The difficulty is that, in the complex world of human experience, there are many possible patterns to be found and many ways of helping the client make meaning of their experience. Precisely which patterns we detect as coaches will depend on the theories, models, and assumptions that we bring to bear on the client’s story. Our theories (like our clients’ theories) are the lenses we use to filter out what is important and what is not. They allow us to understand our world, and they guide both prediction and action. The best coaches are able to take multiple perspectives and select those that are most helpful for the client.

This chapter explores one way of finding patterns in coaching, one set of lenses through which to view the coaching engagement—complex systems theory. In order to do this, I first describe the main features of complex adaptive systems, exploring some of the implications for coaching as we go along. Given that the cases we are discussing are situated in an organizational leadership context, I also consider some ways in which complexity theory informs our approach to leadership. I then explore Stacey’s (2000) metaphor of systems as conversations, and present a model of the coaching engagement that I have found useful in helping me understand what it is I am doing in coaching. Finally, we look at the cases of Bob and Bonita from this systemic perspective.

General Systems Theory was said to be a unifying theory in that the systemic nature of the world can be seen in both the natural sciences, such as biology, physics, and chemistry, and the social sciences such as psychology and sociology. It is not surprising then that a host of systems theories have been developed via parallel streams of research. It is not possible here to outline the differences and similarities between the multitudes of particular theories. Rather we will focus on the branch of systems theories that deals with complex adaptive systems, or complexity theory.

Relatively simple systems, such as a clock or a car, fit the general description of a system, but are not complex adaptive systems in that they are unable to adapt their functioning. They are made up of parts, rather than systems. For example, if we took even a minor cog out of a clock it would cease to function. The principal differentiating feature of complexity theory is that it focuses on systems in which systems members act locally and in so doing produce order at a global level.

There are a number of important features of complex adaptive systems. Many of these are common to all systemic approaches and are important to take into account in coaching.

In scientific terms this is a relatively new approach. The scientific worldview under Newton utilized a predominantly mechanistic and reductionistic approach. It explored and understood the world by breaking it down into its constituent parts and looking for linear chains of cause and effect governing the behavior of those parts (Haines, 1998). The assumption underlying this approach is that the parts of a system have a fundamental nature that is independent of the system in which it is embedded, and which governs their interactions. Logic and the scientific method are the means by which we attempt to access this reality.

We see this linear approach in organizations when we see the traditional “name, blame, and shame” approach to team failures. In this approach, the locus of dysfunction is seen to be the individual’s choices or even the individual’s personality. Blaming the collapse of Barings Bank on the so-called rogue trader Nick Leeson is an example of the reductionist approach.

Reason’s (1990) “Swiss Cheese” model of human error shows clearly that when things go awry (as clearly they did at Barings) typically a host of factors have been at work in an interdependent way to create the outcome. These factors include organizational culture, supervisory practices, time and task pressure, and individual choices. Rawnsley’s (1995) analysis of the Barings collapse clearly shows that there was much more at work than the deviant actions of one individual. That is not to say that Leeson bears no responsibility for the collapse. On the contrary, his actions show a significant failure of adult responsibility for which Leeson was and is responsible. Taking a systems approach, however, means dealing with the system in all its complexity and tailoring one’s coaching to address issues at every level of the system—including the personal. Leeson’s behavior shows that he clearly had some significant issues to work on—as did the rest of the system!

Systems theories see reductionist approaches as incomplete. Rather than focus on the parts of the system in isolation, systems theories focus much more on the relationships between parts. This approach is based on the belief that the parts of the system are essentially interdependent. In other words, their functioning within the system depends on the nature of the relationships between them and the other parts of the system.

This interdependence can be seen in the way we often operate differently when in different groups. Many of us may have had experiences of being in relationships or groups that draw the best (or the worst) out of us. In such a situation, how we respond is not so much about who we are in isolation as it is about the nature of the complex interactions and relationships that exist between us and the system in which we are operating.

The notion of openness refers to the permeability of systems boundaries, or the ability of the system to effectively communicate with, and be affected by, the wider environment. Openness is a dimensional property that is able to change over time; that is, systems can be more or less open or closed. In human systems feedback and communication processes enable the system to remain open to its wider environment. For example, an organization that is unable to hear its customers may be said to have dangerously closed boundaries. A sales team that is unable to effectively communicate with marketing or supply is in trouble. A manager who is unable to seek or hear feedback without defensiveness is facing potential derailment.

