol. 3, Issue 3, London, pp. 196-204. Retrieved 15 March 2019 from: https://search-proquest- com.wwwproxy1.library.unsw.edu.au/docview/232157988/fulltextPDF/D3E27DA74F104D92PQ/1?acco untid=12763

Thaler, R.H. (2015), ‘Misbehaving: The Making of Behavioral Economics’, W.W. Norton & Company Inc., New York, USA.

Thaler, R.H. & Sunstein, C.R (2009), ‘Nudge: Improving Decisions about Health, Wealth and Happiness’, Penguin Books Ltd, Penguin Group Inc, New York, USA.

‘The Facebook Dilemma’ (2018), Frontline Report documentary, Episodes 1 & 2, USA, viewed on 10 January 2019, SBS on Demand, Australia.

Tucker, P. (2016), ‘The US Is Losing at Influence Warfare. Here’s Why’, ‘Threats’ webpage, Defense One website, 05 December 2016. Retrieved 15 October 2019 from: https://www.defenseone.com/threats/2016/12/us-losing-influence-warfare-heres-why/133654/

227 | P a g e 5249168 Cassandra Brooker Masters of Research

Uren, T. (2017), ‘Cyber threats—democracy under fire on two fronts’, The Strategist, Australian Strategic Policy Institute (ASPI) website, posted 6 July 2017. Retrieved 13 February 2019 from: https://www.aspistrategist.org.au/cyber-threats-democracy-fire-two-fronts/

Uren, T. (2018), ‘Cambridge Analytica, Facebook and Russian election interference’, The Strategist, Australian Strategic Policy Institute (ASPI) website, posted 23 March 2018. Retrieved 13 February 2019 from: https://www.aspistrategist.org.au/cambridge-analytica-facebook-russian-election-interference/

US Department of Defense (2011), ‘Joint Publication 5-0: Joint Operation Planning’, of 11 August 2011, pp. 1-264.

US Department of Defense (2014), ‘Joint Publication 3-13: Information Operations’, of 27 November 2012, incorporating Change 1 of 20 November 2014, pp. 11-73.

US Department of Defense (2017), ‘Organizing for Joint Operations’, Chapter IV, Joint Publication 3-0: Joint Operations, 17 January 2017, pp. 102-115.

US Department of Defense (2018), ‘Joint Concept for Operations in the Information Environment’, 25 July 2018, pp. 1-17.

Vaughan, D. (1992), ‘Theory elaboration: the heuristics of case analysis’, Chapter 8 in: Ragin, C.C. & Becker, S.B (1992), ‘What is a Case? Exploring the Foundations of Social Inquiry’, Cambridge University Press, New York, USA.

Waldman, P. (2019), ‘Putin helped Trump in 2016. What is he planning for 2020?’, The Plumb Line: Opinion page, The Washington Post, USA. Retrieved 10 October 2019 from: https://www.washingtonpost.com/opinions/2019/10/08/putin-helped-trump-what-is-he-planning/

Walker, C., Kalathil, S. & Ludwig, J. (2018), ‘Forget Hearts and Minds: Soft power is out; sharp power is in. Here's how to win the new influence wars’, Foreign Policy website, ‘Argument’ webpage, published 14 September 2018. Retrieved 12 March 2019 from: https://foreignpolicy.com/2018/09/14/forget-hearts-and- minds-sharp-power/

Wallis, J. (2019), ‘China and Russia aren’t the same when it comes to information warfare’, The Strategist, Australian Strategic Policy Institute (ASPI) website, 25 September 2019. Retrieved 09 January 2020 from: https://www.aspistrategist.org.au/china-and-russia-arent-the-same-when-it-comes-to-information-warfare/

Ward, T. (2018), ‘The Wahhabi Code: How the Saudis Spread Extremism Globally’, English edition, Arcade Publishing, New York, USA.

