DIALOGIC REGULATION: THE TALKING CURE FOR CORPORATIONS by Michael Cody LL.M., The University of British Columbia, 2006 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Law) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) July 2014 © Michael Cody, 2014 ii Abstract The corporations of our future will be whatever we can collectively imagine and work together to make a reality. Dialogic law and regulation is a generative tool that can build the bridge between the present and an imagined future. Regulators keep people on the bridge by identifying the kinds of dialogues we want corporate actors to have and by encouraging, coaching, and sometimes assisting them to have those dialogues. This approach works because small changes in the way corporate actors talk to and interact with each other can have dramatic effects on the emergent corporate culture. This thesis develops and tests a theory of Dialogic Regulation. The theory assumes that corporate law and regulation is about attaining or maintaining a desired corporate behaviour, the best way to change behaviour is to learn a new one, and learning is a social process that involves dialogue. The model was tested using an experimental game where the rules of the game were treated as proxies for the “law” and the authority figure directing the experiment was treated as a proxy for the corporate “regulator”. The game was called the “Pay-Off” game. Half-way through the game the rules were changed using one of three different regulatory techniques: 1) Rules: a simple rule change, 2) Audit: a rule change combined with an audit and punishment procedure for infractions, and 3) Dialogic: a rule change combined with a dialogic intervention about the rules. Participants were tested not only for their behavioural reactions to the interventions (Compliance to the rules) but also to determine if they learned anything about the rules (Adherence to the rules). The games experiment showed that for simply behavioural outcomes the Audit Based Regulation approach was the most effective. The experiment also showed that there is significant iii promise in a Dialogic Regulation approach if the regulatory desire is to have participants learn. While Dialogic Regulation shows promise, a lot more work needs to be done to refine the application of the theory before it is used in real-life regulatory settings. iv Preface The work presented in this thesis is mine. I identified and designed the research program, performed all parts of the research (including the experiment) and analyzed the research data. The research required ethics approval from The University of British Columbia (BREB No. H12-02737) and Simon Fraser University (Study No. 2012S0958). The case study mentioned in Chapter 6 relating to Suncor’s Firebag facility creative sentence was part of a study conducted by a team from the University of Calgary and Simon Fraser University. I was not a member of the core research team. My participation in the study was extremely limited. I was invited to assist only at the two-day knowledge forum and my participation was limited primarily to the closed-door session on the second day. I have co-authored an article on the case study with Professor Stephanie Bertels and Simon Pek from SFU entitled: “A Responsive Approach to Organizational Misconduct: Rehabilitation, Reintegration, and the Reduction of Reoffense”. It covers topics other than what is presented in this thesis but some of the descriptions of the background of the case are the same. v Table of Contents Abstract .......................................................................................................................................... ii!Preface ........................................................................................................................................... iv!Table of Contents ...........................................................................................................................v!List of Tables ................................................................................................................................ xi!List of Figures .............................................................................................................................. xii!Glossary ...................................................................................................................................... xiii!Acknowledgements .................................................................................................................... xix!Chapter 1: Introduction ................................................................................................................1!Chapter 2: Developments in Regulatory Theory: The Potential of a Complex World ...........8!2.1! Developments in Corporate Regulation ........................................................................... 10!2.2! Developments in Regulatory Theory ............................................................................... 16!2.2.1! Responsive Regulation .............................................................................................. 17!2.2.2! Smart Regulation ...................................................................................................... 19!2.2.3! Self-Regulation and Meta-Regulation ...................................................................... 20!2.2.4! Democratic Experimentalism .................................................................................... 22!2.2.5! New Governance ....................................................................................................... 24!2.2.6! Nodal Governance .................................................................................................... 25!2.2.7! The Similarities of the New Approaches .................................................................. 27!2.3! The Learning Approach to Corporate Regulation ........................................................... 28!2.4! Systems Theory, Organization Theory, and Complexity ................................................ 31!2.5! Conclusion – Learning in a Complex World ................................................................... 36!Chapter 3: How Organizations Learn: The Power of Dialogue ..............................................38!3.1! Organizational Learning and Organizational Development ............................................ 41!3.1.1! Theories of Action .................................................................................................... 42!3.1.2! Single- and Double-Loop Learning .......................................................................... 44!3.1.3! Compliance and Adherence ...................................................................................... 45!3.2! Chaos Theory – Self-Organizing Systems ....................................................................... 56! vi 3.2.1! Surfing the Edge of Chaos – The Myth of Equilibrium ............................................ 57!3.2.2! Emergence and Self-Organization ............................................................................ 59!3.2.3! Learning Complex Tasks and Generating New Ideas ............................................... 61!3.2.4! The Butterfly Effect .................................................................................................. 62!3.2.5! Directing vs. Disturbing Complex Systems .............................................................. 64!3.2.6! Chaos Theory and the Corporation ........................................................................... 67!3.2.7! Chaos Theory, the Corporation, and Dialogic Regulation ........................................ 71!3.2.8! Conclusion ................................................................................................................ 76!3.3! Language, Narrative, and Discourse ................................................................................ 77!3.3.1! Social Constructionism and the Theory of Generativity ........................................... 78!3.3.2! Organizational Discourse Studies ............................................................................. 81!3.4! Dialogic OD ..................................................................................................................... 84!3.4.1! Appreciative Inquiry and Other Dialogic OD Practices ........................................... 89!3.4.2! Social Constructionism ............................................................................................. 90!3.4.3! Generativity ............................................................................................................... 90!3.4.4! Appreciative Approach vs. Problem Solving Approach ........................................... 91!3.4.5! Language, Discourse, and Narrative ......................................................................... 93!3.4.6! Anticipatory Reality .................................................................................................. 95!3.4.7! Positive Affect .......................................................................................................... 96!3.4.8! The Practice of Appreciative Inquiry ........................................................................ 98!3.5! Dialogic Systems Theory and Corporate Law and Regulation ...................................... 103!3.6! Conclusion ..................................................................................................................... 110!Chapter 4: Individual Behaviour: The Importance of Communication ..............................111!4.1! Social Dilemma Experiments and Corporate Law and Regulation ............................... 112!4.2! Introduction to Social Dilemma Experiments ............................................................... 114!4.3! Update on Empirical Evidence from Social Dilemma Experiments (to 2010) .............. 117!4.3.1! The 1995 Sally Meta-Analysis ................................................................................ 118!4.3.2! The 2003 Summary of Social Dilemmas by Colin Camerer .................................. 124!4.3.3! The 2010 Meta-Analysis by Daniel Balliet ............................................................ 126!4.4! Update on Empirical Evidence of “Crowding Out” (to 2010) ....................................... 129! vii 4.5! The Weakness of Social Dilemma Experiments and the Link to Social Constructionism ................................................................................................................................ 134!4.6! The Results of Experimental Games and the Implications for Corporate Law and Regulation .............................................................................................................. 140!4.7! Development of Hypotheses for Corporate Law and Regulation .................................. 146!Chapter 5: Learning the Rules: An Experimental Game Approach to Corporate Law and Regulation .......................................................................................................................152!5.1! Introduction .................................................................................................................... 152!5.1.1! Purpose of the Experiment ...................................................................................... 153!5.1.2! Summary of Experiment ......................................................................................... 153!5.2! Methods ......................................................................................................................... 154!5.2.1! Participants .............................................................................................................. 154!5.2.2! Design ..................................................................................................................... 154!5.2.3! What to Comply With? Unpacking Compliance, Cooperation and Self-Interest, and Contribution and Return Rate ................................................................................. 156!5.2.4! The Pay-Off Game .................................................................................................. 156!5.3! Procedures and Materials ............................................................................................... 160!5.3.1! The Rule Change ..................................................................................................... 163!5.3.2! The Three Experimental Conditions ....................................................................... 163!5.3.3! The Pre-Disposition Survey .................................................................................... 165!5.3.4! The Post-Game Data Collection ............................................................................. 166!5.4! Data and Analysis .......................................................................................................... 167!5.4.1! Compliance Behavioural Measures ........................................................................ 168!5.4.2! Adherence Behavioural Measures .......................................................................... 169!5.4.3! Trust Questions ....................................................................................................... 169!5.4.4! SVO Slider Measure ............................................................................................... 171!5.4.5! Open-Ended Survey Question ................................................................................. 174!5.4.6! Post-Game Interviews ............................................................................................. 174!126.96.36.199! Level of Significance and Significance Testing .............................................. 175!5.5! Results ............................................................................................................................ 175! viii 5.5.1! Participants .............................................................................................................. 175!5.5.2! Behavioural Measure .............................................................................................. 182!5.5.3! Trust Questions ....................................................................................................... 183!5.5.4! SVO Changes .......................................................................................................... 184!5.5.5! Qualitative Data ...................................................................................................... 188!5.5.6! Summary of Adherence Data .................................................................................. 191!5.6! Discussion ...................................................................................................................... 198!5.6.1! SVO Changes in Rules Group ................................................................................ 198!5.6.2! Dialogic Group Changing SVOs in the Wrong Direction ...................................... 198!5.6.3! Conclusion: Dialogic plus Audit may Amplify Results ......................................... 202!5.6.4! Limitations of the Experiment ................................................................................ 203!5.6.5! Avenues for Future Research .................................................................................. 204!5.6.6! Contribution to the Fields of Regulation, Dialogic OD, and Experimental Games 205!5.7! Conclusion ..................................................................................................................... 206!Chapter 6: Dialogic Law and Regulation: A Bridge to an Imagined Future .......................207!6.1! Dialogic Law and Regulation ........................................................................................ 213!6.2! Dialogic Corporate Regulation ...................................................................................... 240!6.3! Continuous Dialogic Regulation – Dialogic Learning Loops ....................................... 243!6.4! Episodic Dialogic Regulation ........................................................................................ 252!6.5! When to Use Dialogic Regulation ................................................................................. 252!6.6! Factors that Increase the Chance of a Successful Dialogic Intervention ....................... 253!6.7! Limitations on the Use of Dialogic Regulation ............................................................. 259!6.8! Suncor’s Firebag Creative Sentencing Project .............................................................. 261!6.8.1! Summary ................................................................................................................. 261!6.8.2! Creative Sentencing ................................................................................................ 261!6.8.3! The Creative Sentencing Process ............................................................................ 264!6.8.4! Discussion of the Dialogic Implications of the Project .......................................... 265!188.8.131.52! Getting Ready .................................................................................................. 266!184.108.40.206! Holding the Dialogic Event ............................................................................. 268!220.127.116.11! Incorporating the Changes ............................................................................... 270! ix 6.9! Template Dialogic Regulation Creative Sentencing Project ......................................... 273!6.9.1! Selection Criteria for Dialogic Intervention ........................................................... 273!6.9.2! Structure of the Dialogic Intervention .................................................................... 275!18.104.22.168! Getting Ready .................................................................................................. 275!22.214.171.124! Holding the Event ............................................................................................ 276!126.96.36.199! Incorporating Changes ..................................................................................... 277!6.9.3! Summary ................................................................................................................. 278!6.10! Conclusion ..................................................................................................................... 278!Chapter 7: Conclusion: Building a Longer Bridge .................................................................280!Bibliography ...............................................................................................................................283!Appendices ..................................................................................................................................304!Appendix A Theories of Regulation ....................................................................................... 304!Appendix B Organizational Analysis ..................................................................................... 305!Appendix C The Stages in the Development of our Understanding of Organizations, Corporations, Corporate Law, and Corporate Regulations .................................... 306!Appendix D Boulding’s System Types .................................................................................. 307!Appendix E Argyris and Schon’s Double-Loop Learning ..................................................... 308!Appendix F Theory-in-Use Characteristics ............................................................................ 309!F.1! Model I ....................................................................................................................... 309!F.2! Model II ..................................................................................................................... 310!Appendix G Diagnostic Vs. Dialogic OD .............................................................................. 311!Appendix H Problem Solving Vs. Appreciative Inquiry ........................................................ 312!Appendix I Organization Theory ............................................................................................ 314!Appendix J Chaos Theory Systems Classification ................................................................. 315!Appendix K Behavioural Approach vs. Dialogic Approach .................................................. 316!Appendix L Cooperation Rates With and Without Punishment ............................................. 317!Appendix M Learning the Rules: Experimental Appendices ................................................. 318!M.1! Regulator Instructions .............................................................................................. 318!M.2! Pre-Game Survey ..................................................................................................... 319!M.3! Audit Procedure ........................................................................................................ 325! x M.4! SVO Slider Measure ................................................................................................. 327!Appendix N Learning Loops .................................................................................................. 329!N.1! Organizational Loops ................................................................................................ 329!N.2! Inter-organizational Loops 1 ..................................................................................... 330!N.3! Inter-organizational Loops 2 ..................................................................................... 331!Appendix O Cynefin Model ................................................................................................... 332!Appendix P Alberta Creative Sentencing Guidelines ............................................................. 333! xi List of Tables Table 1 Behavioural Measures .................................................................................................... 177!Table 2 Compliance and Non-Compliance Measures ................................................................ 181!Table 3 Adherence Measure ....................................................................................................... 182!Table 4 Trust Questions Pre- and Post-Game ............................................................................. 183!Table 5 SVO Categories Pre- and Post-Game ............................................................................ 185!Table 6 SVO Slider Measure – Post-Game Magnitude of Changes ........................................... 188!Table 7 Frequency of Comments on Post-Game Survey ............................................................ 190!Table 8 Frequency of Comments in Post-Game Interviews ....................................................... 190!Table 9 Behavioural Results By SVO Type – Group A ............................................................. 193!Table 10 Behavioural Results By SVO Type – Group B ........................................................... 194!Table 11 Participants Whose SVOs Changed ............................................................................. 195! xii List of Figures Figure 1 The Prisoner’s Dilemma ............................................................................................... 115!Figure 2 Examples of Regulation Approaches ........................................................................... 231! xiii Glossary Adherence Faithful support for a cause, idea, etc.; steady devotion, allegiance or attachment. Used in this study to refer to a participant’s learning of the new regulatory outcome (as opposed to simple compliance where their behaviour changed but they did not learn why the new behaviour is more desirable). Change (Dialogic) Change results from changing the conversations that shape everyday thinking and behaviour through involving more and different voices, altering how and which people talk to each other, and/or by stimulating alternative or generative images to shape how people think about things.1 Compliance To comply with the wishes of another; to do what you are told even though you do not agree. Used in this study to represent a participant’s change in behaviour to comply with the new rules even though they do not agree with them. 1 Bushe & Marshak (2013) infra note 309 at 290. xiv Container A time and space where normal, “business as usual” ways of interaction are suspended, so that different generative conversations can take place.2 Corporate Culture The shared beliefs and assumptions created, maintained and changed through conversations that lead to patterned interactions among the individuals within the corporation. Corporation (Dialogic Definition) The corporation is a socially constructed reality that consists of the patterns of interactions and conversations between the organization members. Dialogic OD Dialogic OD is constructivist in its methodology, tends to view organizations as dialogic or meaning making systems,3 does not assume the existence of a discernable reality but rather multiple versions of reality by organizational participants, and focuses on change in what people think and say, not what they do.4 A planned change process that improves organizational effectiveness by changing collective narratives in order to change collective thinking and action. 2 Ibid at 15. 3 Boje & Arkoubi (2005) infra note 290. 4 Bushe & Marshak (2009) infra note 176 at 353. xv Dialogic Regulation An approach to regulation that argues that the best way to regulate corporations is to have laws that are positive statements about the way we want corporations to be (the “law”) and that the role of corporate regulators is to have ongoing dialogue with corporate participants about how they are working toward those goals (“regulation”). Double-Loop Learning Learning in a way that involves the modification of an organization’s underlying norms, policies, and objectives.5 Experimental Games A controlled experiment or simplified scenario where decision-making can be observed and the variables that might affect decision-making can be varied. A game consists of a set of players, a set of decisions (or strategies), and a set of outcomes (pay-offs) for adopting each set of strategies. Game Theory A body of theory that has created a mathematical language to describe and predict social interaction.6 Generativity [Generativity is] . . . the capacity to challenge the guiding assumptions of the culture, to raise fundamental questions regarding contemporary social life, to foster reconsideration of that which is 5 Argyris (1978) infra note 140 at 3. 6 Camerer (2003) infra note 462 at 465. xvi “taken for granted” and thereby to furnish new alternatives for social action.7 Generative Image Ideas, phrases, objects, pictures, manifestos, stories, or new words with two properties: (1) that allow us to see new alternatives for decisions and actions, and (2) that are compelling and generate change.8 Non-Zero Sum Game A game where one player’s gain does not necessarily mean another player’s loss (and vice versa). The gains and the losses in the game do not always add up to zero. It is possible for both players to win or for both players to lose. Prisoner’s Dilemma The classic game theory game. Two conspirators are arrested and interrogated separately. If one implicates the other, he may go free while the other receives a life sentence. Yet, if both confess, bad fate befalls them. If both stay silent, insufficient evidence will lead to them being charged with and convicted of a lesser crime. It is in each person’s self interest to defect on the other prisoner because they will receive a lesser sentence. If both prisoners stay silent then 7 Kenneth Gergen, “Toward Generative Theory” (1978) 36(11) Journal of Personality and Social Psychology 1344 at 1346. 8 Gervase Bushe “Generative process, generative outcome: The transformational potential of appreciative inquiry”, in D.L. Cooperrider, D.P. Zandee, L.N. Godwin, M. Avital & B. Boland (eds.) Organizational Generativity: The Appreciative Inquiry Summit and a Scholarship of Transformation (Bingley, UK: Emerald Group Publishing Limited, 2013) Advances in Appreciative Inquiry, Volume 4, pp. 89-113. xvii they both go free. Often used to measure the amount of cooperation people are willing to show in experimental game situations. Single-Loop Learning Simple error correction that permits an organization to carry on its present policies or achieve its present objectives. Like a thermostat that feels it is too hot or cold and adjusts.9 Social Dilemma A situation or game where individuals find their own interest in conflict with what is best for their relationships, work organizations, community, nation, and perhaps, most abstractly, their own species.10 Social dilemmas are formally defined by two outcome-relevant properties: (1) each person has an individual rational strategy which yields the best outcome (or pay-off) in all circumstances (the non-cooperative choice, also known as the dominating strategy); (2) if all individuals pursue this strategy it results in a deficient collective outcome – everyone would be better off by cooperating (the deficient equilibrium). System A system is simply a combination of parts that are interrelated.11 Trust A willingness to make oneself vulnerable to another, based on the 9 Argyris (1978) infra note 140 at 3. 10 Baillet (2010) infra note 437 at 39-40. 11 Scott (2003) infra note 107 at 83. xviii belief that the trusted person will choose not to exploit one’s vulnerability (that is behave trustworthily).12 Trustworthiness Unwillingness to exploit a trusting person’s vulnerability even when external rewards favour doing so.13 Zero-Sum Game A game where the sum of all gains by a player or group of players is equal to the sum of all losses for every possible outcome of that game. In a zero sum gain one player’s gain is another player’s loss. For example, in poker your wins equal your opponent’s losses. 12 Blair & Stout (2001) infra note 420 at 1739-1740. 13 Ibid at 1740. xix Acknowledgements I offer my enduring gratitude to the faculty, staff, and my fellow students at UBC Law who have inspired me to continue work in this field. I would like to thank my committee: Bruce MacDougall, Bob Paterson, David Duff, and Gervase Bushe. I owe particular thanks to my supervisor Bruce MacDougall who patiently guided me through this seven-year process and Gervase Bushe who taught me to constantly challenge my own assumptions. I would like to thank SSHRC for the Ph.D. Canada Graduate Scholarships (CGS) Doctoral Scholarship that made this research possible and UBC Law, which provided additional funding through the Four Year Fellowship Scholarship for Ph.D. Students and the Special UBC Graduate Fellowship. I would like to thank the Liu Institute at UBC, which graciously provided financial assistance and a hosting location that allowed Sarah Elder and I to trial some dialogic processes around corporate topics in our Liu CSR Network workshop on the “Road to Rio+20” early in this research program. I would also like to thank the Canadian Business Ethics Research Network for the financial support and opportunity to present most of this work for comment and criticism at various CBERN Ph.D. student events. The following members of the network graciously provided comments on various chapters in this thesis or presentations of the materials held within them: Len Brooks, Jim Cooney, Wesley Cragg, Jeff Frooman, Donna Kennedy-Glans, Dirk Matten, Amy Mullen, Kernaghan Webb, Jan Boon, Sara Elder, Gail Henderson, Hilary Martin, Sarah Pouryousefi, and Bill Woof. xx Finally, I would like to thank my family. My parents have provided moral support throughout this process and my wife, Katerine-Lune, has been remarkably understanding and supportive even when the process extended into the time after the arrival of our first son Thomas. I love you all. 1 Chapter 1: Introduction Carl Jung and Sigmund Freud stand side by side at the railing of a passenger liner on their first trip to New York. They are discussing their ideas on the Talking Cure (Psychoanalysis): Carl Jung: “Take it from me. What you are looking at is the future.” Sigmund Freud: “You think that they know we are on our way bringing them the plague?” Scene from the movie A Dangerous Method (2011) The tension in this exchange between Carl Jung and Sigmund Freud perfectly illustrates the vexingly dichotomous nature of dialogic regulation and other dialogic systems. On the one hand, they offer an exciting paradigmatic shift in understanding that makes all kinds of imagined, and unimagined, futures possible. Dialogic processes can cause the kind of transformational change that we all want to believe is possible and they can do this in the lives of individuals, the operations of organizations and corporations, and the regulatory systems that oversee them. You can stop smoking. You can overcome your issues with your parents or siblings. A failing company can transform itself into a profitable company. Investment bankers can transform their culture. Regulators can create a space where all oil companies are continuously working to improve their environmental practices. The financial markets can transform their culture from one of advisor greed to one of protecting the interests of the investors. Past experiences with dialogic practices show the exciting promise that warrants Jung’s optimism: What we are looking at is the future. But, that future is frightening. One of the underlying features of a dialogic understanding of the world is that it is a complex place that we do not yet fully understand and cannot control. 2 While dialogic processes create containers in which transformational change can occur, it is often very difficult to predict exactly what that change will be. This resonates well with what we would intuitively think. If 100 people were brought into a room and asked “What would this company look like to you if it was at its best?” the result would be very difficult to predict until all the conversations had taken place. In the scene from A Dangerous Method above, Freud knew this about his dialogic practice. While Jung was new to Freud’s methods, Freud had been practising them with his patients for a number of years already and was all too aware of the dangers dialogic practice presented. It is fitting that the producers chose to call the movie “A Dangerous Method” and not its original title “The Talking Cure”. As human beings, we are scared of things that we cannot control. Lack of control especially scares our corporate managers, who in our current culture are supposed to be in “control” of everything all the time. Think back to the Deepwater Horizon Oil Spill and scathing criticism levelled at Tony Heyward, the CEO of a $375 billion corporation, because he did not know what was happening on one specific drill rig in the Gulf of Mexico. Lack of control also scares lawmakers and regulators, because they are in the control business. They make the rules and they enforce the rules. In their world-view, it is difficult to conceive of a situation where a corporation designs its own environmental regulations; the regulator needs to do that, even though the regulator does not understand the business as well as the people who do it every day. Is it possible to overcome this fear of lack of control? Freud’s theory of psychoanalysis was a paradigm shift for psychology. At the time that he conceived it, we were treating psychological patients in a very “command and control” way. We knew (or at least we thought we knew) how they were supposed to act, we asked them to do it, and if they did not we forced them to – even if it meant electrocuting them, isolating them 3 from society, or removing portions of their brain, etc. The paradigm shift occurred when Freud asked the questions: What if we asked them what was going on inside their head? What if we listened to their internal conversations? What if we had a dialogue with them? The result was that everything changed. We no longer prescribe electro-shock therapy for hysteria. Instead, when someone starts to develop psychological symptoms we recommend that they get counselling. Freud created the dialogic revolution for the individual. The theory of Dialogic Regulation offered herein highlights the same kind of paradigm shift in the realm of organizations and corporations and their regulation. In our current regulation of corporations, we are still using command and control regulation. We pass laws to tell them what to do. We watch vigilantly (or not so vigilantly) to see when they break those laws, then we swoop in to punish them with fines, publicly humiliate their leaders, and throw a few bad apples in jail. It is as if we are still stuck in a 19th-century version of psychology. The reality is that these practices are working about as effectively as the old-school psychology practices. We pass more laws, fine more companies, and put more corporate actors in jail every year – but things seem to be getting worse. The problem lies not with the regulatory practices, but in our underlying conception of what corporations are and how they change. In our effort to understand what corporations are we have over-simplified them and we believe that we can change them using overly simple techniques. We have, to a great extent, underappreciated the complexity of what a corporation is. A corporation cannot be changed simply by passing a law. The only way the corporation changes is if someone within that corporation becomes aware of that law, begins to talk about that law, and develops a strategy for getting that law into the daily conversations and routines of some of the people in that corporation. The reason is that individuals and corporations are dialogic 4 systems. They are not mechanical devices, living organisms, a nexus of contracts, or a formalized legal entity. Our internal lives consist of conversations with ourselves and our organizational lives consist of conversations with other people. These conversations are real and complex and it is only by embracing the existence and recognizing the significance of these conversations that we can move forward. The terms “corporation” and “organization” in this thesis have a very specific dialogic meaning. A corporation or organization is a socially constructed reality that consists of the patterns of interactions and conversations between the organization members. This is a pre-existing law and structure definition of corporations and it does not come with the baggage of the existing legal structure or history. The theory of Dialogic Regulation embraces the complexity and non-linear aspects of social systems and the importance of language, conversations, and dialogue in creating change. It asks the same questions of corporations that Freud asked of individuals: What if we asked them what was going on in their management meetings? What if we listened to their internal conversations (their manager meetings)? What if we had a dialogue with them? Dialogic regulation is a trans-disciplinary approach to corporate law and regulation based on insights from law, regulatory theory, corporate theory, systems theory, complexity theory, chaos theory, organizational theory, organizational development, psychology, and social psychology. The model assumes that: the purpose of corporate law is to change or affect behaviour; the best way to change behaviour is to learn a new behaviour; and learning is a social process that involves dialogue. The model hypothesizes that the best way to regulate corporations is to have principle-based laws that are positive statements about the way we want corporations to be (the “law”) and that the role of corporate regulators is to have ongoing dialogues with corporate participants about how they are working toward those goals (“regulation”). 