The degree of openness of a system is important because of the process of entropy. Technically, this is the tendency for all things to move to lower levels of energy—the tendency for things to disintegrate or fragment. Openness enables new energy to be brought into the system. For a system to maintain itself through time it must embody an input-transformation-output process (Haines, 1998). This can be seen in the human body. We must continually bring in resources (food, ideas, etc.) that we transform into energy and action. At a physical level, if we don’t eat, we die. At a psychological level, if we are not stimulated by our world or cannot put back into the world our thoughts, ideas, and other expressions of our individuality (for example, if we are not loved or able to love others), then we begin to disintegrate psychologically. All systems embody some form of transformational process. It is what keeps the system alive. If a system’s boundaries become closed, and they are no longer able to input energy and resources or output production, then the system suffers—it begins to lose energy and fragment. All closed systems eventually run down and die.

Openness enables systems to grow and develop beyond simple maintenance. Growth is a process of the internal elaboration of the system in response to the challenges meted out by the environment (Haines, 1998). The old saying “Necessity is the mother of invention” is apropos here. Feedback and challenge from outside the system are sources of perturbation for the system that stimulate change. Growth is the ability of a system to develop more effective structures and processes in response to the inevitable challenges of life.

More recently, complexity theorists have identified that, rather than seeking steady-state functioning, complex adaptive systems are necessarily marked by a sort of sustainable instability or dynamic equilibrium (Haines, 1998; Stacey, 2000). By this is meant a sort of paradoxical bounded instability that is able to produce new forms of behavior and creative action. Both stability and instability are important. Too much instability and the system becomes chaotic and breaks down. Too much stability and the system becomes closed—it is unable to do anything new or to adapt to its environment—and entropy sets in. This place between chaos and stability is called the “edge of chaos.” It is not a place of balance between stability and instability, but a paradoxical state of unresolvable contradictory forces (Stacey, 2000). The tension between these forces elicits creativity and innovation. According to Stacey, the edge of chaos is created where there is sufficient energy and information flow, diversity, and connectivity between system elements (Stacey, 2000). Most human systems typically function best at the edge of chaos.

We can see the edge of chaos operating in most innovative human behavior. We need a certain paradoxical instability of thinking to create something new. The tension created by the juxtaposition of diverse ideas enables us to make connections where they don’t usually exist, and out of this tension new forms emerge. We see this juxtaposition of diverse ideas in Jørn Utzon’s design for the Sydney Opera House. Utzon connected the idea of sailing with a building for performance of the arts. While instability is necessary for creativity, thinking cannot be completely random or chaotic. There must be sufficient discipline or orderedness to enable the selection of meaningful connections. (It is perhaps this paradoxical bounded instability that explains why so many creative geniuses lead such messy lives!)

In coaching, however, the coach is often called upon to afflict the comfortable! We often seek to increase information flow, energy, and diversity to a level that helps the person move out of stable mind-sets and behaviors so as to create new insights, understandings, and actions.

Creativity and the generation of novel behaviors, ideas, and structures illustrate another important feature of all complex adaptive systems—emergence (Lewin, 1993).

In human complex adaptive systems, human behavior can be seen as an emergent property. We do not go through the world emitting robotic behavior in isolation from the pattern of events and relationships in which we are involved. Rather, we have agency. We respond to events, strive for goals, and introduce novel behaviors in attempts to work within our systems. Human systems are able to adapt in novel ways (as are many other complex systems, like a flock of birds, fisheries, forests, and other environmental systems), because they incorporate complex positive and negative (amplifying and balancing) feedback loops and feed-forward loops that impact on the emergence of behavior (Lewin, 1993; Stacey, 2000).

Feedback and feed-forward loops are critical to the operation of complex adaptive systems in that they provide the nonlinear causality that marks such systems (Senge, 1990; Stacey, 2000; Wheatley, 1999). A systemic approach to coaching requires a familiarity with the workings of feedback and feed-forward loops.