Wardle, C. (2018), ‘Information Disorder: The Essential Glossary’, Shorenstein Center on Media, Politics and Public Policy, July 2018, Research by: Greason, G., Kerwin, J. & Dias, N., Harvard Kennedy School, USA. Retrieved 27 January 2020 from: https://firstdraftnews.org/wp-content/uploads/2018/07/infoDisorder_glossary.pdf?x30563

Wardle, C. (2019), ‘Misinformation Has Created a New World Disorder’, Scientific American, September 2019, Springer Nature America Inc. Retrieved 26 August 2019 from: https://www.scientificamerican.com/article/misinformation-has-created-a-new-world-disorder/

Watts, C. (2018), ‘Messing with the Enemy: surviving in a social media world of hackers, terrorists, Russians and fake news’, Harper-Collins Publishers, New York, USA.

Wieviorka, M. (1992), ‘Case studies: history or sociology?’, Chapter 7 in: Ragin, C.C. & Becker, S.B (1992), ‘What is a Case? Exploring the Foundations of Social Inquiry’, Cambridge University Press, New York, USA.

Wu, T. (2016), ‘The Attention Merchants: The epic scramble to get inside our heads’, Alfred A. Knopf, Penguin Random House LLC, New York, USA. 228 | P a g e 5249168 Cassandra Brooker Masters of Research

Yaworsky, W. (2009) ‘Target Analysis of Shining Path Insurgents in Peru: An Example of US Army Psychological Operations’, Journal of Strategic Studies, August 2009, Vol. 32, No. 4, pp. 651-666. Retrieved 15 March 2019 from: https://www-tandfonline- com.wwwproxy1.library.unsw.edu.au/doi/pdf/10.1080/01402390902987087?needAccess=true

Yin, R.K. (1994), ‘Case Study Research: Design and Methods’, Second Edition, Applied Social Research Methods Series, Vol. 5, Sage Publications, Thousand Oaks, California.

229 | P a g e 5249168 Cassandra Brooker Masters of Research Appendices

Appendix One: Glossary and Acronyms Appendix Two: Key concepts from the systems thinking literature Appendix Three: Key findings from systems literature as it relates to case studies Appendix Four: Key findings from systems thinking literature as it relates to the iWar Appendix Five: Linkages between the three research disciplines’ literature

230 | P a g e 5249168 Cassandra Brooker Masters of Research Appendix One: Glossary and Acronyms

ADF: Australian Defence Force

Antifragile: is used to mean the direct opposite of fragile; a fragile system is one that deteriorates when stressed, an antifragile system grows stronger. [Albino et al. (2016:12)]

Botnet: A botnet is a collection or network of bots that act in coordination and are typically operated by one person or group (see ‘Bots’ below). Commercial botnets can include as many as tens of thousands of bots. [Wardle (2018:3)]

Bots: Bots are social media accounts that are operated entirely by computer programs and are designed to generate posts and/or engage with content on a particular platform. In disinformation campaigns, bots can be used to draw attention to misleading narratives, to hijack platforms’ trending lists, and to create the illusion of public discussion and support. Researchers and technologists take different approaches to identifying bots, using algorithms or simpler rules based on number of posts per day. [Wardle (2018:3)]

Centre of Gravity (COG): a characteristic, capability or locality from which a military force, nation or alliance derives its freedom of action, strength or will to fight. [LWD-1 (2002:14)]

CIMIC: Civil and Military Co-operation

Civcas: Civilian causalities

Civpop: Civilian population

Critical Capabilities (CC): inherent capabilities enabling a COG to function as such. [Grade 3 Operations Battle Book (2006:vii) ]

Critical Requirements (CR): essential conditions, resources and means for a critical capability to be fully operative. [Grade 3 Operations Battle Book (2006:vii)]

CSCC: Center for Strategic Counterterrorism Communications (USA)

Critical Vulnerabilities (CV): a characteristic or key element of a force that, if destroyed, captured or neutralised will significantly undermine the capability of the force and its centre of gravity. A CV is not necessarily a weakness but any source of strength or power that is capable of being attacked or neutralised. [LWD-1 (2002:14)]