5 This thesis explores dialogic regulation in three parts: an extensive literature on dialogic change practices, an experiment to see how those practices work within the context of a game where the rules are changed, and then finally the insights from the literature review and experimental game will be used to offer a framework for a theory of dialogic regulation. In part 1, a literature review will be provided of the materials that are required to gain a good understanding of dialogic systems and dialogic regulation. This literature review is extensive and much larger than normal but it is important because many of the materials reviewed are not normally cited in the legal or regulatory literature. Chapter 2 begins with a literature review of our current understanding of corporate regulation and recent developments in regulation, with an emphasis on illustrating how these new theories embrace complexity and advocate for a learning approach to corporate regulation. The chapter ends with the question: How do organizations learn? In Chapter 3, that question is answered utilizing materials from the field of organizational development (OD), the field focused on studying how organizations change. I begin by summarizing some of the important theories about how individuals and organizations learn and then provide a detailed description of a new emerging field of OD called Dialogic OD that shows the promise to create the transformational changes that corporate regulation desires. In order to place Dialogic OD in its historical and theoretical context, I first provide a literature review of the theories that influenced it the most: chaos theory and the study of non-linear dynamic systems, and post-modern language theory, social constructionism and the theory of generativity. I conclude the chapter by highlighting that a dialogic approach to corporate change assumes that the individual and his or her relations to other individuals in small groups are the key unit of change in an organization. 6 In Chapter 4, I change the focus of the unit of analysis to the individual level and look at the question: How do individuals behave and, more importantly, change their behaviour when relating to other individuals? I do this by reviewing the literature from experimental games and in particular social dilemma experiments. In this review I rely heavily on three meta-analysis articles published in the last twenty years that summarize many of the findings of social dilemma experiments. I then explain how these findings are consistent with the principles of Dialogic OD described in Chapter 3 but conclude that while experimental games taught us a lot about compliance they did not tell us much about adherence or about dialogic processes. Finally, I develop a set of hypotheses around three different and distinct approaches to regulation (rules-based regulation, audit based regulation, and dialogic regulation) predicting how individuals will behave when the rules are changed during an experimental game. These are the hypotheses I tested in the experimental game that I conducted as part of this thesis. In part 2, I provide a summary of the “Pay-Off” game experiment I conducted. I discuss the methods and the results, and I provide a discussion of what occurred. The games experiment showed that for behavioural outcomes the Audit Based Regulation approach was the most effective and most people in this group followed the rules. It also showed that there is significant promise in a Dialogic Regulation approach if the regulatory desire is to have participants learn. In the implications for future research section I caution that even in this controlled experiment there were unintended consequences and that a lot more work needs to be done to ensure that the learning in the dialogic condition is the desired learning. A hypothesis is made that some combination of dialogic intervention followed by an audit process might be the most effective way to use a dialogic process to have the participants learn the rules. 7 In part 3, I use the results from the experiment and the materials from the literature review to develop the basic components of a theory of dialogic regulation for corporations. I talk about the two different components of the theory: the law – the written language and regulation – and the conversations and dialogue that take place about the law. I situate each component within the corporate law literature of the area to differentiate how the dialogic approach is different from the other narratives. I also provide some concrete examples of what would be required to bring dialogic regulation into effect including how to draft laws and what processes would be required to use dialogic interventions effectively in corporate regulatory setting. I then use the novel case of Suncor’s creative sentencing project for its Firebag facility to provide an example of what dialogic regulation could look like. Finally, I critique that case to show the potential pitfalls in dialogic regulation and the inherent tensions of using dialogic processes within the command and control regulatory system. Finally, a conclusion is offered about the prospects of Dialogic Regulation. Its future is promising and exciting, but its application is frightening because the results cannot be predicted. It is the future, however, so let us hope that it is not the “plague” and that instead when a corporation starts to develop symptoms, we can send it for “counselling”. 8 Chapter 2: Developments in Regulatory Theory: The Potential of a Complex World We live in great cities without knowing our neighbors, the loyalties of place have broken down, and our associations are stretched over large territories, cemented by very little direct contact . . . Our schools, churches, courts, governments were not built for the kind of civilization they are expected to serve . . . The world is so complex that no official government can be devised to deal with it, and men have had to organize associations of all kinds in order to create some order in the world. They will develop more of them, I believe, for these voluntary groupings are the only way yet proposed by which a complicated society can be governed. Lippmann (1914)14 Underlying almost every conversation on corporate law and regulation is a set of assumptions about the law and human behaviour, the most common of which goes something like this: pass a law and behaviour will change. This assumption is usually associated with arguments calling for the creation of a new piece of legislation or a new legal right or legal obligation. A slightly more complex set of assumptions looks something like this: pass a law, create a regulator, watch vigilantly, punish when appropriate, and behaviour will change. This set of assumptions is usually associated with calls for more regulatory resources, the creation of a regulator, or the punishment of certain corporate actors. A third assumption also shows up from time to time: forcing corporations to disclose information will change behaviour. This is usually associated with arguments calling for corporate social responsibility reporting, environmental reporting, or reporting on human rights.15 14 As quoted from Rees (2008) infra note 96 at 12. 15 I would like to thank the Canadian Business Ethics Research Network (CBERN) for its financial assistance in preparing this chapter. I would also like to thank the following people for their insightful comments: Bruce MacDougall, Bob Paterson, David Duff, Dirk Matten, and Gervase Bushe. Any errors or omissions remain my own. 9 Most of the time, these assumptions are lurking in the arguments of corporate law and regulation advocates in an unstated, unexplored, and un-challenged state. The problem with these assumptions is that they assume causal chains that are too direct and too simplistic. Laws do not on their own change behaviour. Punishment, on its own, does not change behaviour. Disclosure, on its own, does not change behaviour. These assumptions do not take into account the complexity of corporate law regulation systems, the complex way corporations actually learn new behaviours, or the complex way human beings actually learn new behaviours. This is evident in the reality of our current corporate regulation efforts: we have passed a lot of laws; we have resourced many regulatory agencies; we have watched vigilantly (a lot of the time); we have punished a lot of corporate actors; and corporations have disclosed millions of pages of information. And still, undesired corporate activity continues and actually appears to be increasing in frequency. The only way to transform this failed pattern of corporate regulation is to reconsider the basic assumptions underlying the regulatory system. At least two questions need to be debated: What is corporate law and regulation? And, how do corporations actually learn new behaviour? Recent developments in both regulatory practice and regulation theory have opened up the potential to answer these questions in novel ways that will allow us to develop regulatory systems that are far more effective than previous ones. The common thread to all of these developments is the acknowledgement of complexity in the regulatory system and the movement away from direct causal relationships to complex interdependent causal relationships. In short, the regulation of corporations is a complex task that we do not yet fully understand. However, simply acknowledging complexity allows us to provide different answers to the assumptive 10 questions and to develop an approach where we can explore, from a learning perspective, what corporate law and regulation is and how corporations learn new behaviour. This chapter summarizes some recent developments in corporate regulatory practice and regulatory theory that acknowledge the complexity in the regulatory system. It then provides a framework from organization theory to help understand that complexity and advocates a learning approach to corporate law and regulation that will allow exploratory activities leading to the development of more effective regulatory practices and theories. This chapter is structured in five sections. In section 2.1, developments in the practice of corporate regulation will be summarized. In section 2.2, developments in regulation theory will be discussed. In section 2.3, the learning approach to corporate law and regulation is explored in more detail. In section 2.4, the complexity in the corporate law and regulatory system is discussed with reference to systems theory and the application of complexity to organizations. The conclusion of the chapter advocates for a learning approach to corporate law and regulation and proposes a research agenda focused on understanding how corporations learn new behaviour. 2.1 Developments in Corporate Regulation The number of dramatic market failures in the last decade has caused some corporate regulators and legal scholars to reconsider their traditional command and control approaches to corporate regulation. Command and control regulation is characterized by mandatory rules combined with fines and criminal sanctions for breaches of the rules. In North America, we have been following this approach to corporate regulation for some time. We have a heavy reliance on the passing of laws or texts in which the rules and incentives are laid out and the role of the 11 regulator has traditionally been one of a watchdog vigilantly looking for transgressions and punishing transgressors. Command and control regulation should more accurately be described as “direct” regulation because under this approach the government actually gets involved in regulating the behaviour – the regulators try to do the “rowing”.16 The best examples of this approach were the large specialist regulatory agencies set up by F.D.R. during the New Deal, such as the Securities and Exchange Commission (the “SEC”). In these agencies, experts analyzed problems and then designed universal legislation to restrict behaviour.17 In the case of the SEC, this took the form of voluminous amounts of legislation and rules, backed up by criminal sanctions and large monetary penalties and an enormous regulatory agency to enforce them.18 The command and control approach to corporate regulation does not seem to be working. The outcome of this approach for the SEC was that it ran out of resources to prosecute offences, it could only prosecute the worst cases, and its success rate on those cases was low. In fact, a repetitive cycle of failure has emerged over the last few decades that regulatory scholar John Braithwaite refers to as boom, bust, regulate.19 “Boom” refers to a time of economic prosperity largely driven by businesses that have innovated a new way of avoiding regulation. This is followed by a “bust” where those businesses fail because of a lack of regulation. Finally, the 16 John Braithwaite first applied the terms “rowing” and “steering” to describe regulation approaches. See John Braithwaite, “The New Regulatory State and the Transformation of Criminology” (2000) 40 British Journal of Criminology 222 at 223. 17 See Dorf and Sabel (1998) infra note 68 at 270. 18 See John Braithwaite & Peter Drahos, Global Business Regulation (Cambridge: Cambridge University Press, 2000) at Chapter 9. 19 For a description of this cycle, see John Braithwaite, Regulatory Capitalism: How It Works, Ideas for Making It Better (Cheltenham, U.K.: Edward Elgar, 2008) at Chapter 2. 12 regulators step in to “regulate” the business or market after the fact. In this context, all too often, “regulate” means to identify a few bad apples, punish them to act as a deterrent for future behaviour, and then pass a whole new set of laws to prohibit the now discovered wrong behaviour. The belief that the problems in the market system occur because of a few “bad apples” and lax regulatory oversight is widespread, as evidenced in this quote from Business Week after the Enron collapse: “In many ways, Enron and its dealings with Arthur Andersen are an anomaly, a perfect storm where greed, lax oversight, and outright fraud combined to unravel two of the nation’s largest companies.”20 Unfortunately, there have been too many perfect storms over the last decade, including the tech bubble21, the ImClone insider trading scandal, the corporate fraud scandals (including Enron, Arthur Anderson, Worldcom, Qwest, Adelphia, Global Crossing, Tyco, and others)22, the Hollinger executive pay scandal, the sub-prime mortgage meltdown23, the global financial crisis24, and the BP Deepwater Horizon oil spill.25 The bad apples that were blamed and punished 20 John Byrne et al., “How to Fix Corporate Governance”, Business Week (May 6, 2002) 69. 21 Between 1997 and 2000 technology stocks rose more than 500% before crashing and losing significant amounts of money for individual investors. For a detailed description of what might have caused the tech bubble and who was involved, see John Griffin, Jeffrey Harris, Tao Shu, & Selim Topaloglu, “Who Drove and Burst the Tech Bubble?” (November 15, 2010) Journal of Finance. Online SSRN: <http://ssrn.com/abstract=459803>. 22 For a brief summary of the corporate fraud scandals and the ImClone trading scandal together with summaries of the corresponding European scandals, see Manish Gupta, “Comparative Corporate Governance: Irish, American, and European Responses to Corporate Scandals”, Bepress Legal Series - Working Paper 1310 (February 3, 2006) at 3-6. 23 For a detailed description of the events leading up to the sub-prime mortgage meltdown, see Michael Lewis, The Big Short: Inside the Doomsday Machine (New York: W.W. Norton & Company, 2010). 24 For an account of how the global financial crisis occurred, see Robert Schiller, The SubPrime Solution: How Today’s Global Financial Crisis Happened and What to Do About It (Princeton: Princeton University Press, 2008). 25 For a description of the events leading up to the Deepwater Horizon blowout, see Robert Cavnar, Disaster on the Horizon: High Stakes, High Risks, and the Story Behind the Deepwater Well Blowout (USA: Sterling, 2010). 13 for these scenarios included Frank Quattrone26, Conrad Black27, Martha Stewart28, Kenneth Lay and Jeffrey Skilling29, Bernard Ebbers30, Angelo Mozilo31, Geir Haarde32, and Tony Hayward33, to name a few. The major legislative packages that were passed to solve these problems included 26 Frank Quattrone was an American investment banker who was a key player in the “Hot IT” IPOs of the 1990s. He was charged with interfering into a government probe into Credit Suisse First Boston’s behaviour in allocating IPOs. The charges were eventually dropped. See, Greg Farrell, “Frank Quattrone Gets All Charges Dropped” USA Today (August 22, 2006) online USA Today: <http://www.usatoday.com/money/industries/brokerage/2006-08-22-star-banker_x.htm> (accessed May 4, 2014). 27 Conrad Black was the Chairman and CEO of Hollinger Inc. He was sentenced to six and a half years in jail for misappropriating funds from Hollinger. See “Conrad Black Sentenced to 78 Months in Jail” CBC News (December 20, 2007) online CBC News: <http://http://www.cbc.ca/news/business/conrad-black-sentenced-to-78-months-in-jail-1.64626.> (accessed May 4, 2014). 28 Martha Stewart and her Merrill Lynch broker Peter Bacanovic each received a five-month jail sentence for making false statements to prosecutors about their trading of Imclone Stock. See Constance Hays, “Stewart Sentenced to 5 Months in Prison” The New York Times (July 16, 2004).online New York Times: <http://www.nytimes.com/2004/07/16/business/16CND-MART.html > (accessed online May 4, 2014). 29 Kenneth Lay and Jeffrey Skilling were the Chairman, CEO, and COO of Enron. They were both found guilty of securities fraud and wire fraud and sentenced to an extended period of jail time. Lay died shortly after his conviction. Skilling was sentenced to 24 years in jail. See Alexei Barrionuevo, “Enron Chiefs Guilty of Fraud and Conspiracy” The New York Times (May 7, 2008) online The New York Times: <http://www.nytimes.com/2006/05/25/business/25cnd-enron.html?pagewanted=all&_r=0> (accessed May 4, 2014). 30 Bernard Ebbers was the CEO of Worldcom. He was sentenced to 25 years in jail for accounting fraud after Enron collapsed. See Associated Press, “Ex-Worldcom CEO Checks in For Prison Term” NBC News (September 26, 2006) online MSNBC: <http://www.msnbc.msn.com/id/15011730/ns/business-corporate_scandals> (accessed online May 4, 2014). 31 Angelo Mozilo was the founder and CEO of Countrywide Credit a key sub-prime mortgage lender in the sub-prime crisis. He was charged with securities fraud and insider trading by the SEC. The charges were eventually dropped. See “Angelo Mozilo” The New York Times (October 15, 2010) online The New York Times: <http://topics.nytimes.com/top/reference/timestopics/people/m/angelo_r_mozilo/index.html> (accessed online May 4, 2014). 32 Gier Haarde was the Prime Minister of Iceland during the Global Financial Crisis. He has been charged with negligence for his perceived role in the collapse of Iceland’s three main banks. See ç 33 Tony Hayward was the CEO of BP when the Deepwater Horizon Oil spill occurred. He was famously vilified in the press for being caught on camera saying, “I want my life back.” See “Embattled BP Chief: I Want My Life Back” The Sunday Times (May 31, 2010) online The Sunday Times: <http://business.timesonline.co.uk/tol/business/industry_sectors/natural_resources/article7141137.ece.> (accessed online November 20, 2010). He resigned shortly after his appearance in front of the U.S. Congressional Oversight Committee on Energy. See “BP’s Tony Hayward resigns after being ‘demonised and vilified’ in the US” The Telegraph (July 27, 2010) online The Telegraph: <http//:www.telegraph.co.uk:finance:newsbysector:energy:oilandgas:7912338:BPs-Tony-Hayward-resigns-after-being-demonised-and-vilified-in-the-US.html> (accessed online November 20, 2010). It remains to be seen whether he will face any criminal charges. 14 the Sarbanes-Oxley Act34, the Dodd-Frank Act35 (including the Economic Emergency Stabilization Act36), and the response to the Deepwater Horizon Oil Spill included the Outer Continental Shelf Reform Act and the Big Oil Bailout Prevention Liability Act.37 Even the most casual observer of these events is now able to reasonably conclude that there are not just a few bad apples out there and that there is a systemic problem with corporate and market culture.38 By corporate culture, I mean the basic assumptions, conversations, and patterned interactions among the individuals within the corporate and capitalist market system.39 34 This act was passed in 2002 in response to a wave of corporate scandals including Enron, Tyco, Adelphia, and Worldcom. For a discussion of SOX as a legislative response to corporate scandals, see Gupta (2006) supra note 9 and Jennifer Hill, “Regulatory Responses to Global Corporate Scandals” (2005) 23(3) Wisconsin International Law Journal 367. 35 New York Law Firm Davis Polk issued a 279-page “Financial Crisis Manual” manual in 2009 summarizing the new laws in the United States that resulted from the global financial crisis. See Davis Polk and Wardell LLP, “A Guide to the Laws, Regulations, and Contracts of the Financial Crisis” (September 2009) online: <http://www.davispolk.com/files/Publication/d1ab7627-e45d-4d35-b6f1-ef356ba686f2/Presentation/PublicationAttachment/2a31cab4-3682-420e-926f-054c72e3149d/fcm.pdf> (accessed November 22, 2010) (on file with author). 36 For a description of this legislation, see Steven Davidoff.& David Zaring,“Regulation by Deal: The Government’s Response to the Financial Crisis” SSRN (November 24, 2008) online SSRN: <http://ssrn.com/abstract=1306342> (accessed May 4, 2014). 37 For a brief description of the legislative response to the Deepwater Horizon Oil Spill, see Sally Doremus, “Legislative response to the Deepwater Horizon disaster” Legal Planet (July 6, 2010) online Legal Planet: <http://legalplanet.wordpress.com/2010/07/06/legislative-response-to-the-deepwater-horizon-disaster/> (accessed November 22, 2010). 38 The organizational development (OD) approach to issues discussed later in this thesis would assume that every problem is a system problem. OD practitioner Edgar Schein would always assume that when a client came to him with a “problem person” he would usually find larger systemic problems in the investigation of the problem person. See Jean-Francois Coget, “Dialogical Inquiry: An Extension of Schein Clinical Inquiry” (2009) 45(1) The Journal of Applied Behavioral Sciences 90 at 93. The minute there is a statement that there are a few “bad apples”, an organizational development practitioner would immediately assume that there is some kind of systemic problem. 39 For a classic definition of “corporate culture”, see Charles Hampden-Turner, Corporate Culture: From Vicious to Virtuous Circles (London: Random House, 1990) at 21 quoting from Edgar Schein, “Organisational Culture: What is it and How to Change it” in Paul Evans, Human Resource Management in International Firms (New York: Macmillan, 1990) where it is stated that culture is: “A pattern of basic assumptions invented, discovered or developed by a given group as it learns to cope with its problems of external adaptation and internal integration that has worked well enough to be considered valid, and to be taught to new members as the correct way to perceive, think, and feel in relation to these problems.” 15 It is not what corporate and market actors are doing, or did, which is the real problem – it is the way that they think and relate to each other and to non-market actors that is the problem. The way market participants think affects all of their actions: past, present, and also, most importantly, the future. If the focus of regulation remains only on past behaviour, then regulation will always be one step behind. Corporate regulators in the United States and Australia have realized this and have begun to broaden their roles conceptually as regulators to include the need to change corporate culture. Agencies in both countries have started sending in experts as part of settlement agreements to assist offending corporations to change. In the United States, the Department of Justice (DOJ) has begun sending monitors into corporations under deferred prosecution agreements and the SEC has been sending monitors in connection with reform undertakings.40 In Australia, the Australian Securities and Investment Commission (ASIC) has also been sending monitors in under reform undertakings.41 While the success of these monitors to date has been questionable, it is an important first step for regulators to begin to acknowledge that there may be systemic issues in the corporate system and to envision a broadening of their traditional role into actively trying to change corporate culture.42 40 For a description for the SEC’s reform undertakings, see Christie Ford, “Toward a New Model for Securities Law” (2005) 57 Administrative L Rev 757 at 758. For a description of both the DOJ and SEC’s corporate monitorship programs, see David Hess & Christie Ford, “Corporate Corruption and Reform Undertakings: A New Approach to an Old Problem” (2008) 41 Cornell Intl Law Journal 307. 41 For a description of ASIC’s reform undertakings, see Marina Nehme, “Enforceable Undertakings in Australia and Beyond” (2005) 18 Australian Journal Corporate Law 70. 42 For an assessment of the performance of corporate monitorships to date within the DOJ and SEC, see David Hess & Christie Ford, “Can Corporate Monitorships Improve Corporate Compliance?” (2009) 34(3) Journal of Corporation Law 679. 16 But, changing corporate culture is difficult. Failures may occur in any attempt to change corporate culture. Corporate leaders and corporate counsel have long known that implementing changes in law into the culture of their organization is a difficult and complex task. The frustrating fact of this process is that the outcome is not necessarily linked to the amount of effort put in. Even if a corporation diligently tries to change its culture, it may fail simply because change is a very complex process. Now that regulators are starting to engage in the change process and realize how complex and difficult it is, corporations and regulators can start working together, dialoguing, and learning better ways of being successful. The hope is that if both parties work together, they can learn together, and eventually succeed. 2.2 Developments in Regulatory Theory In parallel to the developments in corporate regulation, regulation theory has, over the past two decades, developed new learning approaches to regulation. The new approaches focus on regulating corporations by building regular and flexible interaction between regulators and corporations based on dialogue where both sides are working together to continuously improve the processes and culture of the corporation.43 These new regulatory theories include responsive regulation, smart regulation, self-regulation combined with meta-regulation, democratic experimentalism, new governance, and nodal governance. Each is briefly outlined below in terms of what it is, how it differs from the command and control approach, and how it differs from the other new approaches. The relative place of each theory in the historical development of regulation theory is depicted in Appendix 43 The term “learning approach” to corporate regulation was coined by Brian Head and John Wright. See Brian Head & John Wright, “Reconsidering Regulation and Governance Theory: A Learning Approach” (2009) 31-2 Law & Policy 192. 17 A: “Theories of Regulation.” At the end of this section, the similarities amongst the new approaches will be discussed. 2.2.1 Responsive Regulation Responsive regulation was developed by Ayres and Braithwaite in the early 1990s as an attempt to bridge the gap between the command and control regulatory theorists and the deregulation theorists.44 Responsive regulation brought sociological insights into the debate. It recognized that it is a complex world out there and no single regulatory solution was sufficient. In the responsive regulation approach, regulation needs to be tailored to the specific situation. Responsive regulation advocates the use of a “responsive regulation pyramid”. The assumption behind the pyramid is that most people want to follow regulation. Therefore, regulators should always begin interaction with a business organization in the least interventionist way (i.e. conversation or dialogue). The presumption is to start at the base of the pyramid and escalate up the pyramid only when more modest forms of intervention or punishment fail. This approach allows cheaper and more respectful regulatory strategies to be used first. As less interventionist modes of regulation fail, the regulator escalates up the pyramid to more and more interventionist modes of regulation. The theory argues that the more clout at the top of the pyramid the more effective the regulator will be at the bottom. The pyramid is shaped to reflect the fact that the number of transgressors who will deliberately contravene the regulations gets increasingly smaller as the severity of the regulatory reaction increases and, therefore, the most interventionist punishments and incentives need only be used with a few parties. The intention of the responsive 44 Responsive regulation was presented in Ian Ayres & John Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (New York: Oxford University Press, 1992). 18 regulation approach was to replace criminal and economic sanctions at the lower part of the pyramid with more responsive approaches consistent with the intentions of the actors. The Ontario Securities Commission uses a version of responsive regulation in its enforcement division.45 Legal scholars Julia Black and Robert Baldwin have taken this theory a step further. They argue that what is required is “Really Responsive Regulation” that takes into account the institutional and contextual circumstances of the corporate environment and engages in feedback and learning loops to change regulation to match the contextual circumstances.46 There are four main differences between responsive regulation and command and control regulation. First, it assumes that most people will follow regulation if given the chance. Second, it acknowledges that the informal processes, self-regulatory structure, and culture within an organization are important and can be leveraged by regulators at the bottom of the pyramid.47 Third, this is one of the first works that started to talk about regulation being “de-centered” or removed from the state as the sole source of regulation.48 Fourth, including the work done by Black and Baldwin, it was one of the first theories to promote learning and feedback loops in the regulatory process. 45 For a review of the responsive regulation practices of the enforcement division of the Ontario Securities Commission, see Keith Marquis, “Responsive Securities Regulation: An Assessment of the Enforcement Practices of the Ontario Securities Commission” Regulatory Governance Institute – Regulation Papers (October 2009) (on file with author). 46 Black and Baldwin developed really responsive regulation over a series of articles, including: Julia Black, “Critical Reflections on Regulation.” (2002) 27 Australian Journal of Legal Philosophy 1; Julia Black, “Constructing and Contesting Legitimacy and Accountability in Polycentric Regulatory Regimes,” (2008) 2 Regulation and Governance 137; and Robert Baldwin & Julia Black, “Really Responsive Regulation” (2008) 71 The Modern Law Review 59. 47 Ayres and Braithwaite (1992) supra note 46 at 5. 48 Ibid at 7. 19 Responsive regulation was the first new approach to regulation that really began to gain popular support, and, not surprisingly, it is consistent with most of the other learning approaches to regulation. All of them can easily fit within its framework. For example, each of the other new approaches can be included in any specific regulatory pyramid.49 2.2.2 Smart Regulation Smart regulation was based on two previous bodies of work: responsive regulation and legal pluralism.50 Legal pluralism posits the idea that law is only one element in a web of constraint on behaviour and that there are many other non-governmental constraints. Under this theory, most regulation is not in the hands of the government but rather in the hands of private sector individuals. Smart regulation is based on the idea that more can be accomplished by harnessing self-regulation within corporations than by using governmental command and control.51 Smart regulation has three key ideas: use multiple regulatory instruments,52 involve as many stakeholders as possible,53 and design regulatory policy to meet specific situations.54 It 49 For an example of this see Braithwaite’s article on using the regulatory pyramid for health care. This pyramid includes self-regulation, meta-regulation, etc. Judith Healy & John.Braithwaite, “Designing Safer Health Care through Responsive Regulation“ (2006) 184(10) Medical Journal of Australia S56-S59. 50 Smart regulation was developed by Neil Gunningham, Peter Grabosky & Darren Sinclair in the following works: Neil Gunningham, and Darren Sinclair “Regulatory Pluralism: Designing Policy Mixes for Environmental Protection,” (1999) 21 Law & Policy 49; and Neil Gunningham, Peter Grabosky, & Darren Sinclair, Smart Regulation: Designing Environmental Policy (Oxford: Oxford Univ. Press, 1998). 51 Ibid at 35. 52 Examples of types of regulatory instruments in this approach include command and control, economic, self-regulation, and voluntarism. 53 Including second party (industry) and third party (community) participants. 54 Gunningham & Sinclair (1999) supra note 52 at 70. 