Data Mining: Data mining is the process of monitoring large volumes of data by combining tools from statistics and artificial intelligence to recognize useful patterns. Through collecting information about an individual’s activity, disinformation agents have a mechanism by which they can target users on the basis of their posts, likes, and browsing history. A common fear among researchers is that, as psychological profiles fed by data mining become more sophisticated, users could be targeted based on how susceptible they are to believing certain false narratives. [Wardle (2018:3)]

Disinformation: Dissemination of explicitly false or misleading information [Benkler et al. (2018:32)]. False information deliberately and often covertly spread in order to influence public opinion or obscure the truth [Merriam-Webster Online Dictionary].

DoD: Department of Defense (USA)

231 | P a g e 5249168 Cassandra Brooker Masters of Research

Effect: The consequence of an action or cause, which impacts on physical, physiological, psychological or functional capabilities. [ADDP 3.14—Targeting, Australian Defence Doctrine Publication (ADDP), edition 3, Operations Series, 12 July 2018.]

Fact-checking: Fact-checking (in the context of information disorder) is the process of determining the truthfulness and accuracy of official, published information such as politicians’ statements and news reports. Fact-checking emerged in the U.S. in the 1990s, as a way of authenticating claims made in political ads airing on television. There are now around 150 fact- checking organisations in the world, and many now also debunk misinformation and disinformation from unofficial sources circulating online. [Wardle (2018:4)]

Gas Lighting: to manipulate someone by psychological means into doubting their own sanity. This may be achieved by manipulating or denying the truth to warp another’s view of the world. Gas Lighting can create self-doubt and self-censorship; and it is being perpetuated repeatedly and successfully through social media. [Singer & Brooking (2018:116)]

IC: Intelligence Community

Information Activities (IA): the integration, synchronisation and coordination of two or more information related capabilities that generate and sustain a targeted information advantage. [ADDP 3.13; ADDP 2.1]

Information Environment (IE): the aggregate of individuals, organisations or systems that collect, process or disseminate information. [ACP 167 – ADFP 2.0.1]

Information Environment (IE): the IE is a heterogeneous global environment where humans and automated systems observe, orient, decide, and act on data, information, and knowledge. [US DoD Strategy for Operations in the Information Environment 2016]

Information Operations (IO): the integrated employment, during military operations, of information related capabilities in concert with other lines of operations to influence, disrupt, corrupt, or usurp the decision making of adversaries and potential adversaries while protecting our own. [US DoD Strategy for Operations in the Information Environment 2016]

Information Warfare (iWar): is a concept involving the battlespace use and management of information and communication technology in pursuit of a competitive advantage over an opponent. [Kiyuna, A. & Conyers, L. ‘Cyberwarfare Sourcebook’ 14 April 2015]

IRA: Russia’s Internet Research Agency

ISIS: “Islamic State of Iraq and al-Sham” or, alternatively, "Islamic State of Iraq and Syria”

Kinetic: Involving the use of forces of dynamic motion/energy to achieve an effect (see ‘effect’) Note: Includes traditional explosive weapons as well as capabilities that can create radiofrequency effects such as continuous wave jammers, lasers, directed energy and pulsed radiofrequency weapons. [ADDP 3.14—Targeting, Australian Defence Doctrine Publication (ADDP), edition 3, Operations Series, 12 July 2018.]

LOAC: Laws of Armed Conflict

Mainstream Media (MSM): traditional or established broadcasting or publishing outlets.

232 | P a g e 5249168 Cassandra Brooker Masters of Research Malinformation: Malinformation is genuine information that is shared to cause harm. This includes private or revealing information that is spread to harm a person or reputation. [Wardle (2018:5)]

Manipulation: Directly influencing someone’s beliefs, attitudes, or preferences in ways that fall short of what an empathetic observer would deem normatively appropriate in context. [Benkler et al. (2018:30)]

Manufactured Amplification: Manufactured Amplification occurs when the reach or spread of information is boosted through artificial means. This includes human and automated manipulation of search engine results and trending lists, and the promotion of certain links or hashtags on social media. There are online price lists for different types of amplification, including prices for generating fake votes and signatures in online polls and petitions, and the cost of down-ranking specific content from search engine results. [Wardle (2018:5)]

Meme: The formal definition of the term meme, coined by biologist Richard Dawkins in 1976, is an idea or behaviour that spreads person to person throughout a culture by propagating rapidly, and changing over time. The term is now used most frequently to describe captioned photos or GIFs that spread online, and the most effective are humorous or critical of society. They are increasingly being used as powerful vehicles of disinformation. [Wardle (2018:5)]

Misinformation: Communication of false information without intent to deceive, manipulate, or otherwise obtain an outcome [Benkler et al. 2018:37]. Misinformation is information that is false, but not intended to cause harm. For example, individuals who don’t know a piece of information is false may spread it on social media in an attempt to be helpful [Wardle (2018:5)].

MoE: Measures of Effectiveness

Non-kinetic: Not involving the use of forces of dynamic motion and/or energy to achieve an effect (see ‘kinetic’). [ADDP 3.14—Targeting, Australian Defence Doctrine Publication (ADDP), edition 3, Operations Series, 12 July 2018]

Propaganda: Communication designed to manipulate a target population by affecting its beliefs, attitudes, or preferences in order to obtain behaviour compliant with political goals of the propagandist [Benkler et al. (2018:29)]. Ideas, facts, or allegations spread deliberately to further one's cause or to damage an opposing cause [Merriam-Webster Online Dictionary].

Propaganda by Deed: using violence as a form of communication [Murray & Blannin (2017)]. This term can also be applied to undertaking soft power or hearts and minds initiatives that provide actions to back up promises.

Propaganda Feedback Loop: A network dynamic in which media outlets, political elites, activists, and public figures form and break connections based on the contents of statements, which progressively lowers the costs of telling lies that are consistent with a shared political narrative, and increases the costs of resisting that shared narrative in the name of truth. [Benkler et al. (2018:33)]

PSYOPS: Psychological Operations, a form of information operation carried out by the military.

QCA: Qualitative Comparative Analysis

ROE: Rules of Engagement

RW: Right Wing 233 | P a g e 5249168 Cassandra Brooker Masters of Research

Satire: Satire is writing that uses literary devices such as ridicule and irony to criticise elements of society. Satire can become misinformation if audiences misinterpret it as fact. There is a known trend of disinformation agents labelling content as satire to prevent it from being flagged by fact- checkers. [Wardle (2018:6)]

SEME: Search engine manipulation effect

Sock Puppet: A sock puppet is an online account that uses a false identity designed specifically to deceive. Sock puppets are used on social platforms to inflate another account’s follower numbers and to spread or amplify false information to a mass audience. The term is considered by some to be synonymous with the term “bot”. [Wardle (2018:6)]

Troll Farm: A troll farm is a group of individuals engaging in trolling or bot-like promotion of narratives in a coordinated fashion (see ‘Trolling’ below). One prominent troll farm was the Russia-based Internet Research Agency that spread inflammatory content online in an attempt to interfere in the U.S. presidential election. [Wardle (2018:7)]

Trolling: Trolling is the act of deliberately posting offensive or inflammatory content to an online community with the intent of provoking readers or disrupting conversation. Today, the term “troll” is most often used to refer to any person harassing or insulting others online. However, it has also been used to describe human-controlled accounts performing bot-like activities. [Wardle (2018:6)]

TSA: Target System Analysis

WoG: Whole of Government

System Definitions: A Glossary [Meadows 2008:4, 13, 95, 187-188]

Archetypes: Common system structures that produce characteristic patterns of behaviour.