20 argues that, in the majority of circumstances, the use of multiple rather than single regulatory instruments and a broader range of regulatory actors will produce better regulation. It also argues that all instruments have strengths and weaknesses so it is best to use them in combinations so that the strength of one can cover over the weakness of the others.55 The Government of Canada launched a smart regulation initiative in 2004.56 Smart regulation differs from command and control in its acknowledgement of the power of non-governmental actors in the regulatory environment. In fact, in smart regulation the regulatory pyramid becomes three-dimensional and the regulator, the industry, and the community each has a face. Any of the three can choose to escalate regulation up the pyramid with their own set of sanctions for the transgressors.57 This approach is different to responsive regulation in that instead of talking about instruments being used in an increasing order of intervention, it focuses on using combinations of instruments and approaches. Therefore, it is concerned with which specific combinations work well together or do not work well together.58 2.2.3 Self-Regulation and Meta-Regulation The self-regulation and meta-regulation model argues for self-regulation inside corporations combined with meta-regulation by state regulators. Christine Parker advocated this 55 Gunningham, Grabowsky & Sinclair (1998) supra note 52 at 15. 56 For a description of this initiative, see External Advisory Committee on Smart Regulation, Smart Regulation: A Regulatory Strategy for Canada: Report to the Government of Canada (Ottawa: External Advisory Committee on Smart Regulation, 2004) and Government of Canada, Smart Regulation: Report on Actions and Plans (Ottawa: Government of Canada, 2005). 57 Braithwaite adopted this idea of a non-governmental actor side to the regulatory pyramid in his 2009 book Regulatory Capitalism when he discussed Qui Tam. See John Braithwaite (2008) supra note 21. 58 Gunningham, Grabowsky, & Sinclair (1999) supra note 52. 21 approach in her 2002 book The Open Corporation.59 The three key ideas to this approach are: a) the open corporation: management decision making should be open to democratic influences because corporations are powerful and their actions can have serious political and social consequences60; b) self-regulation: once corporations are open, we can start to rely on their internal regulation systems to regulate behaviour61; and c) meta-regulation: the duty of the regulator then becomes simply judging the outcomes of the self-regulatory systems against some objective societal standards.62 The key of the whole system is learning. Parker talks about triple learning loops where the self-regulatory professionals, corporate management, and the regulators move forward in waves of continuous improvement. In the triple learning-loop model corporate self-regulatory systems innovate and learn a new way of complying with or exceeding a regulatory requirement, the corporation’s management systems then learn from that innovation, and the regulator, in turn, learns from the corporate management systems.63 This theory is different to command and control because, similar to the other theories, it advocates government stepping back from “rowing” and focusing instead on “steering” regulation.64 It is unique in that it places an extraordinary emphasis on the role that internal 59 Christine Parker, The Open Corporation: Effective Self-Regulation and Democracy (Cambridge: Cambridge University Press, 2002). 60 Ibid at 1. 61 Ibid at Chapter 2. 62 Ibid at Chapter 9. 63 Ibid at 279. 64 Ibid at 41. 22 governance structures within corporations can play in the regulatory process.65 It should be noted, though, that Parker only advocates reliance on self-governing systems if the corporation has become “open” to the influences of other stakeholders. 2.2.4 Democratic Experimentalism Democratic experimentalism seeks to take advantage of local knowledge to encourage local experimentation to find tailored solutions to complex problems.66 The role of government, in this approach, is to “orchestrate” the experimentation process rather than to dictate top-down universal rules. Government does this by encouraging broad participation at the local level involving input from a variety of actors that are affected by the actions and have differing perspectives on the problem, as well as different areas of expertise. This is called “directly deliberative polyarchy”.67 Through experimentation these actors attempt to find the best solution to a problem that takes into account the relevant aspects of that unique situation (e.g. the specific corporation, industry, issue, or geographic area). Any solutions that are developed from this experimentation are understood to be provisional and will be updated based on new knowledge and changing circumstances or societal expectations. This approach was based on the three ideals of the Japanese production system: benchmarking (goals that can be achieved), concurrent engineering (experimenting with 65 While this may look very similar to market-based regulation or deregulation, it is actually quite different because its assumptions are different. First, it clearly states that the corporation is a public entity (not a private one) and is legitimately subject to regulation. Second, it supports self-regulation only after other stakeholders have had a say in the regulation of the corporation. 66 Democratic experimentalism was advocated by Michael Dorf and Charles Sabel. See Michael Dorf & Charles Sabel, “A Constitution of Democratic Experimentalism” (1998) 98 Colum. L. Rev. 267. 67 Ibid at 288. 23 solutions for error-resolution), and learning by monitoring.68 Democratic experimentalism is a centrally coordinated system of parallel local experiments, networked and disciplined through structured information disclosures and monitoring requirements.69 All the experiments are subject to rolling minimum performance benchmarks but, as long as they meet those benchmarks, they are otherwise free to experiment in a continuous and ceaseless effort to improve, learn, and revise to find the best regulatory structure.70 The main difference between democratic experimentalism and the command and control approach is the emphasis it places on specific solutions generated by local actors as opposed to universal legislation created by specialist regulatory agencies. Its main difference from the other approaches is that democratic experimentalism is advocating not only a regulatory approach but also a political approach. It has linked politics and regulation. In all fairness to the authors, this piece was a political piece in its original form. However, for the purposes of regulation theory, the idea of local experimentalism can be separated from democratic principles. For example, China has a long history of regulatory experimentalism without any link to democratic principles. In fact, China used an experimental approach like this to introduce corporations and stock exchanges into their socialist economic system by allowing regional experimentation for long periods of time before collecting the most successful practices into the national corporations and securities laws.71 68 Ibid. 69 Ibid. 70 Ibid. 71 For a brief description of the culture of legal experimentation in China, see Natalie Lichtenstein, “Law in China’s Economic Development: An Essay From Afar”, in Stephen HSU, ed., Understanding China’s Legal System: Essays in Honor of Jerome A. Cohen (New York: New York University Press, 2003) at 274. For a more detailed description 24 2.2.5 New Governance New governance is not a single model or theory but a transatlantic family of governance innovations.72 Each of the innovations associated with it are a move away from command and control regulation toward a new model of collaborative, multi-party, multi-level, adaptive problem solving. The most interesting articles on new governance are Hess and Ford’s articles on how the theories relate to the practices of the SEC and Reform Undertakings in particular.73 In these articles, the authors explain that the SEC came to realize that there were systemic corporate culture problems and that there were not just a few bad actors out there.74 The SEC also realized that part of its mandate was to change or “clean up” those corporate cultures and that meant that they needed to get inside the “black box” that was the corporation to be able to cause real change.75 The SEC’s initial attempt to cause such corporate change involved sending in third party monitors to cause the change. As mentioned previously, this practice has had limited success because the monitors were too often lawyers and not change agents, the focus was too often on rules and codes of conduct, and the change processes were largely de-coupled of how this process was applied to the development of corporate law and securities law in China, see Michael Cody, “Corporate Governance Reform in the PRC: The Layered Approach to Convergence” (2007) 3(4) The Corporate Governance Law Review 366. 72 For a description of the theories that are considered to be included in the term “new governance”, see Bradley Karkkainen, “‘New Governance’ in Legal Thought and in the World: Some Splitting as Antidote to Overzealous Lumping” (2004–2005) 89 Minn. L. Rev. 471. Democratic experimentalism is included within new governance. 73 David Hess’ and Christie Ford’s articles on new governance approaches to corporate regulation include: Christie Ford, “Toward a New Model for Securities Law Enforcement” (2005) 57 Administrative L Rev 757; David Hess and Christie Ford, “Corporate Corruption and Reform Undertakings: A New Approach to an Old Problem” (2008) 41 Cornell Intl Law Journal 307; David Hess and Christie Ford, “Can Corporate Monitorships Improve Corporate Compliance?” (2009) 34(3) Journal of Corporation Law 679; David Hess, “The Three Pillars of Corporate Social Reporting as New Governance Regulation: Disclosure, Dialogue and Development” (2008) 18 Business Ethics Quarterly 447. 74 See Hess and Ford (2008) Ibid at 16. 75 Ibid at 14. 25 as opposed to transformational events for the large majority of the corporate employees.76 However, while the success of this approach has been limited to date, it is a promising example of a new approach to regulation based on the realities of what corporations are and how they change. At this point, new governance is not really a regulatory theory. It is instead a set of innovations that acknowledge that the corporation is a complex social system and some suggested tools that could be of use in adopting such an approach. It is different to command and control in that it acknowledges that in order to change behaviour the regulator may need to become involved in the internal workings of corporations. It is different from the other new approaches because it does not advocate a new system of regulation – rather just some tools for the regulatory toolbox. 2.2.6 Nodal Governance Nodal governance is an elaboration on the information network theory of Manuel Castells.77 It argues that information flows across networks but that it is only transformed into action at “nodes” – places that are organized to turn information into action. This is important to regulation because regulatory networks are information networks characterized by complexity, a plurality of actors, the complex interconnectedness of actors, a multitude of mechanisms to transform information into action, and rapid adaptive change.78 The key is to understand how the 76 See, Hess & Ford (2009) supra note 75. 77 Nodal governance was developed by Peter Drahos, Scott Burris, and Clifford Shaearing. See Scott Burris, Peter Drahos & Clifford Shearing, “Nodal Governance” (2005) 30 Australian Journal of Legal Philosophy 30 and Scott Burris, “Governance, Micro-Governance and Health” (2004) 77 Temple Law Review 335. 78 Drahos, Burris & Shearing (2005) Ibid at 31-35. 26 information on regulation that flows through a network gets translated into regulation. The authors argue that it happens at nodes, or sites within the regulatory or information network where knowledge, capacity, and resources are mobilized to manage a course of events.79 Nodes have four essential characteristics: a) mentalities: a way of thinking about the matters the node has emerged to govern; b) technologies: a set of methods for exerting influence over the course of events at issue; c) resources: to support the operation of the node and the exertion of influence; and d) institutions: a structure that enables the directed mobilization of resources, mentalities, and technologies over time.80 Nodes can be legislatures, government agencies, neighbourhood associations, NGOs, corporations, gangs, etc. Not all nodes are equal. The capacity of a node to influence or regulate depends in large part on its resources, which broadly defined include a wide range of social capital. One example of a node is the pharmaceutical lobby in the United States, who Drahos, Burris and Shearing argue are responsible for causing the TRIPS intellectual property rights to be brought into being at the World Trade Organization.81 Nodal governance theory differs from command and control in the same ways as the other theories. It acknowledges that the state is no longer the sole source of regulation, it acknowledges the complexity in the system, and it has a focus on learning. It is different from the other new approaches because it focuses on exactly how regulation is generated and it does not assume that the process is democratic or that the government is in a position to “steer” it. 79 Ibid at 37. 80 Ibid at 37-38. 81 Ibid at 40-49. 27 Any strong node can steer regulation and that node could just as likely be a corporation (or an association of corporations) as a government agency. 2.2.7 The Similarities of the New Approaches All of the theories discussed above are interesting advances over the command and control regulatory approach and they share certain characteristics that are important for a learning approach to regulation. Their similarities include: i. An Acknowledgement of Complexity: All of the new approaches acknowledge that regulating corporations and markets is a complex thing and that managing a complex system to a desired end state is very difficult.82 ii. A Broader Definition of Regulation: All of the new approaches adopt a broader definition of corporate law regulation that includes far more than just government laws and regulations.83 iii. The Idea of the De-centered State: All of the new approaches acknowledge that the state is not the sole source of regulation and that other parties have a role in regulation: industries, corporations, communities, individuals, etc.84 iv. Tailoring Specific Solutions: Most of the new approaches deal with the complexity in the system by allowing regulatory solutions to be tailored to specific situations.85 82 For example, responsive regulation acknowledges that regulators need a set of tools and different approaches to use to change behaviour; smart regulation acknowledges that different sets of instruments will be more effective with certain regulated corporations and that communities, NGOs, and industries all play a role in the regulatory system; meta-regulation acknowledges that corporations have self regulatory systems that play a large part in the regulatory system; democratic experimentalism acknowledges that no universal regulatory rules are sufficient and that local solutions need to be found to local problems; and nodal governance acknowledges that information on its own is not sufficient and that any party can mobilize itself to turn information into power within the regulatory system. 83 This idea is best illustrated by meta-regulation and smart regulation where the non-governmental portions of the regulatory system are leveraged to obtain regulatory results. 84 Examples of this are Braithwaite’s Regulatory State and Regulatory Capitalism, Black’s polycentric regulatory regimes, Braithwaite and Drahos’ epistemic communities, Christine Parker’s self regulating open corporation, and nodal governance’s non-government nodes. 85 Examples of this include democratic experimentalism and smart regulation’s goal of specific solutions and combinations of instruments. 28 v. Escalating Regulatory Pyramid: All of the new approaches advocate a toolbox of instruments for regulators to accomplish their tasks. These tools are either ordered in an escalating manner or talked about as being used in combination with each other.86 vi. A Focus on the Positive: Implicit in all of these approaches is a movement away from the negative and a commitment to working in a positive fashion on solutions. Rather than focus on punishing wrongdoers, some of the new approaches are focused on preventing the behaviour ahead of time. They are focused on generating “good outcomes”. In some theories, they are also focused on generating those outcomes in the least interventionist way.87 vii. Learning: This is the most important similarity that binds the new approaches to regulation. All of the new theories realize that regulation is simply getting people and corporations to change their behaviour to be consistent with whatever it is that society normatively wants it to be. Therefore, regulation is about causing behavioural changes and the best way for people to change their behaviour is to learn a new one. In almost all of the new theories there is an emphasis on learning – if individuals, groups, corporations, and regulators can all learn together then any normative outcome is possible.88 2.3 The Learning Approach to Corporate Regulation Brian Head and John Wright advocated the concept of a learning approach to regulation in their 2009 article surveying contemporary regulation theories.89 Head and Wright applied each of the surveyed theories to a case study of the gambling industry in Australia.90 They concluded 86 The best examples of this are responsive regulation’s “strategic regulatory pyramid” and smart regulation’s three-sided regulation pyramid. 87 Braithwaite talks about this as flipping “markets in vice” to “markets in virtue”; nodal governance talks about the goal of the regulatory system as being the generation of as many “good outcomes” as possible; and democratic experimentalism allows experimentation and failure without punishment. 88 For example, Parker and Braithwaite refer to this as a triple loop learning, democratic experimentalism is a political system of institutionalized local learning that gets translated to the national level, and nodal governance posits the success of a regulatory network or organization based on its ability to adapt and tap the collective knowledge of members to change and produce more good outcomes. 89 Head & Wright (2009) supra note 45. 90 Head & Wright divided regulation theories into three perspectives: normative theories (that focus on formal institutions, rules, and techniques for enforcement to enhance compliance with public interest goals – for example responsive regulation); descriptive theories (that document the historical, organizational and cultural content of regulatory challenges and matches these with the appropriate mixes of regulatory mechanisms – for example smart 29 that each theory was useful in explaining certain things but that no single perspective was sufficient on its own.91 As a result, they advocated a pragmatic learning approach to regulation where specific designs are implemented for specific issues with monitoring and feedback loops to allow for learning in the regulatory process.92 They argued that regulatory theory needs to remain “flexible” if it is going to assist real world actors in finding solutions to concrete regulatory problems.93 This pragmatist approach was based on the insights of Joseph Rees and his study of the American College of Physicians and Surgeons hospital standardization process.94 In this study, Rees found that the hospital system in the United States might have been the first self-regulatory system “steeped in the pragmatist principles of self-ordering.”95 At the heart of the system was a “critical community of inquirers” – the medical staff that developed ideas and hypotheses that could be verified only “through a process of social interaction.”96 As a result of his findings, Rees argued that regulatory scholars needed to take notice of the growing signs of a pragmatist revival in regulatory landscapes including nuclear energy, chemical manufacturing, natural resource management, and health care.97 The common thread of pragmatism in these regulatory regulation); and post-cultural theories (that trace the ebb and flow of regulatory interactions across sector-specific networks and through nodes of influence – for example nodal governance). 91 Head & Wright (2009) supra note 45 at 212. 92 Ibid at 193. 93 Ibid at 196. 94 Joseph Rees, “The Orderly Use of Experience: Pragmatism and the Development of Hospital Industry Self-Regulation” (2008) 2 Regulation and Governance 9. 95 John Dewey was one of the key drivers behind the process. Ibid at 10. 96 Ibid at 21. 97 Ibid at 12. 30 landscapes is “a spirit of self-correction built into regulatory process by means of a well-developed organizational capacity to learn from experience.”98 These regulatory regimes have a natural skepticism about the habitual way of doing things and construct constant feedback loops to systematically review and change routines.99 The learning approaches offered by Wright and Head and Rees are similar to the learning approaches suggested in some of the other new regulatory theories that were summarized earlier, most notably Christine Parker’s triple loop learning model and Democratic Experimentalism’s “directly deliberative polyarchies.” These future visions of the regulatory system as complex and interdependent learning environments are compelling and exciting. They create a vision of the future where instead of detailed rules that set a minimum standard of behaviour, all of the actors in the regulatory system can be working together in continuous learning loops of improvement to exceed regulatory goals and set new levels of regulatory standards and performance. Regardless of its future potential, the reality is that there has been some serious frustration in putting the ideas contained in the new learning approaches to regulation into practice. Christine Parker has recently written on how the self-regulatory/meta-regulatory system breaks down when the corporation is not “open” to influence from outside stakeholders and when the regulatory agencies are not sure what principles or goals they should include in their meta-regulatory agenda.100 She has also written on the questionable success of corporate 98 Ibid at 11. 99 Ibid at 11-12. 100 Christine Parker, “Meta-Regulation: Legal Accountability for Corporate Social Responsibility?” in Doreen McBarnet, Aurora Voiculescu and Tom Campbell (eds), The New Corporate Accountability: Corporate Social Responsibility and the Law (Cambridge: Cambridge University Press, 2007). 31 monitorships in Australia.101 Christie Ford and David Hess have similarly written on the limited success of corporate monitorships in the United States.102 Keith Marquis has authored a critical study of the OSC’s responsive regulatory approach103, and the smart regulation Initiative in Canada quietly disappeared before its final report was due.104 These trials and failures are to be expected if a pragmatic learning approach to regulation is adopted. Changing corporate behaviour is a complex and difficult task. The biggest lesson from all of these early learning approaches to corporate law and regulation is that we still have a lot to learn about learning. 2.4 Systems Theory, Organization Theory, and Complexity One reason why these new approaches to corporate regulation may be gaining in popularity is that the advances they are offering are consistent with the advancements being made in other disciplines about our understanding of what corporations (and organizations) are. In other words, this movement in corporate regulation and regulatory theory is part of a larger movement in the social sciences accommodating the insights of complexity theory to include more human and social complexity in our theories about what corporations are. In order to effectively regulate a corporation we must have an understanding of what a corporation is. Failed regulatory approaches are often approaches that have misconceived 101 Christine Parker, “Restorative Justice in Business Regulation: The Australian Competition and Consumer Commission’s Use of Enforceable Undertakings” 67(2) Modern Law Review 209. 102 David Hess & Christie Ford (2009) supra note 75. 103 Keith Marquis supra note 47. 104 The Smart Regulation initiative documents can be found online at the Government of Canada: <http://dsp-psd.pwgsc.gc.ca/Collection/CP22-78-2004E.pdf > (accessed May 4, 2014). 32 assumptions about the internal workings of a corporation. For example, if our accepted theory of the corporation conceives it as a “nexus of contracts” of the inputs and outputs of production whose sole purpose is to maximize profits, then it would make sense for us to design a command and control regulatory system with large financial penalties to corporations if they fail to comply. Organization theory is the discipline devoted to answering the question: what is a corporation? Organization theory has a lot to offer the corporate law and regulation field because over the last 40 years it has embraced complexity theory and moved past the “nexus of contracts” theory of the corporation. Unfortunately, mainstream corporate law scholarship has not kept pace. Organizational theory has advanced past the types of theories related to the “nexus of contracts” theory of the corporation because they were not descriptive of what was occurring in real life, largely because they ignore the human and social portions of organizations. Organizational theorist W. Richard Scott has written about the development of organizational theory over the past 50 years and argued that it had developed from closed rational systems theories, to closed natural systems theories, to open rational systems, to open natural systems.105 Appendix B provides a summary of each of these types of theories. Rational systems theories conceive of corporations as formal entities with sets of rules that are designed to achieve specific purposes.106 Rules in these systems are more important than people or culture. The metaphor here is that of a stopwatch. Many economic theories are rational theories. Natural systems theories conceive of corporations as collectivities of human actors each 105 W. Richard Scott, Organizations and Organizing: Rational, Natural, an Open Systems Perspectives (Upper Saddle River: Pearson Prentice Hall, 2007) at 110-113. 106 Ibid at 35-58. 33 with their own dreams, desires, and motivations.107 In these theories, people and culture are more important than rules and these corporations are far more complex than rational corporations. The metaphor here is that of an organism like a tree. Open systems theories acknowledge that it is difficult to draw boundaries around where the corporation ends, especially in an increasingly complex world.108 Who is a part of the corporation? Employees? Creditors? Suppliers? Managers? Directors? Shareholders? Joint venture partners? The community? Open systems theories adopt a more expansive approach to where the boundary of a corporation lies. Scott has argued that organizational theory advanced from rational open theories to natural open theories in the late 1970s.109 In essence, they moved towards incorporating more human complexity into their theories. These same stepped advancements of complexity in theory can also be seen in the new regulatory theories. The natural systems perspective can be found in the acknowledgement of complexity, the focus on individual actors and cultures, and the need to get inside the corporation in order to change it. The open systems perspective can be found in the acknowledgement that the government is not the only source of regulation and that everyone involved in the system has the ability to impact the outcome. These advancements in regulatory theory are similar to the advancements in corporate theory and organization theory. A chart summarizing the timeframe of the advancement from rational to natural systems is attached as Appendix C.110 107 Ibid at 59-86. 108 Ibid at 87-106. 109 Ibid at 113-123. 110 This chart is based on a chart linking the development of corporate theory with organizational theory. See Michael Cody, Social Theories of the Corporation (forthcoming) (on file with author). 34 The development stage can be linked back to systems theorist Kenneth Boulding’s hierarchy of complex systems. This hierarchy organizes types of systems into nine levels with increasing complexity and increasingly indirect causality, from a static simple system at level 1 to an unknowably complex system at level 9. A chart of Boulding’s complex systems hierarchy is attached as Appendix D. Rational systems theories are simpler theories that think of the corporation in a mechanical and rational way. They are good at explaining simple systems in Boulding’s hierarchy, for example level 1 and 2 systems. Natural systems theories are able to explain more complex systems and are level 3 or 4 theories. Open systems theories acknowledge that corporations interact with their environment and are at least level 4 theories. Human social systems, including the corporation, are complex social systems. Boulding argued that they are the most complex systems we are capable of identifying. He classified them as level 8 systems in his hierarchy of complex systems.111 Human social systems are so complex that our theories (our attempts to describe, understand, and predict them) are not complex enough to encompass the whole system. This explains why there are so many different theories that try to explain complex human social phenomenon like organizations, corporations, markets, etc. For example, in my own work on corporate law and regulation, I have identified 23 significant organization theories112, 16 corporate theories113, and 12 regulation theories114. While there are a 111 Boulding defines level 8 systems as “Multicephalus systems comprising actors functioning at level 7 who share a common social order and culture”. Level 7 systems are “Systems that possess self-consciousness and so are capable of using language”. See Cody (forthcoming) supra note 112 at 44. Therefore, human social systems are complex systems of human beings who have the capacity for free will and language interacting with each other. 112 Ibid at 255 and 296. 113 See the Corporate Theory chart in Ibid at 172. 114 See Appendix A. 35 lot of theories, none of them are sufficient on their own because they are not capable of explaining, understanding, and predicting all of the complexity in a level 8 social system. At most, each theory is good at explaining a certain type of organization, a certain practice, or a certain aspect of organizing. The most advanced theories we have to date are probably level 3 or level 4 theories. The problem with applying simple theories to complex systems is that the simple theories often do a good job of explaining a small proportion of the system but a poor job when applied as a general theory for the whole system.115 In the context of regulatory theories, direct regulation and command and control regulation are examples of regulatory approaches suitable for simple systems because they postulate direct causal relationships. Most economic theories of regulation are rational systems (level 3 theories). The new learning regulatory theories are more complex and may be level 3 or level 4 theories. However, all of our current regulatory theories are a long way from being able to explain level 7 or level 8 social systems. When we apply simple theories or solutions to complex systems it often leads to unintended consequences. For example, sociologist Neil Fligstein has argued that the U.S. Anti-Trust laws actually helped create the modern multinational corporation.116 To overcome this weakness, organization theorist Richard Scott has advocated using a layered approach to understanding organizations and using each theory for that portion of the 115 For example, deeming a corporation to be a “person” to allow it to sue and be sued was a great idea and a quick fix until corporations started using their existence as a “person” to argue for rights of free speech. 116 See Neil Fligstein, The Transformation of Corporate Control (Harvard: Harvard University Press, 1990). 36 organization (corporation) it is best at explaining.117 This is simply the same as arguing for a learning approach to regulation. Until we have level 7 or level 8 theories that explain the functioning of complex human social systems, we should all be working together to learn how regulation works. 2.5 Conclusion – Learning in a Complex World A learning approach to corporate law and regulation is the natural way to come to grips with the realization that corporate regulation is a complex task we do not yet fully understand. Changing the behaviour of a complex human social system (like a corporation) is likely to fail more times than it succeeds. Therefore, regulators, corporate actors, and other actors need to work together to experiment and explore the complex interdependencies of the regulatory system. If they fail, they need to resist the urge to punish or reprimand actors (corporate or regulatory) and instead engage again with a different approach. Together the participants in the system can learn how to regulate and change the behaviour of corporations. In this approach, corporate regulators become experts in assisting corporations and other actors in how to learn. Learning is one of the biggest areas of potential in corporate regulation but also one of the biggest issues, because many regulatory scholars do not seem to know very much about how organizations and people learn. In other words, legal and regulatory scholars have not 117 For examples of using the layered approach to understand corporate law and regulation, see Cody (2007) supra note 74 (outlining the layered approach to corporate law convergence); Michael Cody, “Evaluating Australia’s Corporate Law Reform from an Organizational Theory Perspective” (2008) 21(3) Australian Journal of Corporate Law 210 (evaluating Australia’s corporate law reforms using the lens of organization theory); and Michael Cody, “Hostile Takeover Bids in Japan? Using the Layered Approach to Understand Convergence” (2010) 9(1) Richmond Journal of Global Law and Business 1 (understanding the importation of U.S. Takeover laws to Japan using the layered approach). 37 sufficiently leveraged the literature and practice on how complex human social systems (whether corporations, organizations, or otherwise) learn. The key questions for corporate law and regulation going forward are: How do corporations learn? How do people learn? Is there a difference between the two? There are a lot of disciplines that can assist in this effort, including complexity theory, systems theory, chaos theory, organization theory, organizational development, psychology, and social psychology, just to name a few. In particular, organizational development is the discipline that is devoted to understanding how organizations learn and change. It has a rich history of both theory and practice. The organizational development literature and practice can be of great use to regulatory scholars in strengthening the new learning models of corporate law and regulation. It is not enough to argue for a learning approach to corporate law and regulation – we need to understand how the learning takes place. Therefore, regulatory scholarship should add to its agenda: learn about learning. In the next chapter, we explore the question: How do organizations learn? 38 Chapter 3: How Organizations Learn: The Power of Dialogue Most people define learning too narrowly as mere “problem-solving”, so they focus on identifying and correcting errors in the external environment. Solving problems is important. But if learning is to persist, managers and employees must also look inward. They need to reflect critically on their own behaviour, identify the ways they often inadvertently contribute to the organisation’s problems, and then change how they act. Chris Argyris, Organizational Psychologist Changing corporate behaviour is extremely difficult. Most corporate organizational change programs fail: planning sessions never make it into action; projects never quite seem to close; new rules, processes, or procedures are drafted but people do not seem to follow them; or changes are initially adopted but over time everything drifts back to the way that it was. Everyone who has been involved in managing or delivering a corporate change process knows these scenarios all too well. Just how difficult is it to change corporations? It is estimated that: • 75% of all change efforts fail to make dramatic improvements118; • success rates for major change efforts in Fortune 1000 companies range from 20-50%119; • 50-75% of all mergers and acquisitions fail to meet expectations120; • 15% of IT projects are successful121; • 50% of firms that downsize experience a decrease in productivity instead of an increase122; and 118 M.E. Smith, “Implementing Organizational Change: Correlates of Success and Failure” (2002) 15(1) Performance Improvement Quarterly 67. 119 P. Strebel, “Why do Employees Resist Change?” (May/June 1996) Harvard Business Review 139. 120 M. Schraeder & D.R. Self, “Enhancing the Success of Mergers and Acquisitions: An Organizational Cultural Perspective” (2003) 41(5) Management Decision 511. 