Balancing feedback loop: A stabilising, goal-seeking, regulating feedback loop, also known as a “negative feedback loop” because it opposes, or reverses, whatever direction of change is imposed on the system.

Behaviour: is defined as a system’s ‘performance over time’ – with ‘events’ viewed as simply the outputs of a system and its structure.

Bounded rationality: The logic that leads to decisions or actions that make sense within one part of a system but are not reasonable within a broader context or when seen as a part of the wider system.

Dynamic equilibrium: The condition in which the state of a stock (its level or its size) is steady and unchanging, despite inflows and outflows. This is possible only when all inflows equal all outflows.

Dynamics: The behaviour over time of a system or any of its components.

Feedback loop: The mechanism (rule or information flow or signal) that allows a change in a stock to affect a flow into or out of that same stock. A closed chain of causal connections from a stock, through a set of decisions and actions dependent on the level of the stock, and back again through a flow to change the stock. 234 | P a g e 5249168 Cassandra Brooker Masters of Research

Flow: Material or information that enters or leaves a stock over a period of time. Flow direction is depicted using feedback ‘link’ arrows.

Hierarchy: Systems organized in such a way as to create a larger system. Sub-systems within systems.

Interconnections: Interconnections are the relationships that hold the system elements together. They can be physical, informational, emotional, cultural etc. Relationships can be linear or non- linear.

Limiting factor: A necessary system input that is the one limiting the activity of the system at a particular moment.

Linear relationship: A relationship between two elements in a system that has constant proportion between cause and effect and so can be drawn with a straight line on a graph. The effect is additive.

Nonlinear relationship: A relationship between two elements in a system where the cause does not produce a proportional (straight-line) effect.

Paradigms: the deepest set of beliefs within a society, about how the world works; ‘paradigms are the sources of systems’ [Meadows 1999:17-18]

Reinforcing feedback loop: An amplifying or enhancing feedback loop, also known as a “positive feedback loop” because it reinforces the direction of change. These can be either vicious or virtuous cycles.

Resilience: The ability of a system to recover from perturbation; the ability to restore, repair or bounce back after a change due to an outside force.

Self-organisation: The ability of a system to structure itself, to create new structure, to learn, or diversify.

Shifting dominance: The change over time of the relative strengths of competing feedback loops.

Source and Sink: Stocks at the beginnings and ends of flows are called sources and sinks, respectively. They mark the boundary of the system diagram but rarely mark a real boundary, because systems rarely have real boundaries.

Stock: An accumulation of material or information that has built up in a system over time.

Sub-optimisation: The behaviour resulting from a sub-system’s goals dominating at the expense of the total system’s goals.

System: A set of elements or parts that is coherently organised and interconnected in a pattern or structure that produces a characteristic set of behaviours, often classified as its “function” or “purpose.”

235 | P a g e 5249168 Cassandra Brooker Masters of Research Appendix Two: Key concepts from systems thinking literature

Author Key Findings Meadows • One of the leading authorities in systems literature (2008) • Decomposes systems into key components to understand structure and behaviours • Provides understanding into system workings and what type of leverage can be applied to improve underlying behaviours • The decomposition & definitions of system elements assists understanding behavioural patterns, common traps, and applying leverage for change • Gap = predominant use of examples of functional economic systems to demonstrate theories; rather than intangible, complex social systems with non-physical resources, stocks, sources or sinks Stroh (2015) • Supplements Meadows with systems thinking for social change • Expands Meadows’ iceberg theory for complex social systems • Highlights hidden part of iceberg (complex system) – patterns of behaviour, structure, mental models shaping the system, which can be leveraged • These hidden elements unveil root causes of chronic, complex problems - links with Parts 2&3 literature review • Provides insights/solutions for contemporary social problems, encourages change within accessible systems, enables prospective analysis of lessons • Limitation = Applies systems thinking retrospectively to understand why social systems are successful or unsuccessful • Gap = approach not as easily applied to externally analysing adversary systems, which are not transparent, accessible or obvious Gharajedaghi • Analyses components & relationships characterising and defining sociocultural systems, which comprise (2011) elements linked almost entirely by interconnections of information and shared image (culture) • Sociocultural systems’ shared image constitutes its principal bond of members • Analysing the behaviour of information-bonded sociocultural systems presents a different proposition from that of an open social or functional system (eg: Meadows / Stroh) • Integration of components, with individual thoughts & behaviours, into a cohesive whole is an ongoing challenge (links to Parts 2&3 Literature Review) Albino et al. • Champions systems theory to address military strategies against complex adversary systems, albeit from (2016) kinetic targeting perspective Hall & • Recommend a systems-based analytical approach to intelligence analysis, to complement other Citrenbaum assessment techniques (2010) Goble (2002) • Early advocate of applying systems thinking to complex IO • Limitation = Goble‘s experience of the IE in 2002 was not today’s multifaceted, interconnected, 3D, media-saturated battlespace