121 S. Amber, “Defining Success: There Are Lessons To Be Learned When Defining IT Project Success” (2007) quoted from Rothwell et al. (2010) infra note 136 at 21. 122 S. Applebaum, A. Everard & L. Hung, “Strategic Downsizing: Critical Success Factors” (1999) 37(7) Management Decision 535. 39 • Less than 10% of corporate training affects long-term managerial behaviour.123 One organization development OD textbook acknowledges that, “organization change presents one of the greatest challenges in modern organizational life”.124 Because of these difficulties, successful change agents are among the most valuable resources in corporations, and, correspondingly, there are innumerable corporate consultants and change processes that companies leverage to try to create change. These include Change Management125, Program Management126, Lean Manufacturing (TPS)127, Six Sigma128, Good to 123 Bushe (2009) infra note 176. 124 Rothwell et al. (2010) infra note 136 at 21. 125 Change Management is the process of helping a person, group, or organization implement a desired change. A good definition of it is: “a set of principles, techniques, and prescriptions applied to the human aspects of executing major change initiatives in organizational settings. It is not a focus on ‘what’ is driving change (technology, reorganization plans, mergers/acquisitions, globalization, etc.) but on ‘how’ to orchestrate the human infrastructure that surrounds key projects so that people are better prepared to absorb the implications affecting them”. See L. Anderson & D. Anderson, The Change Leader’s Roadmap: How to Navigate Your Organization’s Transformation (San Francisco, Pfeiffer, 2001) at xxviii. Change Management is usually more mechanistic than organizational development approaches. Many business schools offer executive education programs on Change Management that provide managers with a framework and toolkit for managing change. For a description of change management, see Rothwell et al. (2010) infra note 136 at 16-17. 126 Program Management is a methodology for delivering projects. The industry standard for Program Management is now the Program Management Institute (PMI). For a description of Project Management methodologies, see Sebastian Nokes, The Definitive Guide to Project Management. (2nd Ed) (London: Financial Times / Prentice Hall, 2007). 127 Lean Manufacturing is the North American term that refers to the manufacturing process pioneered by Taichi Ono at Toyota. It is also called the Toyota Production System. For a description of Lean Manufacturing, see James Womack & Daniel Jones, Lean Thinking: Banish Waste and Create Wealth in Your Corporation (New York: Free Press, 2003). 128 Six Sigma is a change approach focused on applying basic statistics to business processes to reduce variation in process outputs. A Six Sigma company aims to have 3.4 defects per million in their processes or, in other words, all of their processes run effectively to within 6 standard deviations from the mean. This approach was pioneered by Motorola and has been used by many other Fortune 500 companies. It is often combined with Lean Manufacturing or the Total Quality Manufacturing (TQM) approach. For a description of the Six Sigma approach See Peter Pande Robert Neuman & Cavanagh Roland. The Six Sigma Way: How GE, Motorola, and Other Top Companies are Honing Their Performance (New York, NY: McGraw-Hill Professional, 2001). For a more detailed description of the Six Sigma methodology, see Geoff Tennant, SIX SIGMA: SPC and TQM in Manufacturing and Services (Aldershot, UK: Gower Publishing, Ltd, 2001). 40 Great129, Process Reengineering130, Operational Excellence, and the Balanced Scorecard131, just to name a few. While each of these approaches has been successful in certain situations there is no silver bullet or proven change method that works in all situations. In effect, we are still learning how to learn within corporations. The corporate fascination with change is so prevalent that almost any manager in North America will have been introduced to, or will be a part of, one of the above change processes at some point in their career. For example, in the course of my eight-year tenure with one large corporation, I was trained on project management, an executive course on change management, and a green belt in Lean Manufacturing and Six Sigma methodologies, Good to Great, and the Balanced Scorecard. Consistent with the data presented earlier, most of those programs were unable to accomplish the intended change. Which raises an interesting question about corporate regulation: if corporations are so challenged to change for core business reasons (including profitability or survival), why do we expect them to be able to change in response to changes in the law and regulation? Legal scholars, practitioners, lawmakers, and regulators have long overestimated corporations’ capacity to change. This is especially true of any regulatory theory that purports to rely on the internal 129 Good to Great is a recipe for successful and sustainable change that was created by Jim Collins. See Jim Collins, Good to Great: Why Some Companies Make the Leap . . .and Others Don’t (New York: Harper Collins, 2001). 130 Business Process Reengineering is the analysis and design of workflows and processes within a corporation. See M. Hammer & J. Champy, Reengineering the Corporation: A Manifesto for Business Revolution (New York: Harper Business, 1993). 131 The Balanced Scorecard is a strategic performance management tool that can be used by managers to track the performance of teams and the outcomes of their activities. It is one of the most widely adopted corporate management tools. For more information on the Balanced Scorecard, see Robert Kaplan & David Norton, “The Balanced Scorecard – Measures That Drive Performance” (Jan 1992) The Harvard Business Review 71. 41 governance systems of the corporation as the primary method of regulation.132 That is why the learning approach to corporate law and regulation is so promising. It makes it possible for regulators and corporations to work together on the difficult task of corporate change. In order to be successful it requires legal scholars, lawmakers, and regulators to become more familiar with what corporations are, how they change, and how the law and regulation can assist them in changing. The corporate change approaches outlined above are some of the more scientific and systematic approaches to organization change. They tend to undervalue the role that individuals, individual personalities, and interpersonal conflict have on change processes.133 They also tend to be more of the “quick fix” type of solution. It may be that most corporate change initiatives fail because these approaches fail to take into account the human components of change in corporations. This is where organization development and organizational learning comes in. 3.1 Organizational Learning and Organizational Development Organization Development (“OD”) is the discipline devoted to helping organizations change by teaching them how to learn. It is a difficult discipline to define and describe because it 132 Examples of these types of theories include self-regulation, market-based regulation, and meta-regulation combined with self-regulation (that is not dialogic). 133 Conflict is an often under-emphasized issue with corporate change. Most of the change models talk about managing stakeholders but when significant or transformational changes are happening in a corporation, the power structure is also changing, which inevitably will lead to conflict. In addition, in order to have significant changes in a corporation the people within the corporation need to learn new behaviours. To learn something new requires an individual to unlearn something that exists – which can be a very uncomfortable process. Therefore, effective change processes have to have ways to manage and engage with conflict productively. 42 encompasses such a broad range of practical and theoretical approaches.134 However, most OD approaches share several similarities: • They adopt a long-term approach to change. • They are focused on learning and education. • They are based on the collaborative participation of organization participants in the change process.135 The following two quotes offer example definitions of Organizational Development: [A] process that applie[s] a broad range of behavioral science knowledge and practices to help organizations build their capacity to change and to achieve greater effectiveness. . .136 [A] systemic and systematic change effort, using behavioral science knowledge and skill, to change or transform the organization to a new state.137 There are three OD approaches that are of interest for the purposes of developing the dialogic approach to regulation: Chris Argyris and Donald Schon’s models of individual and organizational learning, Peter Senge’s 5th discipline approach to the learning organization, and, most importantly, the new emerging dialogic OD practices. Each of these approaches offers a different perspective on how corporations learn that is important to dialogic regulation. 3.1.1 Theories of Action Psychologist Chris Argyris and philosopher Donald Schon developed the Theory of Action learning perspective that offers insights into how both individuals and organizations 134 For a discussion of a number of definitions of OD, see William Rothwell, Jacqueline Stavros, Roland Sullivan & Arielle Sullivan, Practicing Organization Development: A Guide for Leading Change (3rd edition) (Pfeiffer, San Francisco: 2010) at 12-16. 135 Ibid at 12. 136 T. Cummings and C. Worley, Organization Development and Change (9th ed.) (Cincinnati: South-Western College Publishing, 2009) at 1. 137 Quoted from Rothwell (2010) supra note 136 at 13. 43 learn.138 The Theory of Action learning perspective acknowledges that there is a difference between what people say and what they do, or their “espoused theory” and what they actually do, their “theory in use”. They argue that every individual has a set of mental maps that tell them how to act in certain situations and it is these maps that guide what they do, rather than the theories or reasons they tell others as rationalizations.139 While most people are aware of the theories they espouse to explain their own behaviour, few are aware of the maps or theories they actually use.140 Argyris and Schon call these two types “Theories of Action”. These theories govern behaviour in implicit ways and they contain assumptions about the self, others, and the environment.141 The “espoused theory” is made up of the words that we use to convey what we do or what we like others to think we do. The “theory-in-use” is the theory that governs what we actually do. Reflection is the process by which individuals engage in thinking about the mismatch between what they say they do (their intentions) and what they actually do (their outcomes). Argyris and Schon argue that personal effectiveness lies in developing the reflective capacity to reduce the distance between the espoused theory and the theory-in-use.142 138 Chris Argyris and Donald Schon wrote three important OD books together: Chris Argyris & Donald Schon, Theory in Practice: Increasing Professional Effectiveness (1974) (San Francisco: Jossey Bass, 1974); Chris Argyris & Donald Schon, Organizational Learning: A Theory of Action Perspective (Addison-Wesley, Don Mills: 1978); and Chris Argyris & Donald Schon, Organizational Learning II: Theory, Method and Practice (1996) (Addison Wesley: Reading, Mass, 1996). 139 Argyris & Schon (1974). Ibid. 140 Chris Argyris, Inner Contradictions of Rigorous Research (New York: Academic Press, 1980). 141 Argyris & Schon (1974) supra note 140. 142 Ibid. 44 3.1.2 Single- and Double-Loop Learning Argyris and Schon also outlined two different kinds of learning: single-loop learning and double-loop learning.143 Single-loop learning is adaptive learning that focuses on incremental change within an existing system. It is about error detection and correction. It solves problems but ignores the question of why the problems arose.144 This kind of error correction permits an organization to carry on its present policies or achieve its present objectives; in other words, it allows people to maintain the current theory-in-use.145 Single-loop learning functions like a thermostat that detects that it is either too hot or too cold and adjusts.146 The criterion for success for single-loop learning is effectiveness.147 Double-loop learning is learning that focuses on transforming the existing way things are done.148 Double-loop learning uses feedback from past actions to question the assumptions underlying current views and the current system structure. Double-loop learning detects and corrects errors in ways that involve the modification of the organization’s underlying norms, policies, and objectives.149 It often involves individuals having to understand how they 143 Note: single- and double-loop learning should not be confused with Christine Parker’s triple learning loop referenced in Chapter 2. Parker’s loops simply involve three different participants. Argyris and Schon’s loops are the depth to which learning takes place – either within the framework of the existing assumptions and system or changing the existing assumptions or system. 144 Peter Senge refers to this as “coping”. See Peter Senge, The Fifth Discipline: The Art & Practice of The Learning Organization (New York: Currency, 1990). Fiol and Lyles refer to this as “lower-level learning”. See C. Marlene Fiol & Margorie Lyles, “Organizational Learning” (1985) 10(4) Academy of Management Review 803 at 807. 145 Argyris (1978) supra note 140 at18. 146 Ibid at 3. 147 Ibid at 29. 148 Senge calls this “generative learning”. See Senge (1990) supra note 136. Fiol and Lyles call this “higher level learning”. See Fiol & Lyles (1985) supra note 136 at 308. 149 Argyris (1978) supra note 140 at 3. 45 themselves contributed to the problem they are trying to correct. It can involve a lot of reflective activities and may require modifications to the current theory-in-use. The difference between single-loop learning and double-loop learning can best be described as the difference between learning a new way to do something and learning a new way to think about something. Single-loop learning is safe and allows individuals to follow routine or some pre-set plan. It is usually present when “goals, values, frameworks and, to a certain extent, strategies are taken for granted”.150 Reflection in single-loop learning is limited to making the strategy more effective. In contrast, double-loop learning “involves questioning the role of the framing and learning systems which underlie the actual goals and strategies”.151 Double-loop learning is more creative, reflective, and, more importantly, risky. It is risky because it often involves questioning the underlying assumptions of a goal or strategy – in a public or group forum. The diagrams of single- and double-loop learning are attached as Appendix E. 3.1.3 Compliance and Adherence The difference between single- and double-loop learning is extremely important for corporate law and regulation because if individuals can engage in double-loop learning related to desired regulatory outcomes then they will have learned not only to change their behaviour but also to change the way they think about behaving. The difference between single-loop and double-loop learning in relation to dialogic regulation will be referred to as the difference between “compliance” and adherence”. Compliance is simply single-loop learning of desired 150 Mark Smith, “Chris Argyris: Theories of Action, Double-loop learning and Organizational learning” (2001) The Encyclopedia of Informal Education. Online: <www.infed.org/thinkers/argyrtis.htm > (accessed November 22, 2010). 151 Ibid. 46 regulatory outcomes and refers to the regulatory participants’ ability to change their behaviour to match the new regulatory outcomes. Compliance, as used in this way, is doing what someone else wants you to do whether you believe it is the right thing to do or not.152 Adherence is the outcome of a double-loop learning process of the desired regulatory outcomes and refers to the participants’ changed way of thinking about the regulatory outcomes, or, in other words, learning why the regulator changed the outcomes and accepting those changed outcomes into their own mental maps. Adherence has a different meaning than compliance. Adherence means support for a cause or idea or faithful attachment and devotion.153 Dialogic regulation argues that adherence is a better regulatory outcome than compliance and that dialogue and dialogic coaching is better at generating adherence than traditional regulatory approaches. Traditional command and control and market-based types of regulation are only designed to coerce (or incent) single-loop learning or changes in behaviour and it is unlikely that they will promote double-loop learning. One way to understand this is to conceive of three different kinds of behaviour modification: coercion, inducement, and persuasion. Coercion is forcing a modification in behaviour through threat of punishment. This is the approach to behaviour modification built into the assumptions of command and control regulation: “do this – or else.” Inducement is behaviour modification through providing incentives or rewards (financial or otherwise) for desired behaviour. This is the approach to behaviour modification built into the assumptions of market-based regulation: “if you research a technology important to the 152 For example, Webster’s defines compliance as: “1) Act or practice of complying; yielding as to a desire, demand, or proposal, 2) a disposition to yield to others.” See Webster’s New Collegiate Dictionary (New York: Merriam Webster, 1959) at 169. 153 For example, Webster’s defines adherence as: “Quality, act or state of adhering;. . . steady or firm attachment; fidelity as to party or principle.” See Webster’s New Collegiate Dictionary (New York: Merriam Webster, 1959) at 11. 47 government, you will get a tax credit.” Persuasion is behaviour modification by getting someone else to adopt your view. This is the assumption behind the new learning approaches to regulation: “we comply with safety regulations because we believe that safety is our number 1 priority.” Both coercion and inducement are relying on external factors to force behavioural change; they are not focused on internally changing the way people think. Only persuasion focuses on internal behaviour modification, and that kind of modification is greatly increased with double-loop learning. Unfortunately, double-loop learning is extremely difficult to accomplish. Argyris has shown, through years of research, that the way individuals act in organizations inhibits double-loop learning – especially when there is something important at stake. The result is that double-loop learning rarely occurs when it is most needed. Argyris and Schon set up two models that described individual theories-in-use that either inhibit or enhance double-loop learning. They referred to them as Model I and Model II.154 They believed that people used these theories-in-use when confronted with problematic situations. Model I involves “making inferences about another person’s behaviour without checking with whether they are valid and advocating one’s own views abstractly without explaining or illustrating one’s reasoning.”155 This theory-in-use is shaped by individual desires to win and not to be embarrassed because exposing our “actions, thoughts, and feelings can make us vulnerable 154 There are similar organizational versions of these models called O(I) and O(II). 155 Amy Edmondson and Bertrand Moingeon, “Learning, Trust and Organizational Change” in E. Easterby-Smith, L. Araujo and J. Burgoyne (eds.) Organizational Learning and the Learning Organization (London: Sage, 1999) at 161. 48 to the reaction of others”.156 It is usually associated with action strategies dominated by unilateral control and unilateral protection of the self and others.157 Model I often leads to deeply entrenched defensive routines at the individual, group, or even organizational level.158 Model I is summarized in Appendix F-1. Argyris has stated that most of the participants in his studies operated from theories-in-use or values consistent with Model I159, but when asked they would usually espouse Model II. Model II is based on an approach that looks to include the views and experiences of participants rather than imposing one’s own view on a situation. In this model, positions are reasoned and open to exploration by others. It is a more dialogic approach to problem resolution that involves shared leadership. OD scholars Edmundson and Moingeon have argued that employing Model II in difficult interpersonal situations “requires profound attentiveness and skill for human beings socialized in a Model I world.”160 Model II is summarized in Appendix F-2. Chris Argyris’ research focused on how to assist organizations in learning how to increase their capacity for double-loop learning, which involves teaching individuals how to move from Model I theories-in-use to Model II theories-in-use. He coined the term “deutero learning” to refer to the process of learning to learn better.161 In this perspective an organization 156 Mark Smith (2001) supra note 152. 157 Ibid. 158 See Chris Argyris, Overcoming Organizational Defenses: Facilitating Organizational Learning (Boston: Allyn and Bacon, 1990) and Chris Argyris, Knowledge for Action: A Guide to Overcoming Barriers to Organizational Change (San Francisco: Jossey-Bass, 1993). 159 Chris Argyris, Strategy, Change and Defensive Routines (Boston: Pitman, 1985) at 89. 160 Edmondson & Moingeon (1999) supra note 157 at 162. 161 This term applies to learning either single- or double-loop learning. 49 is the rules and interactions of individuals who have organized themselves, and organizational learning is changes to those rules.162 Argyris and Schon call this the group’s theory of action, which is “a complex system of norms, strategies, and assumptions” embedded in their processes of interaction. 163 In the case of the corporation, it can be argued that the corporation’s theory of action is the corporation’s culture. This theory of action resides in the thoughts of each individual in the organization and manifests itself in the form of physical images, texts, and maps – for example, organizational charts, corporate procedures, codes of conduct, corporate values, vision statements, etc. Each member of the organization is constantly trying to complete their version of the organizational theory-in-use, because as humans we are all sense-making beings who constantly try to understand the world around us. However, each member’s understanding of the organization theory-in-use is always incomplete.164 For Argyris and Schon organizational learning is “a process mediated by the collaborative inquiry of individual members”165 and organizational learning is a continuous process that is required by all organizations in order to ensure their survival.166 Organizational learning is different to individual learning.167 The difference is one of agency. The individual is the agent of organizational learning.168 Organizations require individuals to exist and 162 Argyris (1978) supra note 140 at 13. 163 Ibid at 15. 164 Ibid at 16. 165 Ibid 166 Ibid at 9. 167 Ibid at 9. 168 Ibid at 19. 50 organizations can only learn through the experience and actions of individuals.169 However, organizations are not simply collections of individuals, nor is organizational learning merely individual learning. There are lots of examples where the organization knows less than the individuals involved. Individuals reaffirm the existing patterns of the organization when their own theories of action are consistent with the organization theory-in-use. Individuals are agents of change when changes in their theories of action run counter to the existing organization theory-in-use. Organizational learning occurs when “individuals, acting from their images and maps, detect a match or mismatch of outcome to expectation which confirms or disconfirms organizational theory-in-use.”170 They continually change the theory in use, which is then recorded in the images and maps of the organization. As a result, organizing is a reflexive inquiry of collaborating individuals.171 Because of the personal and emotional risk involved in that kind of reflective process, conflict plays an important role in organizations that are actively engaged in double-loop learning.172 Therefore, double-loop learning is the process by which groups of managers confront and resolve conflict. If the conflict takes the form of a fight with one side winning all, which is not double-loop learning because neither side emerges from the conflict with a new meaning of the organization, more likely the organization’s dominant theory-in-use will prevail. If they 169 Ibid at 9. 170 Ibid at 19. 171 Ibid at 17. 172 Ibid at 22 where they state that in organizational double-loop learning, “incompatible requirements in organizational theory-in-use are characteristically expressed through a conflict among members and groups within the organization.” 51 engage with each other collaboratively, they can solve the problem and come to a new understanding of what that means for the way they interact with each other. Individuals, organizations, and societies are built to inhibit double-loop learning.173 This creates stability and avoids conflict. We tend to keep our conflicting ideas private, we let failures lie buried, and we do not share our mental maps with others. The result is that so many of our views of others and the organization remain fragmented, incomplete, and often incorrect. All of this limits the possibility for collaborative inquiry and inhibits learning. Learning also gives rise to anxiety because it causes one to shift one’s individual and collective identity, which is existentially threatening. Therefore, it becomes really important to understand how people respond to anxiety and shut down learning processes to make themselves feel comfortable.174 The skills to engage in double-loop learning can be learned. It is possible to intervene in organizations to reduce these inhibitions to learning. Such interventions are focused on decreasing the defensiveness of individuals and groups within the organization.175 They also encourage people to take risks and confront inconsistencies, and they teach people that public testing of assumptions, plans, and strategies is not harmful.176 These types of learning ideas were made popular and available to corporate actors with the publication of Peter Senge’s book The Fifth Discipline in 1990. Senge took a novel approach in combining the psychological work on learning from Argyris and Schon with the emerging 173 Ibid at 4. 174 For a discussion of these defensive behaviours, see Gervase Bushe, Clear Leadership: Sustaining Real Collaboration and Partnership at Work (Boston, MA: Davies-Black, 2009). 175 Argyris (1978) supra note 125 at 139. See also Gervase Bushe, Clear Leadership (2009) supra note 161. 176 Argyris has engaged in this practice for a few decades. 52 thoughts from systems theory to develop an overall systematic approach to organizational learning. He coined his ideal organization the “Learning Organization”. For Senge a learning organization is an organization where “people continually expand their capacity to create the results they truly desire, where new and expansive patterns of thinking are nurtured, where collective aspiration is set free, and where people are continually learning how to learn together.”177 For Senge, a learning organization is an organization that has mastered the five disciplines of a learning organization, which are: 1. The Discipline of Personal Mastery 2. The Discipline of Mental Models 3. The Discipline of Building a Shared Vision 4. The Discipline of Team Learning 5. The Discipline of Systems Thinking The discipline of personal mastery is the discipline of “continually clarifying and deepening our personal vision, of focusing our energies, of developing patience, and of seeing reality objectively.”178 This refers to the ability of individuals in the organization to become better learners. For Senge an “organization’s commitment to and capacity for learning can be no greater than that of its members.”179 Mental models “are deeply ingrained assumptions, generalizations, or even pictures or images that influence how we understand the world and how we take action.”180 These are very similar to the models about learning proposed by Argyris and Schon. Other scholars refer to 177 Senge (1990) supra note 146 at 3. 178 Ibid at 7. 179 Ibid at 7. 180 Ibid 53 these maps as cognitive schemata or percepts.181 A learning organization is able to explore mental models and engage in meaningful dialogue that allows these models to change. The discipline of building a shared vision is important because “[w]hen there is a genuine vision . . . people excel and learn, not because they are told to, but because they want to.”182 Senge agrees with Argyris & Schon that “teams, not individuals, are the fundamental learning unit in modern organizations.”183 He built this into his discipline of team learning. He argued that if teams could not learn, the organization could not learn. For Senge, the discipline of team learning is based on the ability of team members to engage in dialogue. Dialogue has a specific meaning for him: it is “the capacity of members of a team to suspend assumptions and enter into a genuine ‘thinking together’”.184 The word dialogue comes from the Greek word “dialogus”, which meant “a free-flowing of meaning through a group, allowing the group to discover insights not attainable individually.”185 Dialogue can be contrasted to discussion, which is simply the hurling of ideas back and forth at each other with a “winner take all” attitude.186 Finally, systems thinking is the ability to step back from a fragmented linear understanding of a situation and take a holistic and complex view that includes indirect and interdependent causality.187 Senge calls systems thinking the “5th Discipline” because it is the 181 See Bushe (2009) supra note 176 at 7. 182 Senge (1990) supra note 146 at 9. 183 Ibid at 10. 184 Ibid at 10. 185 Ibid at 10. 186 The word discussion has its root in percussion or concussion. See Ibid at 10. 187 Social processes are circular and filled with feedback loops but we tend to think in linear ways, so we still use the term causality in systems thinking. 54 discipline that brings all the other disciplines together. None of the five disciplines on its own is enough – but when they are all drawn together with systems thinking they fuse into “a consistent body of theory and practice” that when used makes organizational learning possible.188 The switch to become a learning organization is a significant shift of mind for organizational participants: a shift from seeing themselves as separate from the world to connected to the world, and from seeing problems as caused by external forces to seeing how they themselves create their own problems. For Senge, a learning organization “is a place where people are continually discovering how they create their reality” and how they can change it.189 Senge shares the view of Argyris and Schon that the primary things that get in the way of organizational learning are conflict, mental maps, and defensive routines. The influence of complexity theory and postmodern language theory can be seen throughout Senge’s work. One significant example is his discussion of the three core learning capabilities for teams inside a learning organization, which Senge describes as: “fostering aspiration, developing reflective conversation, and understanding complexity.”190 He argued that as the world becomes more complex and dynamic we all must work together to become more “learningful”.191 One major drawback of the early attempts to integrate systems theory into organizational learning, including Peter Senge’s approach, was that when it was put into practice the organization tended to be anthropomorphized. For example, Senge’s work led to a practitioner 188 Ibid at 12. 189 Ibid at 12. He refers to this “shift in mind” as metanoia – the Greek word meaning shift of mind. 190 Ibid at xii. 191 Ibid at 4. 55 boom about “learning corporations” and what the corporation needed in order to learn. When the corporation becomes anthropomorphized it is easy to forget that organizations do not learn – people do.192 An organization is not an organism – you cannot point to it. It is more like an ecosystem. It is the system that results from the interaction of all the things you can point at.193 So, if learning involves people, then the focus of organizational learning needs to be down at the individual interaction level (the level of small groups) and not the system level. The anthropomorphization of the corporation can be seen in some of the recent corporate law and regulation initiatives, for example the SEC’s corporate monitorships. The attempts by corporate monitors to change corporate culture have often focused around implementing a new code of conduct for the subject corporation. This is a system level fix that is not focused on individuals. In the Theory of Action learning perspective, that new code of conduct will only exist in the corporation to the extent that it is taken up into each individual employee’s theory of action for the corporation. Often the code of conduct is introduced with a simple training exercise and the signing of the code. This may not be enough because the kinds of changes intended with a change in the code of conduct are only possible in double-loop learning. But double-loop learning is most inhibited under the circumstances in which the corporate monitors are trying to make these changes: meaningful and stressful situations that are existentially challenging to corporate employees. To our knowledge, none of the corporate monitors to date have engaged in any OD practices to assist the employees with double-loop learning while engaged in their monitorships. 192 Ralph Stacey, “Learning as an Activity of Interdependent People” (2003) 10(6) The Learning Organization 325. 193 Ibid at 238. 56 Therefore, the question remains: How do you cause the kind of transformational organizational change that requires double-loop learning? It has been an elusive goal for many OD practitioners and approaches. However, a developing movement in OD called Dialogic OD has the potential to cause these kinds of transformational changes by leveraging important insights from two additional intellectual movements in the social sciences to cause change in organizations. OD theorist Gervase Bushe has identified those two intellectual movements as chaos theory’s understanding of dynamic non-linear systems and the postmodern focus on the importance of language and discourse.194 In the next two sections, I will summarize the influence of these movements on Dialogic OD and then describe Dialogic OD. 3.2 Chaos Theory – Self-Organizing Systems In modern science, the term “Chaos Theory” is used to refer to the study of complex, non-linear, and dynamic systems. Chaos theory emerged from the study of non-linear systems and the ability of computers to model non-linear equations over millions of interactions.195 Then people began making the link between physical non-linear systems, like the weather, and living systems. The application of systems theory to living things is called complex adaptive systems theory (CAS). 196 CAS has four basic principles: 1. Complex adaptive systems are at risk when they are in equilibrium because equilibrium is a precursor to death; 2. Complex adaptive systems exhibit the capacity for self-organization and emergent complexity; 194 Bushe & Marhsak (2009) infra note 283. 195 A lot of work came out of the New Mexico Santa Fe Institute. 196 In the context of organizational development and corporate change CAS is often associated with the work that Arie de Gues did at Royal Dutch Shell. His most recognized book evidencing a natural systems approach is Arie de Gues, The Living Corporation (London: Nicholas Brealey, 1997). 57 3. Complex adaptive systems tend to move toward chaos when confronted with a complex task; and 4. Complex adaptive systems cannot be directed only disturbed.197 While CAS is a natural systems perspective, these four principles can also be applied to dialogic systems. OD and business scholars have leveraged its insights to understand how organizations change. These four principles will be applied to the corporation as a dialogic system in four sections below: surfing the edge of chaos; emergence, including a discussion of the “butterfly effect”; learning complex tasks; and disturbing complex systems. 3.2.1 Surfing the Edge of Chaos – The Myth of Equilibrium Richard Pascale is a business scholar who uses complexity and chaos theory to understand business organizations. In 1999, he wrote an important article called “Surfing the Edge of Chaos.”198 He argued that organizations exist in one of three states: organized, self-organizing, and chaotic. A diagram of these states is included as Appendix K. An “organized” organization is an organization that is in equilibrium – one where everyone knows what to expect all (or most) of the time. In chaos theory, an “organized” organization is one that is in trouble because it will have a hard time learning and generating the new ideas that are crucial for its survival. A “self-organizing” organization is one with a certain amount of emergent qualities. Centralized organizational patterns are present but they are more like guidelines and individuals are allowed to organize themselves. This state allows for the generation of novel organization 197 Richard Pascale, “Surfing the Edge of Chaos” (1999) Spring Sloan Management Review 83 quoted from Ralph Stacey, et al., Complexity and Organization: Readings and Conversations (New York: Routledge, 2006). This article was later turned into a book that was important for the formation of Dialogic OD. See Richard Pascale, Mark Milleman & Linda Gioja, Surfing the Edge of Chaos: The Laws of Nature and the New Laws of Business (New York: Three Rivers Press, 2000). 198 Ibid. 58 patterns, the generation of new ideas, and promotes the learning that is necessary for the organization to survive. A “chaotic” state is problematic for an organization because without any structure it will cease to exist. Pascale’s term “Surfing the Edge of Chaos” refers to the delicate balance that is required to maintain an organization in a self-organizing state without tipping over into chaos.199 Corporate theory, corporate law, and economics assume that organizations, markets, and economies are naturally in an equilibrium state. This idea is really just an assumption and it has never been proven to be true.200 In fact, as we learn more about complex social systems comprised of interdependent human actors interacting, the more we realize that corporations, markets, and economies are anything but in equilibrium and we do not want them to be. When conceived of as complex, dynamic and non-linear systems, most corporations are self-organizing systems that must constantly learn and innovate or else they fail in the same way complex adaptive systems do. This is one of the well-known paradoxes of corporations. On one hand, they need to have a vision, value system, and culture that creates strong bonds amongst corporate participants. On the other hand, they need to create dis-equilibrium in order to promote uncertainty, learning, and innovation in order to survive. It is a delicate balancing act and straying too far in either direction may result in the failure of the organization. OD scholar Gervase Bushe calls this “learning while performing” and it is the holy grail of OD and most corporate management teams. 199 For the description of surfing the edge of chaos see Ibid at 67-72. 200 This assumption is part of the dominant narrative in philosophy that social systems are stable with periods of change. This assumption is now under attack from many different sources that contend that social systems are ever-changing with periods of stability. 59 “Surfing the edge of chaos”201 is very important from an organizational perspective because when an organization is in that state it is at its peak performance in three major components of organizational success: engaging the processes of self-organization and emergence, leveraging sensitive dependence and changing the initial conditions of its organizing patterns (or what has popularly become known as the “butterfly effect”), and learning and generating new ideas. 3.2.2 Emergence and Self-Organization In chaos theory, emergence is the capacity of complex non-linear systems to have an orderly state emerge out of a chaotic state.202 This idea was first postulated by chemist Ilya Prigogine when he argued that order could appear out of chaos in a seemingly natural and inexorable manner without the benefit of an external organizer.203 He based this idea on his observation of how chemicals acted in a self-organizing way – for example, the way molecules act in the boiling water of a teapot: at first they move around frantically in seemingly random patterns until they hit the boiling point, when they all organize into stable and repeated patterns of movement. Emergence is best illustrated by describing an experiment conducted by geneticist Stuart Kaufman at the Santa Fe Institute. Kaufman was interested in discovering how individual genes that execute their instructions simultaneously fall into regular patterns that allow the replication 201 For a more detailed description of this concept, see Shona Brown & Kathleen Eisenhart, “The Art of Continuous Change: Linking Complexity Theory and Time-Paced Evolution in Relentlessly Shifting Organizations (Mar 1997) 42(1) Administrative Science Quarterly 1. 202 Pascale (1999) supra note 200 at 58. 203 See Ilya Priorogine and Isabelle Stengers, Order Out of Chaos: Mans New Dialogue with Nature (New York: Bantam, 1984), 60 of a species. To investigate this he designed a simple replication of a genetic system. He had 100 light bulbs. They all had instructions to turn on or off independently according to their own instructions. No governing system existed and so his hypothesis was that the random behaviour of the light bulbs would settle into random patterns. The results of the experiment were astonishing. Within a few minutes the system always settled down into a few more or less orderly states. This is emergent complexity – orderly systems that arise out of chaotic states where independent nodes are all operating according to their own instructions. Another example of how to explain emergence is to use the concept of fractals from geometry.204 A fern has a simple set of initial rules on the construction of its body – its genetic code. As it repeats the set of instructions on a strand of a leaf, a leaf, a branch, or the overall plant, a complex system emerges that resembles the initial structure specified by the genetic instructions. The single pattern of the initial genetic instructions is repeated at ever-greater levels of complexity. The complete fern is the emergent complexity from the replication of the initial set of instructions at increasing levels of complexity, with higher levels of complexity emerging from lower levels of complexity. OD practitioner Harrison Owen has used these insights from chaos theory to create the self-organization hypothesis: “All human systems are self-organizing and naturally tend toward high performance provided that the essential preconditions are present and sustained.” The implicit idea in this hypothesis is that self-organization is the key to high performance. But how 204 This analogy was taken from an article by Margaret Wheatley, see Margaret Wheatley, “Chaos and the Strange Attractor of Meaning” in Ralph Stacey, et al, Complexity and Organization: Readings and Conversations (New York: Routledge, 2006) at 101. 61 does emergence work? In human social systems (like corporations) the process of emergence works in a predictable pattern:205 1. The organization moves out of an equilibrium state because of an internal or external event (usually a complex problem it is unable to solve with current patterns of organization). 2. There is a breakdown of existing structures and events occur that sever possibility of ever going back to the previous equilibrium state. 3. There is a period of experimentation with new organizing patterns. 4. Order re-emerges in the system. The key to this process is that organizations learn when they are confronted with complex new situations and in order to learn they are required to confront and consider the assumptions they have about the way they organize or do business. In order for this to work the social system needs to have rich networks for communications to flow. 3.2.3 Learning Complex Tasks and Generating New Ideas Corporations that surf the edge of chaos learn better because they are leveraging the power of self-organization and emergence. That is because the most meaningful learning (double-loop learning) involves the challenging of assumptions and plans.206 In an “organized” organization this kind of challenge is often not accepted or there are programmed responses or feedback loops designed to stop this kind of questioning from occurring. In a “self-organizing” organization these kinds of conversations are often normal. As an organization is challenged with 205 When chaos theory ideas have been applied to small groups it has generated interesting results where systems had transformative change – there appears to be a common set of events. For examples see Priorogine & Stengers (1984) supra note 206 and C. Smith and G. Gemmill, “A Dissipative Structure Model of Organization Transformation (1985) 38(3) Human Relations 751. 206 This statement is based on Chris Argyris and Donald Schon’s theory of action learning model, which was described in detail earlier in this chapter. See note 140. 62 complex tasks they move towards chaos because the normal responses and procedures in the organization cannot find a suitable solution for the task. A new solution is required and that often means changing the basic ways that the organization functions. One way to understand the underlying organizational assumptions from a chaos theory perspective is to conceive of them as the initial conditions of the system – the pattern of organizing that then gets repeated as a fractal. For example, if the rules of interaction or cultural fractal contain distrust, self-interest, and a preference for competition it is almost certain that the emergent organization or social system will also contain those characteristics – often in an amplified way. In contrast, if the fractal contains trust, dialogue, and temperance the emergent social system may be different.207 3.2.4 The Butterfly Effect Complex non-linear systems are extremely sensitive to variations in their initial set of conditions. Tiny variations in the initial conditions can be amplified through repetition and cause unpredictable and disproportionate outcomes in the system. This property of non-linear systems is called “sensitive dependence”. It was made famous by Lorenz, who labelled it the “Butterfly Effect”. In 1963, Lorenz was working on a computer model that predicted weather patterns. In setting up the model he accidentally entered an initial variable as .506 instead of .506127. The result was a completely different weather pattern than the one generated using the full number.208 Lorenz coined the metaphor “butterfly effect” to explain this sensitivity to initial conditions. The metaphor goes something like this: Does the flap of a butterfly’s wings in Brazil set off a tornado 207 It is interesting to note that this is not a new idea at all. In fact, this is one of the oldest ideas we have – and many of the word’s religions are based on this idea. 208 He published this finding that complex non-linear systems are extremely sensitive to initial conditions in his 1963 paper, see Edward Lorenz, “Deterministic Nonperiodic Flow” (March 1963) 20(2) Journal of the Atmospheric Sciences 130–141. 63 in Texas? The answer is yes.209 Sensitive dependence states that complex non-linear systems (like organizations and social movements) are extremely sensitive to the initial sets of conditions (the conversations people have that replicate the organizational culture). Small changes in the initial set of conditions can have dramatic non-proportional (non-linear) effects on the emergent systems. In the context of organizations, sensitive dependence means that if an organization is surfing the edge of chaos, small changes in the way people interact with each other or in the conversations they have with each other can have dramatic and transformative effects on the organization. The “tipping point” or “bifurcation point” is the point at which enough small changes have happened that the system switches over and a new pattern emerges. Complex non-linear systems are not random. They still follow deterministic laws. They are in effect path dependent and future states depend on prior states.210 However, because of their sensitivity to initial conditions it becomes very difficult to predict long-term outcomes in complex non-linear systems because each component of the system is caught in a complex non-linear feedback loop. Each time the component engages in a feedback loop it can carry out the initial set of conditions or it can vary them. If the initial conditions are repeated it leads to stable (and predictable) outcomes. If the initial conditions are varied it can lead to unstable (and 209 In fact, Lorenz did not refer to a butterfly at all but rather a seagull. The more elegant butterfly was developed through later speeches. For a description of Lorenz’s contribution to chaos theory and the butterfly effect, see Tim Palmer, “Edward Norton Lorenz” (2008) 61 (9) Physics Today 81. For Lorenz’s papers containing the ideas that would lead to the butterfly effect, see Edward Lorenz, “Three approaches to atmospheric predictability” (1969) 50 Bulletin of the American Meteorological Society 345 and Edward Lorenz, “Atmospheric predictability as revealed by naturally occurring analogues” (1969) 26 Journal of the Atmospheric Sciences 636. 210 The concept of path dependence has featured prominently in corporate law scholarship in the debate on convergence of corporate governance. See Lucian Bebchuk & Mark Roe, “A Theory of Path Dependence in Corporate Ownership and Governance” (1999) 52 Stanford Law Review 127. 64 unpredictable) outcomes. The feedback loops are the deterministic structure of the system. The ability to vary the initial set of conditions is the non-deterministic property of the system. The interaction of these two effects is referred to as “bounded instability”. Short-term predictions of chaotic systems are possible because the ability for the variation in the short term will most often be limited. For example, weather predictions are usually good up to about a week. 3.2.5 Directing vs. Disturbing Complex Systems The consequence of applying the lens of chaos theory to corporations is that the difficulty in predicting the end result of an intervention or change in a complex non-linear system becomes understandable. In every conversation the current patterns may be reinforced or they may be changed. The result of all those conversations just emerges. Emerging complexity creates multiple futures. Chaos theorist Richard Pascale said it this way: One consequence of emerging complexity is that you cannot see the end from the beginning. While many can readily acknowledge nature’s propensity to self-organize and generate more complex levels, it is less comforting to put oneself at the mercy of the process with the foreknowledge that we cannot predict the shape that the future will take. Emerging complexity creates not one future but many.211 It also means that it is very difficult to direct complex systems because there are weak cause and effect relationships. Greater precision is neither sought nor possible. This idea, when applied to corporate law and regulation, calls into question the whole idea of command and control regulation that is attempting to direct corporations towards specific regulatory outcomes. The idea also calls into question the notion of managerial control and strategic planning because it may not be possible to plan and control activities in a corporation. 211 Pascale (1999) supra note 200 at 65. 65 Harrison Owen has presented a few examples of how control and planning are really just illusions in corporations. His first example is that of the corporation’s organizational chart – how come it always seems to be out of date and should not be trusted? His second example illustrates further the limits of the formal system within a corporation: he uses the example of a labour union’s “work to rule” campaign where workers are only doing what the rules say they should do. In this situation, management should be happy – but they rarely are. That is because “if we actually did business the way we say we did business, we would be out of business.”212 Owen goes so far as to argue that control is really the enemy of high performance. In his words, the only way to make sure his OD practice Open Space will not work is for someone to take control.213As a result of these realizations there are new types of corporate planning called improvisational planning and leadership within corporations being developed.214 The result of attempts at planning and control in self-organizing systems is usually unintended consequences. This is because the attempt at control is unable to understand the full complexity of the system and while it may cause the effect desired in the direct relationships, it will usually have counter-effects in other areas not considered when the control was designed or planned. Pascale provides two examples of attempted interventions in complex systems that had dramatic unintended consequences. In the first, the Forest Service in Yellowstone Park attempted to eliminate forest fires by putting them out every time they happened. In effect, they wanted to maintain the ecosystem of the park in an equilibrium state. The result was that 100 years of dead 212 Harrison Owen, Wave Rider – Leadership for High Performance in a Self-Organizing World (San Francisco: Berrett-Koehler, 2008) at 100. 213 Ibid at 130. 214 Ibid at 4. For a leading example see Henry Mintzberg’s work on emergent strategy: Henry Mintzberg & James Waters, “Of Strategies, Deliberate and Emergent” (1985) 6(3) Strategic Management Journal 257. 66 material accumulated until eventually the fire that erupted could not be put out and living things and top soil that otherwise might have survived was destroyed.215 In the second example, the U.S. Fish and Wildlife Service tried to control the coyote to protect sheep and cattle ranchers. They spent $3 billion over 100 years for a variety of measures, including bounty hunters, traps, poison, and genetic technology. The result was that the modern-day coyote is 20% larger and significantly smarter than its predecessors and can be found in 49 of the 50 states instead of the 12 states that were its traditional habitat.216 Emmanuel Ogbonna and others have studied the impact of unintended consequences on organizational interventions.217 They argue that there will always be anticipated consequences and unanticipated consequences of every managerial action and that unintended results come from the divergences in the ways individuals intervene in or take up the managerial action. 218 In a case study in culture change initiatives in eight companies where they were looking for unintended consequences, they found that “in each company, the desired change had been undermined by at least one unintended consequence, which was accepted by members to have either slowed or even stopped the change programme.”219 The conclusion of the study was that practitioners “should be wary of culture change programmes or models that promise totally predictable change, and should embrace guidance that appreciates and incorporates unpredictable 215 See Pascale (1999) supra note 200 at 59-60. 216 Ibid at 69-70. 217 See Lloyd Harris & Emmanuel Ogbonna, “The Unintended Consequences of Culture Interventions: A Study of Unexpected Outcomes” (2002) 13 British Journal of Management 31; and Emmanuel Ogbonna & Barry Wilkinson, “The False Promise of Organizational Culture Change: A Case Study of Middle Managers in Grocery Retailing” (2003) 40(5) Journal of Management Studies 1151. 218 Harris & Ogbonna (2002) Ibid at 35-36. 219 Ibid at 37. 67 effects.”220 A second case study of attempted culture changes in the grocery retailing industry showed the same results for change initiatives aimed at corporate managers who, the authors hypothesized, should have been more accepting of culture change processes.221 The conclusion of that study included the following: “we find it difficult to accept any notion that changing the organizational context would be easy, or indeed would be considered suitable for systemic pursuit.”222 In the realm of corporate regulation there are many examples of unintended consequences. For example, sociologist Chalmers Johnson has argued that the United States’ attempt to legislate away the Zaibatsu in Japan after the Second World War just led to the creation of the Kieretsu223 and sociologist Neil Fligstein has argued that U.S. anti-trust laws attempt to break monopolies of trust power in the U.S. triggered the creation of the large multi-national corporations.224 3.2.6 Chaos Theory and the Corporation The application of chaos theory concepts to the corporation has a long history.225 One of the first was Peter Viall in 1975 in his article “Towards a Behavioral Description of High Performing Systems.” His ideas led to the famous work by Peters and Waterman, In Search of Excellence, where they argued that excellence kept showing up in organizations – just not where 220 Ibid at 46-47. 221 Ibid. 222 Ogbonna & Wilkinson (2003) supra note 220 at 1174. 223 See Chalmers Johnson, MITI and the Japanese Miracle (Stanford: Stanford University Press, 1982). 224 See Fligstein (1990) supra note 118. 225 These historical notes are summarized from Harrison Owen, supra note 215 at 26-40. 68 it was expected and not according to plan. Their book and the OD practice that resulted from it called Operational Excellence led to a revolution in management practice and theory. Jerry Collins also found that excellence occurs not according to plan and where it is least expected in a popular recent study, described in his book Good to Great. After studying a large sample of Fortune 500 companies to determine what made great companies become great while their competitors floundered, Collins’ team identified what they called Level 5 Leadership as one of the characteristics of great companies.226 Collins argued that the way to identify these leaders in an organization is to look for excellence where no one is taking credit for it. In a more recent example, the properties of self-organizing systems were used by AT&T during the preparation for the 1996 Olympics when they used Harrison Owen’s Open Space Technology to fast track 10 months of design and planning for their pavilion in Olympic Village into a 2 day contractor summit. 25 contractors came to the summit with lots of difficult history and a blank page to design from. Open Space, developed by Harrison Owen, is an example of a self-organizing dialogic process that leverages the insights of chaos theory. It involves collecting a large group of people in an empty room, with no agenda, and letting them do whatever they want. The rules are simple: anyone can suggest a topic and become the leader for that topic in a breakout session. Four principles then apply: 1) whoever comes are the right people; 2) whatever happens is the only thing that could have; 3) whenever it starts is the right time; and 4) when it’s over, it’s over. One 226 Level 5 leaders are leaders who “build enduring greatness through a paradoxical blend of personal humility and professional will.” See 131.(2001) supra note at 20. 69 law also applies: the law of two feet. If you are not contributing or getting value where you are, use your two feet to go somewhere else.227 In order for Open Space to work a few characteristics need to be present: 1. Need a real issue – something people care about. 2. Voluntary self-selection – people come because they cared to come. 3. High levels of complexity – a situation that is so complex no one person can figure it out. 4. High levels of diversity in participant group. 5. Presence of passion and conflict combined with urgency.228 The outcome of Open Space is a community of people that are drawn together in a “nexus of caring.” Owen describes the magic of Open Space in the following way: “When caring people gather around something they care about there is a high likelihood that useful things will happen.”229 But all of this may be hard to believe for some because it is so far removed from conventional thinking about the way to manage corporations and to run change initiatives in corporations. OD practitioner and theorist Peggy Holman stated this the best when she said that you need to fall flat on your face in a change effort to understand the power of emergence.230 Holman’s work is focused on how to leverage the capacity of corporations to self-organize and exhibit emergence. She argues that there are two types of change in an organization: small 227 See Holman et al. (2007) infra note 384 at 135. For a guide on how to run an Open Space event see Harrison Owen, Expanding Our Now: The Story of Open Space Technology (San Francisco: Berrett Koehler, 1997) and Harrison Owen, Open Space Technology: A User’s Guide (2ed) (San Francisco: Berrett Koehler 1997). 228 Owen (2008) supra note 215 at 69-70. 229 Ibid at 76. 230 Peggy Holman, Engaging Emergence: Turning Upheaval Into Opportunity (San Francisco: Berrett Koehler, 2010) at xi. 70 incremental change with foreseeable outcomes, or spontaneous transformational change that occurs with emergence.231 Holman also argues that it may be possible to create “applied emergence” or actually create the conditions for emergence in an organization. She argues that in an emergent change no one is in charge and simple rules can engender complex behaviour. Peggy Holman’s argument is that to cause emergent change you simply need to change the rules of interaction. By interaction she means the social interactions between the organizational participants and she is using an expansive use of the word “rules” here that includes not just the formal rules of the organization but also the informal ones. In a corporation people follow simple rules to organizational assumptions. In order to do the least to cause the greatest change and benefit you just need to focus on changing those organizational rules and assumptions. For Holman, emergent change processes are “methods that engage the diverse people of a system in focused yet open interactions.”232 She uses the phrase “designing conversations that matter” to describe this.233 Her model of change is very similar to the model of change for self-organizing systems in CAS: 1. Disruption: Change starts with disturbance – a new complex problem that the corporation is unable to solve. 2. Differentiation: Accentuate the differences that matter among people. Things start changing while in a state of chaos. 3. Coherence: A new understanding or system emerges.234 This is the model of change within corporations adopted here for the dialogic regulation model. 231 Ibid. 232 Ibid at xi. 233 Ibid at 47. 234 Ibid at 10-18. 71 3.2.7 Chaos Theory, the Corporation, and Dialogic Regulation The insights of chaos theory when applied to the corporation as a complex human social system hypothesize that the initial set of rules (or fractal structure) of the corporation are the daily interactions, dialogues, and conversations amongst corporate actors.235 Corporate culture then is the context within which daily interactions are made possible and the emergent social system that is the result of all of those daily interactions. If the emergent corporate culture is not desirable the root cause of that problem probably lies in the interactions, conversations or dialogues happening at the small group level within the corporation. It also means that to change the corporate culture by leveraging the emergent properties of the corporations one needs to change the daily interactions of the corporate participants because any small change in those initial conditions can be repeated and taken up by many individuals and then emerge as a property of the overall corporate culture. How to change it in a desired direction is the question that Dialogic OD takes up. At some point, if small changes are repeated enough times a bifurcation or “tipping” point will be reached and a new culture will emerge. In organizations, there are expected institutionalized interactions, which take the form of roles and scripted relationships between corporate actors. These institutions can be thought of as the bounded instability of the system. Corporate actors have no choice but to engage in the non-linear feedback loops which these institutions constitute. For example, corporate managers have to go to the weekly managers’ meeting, do annual performance reviews with all of their direct reports, etc. To the extent that there are institutionalized feedback loops or scripted interactions within which the corporate actors operate, the corporation social system is deterministic. Each 235 Wheatley (2004) supra note 207 at 110. 72 time a corporate actor engages in these feedback loops, however, that agent is free to vary, ignore, or alter the institutional arrangements. Corporate actors still have the capacity for freedom of choice. For example, what they say or do at the weekly managers’ meeting or in their performance review meetings with their employees is their choice. Depending on the extent to which corporate actors change the rules or scripts, stable or unstable outcomes are possible. Complexity theorist Ralph Stacey refers to this as “transformative causality”236 because cause and effect links are circular and can lead to unexpected outcomes.237 In his words, patterns of interaction between human agents either reconstitute themselves though repetition or transform and evolve. If they evolve they can get amplified if many people take them up. Small immeasurable changes in patterns of interaction can escalate into major changes in the system but the direct causal relationship is lost in the complexity of what happens.238 The idea that cultural change can happen simply by changing conversations probably seems a bit naive and hard to believe. So, I have developed a simple participatory exercise that can be completed in fifteen minutes that illustrates all of the basic ideas of emergence and corporate regulation firsthand for audiences. I use this exercise at conferences and workshops when I am talking about how changing conversations can change culture. The exercise is based loosely on the “Helium Stick” exercise used by consultants as a team building exercise. The exercise is perfect for illustrating emergence because it recreates a self-organizing system with 236 Ralph Stacey, Complexity and Organizational Reality: Uncertainty and the Need to Rethink Management After the Collapse of Investment Capitalism (Milton Park: Routledge, 2010) at 17. 237 Stacey (2006) supra note 207 at 79. 238 Ibid at 82. 73 initial conditions, emergence, and an attempt at control, unintended outcomes, and the learning of a complex task. The exercise works like this. You need at least eight people and a light stick like a broom handle. You get the people to stand in two lines facing each other with the arms out at chest height. Each person should make a “gun” sign in each hand and extend his or her hands into the middle. The stick then gets laid across everyone’s fingers so that their fingers are underneath the stick. At this point you have a system in equilibrium. Everyone is standing facing each other, the stick is flat, and everyone’s fingers are in contact with the stick and holding it up at chest height. We normally proceed at this point by explaining the system we just created by telling the team that they are a corporation and the stick represents their environmental performance. If the stick is on the ground they are having no impact on the environment. If the stick gets to shoulder height they will have an environmental catastrophe like the Deepwater Horizon. Everyone is asked to acknowledge that the stick is in equilibrium and then it is removed for the next few minutes (it can have a tendency to act like helium during the instructions that follow). The next step is to create the initial conditions in the culture of their organization. This is when the team is told that there is one rule in the corporation – your fingers must always be in contact with the stick. Next, the attempt at control is introduced. The facilitator states that they are the environmental regulator and that the environmental performance of the corporation is unacceptable. The regulator then instructs the team that a new law has been passed and that they need to put the stick on the ground. Simple? Understood? Great. The stick is then put back on people’s hands and the regulator watches diligently. Every time someone’s fingers are off the stick they remind them to “keep your fingers in contact with the stick”. The result – without exception – is that the stick rises, usually very fast to above 74 shoulder height. At this point we take the stick away and we say, “You failed. What happened? Were we not clear enough? Did you not hear us? We want the stick on the ground. Let’s try again.” People are usually surprised and willing to try again. If it is done quickly before people are allowed to start a “self-organization” process, the same result will occur. If they do start self-organizing – we usually play a trick on them after the second failure by stating that as regulators we need to punish the corporation and so we are going to put the CEO in jail, and we always pick the person who was beginning to organize people – it has the effect of slowing down the self-organizing process. The exercise gets really interesting at this point because the only way to solve the problem is for the team to start challenging its own assumptions and to self-organize under new assumptions. Often they start by setting a count to lower the stick or agree to bend their knees to lower at the same time – and it usually involves someone taking the lead. While they will make progress doing this, it is not the solution to the problem because the stick can never get on the ground while their fingers are under it. Inevitably, after about ten minutes, someone will ask one of the two crucial questions: “Why do our fingers always need to be in contact with the stick?” or “Why are we holding our hands like guns? Why don’t we just grab the stick in our hands and put it on the ground?” This solution came via a demonstration at the Canadian Business Ethics Research Network and it is a brilliant solution. That team was able to put the stick on the ground within two seconds after struggling with the exercise for the previous ten minutes after having lost their CEO member to jail in the corner. The CEO continued to try to organize them by lobbing instructions to the team from jail. If we want a different outcome we should just question our own assumptions, change our behaviour, and achieve the future we want. 75 The exercise is also interesting at this point for what it shows us about the behaviour of the regulator. If the regulator is vigilant and active and does not provide the team the time to organize, the same result will keep happening over and over. More rules, laws, or re-stating the rules may not be helpful. Putting the CEO in jail, punishing and humiliating them, just shuts down conversation, slows down the self-organizing process, and usually prolongs the time until the solution is found. When the regulator steps back and asks the team “Do you understand the objective?” “Is everyone honestly trying to put the stick on the ground?” and the key one “What is stopping you from doing that?” and when these questions are combined with some time between attempts, the team can usually solve the problem very quickly. This exercise illustrates all of the concepts introduced in this section on self-organizing systems and it also provides a quick insight into the change to regulation that will be required if corporations are considered to be self-organizing systems. To summarize the insights of the exercise: • It creates a human social system in an equilibrium state. • It sets initial conditions: hands in “guns” and fingers in contact with the stick. • It introduces a new complex task to the system that cannot be solved with the current sets of organizing principles. • It involves an attempt at control in a complex system (the regulatory outcomes). • That attempt at control leads to unintended consequences – the stick goes up instead of down. • The system goes into chaos and confusion until someone starts to ask the questions about the assumptions built into the initial conditions. • Once the initial conditions are changed, the result is a dramatic and transformative effect on the system and the desired regulatory outcomes are achieved easily. 76 3.2.8 Conclusion To summarize, in a chaos theory perspective of the corporation, the corporation is a self-organizing system: a complex human social system that is non-linear and dynamic.239 Therefore, it shares many of the characteristics of other self-organizing systems: 1. Corporations exhibit emergent qualities. 2. They are sensitive to initial conditions. 3. They are replete with feedback loops (both negative and positive). 4. There is no proportionality between cause and effect. 5. More complex levels of organization arise out of lower levels of complexity organization – for example actions and outcomes may arise out of corporate culture. 6. The patterns and content of the conversations and interactions between system participants are the initial conditions of the system. 