236 | P a g e 5249168 Cassandra Brooker Masters of Research Appendix Three: Key findings from systems literature relating to case studies

Author Benefits to Case Study Analysis Meadows • Stock flow changes, feedback mechanisms, hierarchies, behaviours driving events, system resilience & leverage = (2008) linked to case studies • Complex systems’ ability to learn, diversify & evolve (ISIS) • ‘Success to the Successful’ principle (ISIS) • Higher meta-resilience achieved by self-organising systems (ISIS) • Psychological /political tendencies to blame external influences for causes of system problems (Clinton) • Competing goals of sub-systems (Clinton) • ‘Drift to Low Performance’ archetype (Clinton) • Intrinsic systems problems, undesirable perceptions, and behavioural characteristics of system structures that produce them (Clinton) • Gap = Analyses physical systems, with functional goals, tangible stocks with balancing feedback, visible capping/checking occurs

Stroh (2015) • Social problems of systems directly correlate to behaviours observed in the case studies • Reactive ‘quick fixes’ and ‘fixes that backfire’ (Clinton) • Lack of investment in long-term, core solutions (Clinton) • ‘Accidental Adversaries’ (Clinton) • ‘Competing Goals’ and conflicting messages of sub-systems (Clinton) • Unintended ‘Escalation’ with an adversary (Clinton) • Self-created limits on ‘Growth or Investment’ (Clinton) • ‘Drifting Goals’ where tolerance is developed for low performance (Clinton) • External system events, eg: exploitation of social media (undermined Clinton, strengthened ISIS) •‘Tragedy of the Commons’ - integrity of the internet (both Stroh & Meadows) • Limitation = Virtuous & Vicious feedback loop theory more difficult to apply neatly to the declining influence stock of Clinton campaign

Gharajedaghi • Sociocultural systems analysis lends well to analysing the ISIS case study (2011) • Inertia and resistance to change discussion, linked to Muslim society - makes cultures resilient and sustainable (ISIS) • ISIS remain viable by questioning assumptions, being self-organising & self-evolving • Any system surviving initial attempts at suppression & resonates with pre-existing frustrations, sees an iterative process of deviation amplification (ISIS able to produce structural changes in its less developed social system) • Conversely, extreme ideologies have proven to be major obstructions to the viability of a social system (ISIS’ balancing feedback loop) Albino et al. • Discusses alternative system responses to attack - depending whether a system is fragile, robust, resilient, anti-fragile (2016) or anticipatory • ‘Robustness’ & ‘resilience’ characterise systems able to tolerate stress / recover from shock, so targeting them is wasted effort (ISIS) • A ‘fragile’ system deteriorates when stressed (Clinton?) • ‘Anti-fragile’ system grows stronger, must be opposed by other dimensions (ISIS) • ‘Anticipatory’ responses are a key capability of complex systems and the ideal (iWar) • Limitation = Concerned primarily with addressing weaknesses in current military strategic thinking surrounding kinetic targeting