7. Small changes in the initial conditions can have dramatic non-proportional effects on the resulting system.240 If these characteristics are true, it means that we have been approaching corporate law and regulation with the wrong approach and from the wrong perspective and we will have to reconsider and redesign our attempts to generate regulatory outcomes by focusing on the patterns of interactions of the individuals in the corporation and leveraging the corporation’s self-organizing properties. 239 There is no settled definition of a chaotic system. However, there are three properties that are generally associated with chaotic systems: 1) sensitivity to initial conditions; 2) topologically mixing (components of the system blend into each other over time); and 3) periodic orbits are dense (components come into contact with each other on a regular basis). These three properties are what allow small changes in the initial conditions to spread in a non-linear fashion. The corporation, as a human social system, exhibits all three of these properties. 240 This list closely mirrors Ralph Stacey’s list for the application of chaos theory to organizations: 1) fractal structure – irregular forms are scale dependent; 2) recursive symmetries between scales and levels – repeat a basic structure or fractal at different levels; 3) sensitive to initial conditions – small changes in the system send the system in a wildly different direction; and 4) replete with feedback loops. Systemic behaviour is the emergent outcome of multiple chains of interaction. See Stacey (2006) supra note 207 at 255. 77 3.3 Language, Narrative, and Discourse There are a lot of similarities between chaos theory, which focuses on the chaotic properties of systems, and dialogic systems theory, which focuses on the chaotic forces of language. It should come as no surprise then that language, narrative, and discourse analysis should have similarities with chaos theory and this pattern is emerging more and more all the time. For example, Harrison Owen (the inventor of Open Space Technology), one of the leading OD practitioners in the application of chaos theory to corporations, had a background in narrative analysis.241 The benefit of the link between narrative analysis and chaos theory is that it highlights the fact that ideas about chaos are not new and that “serious thinking about chaos and order has been ongoing for about three or four thousand years.”242 In fact, Owen has argued that “the eternal dance of chaos and order” is the fundamental social process and that chaos, confusion, and conflict are essential to living.243 The idea of the importance of language came from the postmodern thinkers, including Foucault, Derrida, and Wittgenstein.244 These thinkers rejected the idea of objective truth and global meta-narratives and they argued that language was important because, among other things, we cannot imagine something that we do not have the words to describe245 and our 241 This similarity will be seen again in the law section. The leading thinker of dialogic laws was Robert Cover, whose background was also in analyzing scripture. 242 Owen (2008) supra note 215 at xiv-xv. 243 Ibid at xvi. 244 For examples of the work of the post-modern thinkers, see Michel Foucault, Discipline and Punish (New York: Pantheon, 1977), Jacques Derrida, Of Grammatology (Baltimore & London: Johns Hopkins University Press, 1997, corrected edition, trans. Gayatri Chakravorty Spivak), and Ludwig Wittgenstein, Tractatus Logico-Philosophicus (London: Kegan-Paul, 1922). 245 Wittgenstein wrote; “The limits of my language mean the limit of my world”. See Wittgenstein (1922) Ibid at 5.6. 78 language limits our ability to think of possibilities by not taking into account the other side of meaning.246 These postmodern thoughts are beginning to have a larger effect on many disciplines including, among others, sociology, law, and organizational development. Within organizational development these thoughts have had an impact on two particular theories that form the theoretical basis of Dialogic OD: social constructionism and the theory of generativity. The application of these two theories to organizations is referred to as Organizational Discourse Studies. These theories are summarized below. 3.3.1 Social Constructionism and the Theory of Generativity Social constructionism posits that social reality is constructed by the interaction of the participants.247 In a social constructionist view, nothing has inherent meaning in any objective way, it only gains meaning in the context of a particular social group who derive and maintain its meaning through social interactions.248 Social constructionism can be applied to things and to beliefs.249 For example, in the world of corporate regulation, the significance of the thing “tree” depends on the social group to whom you are speaking or you are a part of. Corporate managers in the forest sector could conceive of it as a “revenue generating unit” or simply “lumber”, 246 See the concept of deconstruction from Jacques Derrida, See Derrida (1997) supra note 247. 247 For a description of social constructionism, see Peter Berger & Thomas Luckmann, The Social Construction of Reality: A Treatise in the Sociology of Knowledge (Garden City, NY: Anchor Books, 1966); John Searle, The Construction of Social Reality (New York: Free Press, 1995); and Kenneth Gergen, An Invitation to Social Construction (2ed) (Thousand Oaks: Sage, 2009). 248 There are two different strands of social constructionism: weak-form social constructionism accepts that there are some physical and certain things in the world that are not socially constructed, that socially constructed things and beliefs build upon; and strong-form social constructionism argues that nothing exists until it is talked about. For a discussion of weak- and strong-form social constructionism, see Searle (1995) supra note 250 at 56. He refers to facts that exist without social construction as “brute facts”. See also Stephen Pinker, The Blank Slate: The Modern Denial of Human Nature (Penguin: New York, 2002) at 202. 249 Ian Hacking, The Social Construction of What? (Cambridge: Harvard University Press, 1999). 79 economists could conceive of it as a “carbon off-set”, environmentalists as natural living thing, or First Nations people might consider it a spiritual entity. Similarly, beliefs such as “the corporation” are also socially constructed and depend on what social actors think corporations should be at any given time in any given group. For example, in previous work I have shown how the understanding of what a corporation is, for the purposes of the law, has changed over the last one hundred years to encompass at least a dozen or so theories of the corporation, each linked to the context of society at the time the understanding was created.250 Over a hundred years ago, a corporation was thought to be a body politic or organization of people.251 Now it is thought of as a nexus of contracts that govern the inputs and outputs in the production process.252 In this thesis, it is argued that the corporation is a socially constructed reality that consists of the patterns of interactions and conversations between the organization members. A social constructionist view argues that the patterns of organizing within corporations are not dependent on biological or physical reality but are constructed simply from the interactions of the participants.253 This is a dramatically different view than the economic view of the corporation that asserts that corporations exist in their current form because they are objectively the most efficient way of organizing economic activity.254 250 For a description of this history, see Cody (forthcoming) supra note 112 at Chapter 3. 251 The classic definition of the corporation as a body politic comes from Stewart Kyd. See Stewart Kyd, A Treatise on the Law of Corporations (New York: Garland Publishing, 1978) at 13. 252 Ibid. 253 See David Cooperrider & Diana Whitney, “A Positive Revolution in Change” in David Cooperrider, P. Sorenson, D. Whitney & T. Yeager (eds), Appreciative Inquiry: An Emerging Development for Organizational Development (Champaign: Stipes, 2001) at 15. 254 This is consistent with the “nexus of contracts” theory of the firm and the work of economic historian Alfred Chandler. See Alfred Chandler, The Visible Hand: The Managerial Revolution in American Business (Cambridge: Belknap Press, 1977). 80 The social constructionist perspective can support anything from extreme nihilism to extreme generative capacity. It can support nihilism because it can be used to argue that nothing is real and the world only exists to the extent that we create it. It can also be incredibly generative because it can be used to argue that we can create any future we want.255 In fact, when it comes to the structure of corporations, the social constructionist view allows the possibility that “the only limitation to how people organize is their imagination and collective agreement about what is expected and possible.”256 Therefore, to change the corporation, we need only envision a new type of corporation and work together to make it a reality. The idea of generativity used in this thesis was first presented in 1978 by social psychologist Kenneth Gergen.257 He argued that the positivist-empiricist approach to social science, with its pre-eminent focus on facts, the demand for verification of theory, the assumption of temporal irrelevance, and the commitment to objective status for the scientific researcher, limited the ability of contemporary social science to generate new theories and ideas that had the capacity to transform social life. He argued that there was a difference in the approaches between the American social psychologists engaged in “stimulating research within an elite, professional circle” and the approach of European social theorists who “challenged the 255 See David Cooperrider, Frank Barrett & Suresh Srivastva, “Social Construction and Appreciative Inquiry: A Journey in Organizational Theory” in D. Hosking, P. Dachler, and K. Gergen, Management and Organization: Relational Alternatives to Individualism (Brookfield, VT: Ashgate, 1995). 256 David Cooperrider and Suresh Srivastva, “Appreciative Inquiry in Organizational Life” in R. Woodman & W. Passmore, (eds.), Research ion Organizational Change and Development – Volume I (Stamford: JAI Press, 1987) at 129. 257 Kenneth Gergen, “Toward Generative Theory” (1978) 36(11) Journal of Personality and Social Psychology 1344. 81 assumptive bases of social life”.258 In Gergen’s words, “[Generativity is] . . . the capacity to challenge the guiding assumptions of the culture, to raise fundamental questions regarding contemporary social life, to foster reconsideration of that which is ‘taken for granted’ and thereby to furnish new alternatives for social action.”259 3.3.2 Organizational Discourse Studies Within organizational behaviour (OB), the application of the postmodern concepts of language and study of the implications of social constructionist theory in understanding how language, discourse, and narrative within organizations contribute to organizational change is referred to as organizational discourse studies (ODS).260 ODS conceptualizes organizations not as machines or living systems but “more like an ongoing conversation or dialogic system”261 where the reality of the organization is continually created by the interactions among the organization’s actors.262 In this approach, organizational change is driven by changing the discourse in the corporation, in its many forms.263 For example, changes can be made to how conversations unfold, what narratives define the way things are done, and what new ideas or 258 Ibid at 1345. 259 Ibid at 1346. 260 For examples of Organizational Discourse Studies, see Mats Alvesson & Dan Karreman, “Varieties of Discourse: On the Study of Organizations through Discourse Analysis” (2000) 53(9) Human Relations 1125 and D. Grant, C. Hardy, C. Oswick & L. Putnam, “Introduction – Organizational Discourse: Exploring the Field” in and D Grant, C. Hardy, C. Oswick & L. Putnam (eds) The SAGE Handbook of Organizational Discourse (London: Sage, 2004). 261 This definition comes from Marshak, Grant and Floris. See Robert Marshak, D. Grant & M. Floris, “Discourse and Dialogic organization Development” in in D Cooperrider, D. Zandee, L. Godwin, M. Avital & B. Boland (eds.) Organizational Generativity: The Appreciative Inquiry Summit and a Scholarship of Transformation (Advances in Appreciative Inquiry, Volume 4, pp. 89-113) (Bingley, UK: Emerald Group Publishing Limited, 2013) at 4. 262 Ibid at 3. 263 For the full form of this argument, see Frank Barrett, G. Thomas, & S. Hocevar, “The Central Role of Discourse in Large Scale Change: A Social Construction Perspective” (1995) 31 Journal of Applied Behavioral Science 252. 82 conversations might enable new ways of thinking to emerge.264 There are four concepts that are important to understanding ODS: discourse, text, context, and conversations.265 Discourse is “a set of inter-related ‘texts’ that along with the related practices of text production, dissemination and consumption brings an idea or way of thinking into being.”266 Texts are forms which convey content or meaning and include speech, documents, pictures, gestures, and symbols.267 All discourses are dependent on their temporal, historical, and social context. For example, while the thumbs-up hand gesture is positive in Western cultures it has a more negative meaning in Middle-Eastern, African, and South American cultures. Discourse takes place through conversations. A conversation is defined as “a set of texts that are produced as a part of a dialogue between two or more people.”268 This means it is possible to cause organizational changes by making changes to the everyday conversations of the organizational actors.269 Applying ODS theory to OD, ODS theorists Marshak, Grant and Floris recently wrote an article outlining the main implications of the field to understanding how change occurs in organizations. In that article they argued that: • Organizational discourse (and discursive processes) play a central role in the continuous and iterative social construction of organizational reality.270 • There is a diversity of discourses latent in any organizational situation.271 264 Marshak et al. (2013) supra note 266 at 5. 265 Ibid at 7. 266 Ibid. 267 Ibid. 268 Ibid. 269 Ibid. 270 Ibid at 17. 271 Ibid at 9. 83 • Organizational discourse includes more modalities than just text and speech and can include things like visual representations, gestures, and symbols, etc.272 • Power has an impact on organizational discourse by favouring certain dominant discourses over others.273 Organizational change involves conflict because there is tension between two different discourses.274 • Changing the existing dominant organizational discourse will lead to organizational and behavioural change.275 • Discourse operates at multiple social and psychological levels simultaneously that impact how actors think and act: intrapersonal (internalized stories and beliefs), personal (how individuals use language, stories, gestures, etc.), inter-personal and group (direct interaction among organizational actors), organizational (the dominant thinking and organizational practices e.g. mission statement, values etc.), and socio-cultural (standard ways to refer to phenomena at the societal level, e.g. the market, social responsibility, etc.).276 Because the discourse at any level is linked to and informed by the discourse at other levels, multiple levels of discourse must be affected at the same time in order to cause change. • Change is an ongoing iterative process and not an episodic process.277 The similarities between chaos theory and ODs are notable. Both argue that organizations are complex non-linear systems created from basic organizing patterns that are the everyday interactions the organization actors have with each other. Both also posit that organizational change can occur by simply changing those everyday patterns of interaction. ODS takes the analysis a step further, though, and argues that it is the discourse that needs to change: the speech, the texts, the pictures, images, and symbols. Dialogic OD is based on these common similarities. 272 Ibid at 20. 273 Ibid at 6. 274 Ibid. 275 Ibid. 276 Ibid at 10-12. 277 Ibid at 22. 84 3.4 Dialogic OD In 2009, Gervase Bushe and Robert Marshak wrote an important article arguing that a new dialogic organization development practice has emerged in the last 25 years.278 They argued that these new dialogic practices are not just developments in previous open systems theories of organization development,279 but departures based on different underlying assumptions about people, social systems, and change. Bushe and Marshak compared these new practices, which they referred to as “Dialogic OD”, to traditional organization development practices, which they referred to as “Diagnostic OD”. Traditional Diagnostic OD is positivist in its methodology and tends to view organizations as if they were living systems.280 It assumes the existence of a discernable objective reality that can be investigated or researched to produce data. The data is then used to compare a given team or organization to a prescriptive model of what a “healthy” team or organization looks like. Any deficiencies or problems with the current system are identified and problem-solving skills are engaged to bring the team or organization into the desired state. This Diagnostic OD approach is focused on changing people’s behaviour.281 It is also focused on finding problems and proposing solutions, so the tone of the approach can often be negative. The kinds of statements that are associated with this approach are: you can only change what you can 278 See Gervase Bushe & Robert Marshak, “Revisioning Organizational Development: Diagnostic and Dialogic Premises and Patterns of Practice” (2009) 45(3) The Journal of Applied Behavioral Sciences 348; and Gervase Bushe and Robert Marshak, “Further Reflections on Diagnostic and Dialogic Forms of Organizational Development” (2009) 45(2) The Journal of Applied Behavioral Sciences 378. 279 I explain the term “open systems theory” in the section on complexity that follows this section. 280 In my previous work I refer to this approach using W.R. Scott’s terminology: natural open systems perspective. 281 Bushe & Marshak (2009) supra note 283 at 355. 85 measure; diagnosis should precede plans and actions; organizing is a problem to be solved; change behaviour and a change in thinking will follow.282 In contrast, Dialogic OD is based on the ideas of social constructionism, postmodernism, complexity theory, and linguistic and narrative approaches to organizations.283 Bushe and Marshak refer to it as “post-positivist”.284 Dialogic OD is a planned change process that improves organizational effectiveness by changing collective narratives in order to change collective thinking and action. It views organizations as dialogic or meaning making systems.285 It does not assume the existence of a discernable reality but instead the existence of many versions of reality – one for each organizational participant. Dialogic OD focuses on changing what people think and say, not what they do.286 In this way, Dialogic OD practices are more conducive to double-loop learning. Rather than attempting to diagnose and manage change, dialogic approaches to OD focus on generating new ideas that will self-organize change towards the desired end state. Instead of facts, the key data in this approach are people’s narratives of the potential future. David Cooperrider’s Appreciative Inquiry approach is a good example of this type of Dialogic OD. Other examples include Clear Leadership, Open Space, and World Café.287 These Dialogic OD practices did not emerge from any grand theory; instead, they emerged from 282 For an example of this see Gwyn Bevan & Christopher Hood, “What’s Measured is What Matters: Targets and Gaming in the English Public Healthcare System” (2006) 84(3) Public Administration 517. 283 Bushe & Marshak (2009) supra note 283 at 349. 284 Ibid at 349. 285 D. M. Boje & Al Arkoubi, “Third Cybernetic Revolution: Beyond Open to Dialogic Systems Theories” 4(4) Tamara Journal 139. 286 Bushe (2009) supra note 176 at 353. 287 Each of these dialogic approaches, together with AI, is summarized in the next section on social constructionism. 86 what worked in practice.288 The differences that Bushe and Marshak identified between Diagnostic OD and Dialogic OD are summarized in Appendix G. The organization in the dialogic OD approach is a socially constructed reality that consists of the patterns of interactions and conversations between the organization members.289 Dialogic OD approaches adopt conversational approaches to working with people, groups, and larger social systems in efforts to cause change. Instead of diagnosing problems, they build narratives, stories, and conversations “that aid in the establishment of more effective or just patterns of organizing.”290 For example, Gervase Bushe’s Clear Leadership approach to collaborative learning assumes that there are multiple realities in any group and that attempting to agree on one version is counterproductive.291 Change happens in groups when people become “aware of the variety of stories people have about themselves and each other and understand their own part in creating unproductive patterns of interaction”.292 Bushe and Marshak explain the importance of dialogue and conversation to organizational change in the following quote: “What these newer forms of OD also have in common is a search for ways to promote more effective dialogue and conversation and a basic assumption that it is by changing the conversations that normally take place in organizations that organizations are ultimately transformed”.293 288 Bushe & Marshak (2009) supra note 283 at 349. 289 Ibid at 360. 290 Ibid at 353. 291 Bushe (2009) supra note 176. 292 Bushe & Marshak (2009) supra note 283 at 353. 293 Ibid at 360. 87 In their 2009 article, Bushe and Marshak proposed four characteristics for Dialogic OD: 1. Change comes from changing the everyday conversations that take place in the system (who has conversations, how they have the conversations, the skills they bring to the conversations, and/or what the conversations are about). 2. There may or may not be a data collection phase. If there is, it is not about discovering an objective reality or set of facts but to discern the alternate realities of the organizational participants. 3. The aim is to generate new ideas, images, stories, narratives, and socially constructed realities that affect how people in the system think and act. The focus is not on behaviour but on the intersubjective reality. 4. The focus is on collaboration and participation of the people in the organization to allow them to make informed choices.294 In contrast, Diagnostic OD change processes that are designed to change organizational culture often have negative unintended consequences. 295 This is because people resist being told what to do. Dialogic approaches to change encounter less resistance because there is no attempt to change behaviour without the consent of those who must change – just a challenge to change narratives, symbols, and communications to co-construct a more valued future. Change in a Dialogic OD approach becomes self-organizing, or, in other words, the participants collectively identify a desired end state and figure out how to get there themselves without being told exactly what to do. Dialogic approaches elicit new thinking in people that allows them to change themselves. The Dialogic OD approach is empowering to individuals, it is collaborative, it 294 Ibid at 356-359. 295 For a study of the unintended consequences of culture change initiatives, see Harris & Ogbanna (2002) supra note 220. 88 increases awareness about a social system in order to change it, and it develops and enhances organizations in humanistic ways.296 Since the 2009 article, Bushe and Marshak have further refined their understanding of Dialogic OD. In 2013 they identified three concepts Dialogic OD is based on: discourse, emergence, and generativity.297 They have also offered insights as to what they believe are the characteristics of an OD practitioner’s “Dialogic Mindset”298: • Groups and Organizations are Continuously Self-Organizing: Change is a part of the continuous process of self-organizing. Organizations are not static entities that go through periodic episodic changes – they are in constant flux as organizational reality is constantly created by the interactions of the organizational members. Instead of planning a specific change, dialogic OD practitioners help foster the conditions that lead to new ways of thinking.299 • Organizations are Meaning Making Systems: Organizational reality is a social construct that emerges through dialogic processes. How things are talked about by organizational actors is the most significant factor in shaping how people think about any given situation.300 • Language Matters: Language does more than convey information. It creates, sustains, and transforms social experience.301 • Structure Participative Inquiry and Engagement to Increase Differentiation: Narratives are stories that are shared by a group of people to explain how things are. In any organization there are a variety of narratives. The role of the practitioner is to help people understand that in any situation there are a number of narratives, learn what the consequences of their own narratives are, and recognize which narratives are dominant or suppressed.302 296 Bushe & Marshak (2009) supra note 283 at 357. 297 Gervase Bushe & Robert Marshak, “Dialogic Organization Development” in B. Jones & M. Brazzel (eds) The NTL Handbook of Organization Development and Change 2nd Ed. (forthcoming) at 2. 298 Bushe and Marshak, “Dialogic Mindset” Article. Ibid. 299 Ibid at 293. 300 Ibid. 301 Ibid. 302 Ibid. 89 • Transformational Change is more emergent than planned: Transformational change cannot be planned – attempts to plan are more obstacles and impediments than resources to transformational change. No top down hierarchical planned changes. • Dialogic conditions that lead to change include most or all of the following: o Disruption to prevailing social conditions. o Creating a “container” that provides the right ingredients and space for participants to inquire together to allow new possibilities to emerge. o Emphasizing generativity rather than a problem solving approach. o Inviting the whole person to the conversation including emotions and not just the rational side.303 3.4.1 Appreciative Inquiry and Other Dialogic OD Practices A good example of a widely used and accepted dialogic OD practice is Appreciative Inquiry (AI). It has been described as a “positive revolution in change” management that focuses on using positive language, framing, and dialogue as an intervention technique to assist groups in finding transformative solutions to difficult situations. AI arose out of the work of David Cooperrider, Frank Barrett, and Suresh Srivastva at Case Western in the 1980s.304 At the time Cooperrider was a Ph.D. student in the Organizational Behavior program at Case Western and he was engaged in collecting data on the problems in an organization.305 He realized that simply asking questions about the problems in the organization changed the conversation about the organization to how problematic things were. In reality, the organization was doing very well and he decided to celebrate that. So, he changed his approach and instead of asking about problems he became excited about the organizational processes that gave life and vitality to the 303 See Gervase Bushe & Robert Marshak, “The Dialogic Mindset” (forthcoming) 22 Research in Organization Development and Change. 304 For a detailed description of the history and theory of AI, see Gervase Bushe, “Foundations of Appreciative Inquiry: History, Criticism, and Potential” (2012) 14(1) AI Practitioner 8, and Gervase Bushe, “Appreciative Inquiry: Theory and Critique” in D. Boje, B. Burnes and J. Hassard (Eds.) The Routledge Companion to Organizational Change (Oxford, UK: Routledge, 2012) 87-103. 305 The Cleveland Clinic. 90 organization, something he later called an “appreciative analysis”.306 He decided to focus instead on these positive “life giving” properties of the system. In 1987, Cooperrider and Srivastva outlined three main theoretical bases of AI: social construction, generativity, and adopting an appreciative approach instead of a problem solving approach.307 3.4.2 Social Constructionism At the core of AI and the other new Dialogic OD practices is the idea from social constructionism that social reality is constructed by the interaction of the participants. Therefore, to change the corporation, we need only envision a new type of corporation and work together to make it a reality. This is a bold statement but we are going to hold to it and repeat it throughout this thesis because of the concept’s significant generative power. 3.4.3 Generativity Influenced heavily by Gergen’s work on generativity, Cooperrider thought that organization theory would benefit from a new generative metaphor of human “organization as a mystery and miracle” that can never be fully comprehended. The language we use also limits the extent of the generativity of any endeavour that we undertake. For example, OD theorists David Cooperrider and Frank Barrett have stated, “. . .we live in worlds that our questions create. . .”308 306 Bushe & Marhsak (2010) supra note 283 at 3. 307 See David Cooperrider & Suresh Srivastva, “Appreciative Inquiry in Organizational Life” (1987) 1 Research in Organizational Change and Development 129. For a summary of the theoretical bases for AI and the critiques of AI, see Gervase Bushe, “Appreciative Inquiry: Theory and Critique” in D. Boje, B. Burnes, and J. Hassard (eds), The Routledge Companion to Organizational Change (Oxford, U.K.: Routledge, 2011) 87-103. 308 David Cooperrider and Frank Barrett. “An Exploration of the Spiritual Heart of Human Science Inquiry” (2002) 3(3) Reflections 56 at 58. 91 This may have been the beginning of the difference between dialogic and diagnostic OD practices. Dialogic OD practices are based on eschewing the preeminence of facts, the verification of theory, and the objective status of the scientific researcher. Instead they allow researchers to take an active and participative role in facilitating the generation of new ideas that can transform social systems.309 AI is generative when it generates “one or more new ideas arise that compel people to act in new ways that are beneficial to them and others.”310 3.4.4 Appreciative Approach vs. Problem Solving Approach AI is also very different from a problem solving approach.311 Cooperrider and Srivastva argued that problem solving as a tool for social change did a very poor job and that it might actually be counterproductive. This conclusion was based on the belief that “through our assumptions and choice of method we largely create the world that we discover.”312 In other words, the questions we ask determine the answers we generate. 313 Even the “most innocent question evokes change.”314 By simply asking questions about what problems exist the researcher can perpetuate the problems. For example, Bushe has stated that “questions about 309 Bushe (2010) supra note 283 at 3. 310 Ibid at 2. 311 Cooperrider and Srivastva referred to the problem solving approach as “sociorationalist”. See Cooperrider & Srivastva (1987) supra note 313. Cooperrider and Whitney referred to this as “The Principle of Simultaneity” in their (2001) article. See Cooperrider & Whitney (2001) supra note 256 at 15. Bushe refers to this as “Inquiry as Intervention”. See Bushe (2010) supra note 283 at 4. 312 Cooperrider & Srivastva (1987) supra note 313 at 129. 313 Cooperrider &Whitney (2001) supra note 256 at 15. 314 For an argument supporting this statement, see Kenneth Gergen’s critique of the method of Social Psychology, Kenneth Gergen, “Social Psychology as History” (1973) 26(2) Journal of Personality and Social Psychology 309, where he argues that scientific knowledge generated by social scientists actually influences the phenomenon it is meant to passively describe. This idea has been applied to economics with the label “performativity”. For an example of performativity, see Mackenzie (2006) infra note 521. 92 conflict create more conflict.”315 Cooperrider and Diana Whitney similarly believed that “human systems grow in the direction of what they persistently ask questions about.”316 Inquiry and change are simultaneous.317 This is a major difference from the diagnostic approach, which advocates inquiry before change. By acknowledging that inquiry itself is an intervention into the organization, the nature of the inquiry made becomes paramount. One of the most “impactful things a change agent or [researcher] can do is to articulate questions.”318 By adopting a positive and generative approach to inquiry the possibilities for transformational change are increased. For example, a question like “Why are corporations not more socially responsible?” might end up with a solution that proposes changes to the existing system to make them prove they are being socially responsible, including implementing corporate social responsibility reporting (a problem-solving approach). In contrast, a question like “How can businesses become agents of world benefit?” has the possibility to generate new ideas beyond the scope of the existing system (an appreciative approach). Problem-solving approaches often lead to what Cooperrider calls “deficit discourse”.319 A chart summarizing the difference between a problem-solving approach and an appreciative approach is attached as Appendix H. In 2001, Cooperrider and Diana Whitney built on the theoretical bases of AI offered by Cooperrider and Srivastva and outlined five theoretical principles that are at the core of AI.320 315 Bushe (2011) supra note 313 at 4. 316 Cooperrider & Whitney (2001) supra note 256 at 3. 317 Ibid at 15. 318 Ibid (2001) at 5. 319 Ibid (2001) at 20. 320 See Ibid (2001) at 14-21. The five principles were: 1) the Constructionist principle (social realities are constructed by social participants); 2) the Principle of Simultaneity (inquiry itself can cause change in an 93 Other authors have also proposed additional principles.321 Gervase Bushe summarized ten of these principles in his 2012 article on AI theory and critique.322 In addition to the social construction, generativity, and problem solving323 principles outlined above, three other principles are particularly relevant for dialogic regulation: discourse and narrative, anticipatory reality, and positive affect.324 3.4.5 Language, Discourse, and Narrative Language, discourse, and narrative are important in AI. Language is important because we cannot imagine something that we do not have the words to describe325 and our language limits our ability to think of possibilities by not taking into account the other side of meaning.326 The language we use also limits the extent of the generativity of any endeavour that we organization and the questions we ask can determine the answers we get); 3) the Poetic Principle (the organization is a book that is constantly being co-authored with the stories and narratives of the participants); 4) the Anticipatory Principle (the only limit on what future organizations can be is our collective imagination and ability to work together to make it happen); 5) the Positive Principle (human beings are more effective at causing change when it is approached in a positive way). 321 See Gervase Bushe, “Five Theories of Change Embedded in Appreciative Inquiry” in Cooperrider, Sorenson, Whitney and Yeager (eds), Appreciative Inquiry: An Emerging Direction for Organization Development (Champaign, Il: Stipes, 2001) at 117-127. 322 The ten principles outlined by Bushe are: 1) Inquiry as Intervention; 2) Generativity; 3) Discourse and Narrative; 4) Anticipatory Reality; 5) Positive Affect; 6) Building on Strength,; 7) Stakeholder Engagement; 8) Working with Self-Organizing Processes; and 9) Life Giving Properties of Social Systems. See Bushe (2011) supra note 3013 at 3-13. 323 Bushe refers to this as “Inquiry as Intervention”, Bushe (2011) supra note 313 at 4. 324 The language used to describe the 5 principles is the language from Bushe (2011) supra note 313 at 4-9. 325 Ludwig Wittgenstein wrote: “The limits of my language mean the limits of my world”, see note 248. 326 See Derrida (1997) supra note 247. 94 undertake. As an example of this, Cooperrider and Whitney stated: “[w]e create the organizational worlds we live in”. 327 Narrative and discourse are important because in dialogic OD practices organizational life is expressed in the stories people tell each other every day.328 Cooperrider and Whitney posed the “Poetic Principle,” which states that the corporation is an open book whose story is constantly being co-authored by corporate participants. Organizations make themselves understandable to their members and stakeholders through the stories they tell.329 AI advocates “that organizations consist of multiple stories and perspectives and seek to ensure that no particular story is considered more significant than another”330 because the marginalized voices in an organization are often where innovations reside. Bushe has stated that corporations change when people change and groups change when they change their assumptions through a group learning process and their stories change. Ludema has gone a step further to argue that the “collection, telling, and re-telling of people’s “best of” stories results in a wave of countervailing micro-narratives that combine, over time, to change the prevailing macro-narrative of the organization.”331 327 Cooperrider & Whitney (2001) supra note 256 at 12. 328 Bushe (2011) supra note 313 at 5. 329 James Ludema, “Appreciative Storytelling: A Narrative Approach to Organization Development and Change” in Fry et a. (eds) Appreciative Inquiry and Organizational Transformation: Reports from the Field (Westport, CT Quorum, 2002) at 239-261. 330 Bushe (2011) supra note 313 at 5 citing Diana Whitney, “Postmodern Principles and Practices for Large Scale Organization Change and Global Cooperation” (1996) 14(4) Organization Development Journal 53. 331 Ludema (2002) supra note 337. This is an example of an emergent property in a social system and it will be discussed in more detail in the section on Chaos Theory. 95 3.4.6 Anticipatory Reality Anticipatory reality refers to the fact that we bring into reality the futures that we imagine.332 For example, think of the hand-held communication devices in Star Trek and then look at their similarity to the very first Motorola flip-phones. The key here is to adopt a possibility-centric approach to organizational change as opposed to a problem-centric approach. Boyd and Bright have argued that the problem-centric approach creates a deficit discourse that makes organizational participants wary of consultants and change and more likely to be defensive and resist change interests and to be more focused on their self-interest than the common good.333 Bright and Cameron recently confirmed that human social systems do move towards the affirmative images that they create of the future.334 The idea of anticipatory reality is based on the work of philosopher Martin Heidegger. It means that, with regard to corporations, the corporations of our future will be limited only by our collective imaginations, what we want, what we believe to be possible, and our ability to work together to make them a reality. There is a lot of research that supports the anticipatory principle, including placebo studies in medicine, where people are cured because they think they are getting the drug,335 the Pygmalion dynamic in classrooms, where the smartest child in the 332 See Martin Heidegger, Being and Time (New York: SCM Press, 1962) (trans. John Macquarrie & Edward Robinson). 333 Neil Boyd & David Bright, “Appreciative Inquiry as a Mode of Action Research for Community Psychology” (2007) 35(8) Journal of Community Psychology 1019. 334 See David Bright & Kim Cameron, “Positive Organizational Change: What the Field of POS Offers to OD Practitioners” in Rothwell et al. (eds) Practicing Organization Development: A Guide for Managing and Leading Change (3rd Ed) (San Francisco: Pfeiffer-Wiley, 2009) at 397-410. 335 See Lee Jussim, “Self-Fulfilling Prophecies: A Theorectical and Integrative Review” (1986) 93 Psychological Review 429. 96 classroom is whoever the researcher tells the teacher is the smartest child in the classroom,336 and the importance of a positive inner dialogue to personal and relational well being.337 The simplest way to state the anticipatory principle is that what we believe will happen shapes our perceptions and actions, making it more likely to happen.338 3.4.7 Positive Affect Positive Affect refers to the ability to build rapport between people that supports and sustains change processes. Creating good feelings among people assists in getting change going because people experiencing positive feelings are more flexible, creative, integrative, open to information, resilient, able to cope with adversity, have an increased preference for variety, and accept a broader array of behavioural options.339 Bushe has found that positive affect has the greatest impact on what he called “pre-identity” systems or groups where the majority of members do not identify with the group, because positive affect can allow formation without going through the storming phase.340 Once a group is formed (a “post-identity” group where the majority of the members already identify with the group) positive affect will not have as much of 336 See Robert Rosenthal & Lenore Jacobsen, Pygmalion in the Classroom: Teacher Expectation and Pupil’s Intellectual Development (London: Crowne House Pub Limited, 2003). 337 For example, see Sarah Pressman & Sheldon Cohen, “Does a Positive Affect Influence Health?”(2005) 131(6) Psychological Bulletin 925. 338 Bushe classifies this as a separate theoretical principle called Building on Strength. He argues that it is more about getting more of what we pay attention to. See Bushe (2011) supra note 313 at 7. 339 See Bushe (2011) Ibid.at 21. Bushe provides the following studies as examples to support this statement: Alice Isen, “Positive Affect and Decision Making” in M. Lewis and J.M. Haviland-Jones (eds.) Handbook of Emotions (NY: Guildford, 2000) at 417-435; Barbara Lee Frederickson, “The Broaden-and-Build Theory of Positive Emotions” in M. and I. Csikszentmihalyi (eds.), A Life Worth Living: Contributions to Positive Psychology (NY: Oxford University Press, 2006) at 85-103; Barbara Lee Frederickson, “The Role of Positive Emotions in Positive Psychology: The Broaden-and-Build Theory of Positive Emotions” (2001) 56 American Psychologist 218-226. 340 Gervase Bushe, “Appreciative Inquiry with Teams” (1998) 16(3) Organization Development Journal 41. 97 an impact.341 Positive affect will also have the biggest impact in organizations where there is little of it.342 Positive affect is also very fragile, because the human mind has a predominant propensity to notice and store the negative.343 For example, relationship psychologists John and Julie Gottman have found that among partners in happy relationships the ratio of positive to negative comments about each other in simple conversation was 20:1!344 That means that to maintain a positive feeling the positive affect has to be overwhelmingly positive. Bushe has argued that this may not be possible in sustained change initiatives and that AI is not just all about being positive.345 Social systems are built on paradoxes and oppositions – they must have room for both the positive and the negative.346 Systems can move forward and learn from both positive conversations and critical conversations. Bright, Powley, Fry, and Barrett have argued that negative organizational experiences can be inquired into in a generative way if the assumption is made that there is an image of unmet positive expectation in every negative experience.347 Bushe has argued that AI is not just action research with a positive question.348 The key to AI is not about being positive – being generative is far more important.349 341 For a description of the effects of positivity on pre and post identity systems, see Gervase Bushe, “Meaning-Making in Teams: Appreciative Inquiry with Pre-Identity and Post-Identity Groups” in Fry, Barrett, Seilling, and Whitney (eds) Appreciative Inquiry and Organizational Transformation: Reports from the Field (Westport: CT Quorum, 2002) at 39-63. 342 See Bright & Cameron (2009) supra note 342. 343 See Roy Baumeister, Ellen Bratslavsky, Catrin Finkenauer & Kathleen Vohns, “Bad is Stronger than Good” 5(4) Review of General Psychology 323. 344 See John Gottman & Julie Gottman, The Art and Science of Love (Seattle: Gottman Institute, 2009) at 4. 345 Gervase Bushe, “Appreciative Inquiry Is Not (Just) About the Positive” (2007) 39(4) OD Practitioner 30. 346 Boje & Arkoubi (2005) supra note 290 at 12. 347 David Bright, Edward Powley, Ronald Fry & Frank Barrett, “The Generative Potential of Cynical Conversations” in Zandee, Cooperrider, and Avital, Generative Organization: Advances in Appreciative Inquiry (Vol. 3) (Beingley, England: Emerald Publishing, 2010). Pamela Johnson has also made the argument that AI needs to evolve out of the 98 3.4.8 The Practice of Appreciative Inquiry The practice of AI has four phases: Discovery, Dream, Design, and Destiny. This is referred to as the “4-D model”.350 A diagram of the 4-D Model is attached as Appendix I. The model starts with the Discovery phase, where participants discuss the best of “what is” concerning the object of inquiry. For example, if the inquiry is into corporate social responsibility, participants might inquire into the best examples, experiences, or programs they can find that reflect socially responsible corporate behaviour.351 Most often this takes the form of cascading interviews where participants are interviewed for their own “best of” experiences and then they become interviewers who ask others about their “best of” experiences.352 The Dream Phase follows, in which the participants are asked to do three things: imagine their group, organization, or community at its best, identify the common aspirations of the group participants, and symbolize these aspirations in some way.353 In the Design Phase the participants take their deficit discourse because it is only through the deficit discourse that the affirmative statement can be created. See Pamela Johnson, “Transcending the Polarity of Light and Shadow in Appreciative Inquiry: An Appreciative Exploration of Practice” in Zandee, Cooperrider and Avital (eds.) Generative Organization: Advances in Appreciative Inquiry (Vol. 3) (Bingley England: Emerald Publishing, in press). 348 See Gervase Bushe “Generative process, generative outcome: The transformational potential of appreciative inquiry”, in D.L. Cooperrider, D.P. Zandee, L.N. Godwin, M. Avital & B. Boland (eds.) Organizational Generativity: The Appreciative Inquiry Summit and a Scholarship of Transformation (Bingley, UK: Emerald Group Publishing Limited, 2013) Advances in Appreciative Inquiry, Volume 4, pp. 89-113. 349 For a description of the circumstances under which AI is generative, see Ibid at 90 where he states: “. . . but simply a focus on the positive, without a focus on the generative, will likely not produce much change at all.” 350 The 4-D Model was first outlined in 2001 by Cooperrider and Whitney, see Cooperrider & Whitney (2001) supra note 256 at 7-14. For a good summary of the 4-D Model, see Gervase Bushe (2012) supra note 310 at 2-3. 351 This was actually done at the meeting that created the Global Compact. 352 Bushe (2012) supra note 310 at 2. The innovation to have participants interviewing each other came from John Carter at the Gestalt Institute. See John Carter & Pamela Johnson, “The Roundtable Project” in C. Elliot, Locating Change: An Introduction to Appreciative Inquiry (Winnipeg: International Institute for Sustainable Development, 1999) at 255-279. 353 Ibid. 99 common dream and develop a concrete proposal for the new state of the group, organization, or community. Finally, in the Destiny354 phase all of the new ideas are brought into action. This phase was initially called Delivery but the name was changed because Delivery evoked images of traditional change management. The goal here is more improvisational implementation where widespread understanding of the common dream allows the authorization of individuals to take whatever actions are necessary to make the dream a reality.355 Leadership’s role in this scenario then becomes “tracking” for wanted behaviour and “fanning” or rewarding desired behaviour to elicit more of it.356 In this approach, leaders cannot control change – they only unleash it. Some practitioners have modified the model to include a fifth phase that comes first: Define.357 Cooperrider and Whitney referred to the Define step as the “Affirmative Topic Choice” and believe that it is the most important part of any Appreciative Inquiry.358 The “Affirmative Topic” should be a generative metaphor that creates a “compelling enough image that it evokes new thoughts, conversations, and actions”.359 Barrett and Fry, among others, have 354 Cooperrider now calls this stage the Deployment stage. 355 Bushe has offered the following four-step model of Improvisational Destiny: 1) organizational members have a collective sense of the changes they want to make; 2) they believe that they don’t need permission to act but instead are encouraged to take whatever action they deem necessary to achieve the design; 3) a “launch event” creates conditions for people to take voluntary, visible action toward the change objectives; and 4) the leader’s role is to watch what happens and support and amplify what they want more of. See Bushe (2011) supra note 313 at 25. For a description of the improvisational implementation process, see Frank Barrett, “Creativity and Improvisation in Jazz and Organizations: Implications for Organizational Learning” (1998) 9 Organization Science 605; Gervase Bushe & Anique Kassam, “When is Appreciative Inquiry Transformational? A Meta-Case Analysis” (2005) 41(2) Journal of Applied Behavioral Science 161 at 9-10 arguing that the improvisational approach is more generative because six of seven generative studies in the meta-analysis used the improvisational approach. 356 Bushe (2009) supra note 176 at 218-231. 357 AI models with the Define Phase included are referred to as 5-D Models. 358 Cooperrider & Whitney (2001) supra note 256 at 5. 359 Bushe (2010) supra note 310 at 16. 100 argued that engaging the right people and identifying a topic that is of interest to the organization and compelling to stakeholders is critical to success.360 An AI summit is one of the more frequent used AI practices.361 An AI summit is a meeting of anywhere between 30 people and tens of thousands who are brought together over a period of 3-4 days to (1) talk about the organization’s strengths, (2) envision opportunities for positive change, (3) design the desired changes, and (4) implement and sustain the changes and make them work.362 AI has been used successfully at many corporations including GTE363, Avon (Mexico)364, Hunter-Douglass,365 Nutrimental Foods,366 and Roadway.367 In the late 1990s, Cap Gemini Ernst 360 As referenced in Bushe (2010) supra note 310 at 3 and Frank Barrett & Ronald Fry, Appreciative Inquiry: A Positive Approach to Building Cooperative Capacity (Chagrin Falls: Taos Institute, 2005). 361 Bushe (2011) supra note 313 at 3. For a description of the emergence of the AI Summit as a practice, see David Cooperrider & Diana Whitney, “The Appreciative Inquiry Summit: An Emerging Methodology for Whole System Positive Change” (2000) 32(2) OD Practitioner 13. 362 See Holman et al. (2007) infra note 384 at 201. For a complete description of how to hold an AI Summit, see James Ludema, Diana Whitney, Bernard Mohr & Thoman Griffin, The Appreciative Inquiry Summit: A Practitioner’s Guide for Leading Large Group Change (San Francisco: Berrett-Koehler, 2003). 363 For a description of the GTE process, see Diana Whitney, David Cooperrider, D.L. Garrison, M.E. & J.P Moore, “Appreciative Inquiry and Change at GTE: Launching a Positive Revolution” in Fry, Barrett, Seiling & Whitney (eds) Appreciative Inquiry and Organizational Transformation: Reports from the Field (Westport: CT Quorum, 2002) at 165-180. 364 M. Schiller, “Imagining Inclusion: Men and Women in Organizations” in Fry, Barrett, Seiling and Whitney (eds.) Appreciative Inquiry and Organizational Transformation: Reports from the Field (Westport: CT Quorum, 2002) at 149-164. 365 Amanda Trosten-Bloom, “Creative Applications of Appreciative Inquiry in an Organization-Wide Culture Change Effort: The Hunter-Douglass Experience” in Fry, Barrett, Seiling & Whitney (eds) Appreciative Inquiry and Organizational Transformation: Reports from the Field (Westport: CT Quorum, 2002) at 181-210. 366 Nutrimental foods engaged 750 employees in two AI summits, which led to outstanding productivity and financial results. See Ilma Barros & David Cooperrider, “A Story of Nutrimental in Brazil: How Wholeness, Appreciation, and Inquiry bring out the Best in Human Organization” (2000) 18(2) Organization Development Journal 22. 367 Roadway is a U.S. Trucking firm that has used AI to improve management/employee relations and improve performance. See Ludema et al. (2003) supra note 370. 101 & Young adopted AI as the core of its human capital consulting practice.368 Wal-Mart has also used it to support their goal to become a model of sustainable enterprise. Wal-Mart’s CEO Lee Scott wanted to engage large numbers of employees inside Wal-Mart and large numbers of stakeholders outside of Wal-Mart, particularly suppliers, in the process. Wal-Mart used an AI summit to accomplish this. It was a whole system intervention where over 500 people from a wide variety of industries met with technical experts to create the Wal-Mart sustainability index that measures the sustainability of Wal-Mart 65,000 suppliers.369 This is only one of many instances where Walmart used AI summits to further their sustainability agenda. AI has also been used to engage in the corporate social responsibility debate. In 2004, the United Nations used an Appreciative Inquiry Summit as the structure for its Global Compact Leaders Summit.370 UN Secretary Kofi Anan inspired the Global Compact when he met with world business leaders at Davos at the World Economic Forum. The Global Compact is a set of 10 principles in the areas of human rights, labour, the environment, and anti-corruption, which corporations are asked to embrace, support, and enact.371 The Global Compact now has over 1,700 participating businesses.372 One of the outcomes of the 2004 AI Summit was the addition 368 Bushe (2011) supra note 313 at 9. 369 For a description of the Wal-Mart AI Summit, see Bushe (2011) supra note 313 at 37-38. 370 For the outcomes of the AI Leaders Summit, see United Nations, “The Global Compact Leaders Summit Final Report” (2004) online United Nations: <http://www.unglobalcompact.org/docs/news_events/8.1/summit_rep_fin.pdf > (accessed January 4, 2010). 371 The Global Compact 10 principles can be found at online: <http://www.unglobalcompact.org/AboutTheGC/heTenPrinciples/index.html> (accessed May 4, 2014). 372 See McKinsey & Company, “Assessing the Global Compact” (May 4, 2004) at 5 online: <http://www.unglobalcompact.org/docs/news_events/9.1_news_archives/2004_06_09/imp_ass.pdf > (accessed January 4, 2011). 102 of the 10th principle regarding anti-corruption. This is what Kofi Anan had to say about the Global Compact AI process in a letter to David Cooperrider after the Summit: I would like to commend you more particularly for your methodology of Appreciative Inquiry and to thank you for introducing it to the United Nations. Without this, it would have been very difficult, perhaps even impossible, to constructively engage so many leaders of business, civil society, and government.373 Shortly after the 2004 Global Compact Leaders Summit, the Academy of Management, the United Nations Global Compact, and the Case Weatherhead School of Management held a world forum entitled “Business as an Agent of World Benefit: Management Knowledge Leading Positive Change.”374 Business as an agent of world benefit is a generative image created to make people think about new ideas and new ways of engaging corporate enterprise and corporate innovations for world benefit. The forum involved many stakeholder groups, including corporate participants. There were 400 delegates and over 1000 virtual participants. AI is not the only dialogic OD practice. In fact, The Change Handbook by Peggy Holman et al. outlines over 60 ways to construct meetings to assist large groups of people to engage in dialogue.375 World Café and Open Space are examples of other methodologies. World Café is based on the idea that the knowledge to solve all of our issues already lies within our collective knowledge. It is a dialogic process designed to draw out that collective knowledge. In the process, groups of four people sit at a table and talk about a specific question or issue. As they 373 Personal correspondence from Kofi Annan. Online: <http://appreciativeinquiry.case.edu/intro/un/KofiAnnan.pdf> (accessed January 4, 2011). 374 See “Forum Overview - Business as an Agent of World Benefit: Management Knowledge Leading Positive Change” online: <http://www.bawbglobalforum.org/docs/WEBPDFForumOverview.pdf.> (accessed December 26, 2010). There was another Global Forum in 2009. 375 See Peggy Holman, Tom Devane, & Steven Cady, The Change Handbook: The Definitive Resource on Today’s Best Methods for Engaging Whole Systems (San Francisco: Berrett-Kohler, 2007). 103 talk they write down key ideas and insights. After 20 minutes people are invited to switch tables. One host stays at the table and shares the key insights or ideas from previous groups with newcomers. This process repeats itself until desired, often with new questions that build on each other as the café progresses. As people dialogue and become more connected, latent collective knowledge becomes apparent.376 How much of a difference does this type of change intervention make? The truth is that there is not a lot of empirical data that proves this approach works. But, that is because dialogic processes are not about measurement and dialogic researchers believe that any measurement can affect the change process. Bushe has stated that: “[w]hen successful, AI generates spontaneous, unsupervised, individual and group organizational action toward a better future.”377 Bushe has also argued that AI is transformational when 1) there is a focus on changing what people think instead of what people do, and 2) there is a focus on supporting self-organizing change processes that flow from new ideas rather than leading implementation of centrally or consensually agreed upon changes.378 3.5 Dialogic Systems Theory and Corporate Law and Regulation The dialogic approach to organization development outlined above is very important for corporate law and regulation. I have argued in previous work that the key to designing effective corporate laws and corporate regulation is to first explore the answers to two questions: 1) what 376 See Holman et al. (2007) Ibid at 179. For more information on how to run a World Café event, see Juanita Brown, David Isaacs & the World Café Community, The World Café: Shaping Our Futures Through Conversations that Matter (San Francisco: Berrett-Koehler, 2005). 377 Bushe (2007) supra note 353 at 30. 378 See Bushe & Kassam supra note 363 where the authors found that 7 out of 20 AI processes in the meta-case analysis were transformational. See also Bushe (2011) supra note 313 at 4. 104 is a corporation? and 2) what is the purpose of corporate law?379 I argued that the answer to the first question determines the answer to the second. For example, if a corporation is a “nexus of contracts” that represents all of the contracts between inputs that together produce goods and services, then the purpose of corporate law will likely be to act as a standard form contract among the inputs that offers efficiency gains and establishes their rights.380 Similarly, if the answer to the first question is that the corporation is an entity with relationships with many stakeholders, then the purpose of corporate law will be to organize the interests of the various stakeholders.381 I also argued that organization theory is of primary importance to corporate law because organization theory is the discipline concerned with answering the question what is an organization, and by implication a corporation? I argued that corporate law has fallen behind organization theory because the still dominant theory of the corporation in corporate law – the “nexus of contracts” theory of the corporation – is a rational systems theory, while organization theory has, since the late 1970s, moved on to work more with natural open systems theories.382 After analyzing the current corporate theories through this lens of organizational theory, I applauded recent attempts by corporate theorists like Lynne Dallas (Power Coalition Theory) to create level 4 or level 5 natural open systems theories of the corporation for corporate law.383 I argued that these theories were attempting to keep pace with the understanding of what 379 See Cody (forthcoming) supra note 112 at 3-4. I refer to a theory that answers both of these questions as a “Corporate Theory”. 380 Ibid. 381 Ibid. 382 See Chapter 1 pages 32-33. 383 For Boulding’s hierarchy of systems, see Cody (forthcoming) supra note 112 at 44 and the chart attached as Appendix D. For a description of Power Coalition Theory, see Cody (forthcoming) supra note 112 at 151-157. 105 corporations are in the other social sciences. However, I criticized Power Coalition theory for being too focused on the conflict perspectives within organization theory and offered the building blocks of a more balanced natural open systems perspective theory of the corporation called The Social Theory of the Corporation.384 The adoption of such a theory would have a profound effect on corporate law and regulation and would provide the following two answers to the corporate theory questions: 1) the corporation is a human organization that provides meaning and stability to social life, and 2) the purpose of corporate law is to facilitate the formation, survival, and evolution of corporations and to provide stability to society, markets, and corporations by managing the relationships among corporate participants in a way that is consistent with the normative views of the society in question.385 I see now that this argument did not go far enough. I was trying to progress a level 3 or 4 theory to become a level 4 or 5 theory. The question I should have been asking was: “is there a way to develop a level 8 theory?” The Dialogic OD practices described above hold within them the potential to solve this issue because they can allow us to start to explain, understand, and predict how complex social beings with free choice interact with each other in complex social systems through the use of language, symbols, dialogue etc. The dialogic OD practices outlined in the previous section are all linked to dialogic views of systems because they focus on the collaborative properties of systems, acknowledge the multiplicity of languages, realities, and perceptions within a social system, and have at their core the goal of improving the self-reflexive characteristics of social systems – or the capacity to learn. Double-loop learning is a self-reflective dialogic process. A 384 Ibid Chapter 6: “The Social Theory of the Corporation”. 385 Ibid at 350 and 358. 106 “dialogic theory” of the corporation could transcend a level 4 or 5 system theory and become a level 7 or even level 8 theory of the corporation. 386 Organization theorists Boje and Arkoubi have made this argument as it pertains to organization development and dialogic systems theory. They have argued that a dialogic systems theory is possible and that it may be the 3rd cybernetic revolution. The 1st cybernetic revolution was mechanistic systems (or rational systems) and the 2nd cybernetic revolution was based on the law of requisite variety and links the variety in organizations to the variety in the environment (or open systems). The 3rd cybernetic revolution is about the “chaos of language variety” (or dialogic systems). They acknowledge the importance of such a theory to the development of systems theory in general and particularly to organizations: Third cybernetics takes us beyond open systems theory (level 4) in Boulding’s (1956) nine orders of complexity model. The reason is that lower order system levels [1-5 are] . . . fixated upon sign, upon unified language representations and metaphorizations. . . . At more complex orders, Boulding argues, similar to Bahktin, that sign-representation gives way to more multi-languaged ways of envisioning human systems: image (level 6), to symbol (level 7), to social networks engaged in history and self-reflexivity (level 8). . .387 Boje and Arkoubi use Bahktin’s dialogism theory of language as the basis for their dialogic systems theory. Bahktin argued that language determines our understanding of systems.388 Systems are not just dialogue between players. They are dialogic in their language forces – they are reflexive and interactive. They contain both orderly language forces and chaotic 386 Boje and Arkoubi argue that there were three earlier attempts to move beyond level 4 systems: Pondy (1976), Chomsky (1975), and Cooper (1989). 387 Boje & Arkoubi (2005) supra note 290 at 3. 388 Stacey refers to this as Second Order Complexity – that our systems become as complex as our understanding of them allows. 107 language forces. Bahktin referred to this as heteroglossia. Boje argued that heteroglossia raises two important challenges to standard systems theory based on unified language: 1) there is no unifying single language of system, and 2) there is no system language that is independent of context. Boje and Arkoubi conclude that “a dialogic systems theory, therefore, is about the chaos of language variety.” Boje and Arkoubi describe dialogic approaches as “a form of inquiry into differences.”389 A dialogic system is the antithesis of a mono-theory or mono-logic system, it is “an opposition of multiple philosophical views”.390 Dialogic systems theory is by definition a multi-paradigm theory. It is about building bridges between paradigms by allowing people to internalize other paradigms through dialogue.391 It requires “bridgers”: specialists in several fields who can master different paradigmatic languages and internalize them all to create new worldviews.392 The addition of each paradigm changes the overall system through dialogic language processes. This is a difficult task because people have a preference for simple explanations and single languages.393 In effect, dialogic regulation, as proposed in this thesis, is just such a multi-paradigm “bridge” and the dialogic regulation process built on collaborative learning loops is an institutionalization of that multi-paradigm dialogic process. Boje and Arkoubi propose three components of dialogic theory: cognition, axiology, and emotions. They outline the constituents of dialogic theory as “[d]ialogue between multiple 389 Boje (2005) supra note 290 at 9. 390 Ibid at 9-10. 391 Ibid at 7-8. 392 This is based on Pondy and Boje’s work on “bridgers”. See L. Pondy & David Boje, “Bringing Mind Back In” in Evan (ed.) Frontiers in Organization and Management (New York: Prager, 1981) 82-101. 393 This is evidenced in the following quote from Dostoevsky: “But Man is so partial to [monolithic] systems and abstract conclusions that he is willing to distort the truth deliberatively, close his eyes, and plug up his ears, all to justify his [mono] logic” quoted from Boje & Arkoubi (2005) supra note 290. 108 consciousnesses and identities, appreciative and collective inquiry, collaborative learning, tendency toward permanent liberation and transformation, reflexivity and toleration.”394 They also argue that dialogic systems need AI, which builds on positives, and depreciative inquiry, which helps with the rethinking of a dominant story.395 Boje and Arkoubi argue that in the corporate world it was perhaps the focus on competition and the unwillingness to look at collaboration that has held back the move into more complex systems models.396 Complexity theorist Stacey shares the excitement about the possibility of dialogic systems or systems built on the triumvirate of postmodernism, complexity theory, and chaos theory.397 He calls them a potential Kuhnian scientific revolution.398 This trio of advances in understanding can be seen through each discipline outlined in this literature review. They apply to regulation theory, organization theory, OD practices, corporate theory etc. A dialogic corporation is an ongoing conversation. Corporate culture is the shared beliefs and assumptions, created, maintained, and changed through conversations that lead to patterned interactions among the individuals within the corporation. Stacey sees the corporation as the patterns of interactions between people in the “here and now”.399 These patterns are co-created as people interact. Those interactions can be either consensual or conflictive.400 Change inside the 394 Ibid at 10. 395 Ibid at 12. 396 Ibid. 397 Stacey (2006) supra note 207 at 33. 398 See Thomas Kuhn, The Structure of Scientific Revolutions (Chicago, IL: Chicago University Press, 1962). 399 Ralph Stacey & Douglas Griffin (eds.), A Complexity Perspective on Researching Organizations: Taking Experience Seriously (New York: Routledge. 2005). 400 Ibid at 3. 109 organization emerges from personal and group development. In this view, there is no separate “organizational learning system” and the organization itself is not an anthropomorphized entity. Each person interacts with the others and it is the “net effect” of all of these interactions that creates the organization. It does not exist separate from those human agents; it is simply an understanding that they hold in their heads. In this way, organizations are self-organizing and emergent. “Self-organizing” means that human agents on the local level interact with each other and create their own rules of interaction. “Emergent” means that the higher-level system or global system that emerges is a property of those patterns of interactions among the human agents but is more than the sum of those interactions.401 No one is designing the overall system or controlling the evolving patterns of society – they simply emerge as spontaneous choices of individuals and the amplification of small differences in interaction between one present and another.402 This is why system level interventions like TQM, Lean, and Six Sigma often do not work or have unintended consequences as they are internalized by each individual into their patterns of interaction. While no one has yet formulated a dialogic theory of society or the corporation, the possibility now exists and it is just a matter of time. For the purposes of this thesis, prior existence of such a theory is not required; the fact that it is a possibility is enough to support the arguments being made. A dialogic systems perspective theory of the corporation answers the two questions of corporate law in the following way: 1) the corporation is the patterns of dialogue between the participants, and 2) the purpose of corporate law is to assist and promote generative 401 Ibid. 402 Ibid at 19. 110 dialogues within corporations to achieve the normative principles and goals set for corporations by society. A more developed dialogic approach to regulation is offered later in this thesis. 3.6 Conclusion In this chapter, I explored the question: How do organizations learn? I answered this question by focusing on the literature and practice in OD and tracing its understanding of organizational change up to the current emergence of Dialogic OD in the field. Dialogic OD is based on insights from post-modern language theory and complex non-linear dynamic systems. At the heart of Dialogic OD is that organizational change occurs when everyday conversations change. There are many Dialogic OD practices that have a history of success in causing this kind of change. These change efforts focus on individuals and the way they relate to other individuals in the organization. Therefore, the next question that is important to Dialogic Regulation is: How do individuals change their behaviour in organizational contexts? This question will be addressed in the next chapter. 111 Chapter 4: Individual Behaviour: The Importance of Communication “[O]ne of the most important lessons of trust is that cooperation is not always best promoted by promising rewards and threatening punishments. To the contrary, attempts to employ external incentives can often reduce levels of trust and trustworthiness within the firm by eroding corporate participants’ internal motivations.” - Margaret Blair and Lynn Stout (2001) This chapter will focus on exploring how individuals act in organizational settings, specifically within corporations. In corporate law and regulation, the traditional approach to understanding how individuals behave in corporations is dominated by an economic model of rational self-interest. In that model, the way to change the behaviour of a corporate actor is to change his or her external incentives through rewards or punishments. This model is focused on behaviours and external circumstances. In contrast, the dialogic understanding of the individual introduced in the last chapter would argue that behaviour in a corporate setting is significantly influenced by the individual’s world-view as it is created on an ongoing basis by his or her interactions with other individuals in the corporation.403 This model is focused on mindsets and internal circumstances. This chapter will explore the differences between these approaches by looking at the literature and results from social dilemma experiments to understand which model is more effective at explaining how individuals act in organizations and what hypotheses we can generate from those experimental results that could be useful to a dialogic approach to corporate law and regulation. In section 4.1, I place the social dilemma experiments in context by reviewing a 403 John Inman & Tracy Thompson, “Using Dialogue Then Deliberation to Transform a Warring Leadership Team” (2013) 45(1) OD Practitioner 35. 