Hall & • Systems approaches to intelligence analysis, which can be readily applied to case studies. Citrenbaum • Decomposition enables insight into attributes, characteristics, interactions, and behaviours of adversary systems - (2010) links to Parts 2&3 Literature Review • Systems thinking enhances complementary intelligence analysis techniques (ISIS) Goble (2002) • Early champion of using systems thinking for PSYOPs (ISIS) • Models of adversary’s decision systems are integral to effective IO targeting (ISIS) • ‘Complexity science’ must be used to predict /measure IO effects, in same way TSA predicts & measures kinetic weapon effects (ISIS)

237 | P a g e 5249168 Cassandra Brooker Masters of Research Appendix Four: Key findings from systems literature relating to the iWar

Author System Thinking Benefits to iWar

Meadows • Provide tools, definitions and methodology for practical iWar analytical systems linking (2008) • Systems thinking theories and methods already being applied for kinetic target systems analysis Stroh (2015) • Provide tools, definitions, methodology for practical iWar analysis of complex neutral, friendly and enemy social systems in the iWar • Social problems covered by Stroh’s systems theories can be readily applied to Western Democratic political, economic, civilian and military systems operating in the iWar - for improved system functioning, resilience, effectiveness and changing system behaviours • Both Stroh & Albino et al. (2016) demonstrate how complex social systems, such as ISIS, are able to strengthen themselves through feedback mechanisms, using setbacks as a source of inspiration for future responses Gharajedaghi • Complements Stroh in his theories for conducting systems thinking analysis for complex friendly and (2011) enemy sociocultural systems • Focus on cultural aspects & behaviours - that link Part 2 & 3 literature review for ensuring a holistic systems analysis for iWar problems Albino et al. • Criticises current doctrinal approaches focused on applying force; champions systems thinking (2016) • “Traditional perspectives of military strategy implicitly assume fragility”, which undermines targeting • Complements Goble (2002): military planners must consider non-conventional capabilities contributing to adversary centre of gravity • Systemic evolutions & adaptations are infrequently adopted by Western military strategic planners to enable better responses to the dynamic and unpredictable nature of iWar Hall & • Combines traditional doctrinal intelligence analysis techniques with innovative approaches and systems Citrenbaum thinking (2010) • Champions the conduct ‘decomposition’ at beginning of the intelligence process • ‘Recompose’ the data and fuse with intelligence, synthesise into whole-of-system knowledge to drive collection & effective targeting Goble (2002) • MoE could be established by collecting intelligence on specific links and nodes within the system to determine impacts on functionality (NB: current intelligence & targeting processes focus on nodes only) • Linear thought processes, prevalent in the military, lack understanding of non-linear effects of non- kinetic influence • Limitation = recommends establishing military relationships with complexity (systems) scientists / think tanks = raises ethical issues of conducting human terrain analysis by civilians for targeting purposes & outsourcing systems thinking means knowledge is lost / siloed • Benefit = generation of more academic literature on the subject, presently a clear gap

238 | P a g e 5249168 Cassandra Brooker Masters of Research Appendix Five: Linkages between the three research disciplines’ literature

Part 1: Systems Thinking Part 2: Effective Influence Part 3: Behavioural theory

1) Success to the Successful • All influencing tactics in Table 5 reinforce ‘Success to the • Links to behaviours in Table 8, depends on stocks Successful’ • Exploits heuristics of System 1 • ISIS’ reinforcing feedback of influence = virtuous cycle • ‘Appeal to Authority’ cognitive bias

2) Shifting the Burden - to the • Influence activities can nudge systems into choosing easy • Addictive / Quick Fix Intervenor solutions, reassign blame, avoid critical thinking • Reactive System 1 decision-making / shortcuts • Limited attention spans / computational capacity

3) Accidental Adversaries • Emotional & social contagions • Linked to competitive behaviours / biases • Disinformation / fake news • System 1 thinking, present bias • eg: Clinton campaign competing with own sub-systems • Not consider consequences of behaviour

4) Fixes that Backfire/quick fixes, • Biased/swayed decision-making • System 1 thinking Unintended consequences • Addictive solutions

5) Drifting Goals / Drift to Low • Social & Emotional Contagions • Negative bias in perceiving past performance Performance • Allowing performance standards to be influenced by past • Sets up reinforcing feedback loop of eroding goals performance, eg: Clinton campaign • Linked to normative behaviours & emotions

6) Competing Goals – related to • Related to ‘Accidental Adversaries’ • Related to ‘Accidental Adversaries’ accidental adversaries • Linked to ‘Quick Fixes’ above • Linked to ‘Quick Fixes’ • Sub-optimization • Negative behaviours, limited cognition, System 1

7) Escalation – vicious cycles • eg: Clinton Campaign reacting to other systems’ outputs • System 1, reactive behaviour or external events

8) Limits to Growth – Balancing • Media bias, covert influence, audience susceptibility • Exploit ‘System 1’, non-critical thinking and Reinforcing feedback loops • Importance of controlling the narrative • Play on biases, emotions, contagions, social norms • Disinformation / fake news

9) Vicious / Virtuous circles – • All influencing tactics (Table 5) feed either vicious or • Links to all behaviours listed in Table 8, depending on (also see loops below) virtuous cycles the stocks and flows targeted • Stocks & Flows • Influence impacts on external system behaviours & • Behavioural theories provide insight into the features internal stocks of a system impacted upon by external pressures, • eg: influence stocks and information flows (Fig 2) such as interconnections, stocks, flows

10) Balancing process • All influence tactics (Table 5) provide balancing feedback, • Feedback from vicious & virtuous circles drives • Balancing feedback loop esp. addictive content & feedback mechanisms reactive behaviours / how it is interpreted • Reinforcing feedback loop • Effective influence forces complex social systems to • Behavioural responses to feedback influences stocks, – vicious or virtuous circles adapt behaviour - based on feedback flows, interconnections of complex social systems

11) Tragedy of the Commons • Russian / ISIS exploitation of internet, accesses • Exploits peer group pressure, social norms • All influencing tactics (Table 5) enabled by uncontrolled, • ‘Homophily’ / ‘Confirmation Biases’ unregulated, global internet commons • Exploits heuristics of System 1, emotions, biases, cues

12) Growth / Underinvestment • Controlling the narrative (in own organisation) • Self-created limits, System 2 heuristics

Linear and Nonlinear • Effective influence exploits linkages of target systems • Understanding behavioural impacts on the relationships (interconnections) • Examining interconnections helps identify vulnerabilities, interconnections within a system is important for between two elements / nodes in weaknesses &discord in own system & strengthen effective influencing or targeting a system against enemy influence • Hard & soft communications links • Systems analysis allows target links to be exposed, manipulated, weakened, leveraged

Resilience: The ability of a system • Misguided targeting can strengthen anti-fragile systems • Behavioural theories assist in identifying system to recover, restore or repair after • Level of resilience, characteristics & attributes determine characteristics i.e. level of fragility, mental models a change, due to an outside force. how to effectively target/influence system • The strengths & vulnerabilities of links & flows determines strength & behaviours of target system

Bounded Rationality - impacts • All influencing tactics are designed to confuse, mislead or • Humans misinterpret information, misperceive risk, system functioning manipulate target audience systems focus on current events/outputs rather than long- term behaviour or structural problems

Mental Models – hidden part of • To effectively influence vital system behaviours must first • Cognitive biases in a system allows the exploitation or the iceberg (Fig 1) be identified that will drive required changes manipulation of that system’s behaviours • Social systems are influenced by observing behaviour of • To change systems, you must change behaviours & others (social /emotional contagions) understand how they think

239 | P a g e 5249168 Cassandra Brooker Masters of Research END

240 | P a g e 5249168 Cassandra Brooker Masters of Research