112 pivotal article in the legal literature that explained the usefulness of social dilemma experiments to corporate law and regulation. Section 4.2 provides an introduction to social dilemma experiments. Sections 4.3 and 4.4 describe the current state of the empirical evidence from social dilemma experiments and from “crowding out” experiments exploring what crowds out internal motivation and who this is most likely to happen to. In section 4.5, the weaknesses in the experimental results are discussed using a social constructionist lens. Finally, in section 4.6 I develop a set of hypotheses from the information provided in this chapter that predict how individuals might respond to regulation or rule changes in regulatory settings and in particular where dialogic techniques are used to introduce the rule changes. Those hypotheses will then be tested in the experiment described in the next chapter. 4.1 Social Dilemma Experiments and Corporate Law and Regulation In 2001, Margaret Blair and Lynn Stout published an important article titled “Trust, Trustworthiness and the Behavioral Foundations of Corporate Law”, in which they argued that the empirical evidence from social dilemma experiments called into question the dominant law and economics approach to corporate law and regulation.404 In the article, they diverged from the traditional economic analysis of corporate law and its assumption that people always act in their own self-interest and hypothesized that corporate participants cooperate with each other to a much greater degree than can possibly be explained by legal or market incentives.405 They argued that this cooperation was caused not by external constraints but by the internal constraints 404 Margaret Blair & Lynn Stout, “Trust, Trustworthiness, and the Behavioral Foundations of Corporate Law” (2001) 149(6) University of Pennsylvania Law Review 1735. 405 Ibid at 1738. 113 of trust and trustworthiness.406 To support their argument they cited the then current set of empirical evidence from social dilemma experiments that showed people trust other people a lot more than the rational choice economic model predicts and that they do not trust randomly. Social context was presented as the most important factor in creating or destroying trust. 407 They argued that by manipulating social context experimenters could “produce everything from nearly universal trust to an almost complete absence of trust among subjects in social dilemmas.”408 Blair and Stout argued that these behavioural findings had importance for social institutions where cooperation among participants is necessary, like the business corporation. They argued that we need to understand these findings to create effective corporate law and regulation and they also voiced a strong warning that failing to heed the results of the social dilemma experiments might lead to the erosion of internal trust and trustworthiness in corporate participants.409 Those were sage words in 2001. It is now 2014, and there is little need to enumerate the numerous corporate governance scandals that have occurred in the decade since those words were written. We are all too familiar with them.410 To make matters worse, trust seems to be at the heart of most of these scandals, so much so, in fact, that at least one prominent corporate law and regulation scholar, Tamara Frankel, is lamenting the loss of trust in North 406 Ibid. 407 Their primary source of empirical evidence was a meta-analysis of social dilemma experiments conducted by David Sally in 1995. See David Sally, “Conversation and Cooperation in Social Dilemmas: A Meta-Analysis of Experiments from 1958 to 1992” (1995) 7 Rationality and Society 58. 408 Ibid at 1738-1739. 409 Ibid at 1739. 410 The list includes the dot.com collapse, the auditing scandals (Arthur Anderson, WorldCom etc.), insider trading scandals (Martha Stewart), the Sub-Prime Mortgage meltdown, the Global Financial Crisis, and the BP Deepwater Horizon Oil Spill. See Chapter 1. 114 American business and corporations411 and prominent business culture expert Geert Hofstede has commented that U.S. business leaders are “greedy, short-term gain oriented, and out for power.”412 Did we, as Blair and Stout warned, destroy trust through the way we structured our laws and/or the way we regulated corporations? It is difficult to say – but the empirical evidence from ongoing social dilemma experiments suggests that we might have. The evidence from those experiments is now overwhelming and shows exactly what Blair and Stout argued in 2001: people cooperate more often than economic models predict, and the use of external incentives (rewards and punishments) can erode or “crowd out” internal incentives to cooperate (including trust). Furthermore, we now know which factors are most likely to affect cooperation rates: not just social context but also culture, age, instructions from authority, and, most importantly, communication. 4.2 Introduction to Social Dilemma Experiments Social dilemma experiments are experimental games that are used to predict social norms and social preferences.413 They are termed “social dilemmas” because they encompass “societal problems that arise because we often assign higher priority to our own short-term interests than to the interests of others or other longer term considerations.”414 Examples of social dilemma 411 For example, see Tamar Frankel, Trust and Honesty: America’s Business Culture at a Crossroad (New York: Oxford University Press, 2006). 412 Geert Hofstede, “American Culture and the 2008 Financial Crisis” (2009) 21(4) European Business Review 307. 413 Or how people rank different allocations of benefits (pay-offs) or bundles of benefits (for example food, money, time, prestige etc.). See Camerer & Fehr (2004) infra note 432 at 2. 414 Wim Liebrand, David Messick & Henk Wilke, Social Dilemmas: Theoretical Issues and Research Findings (Tarrytown, New York: Pergamon Press, 1992) at vii. 115 experimental games include the Prisoner’s Dilemma Game, the Public Goods Game, the Ultimatum Game, the Dictator Game, the Trust or Investment Game, the Gift Exchange Game, and the Third Party Punishment Game.415 The prisoner’s dilemma is the classic example of a social dilemma experimental game. In this experiment, two suspects are arrested by police. They are separated and not allowed to talk to each other. The police have insufficient evidence to convict either prisoner so they offer each prisoner the same deal. If the prisoner agrees to testify against his or her partner (defects) and their partner stays silent (cooperates), the defector goes free and the silent partner gets a ten-year jail sentence. If they both defect, they each get a five-year jail sentence. If they both stay silent, they each get one year in jail for a minor charge. Each player must decide what to do. They are assured that the other suspect will not know their decision until the end of the investigation. The prisoners’ dilemma is represented in Figure 1: Figure 1 The Prisoner’s Dilemma 415 For a great summary of game theory and the specific games see Colin Camerer & Ernst Fehr, “Measuring Social Norms and Preferences Using Experimental Games: A Guide for Social Scientists” in Joseph Henrich, Robert Boyd, Samuel Bowles, Colin Camerer, Ernst Fehr & Herbert Gintis, Foundations of Human Sociality: Economic Experiments and Ethnographic Evidence from 1Fifteen Small-Scale Societies (Oxford, U.K.: Oxford University Press, 2004) at 55. 116 In the prisoner’s dilemma game, economic or rational choice theory predicts that each suspect’s only concern is maximizing their own utility (pay-off) without concern for the other player’s utility. Therefore, it predicts that the equilibrium for this game is that both players defect on each other even though they would both be better off if they remain silent. However, the experimental evidence does not support the theory. In one-shot prisoner’s dilemma games people cooperate about half of the time.416 The other social dilemma experiments, including the Trust Game and the Ultimatum Game, are based on roughly the same type of experimental game structure. They are controlled experiments of simplified scenarios where decision-making can be observed and the variables that might affect decision-making can be varied. Usually they are matched to well-defined mathematical models. A game consists of a set of players, a set of decisions (strategies) available to players, and a set of expected outcomes (pay-offs) for adopting each combination of strategies. Pay-offs are usually provided in financial incentives and players’ decision-making (strategies) is manipulated by changing the pay-offs in the game. A classic example of a prisoner’s dilemma experiment was run by Robert Axelrod in the 1980s and outlined in his book The Evolution of Co-Operation.417 Axelrod ran a computer tournament of players using the prisoner’s dilemma game to study how cooperation evolved in repeated games. He asked other game theorists to provide strategies for the computer players to use in the tournament. The winning strategy was a simple strategy called “tit-for-tat” where the 416 See Sally (1995) supra note 423. In fact, the mean cooperation rate is about 47%. 417 See Robert Axelrod, The Evolution of Cooperation (New York: Basic Books, 1984). Axelrod realized that this analysis was too simple and so spent a number of years trying to add complexity to his model to determine whether the results would hold in more complex situations. For a summary of his further work see Robert Axelrod, The Complexity of Cooperation: Agent Based Models of Competition and Collaboration (Princeton: Princeton University Press, 1997). 117 player cooperates on the first move and then subsequently echoes the behaviour of the other player on the previous round. This experiment showed that game theory predictions are also not true in iterated prisoner’s dilemma games. Axelrod’s experiment had the structure of a classic game theory experiment. There were only two players, the situation was a zero-sum game (if one person gains the other loses the same amount), the players were isolated, and there was no communication allowed (in fact there were no real players). This structure is obviously not realistic because real life is much more complex than that. There are often more than two people, communication is almost always a part of the scenario, there are non-zero-sum games etc. Cognizant of these facts, experimental game experimenters began running more complex experiments to see what happens in more complex scenarios. The results of those experiments are captured in the meta-analyses discussed in the next part. 4.3 Update on Empirical Evidence from Social Dilemma Experiments (to 2010) When Blair and Stout wrote their article in 2001, they relied on a meta-analysis of social dilemma experiments from 1958-1992 that was written by David Sally in 1995.418 Since then, there have been two more important meta-analyses of social dilemma experiments published: one by behavioural economist Colin Camerer in 2003419 and a recent meta-analysis on the effect of communication in social dilemmas by psychology researcher Daniel Balliet in 2010.420 The 418 Sally (1995) supra note 423. 419 See Camerer (2003) infra note 462 at Ch2 - “Dictator, Ultimatum and Trust Games” (43-117). This chapter summarizes the results of 31 ultimatum games, 11 Dictator games, and a sample of Trust Games. 420 This article reviewed prisoner’s dilemmas, public goods games, and resource dilemmas. See Baillet, Daniel, “Communication and Cooperation in Social Dilemmas: A Meta-Analytic Review” (2010) 54 Journal of Conflict Resolution 39. 118 results of each of these meta-analyses will be presented below, together with an accompanying analysis of its findings for corporate law and regulation. Meta-analysis is the best way to test the social dilemma literature because there are so many studies and the outcomes vary considerably depending on the way the studies are structured and conducted. The findings of the social dilemma experiments are overwhelming: people cooperate a lot more than the rational choice model predicts and a small set of factors significantly affect cooperation rates including, but not limited to, social context. 4.3.1 The 1995 Sally Meta-Analysis In 1995, David Sally did a meta-analysis of 25 years of prisoner’s dilemma and social dilemma experiments to determine whether the evidence from the experiments matched the economic theory approach to behaviour.421 This is the meta-analysis that was relied upon by Margaret Blair and Lynn Stout in their 2001 article. In this meta-analysis Sally analyzed 37 studies and found that the results were usually inconsistent with a model of pure self-interest. Instead of the prediction of pure self-interest, the mean cooperation rate in the studies was 47.4%.422 In the individual experiments, a number of variables were shown to affect cooperative behaviour: the instructions given, the presence of an authority figure, the normative significance of the language used, repetition, the pay-off matrix, anonymity vs. face-to-face interaction, group identity, and communication. Sally tested all of these variables in a meta-analysis to see which 421 See Sally (1995) supra note 423. 422 Cooperation was defined as the “percentage of total choices made in an experiment that benefit the overall group at the expense of the individual deciding”. The standard deviation was 23.7%. See Ibid at 62. 119 were significant when the results of all the studies were taken into account.423 His regression analyses did “not support the view that thousands of subjects in tens of experiments over three decades were motivated solely by self-interest in their own individual pay-offs.”424 In other words, the variables that rational choice theory would predict as significant were not or were only weakly correlated.425 The most important variables that affected behaviour were the instructions given to the participants and whether the participants were allowed to communicate with each other.426 Where participants were instructed to cooperate with each other, cooperation went up by 36%.427 Similarly, where participants were allowed to communicate with each other through discussion between rounds, cooperation went up by 40%.428 One hypothesis that can be made regarding these results is that language (instructions) and communication may be more important in regulating corporate behaviour than the incentives and punishment predicted by economic theory.429 Blair and Stout saw this potential in Sally’s results. They argued that the experimental evidence called into question the dominant law and 423 The list of possible variables that Sally tested included: subject characteristics, instructions, repetition, payoff matrix, anonymity, group identity, and communication. 424 Sally (1995) supra note 423 at 75. 425 The variables that promoted self-interest were that temptation to defect was great, there was no money at stake (pay-off), or the group size was large. See Ibid at 86. 426 Ibid. 427 Ibid. 428 Ibid at 78. 429 Sally’s results were consistent with the results of other summaries of experimental games published at around the same time. For examples of other summaries see John Ledyard, “Public Goods Experiments” in Kegel & Roth (eds.), Handbook of Experimental Economics (Princeton: Princeton, 1995); Andrew Coleman, Game Theory and Its Applications in the Social and Biological Sciences (Oxford: Butterworth-Heinemen, 1995); Douglas Davis & Charles Holt, Experimental Economics (Princeton: Princeton University Press, 1993); and Martin A. Nowak, et al., “Fairness Versus Reason in the Ultimatum Game” (2000) 289 Science 1773. 120 economics approach to corporate law and corporate regulation.430 That approach to corporate law, which has at its centre the “nexus of contracts” theory of the firm, assumes that each corporate actor wants to maximize their own individual gains and will do so unless legal rules and/or market incentives and punishments keep their behaviour in line. Blair and Stout argued that the results of the social dilemma experiments show that corporate participants often engage in cooperative behaviour in the absence of law, market incentives, or punishment,431 and that they do so because of internal incentives.432 Blair and Stout argued that Sally’s findings had five important conclusions for corporate law: 1. Cooperative behaviour is an empirical reality. Individuals in social dilemma experiments do not always act in their own self-interest and exhibit far more cooperative behaviour than can be explained by economic theory. 2. Different individuals vary in their willingness to cooperate (or in their ability to trust and to be trustworthy) in new situations. 3. To some degree these individual differences reflect past experiences, suggesting that trust may be a learned behaviour. This is referred to as a predisposition to cooperate. 430 Blair & Stout (2001) supra note 420. 431 Ibid at 1745. 432 Blair and Stout actually used the language of “internal constraints” to describe the phenomenon of trust and trustworthiness. See Ibid at 1737 where they state: “corporate participants often cooperate with each other not because of external constraints but because of internal ones”. The use of this kind of language to describe trust and trustworthiness exposes a lot about the authors’ assumptions. Their reference to trust and trustworthiness as a “constraint” is interesting and shows that they are still working under the assumption that trust acts as a deterrent to the natural state of self-interested behaviour. This subtle use of language may be damaging in and of itself, see the discussion on performativity later in this Chapter starting on page 138. Similarly, their reference to the constraints as internal is in direct opposition to the economic view of external constraints, which are simply incentives and punishments. From a dialogic point of view trust and trustworthiness are not simply internal phenomena but social phenomena that depend on a lot of external effects like group membership, culture, education, training, reciprocity, dialogue, etc. 121 4. Trust is a socially contingent behaviour – it depends significantly on individuals’ perceptions of others’ expectations. 5. Economic payoffs are not irrelevant. When the personal costs of cooperation become too high, people stop cooperating.433 This last statement needs to be restated a little because it is not exactly what the experimental evidence showed, nor what Blair and Stout were referring to. What they were referring to was the opposite finding that when the economic incentives of non-cooperative behaviour are increased to a very high level, people are tempted to engage in non-cooperative behaviour.434 The simplest way to summarize the finding is that in complex social situations people tend to engage in cooperative behaviour unless there is a sufficient economic benefit to act in their own self-interest. The experimental game literature shows that cooperation blossoms when it finds the right social circumstances. What is the role of law in creating those circumstances? Blair and Stout said it well: Relaxing the assumption that people are always self-interested in favor of the more realistic claim that people have a capacity for socially contingent, other-regarding behavior opens new channels for analyzing a wide variety of relationships in which the law seeks to encourage cooperation and discourage opportunism. These include not only relationships within families and among citizens in the broader community but also business relationships like partnerships, relational contracts, and (our focus here) incorporated firms.435 433 This list is provided in two different places in the article. See Ibid (2001) at 1742 and 1761. 434 They are actually referring to the opposite. See Ibid (2001) at 1774 where they state: “as the personal cost associated with cooperating rises (that is, as players’ expected gains from defection increase), cooperation rates begin to decline.” [emphasis added]. 435 Ibid at 1808. 122 The other really important idea from Blair and Stout’s article is that rewards and punishment only work if the corporate situation is transparent436 but trust and trustworthiness (or cooperation) can work even when the situation is opaque.437 One of the examples they provide to illustrate this point is when a manager refrains from stealing: is it because they are afraid that they will be caught and fired or jailed (punishment) or because they are trustworthy (learned behaviour)?438 In the first situation posited by Blair and Stout, the manager refrains because of the fear of punishment, which requires that their behaviour be transparent so that there is the potential that they will be caught. If it becomes clear that no one is vigilantly watching it is possible that the behaviour will occur. In the second situation, the manager has learned not to steal and does not steal because the manager’s worldview no longer has a place for “stealing”. This creates the opportunity to conceive of a corporate law and regulation approach that is focused on language and communication and has as its goal the learning of regulatory outcomes by corporate participants. Blair and Stout’s most troubling conclusion was that they felt law was of limited importance in promoting cooperation within firms.439 This conclusion was probably the result of them approaching the issue from a law and economics perspective. It is true that the law and economics approach to corporate law is of limited importance in promoting cooperation within a 436 Ibid at 1740: “But markets and law work best when the situation is transparent and opportunistic behavior can be detected and punished. Trust can work even when the situation is opaque. As a result, business firms that cultivate and support trust can enjoy a competitive advantage over those that do not.” 437 This begs the question of the importance of transparency as the key feature of securities and corporate law as we move away from law and economics and towards dialogic regulation. 438 Blair & Stout (2001) supra note 420 at 1741. 439 See Ibid at 1744, where they state: “Finally, we consider how trust highlights the potentially limited importance of law in promoting cooperation in firms.” 123 firm. But, there are other approaches to law that could play a large and prominent role, like the “learning approach” mentioned in Chapter 2. Another example is Lynn Stout’s behavioural approach to corporate law and how it handles the issue of how best to motivate corporate directors to serve the best interests of the corporation.440 She concluded that the outcomes of the social dilemma experiments had important implications for how we select, educate, regulate, and compensate corporate directors. Her recommendations for dealing with the motivation of directors included creating the right social context for cooperation, choosing directors with a predisposition to cooperate, and removing the incentives that create self-interested behaviour (for example, stock options). All of these suggestions are consistent with a “learning” approach to corporate law and regulation rather than an economic approach. At this point in the social dilemma literature there was still one experimental finding that was counter-intuitive: in repeated prisoner’s dilemma games cooperation rates declined until they hit almost zero after about ten repetitions of the game. This result made no sense, because if it were true, then how did organizations and human society evolve? In 2001, Fehr and Gachter wrote an article offering a solution. They argued that cooperation does not decline in repeated games when there is an opportunity for punishment. They ran an experiment and found that when participants are allowed to set aside some of their money to punish cheaters cooperation actually starts to increase in repeated games. The idea of punishment as a solution to the declining cooperation is the result of the assumptions built into game theory and its economic models. For example, Hobbes said, “Covenants without the Sword, are but Words, and of no 440 Lynn Stout, “On the Proper Motives of Corporate Directors (Or Why You Don’t Want Homo Economicus to Join Your Board” (2003) 28 Delaware Journal of Corporate Law 3. 124 strength to secure a man at all.”441 This finding was troubling, however. Is it really the case that we only cooperate with each other because of the fear of punishment or is it possible that something else is going on? The answer to that question did not come until 2010. 4.3.2 The 2003 Summary of Social Dilemmas by Colin Camerer In 2003, behavioural game theorist Colin Camerer provided a summary of a broader range of social dilemma experiments and a broader range of potentially significant variables. His findings mirrored the findings of Sally but he also identified some additional emerging patterns, including the fact that culture and age also affect cooperation rates. He had similar findings to Sally in that he found that people cooperate in one-shot PD games about half the time and contribute about half their endowments in public goods games.442 These cooperative results in about 50% of the cases were mirrored through the other experiments including Dictator Games,443 Trust Games,444 and Ultimatum Games.445 He also found that pre-play communication was the variable that raised cooperation the most, and non-cooperation could be generated by significantly increasing the pay-offs for non-cooperation.446 441 Thomas Hobbes, Leviathan (New York: C.B. Macpherson, 1968). 442 Colin Camerer, Behavioral Game Theory: Experiments on Strategic Interaction (Princeton, N.J.: Princeton University Press, 2003) at 46. 443 In Dictator Games mean offers range between 15 and 50%. Ibid at 57-58. 444 In Trust games, players risk about half their investment and earn essentially nothing for their investment (they get back what they invested). Ibid at 467. 445 In Ultimatum games, people usually offer 30-50% of their money. The mode and median offers are 40-50 percent and the mean offers are 30-40 percent. There are hardly any offers that are really low (0-10%) or very fair (51-100%). Offers below 20% are rejected about half of the time. See Ibid at 56. 446 Ibid at 46. 125 He also found results similar to those proposed by Fehr and Gachter in that when games are repeated cooperation and contribution decline until they reach almost zero unless there is a way to police the non-cooperators (in economic game theory language it is called punishment), in which case it rises over time to about 60%.447 Camerer found three variables that had significant effects on the behaviour in games: culture448 (significant differences – countries have different sharing norms, e.g. Roth et al. and Henrich), instructions (mild effect),449 and age (kids under 7 are more self-interested).450 Camerer explored a number of other variables that ended up having no effect, an inconclusive effect, or very small effect, including repetition,451 stakes,452 anonymity,453 gender,454 race,455 and academic major.456 Camerer’s finding that culture affects cooperation rates is consistent with the findings of earlier studies that argued that social context has the largest effect on cooperation rates. His finding that the instructions given affect cooperation rates is also consistent with previous studies that showed that instructions and language had an effect on cooperation rates. The interesting 447 See Ernst Fehr & Simon Gachter, “Cooperation and Punishment in Public Goods Experiments” (2000) 90(4) American Economic Review 980 at 989. 448 Camerer (2003) supra note 462 at 68. 449 Ibid 450 Ibid at 65. 451 Which makes no difference unless the simulation is seeded with computer self-interested players. Ibid at 59. 452 Very large changes in pay-off stakes have a modest effect. See Ibid at 63. 453 Stakes are sometimes lower dictator allocations but not ultimatums. See Ibid at 63. 454 Gender had no effect; it seems to interact with other variables. See Ibid at 64. 455 Race was inconclusive and may interact with other variables e.g. culture, see Ibid at 65. 456 Inconclusive – results go both ways. See Ibid. 126 finding in Camerer’s results is that of age. The fact that children and adults have significantly different cooperation rates may be evidence that learning is going on. Camerer found that children start out self-interested and act that way until about age 7.457 By adulthood people in North America cooperate about half the time and contribute about half in public goods games.458 There is some evidence that educational background may make a difference to the way people act because some studies show that individuals with economics degrees act more in their own self-interest.459 All this evidence points to a learning approach to corporate law and regulation being possible. People can learn to be trusting, trustworthy, and cooperative. 4.3.3 The 2010 Meta-Analysis by Daniel Balliet In 2010, Daniel Balliet wrote an important article that might have the potential to solve the issue of ongoing cooperation in social dilemma experiments. He discovered that cooperation rates in repeated games with large groups did not decline where communication was allowed, and punishment did not need to be involved in order for this to happen. This solves the riddle of why society and organizations evolved cooperative patterns without punishment. So, Fehr and Gachter may not be correct – simple communication between participants might preserve cooperation rates. Balliet’s article was a meta-analysis of 45 studies on communication in social dilemmas.460 As was shown above, it is well known that communication increases cooperation in 457 Camerer (2003) supra note 462. 458 Ibid. 459 See the discussion on page 139 of this Chapter. 460 Baillet (2010) supra note 437. 127 social dilemmas. Balliet took that analysis a step further to look at social dilemma experiments461 and checked the effect of the type of communication (face-to-face, oral, or text), the timing of the communication (before or during the game), and the size of the group. He found a large positive effect between communication and cooperation. The effect is moderated by the type of communication, face-to-face discussion having a stronger effect than e-mail or written messages.462 He hypothesized that face-to-face communication was more effective for three reasons: it is more dynamic and fluid than electronic communication and allows the participants to more accurately address the issues that come up in social dilemmas; it involves the ability to see each other and give and receive social cues; and it allows a space for social norms and promise keeping.463 Baillet found that the positive effect of communication on cooperation was stronger in larger as opposed to smaller groups.464 However, the timing of the communication did not seem to matter. For example, ongoing communication did not matter as long as people had a chance to communicate before the game began.465 461 Which included PD, public goods, and resource goods. 462 Baillet references the following studies: Nathan Bos, Darren George, Judith Olson, & Gary Olson, “Being There Versus Seeing There: Trust Via Video” Unpublished manuscript. University of Michigan, Ann Arbor; Norman Frolich & Joe Oppenheimer, “Some Consequences of e-mails. Face-to-face communication in experiment” (1998) 35 Journal of Economics Behaviour and Organization 389; Azi Lev-On, Alex Chavez,& Christina Bicchieri, “Group and Dyadic Communication in Trust Games” Unpublished Manuscript. University of Pennsylvania. Philadelphia PA; Rocco, Elena, “Trust Breaks Down in Electronic Contexts But Can be Repaired by Some Initial Face to Face Contact” Unpublished Manuscript. University of Michigan. Ann Arbor, MI; Rocco, Elena, & Massimo Warglien, “Computer Mediated Communication and the emergence of “Electronic Opportunism” (1996) Unpublished Manuscript. University of Trento, Trento Italy. 463 Baillet (2010) supra note 437 at 48. 464 Note: the largest group size in the meta-analysis was nine people. See Ibid at 52 for a reference to this limitation. 465 Ibid at 48. The studies that made this finding were: Jeanette Brosig, Axel Ockenfels & Joachim Weimann, “The Effect of Communication Media on Cooperation” (2001) 4 German Economic Review 217 and Robert Radlow & Marianna Weidner, “Unenforced Commitments in ‘Cooperative’ and ‘Non-Cooperative’ Non-Constant Sum Games” (1966) 10 Journal of Conflict Resolution 497. 128 The fact that ongoing communication does not significantly increase cooperation is puzzling. One hypothesis is that initial communication creates a norm of reciprocity and that ongoing communication will only increase that if the communication is of the correct type. To use the language of OD theorist Gervase Bushe, if the ongoing communication is not “clear” then it may actually contribute to non-cooperation or the falling apart of the cooperation.466 However, if the communication is “clear” then we should see a significant benefit to cooperation from ongoing communication. There is evidence from the experiments in support of this hypothesis: where there are a greater number of commitments or promises from group members, then the individuals in that group are more likely to cooperate467; and where the communication involves sincere signalling of cooperative intentions that are followed by cooperative behaviour then ongoing communication has a positive effect on preserving cooperation rates.468 Again, these experimental results point to the fact that the traditional economic approach to corporate law and regulation centred on external incentives like punishment is not supported by the empirical evidence and that a different approach needs to be developed to accommodate these results. That approach could leverage participants’ internal motivators by utilizing language and communication and assisting corporate actors and corporate organizations to engage in learning. 466 Bushe refers to this as “interpersonal mush”. See Bushe (2009) supra note 176 at Chapter 1. 467 John Orbell, Alphone van de Kragt & Robyn Dawes, “Explaining Discussion-Induced Cooperation” (1988) 54 Journal of Personality and Social Psychology 811. 468 See Lubell, Mark, Vicken Hillis, William Baum, Richard McElreath & Peter Richerson, “Group Size and Sincere Communication in Experimental Social Dilemmas” (2008) Unpublished Manuscript. University of California Davis. Davis, CA [on file with author]. In this scenario, cooperation does decline but it does not decline to zero. 129 4.4 Update on Empirical Evidence of “Crowding Out” (to 2010) In response to Sally’s 1995 meta-analysis, Blair and Stout proposed that there were “dangers in the contractarian approach by suggesting that an excessive emphasis on external sanctions . . . may not only be ineffective but counter-productive, serving to undermine trust and trustworthiness in the firm.”469 Trust, they stated, “is not always best promoted by promising rewards and threatening punishments. To the contrary, attempts to employ external incentives can often reduce levels of trust and trustworthiness within the firm by eroding corporate participants’ internal motivations.”470 There have been five significant sets of experiments that show conclusively that internal incentives do get “crowded out” by the use of external incentives in certain circumstances. First, experimental economist Sam Bowles found in a series of experiments that external incentives can actually “crowd out” internal incentives and prosocial behaviour.471 After reviewing 41 behavioural experiments Bowles concluded that “incentives that appeal to self-interest may fail when they undermine the moral values that lead people to act altruistically or in other public spirited ways.”472 He suggests that economic incentives may be counterproductive when they 469 Blair & Stout (2001) supra note 420 at 1736. 470 Ibid at 1739. 471 See Samuel Bowles, “When Economic Incentives Backfire” (March 2009) Harvard Business Review 22; Samuel Bowles, “Policies Designed for Self-Interested Citizens May Undermine ‘The Moral Sentiments’:’ Evidence from Experiments” (2008) 320 (5883) Science 1605; Samuel Bowles & Sung-Ha Hwang, “Social Preferences and Public Economics: Mechanism Design When Social Preferences Depend on Incentives” (2008) 92 Journal of Public Economics 1820; Robert Boyd, Herbert Gintis
UBC Theses and Dissertations
Dialogic regulation : the talking cure for corporations Cody, Michael 2014
Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.
- 24-ubc_2014_september_cody_michael.pdf [ 8.82MB ]
- JSON: 24-1.0077782.json
- JSON-LD: 24-1.0077782-ld.json
- RDF/XML (Pretty): 24-1.0077782-rdf.xml
- RDF/JSON: 24-1.0077782-rdf.json
- Turtle: 24-1.0077782-turtle.txt
- N-Triples: 24-1.0077782-rdf-ntriples.txt
- Original Record: 24-1.0077782-source.json
- Full Text