Amplifying or positive feedback describes a system dynamic where elements of the system reinforce the behavior of the system such that the behavior grows. Vicious and virtuous cycles are examples of amplifying feedback. For example, the compounding of interest in a bank account is a form of amplifying feedback (and a virtuous cycle for the owner of the bank account!). Anxiety may also create amplifying loops with respect to defensive behavior. The more anxious a management team feels, the more defensive they become (a vicious cycle).

The action of amplifying loops is moderated by negative (balancing or regulatory) loops. These function to reduce system growth, or focus it at a specific level. For example, commercial fishing creates a balancing loop that reduces the growth of fish stocks. Other common examples of balancing loops include price regulating demand and aggressive criticism curtailing open exchange of information.

Human systems also incorporate feed-forward loops. These are the ability to anticipate the future in a way that causes the emergence of behavior in the present. For example, investors may anticipate an interest rate rise and decide to shift their investments from property to the stock market. Like feedback loops, feed-forward loops can be amplifying or regulatory (Stacey, 2000). For example, in team settings, patterns of past responding can lead team members to anticipate either positive or negative reactions when they put forward novel ideas. Anticipating positive reactions to new ideas is likely to increase creativity and innovative behavior. Creativity and innovation will be reduced if the anticipated response is negative.

Much has been written on the process of identifying and mapping different patterns of feedback loops. In the interests of brevity, the reader is referred to several sources for a fuller description (see Argyris, 2004; O’Connor & McDermott 1997; Senge, 1990; Senge, Ross, Smith, Roberts, & Kleiner, 1994).

Feedback is important in coaching at a number of levels. The nature, quality, and timeliness of feedback are critical for good coaching, regardless of whether the coach is operating from a systemic perspective. Here feedback is used in its broadest sense, as information flow that links action and outcome. Feedback functions to help the coachee identify goals and changes, track progress, and adjust behavior. From a systemic perspective, the coach’s role is to help the client become sensitive and responsive to the many forms of feedback produced by the system. For example, at an individual level this might involve building sensitivity to others’ reactions or using 360-degree feedback. At a team level, this might involve helping the client identify where amplifying and regulatory loops are operating in a way that subverts the achievement of the client’s or organization’s goals.

Systemic coaching also uses feed-forward loops when it seeks to explore the potential impacts that may be created by any changes the client makes. Take the case of a manager seeking to increase productivity in her team. There are multiple avenues open to the manager to influence the team—for example, she may institute bonus incentives, or she may introduce a policy of sacking the lowest-performing team members each year (as has been tried in some multinationals). Both of these strategies are likely to have unintended side effects. One side effect of a bonus system might be to focus people’s efforts on achieving metrics at the expensive of genuine improvements in productivity (a pattern not dissimilar to governments reclassifying underemployed workers as employed so as to improve employment figures). Sacking the poorest performers regardless of actual performance is likely to create an atmosphere of hostile competition in which team members seek to secure their own jobs by withholding information and assistance from other team members for fear the others will be too successful.

Both of these are examples of unintended system patterns that function as negative or regulatory loops with respect to productivity. The coach may assist the manager in examining the potential effects of implementing such strategies. It may be that the manager can redesign the strategy so as to minimize negative outcomes. It may be that, once the potential feedforward or feedback loops are discovered, an entirely new strategy is called for. Whatever the outcome, by discussing feedback and feed-forward, the coaching conversation has moved the discussion to a place of tension. That tension is evidence of the edge of chaos, and is the place where creativity can emerge.

Perhaps the easiest way to think of attractors is to imagine a three-dimensional model of a landscape—say a large relief map such as you might find in the visitors’ center of a national park. Now imagine throwing a marble into the model. It will tend to travel downhill, into the valleys until it reaches the lowest local point. That valley or the area bounded by that set of hills is an attractor—when the marble comes into the field of its influence, it attracts particular types of behavior from the marble. Attractors are more or less stable places in the landscape. If you somehow disturb the marble as it is sitting in the valley, it may move around the valley floor a bit, but it will tend to stay somewhere in the valley. Moving the marble to a new valley takes a lot more energy. You need to either change the shape of the mountains or push the marble up and over and into another valley. But remember that attractors can be powerful. You may end up like Sisyphus, putting in enormous effort for little result.

Attractors are like the valleys. They exert forces that shape behavior in systems. In organizations, remuneration and bonus systems act as attractors. People, mental models, rules, assumptions, events, and even objects can elicit more or less stable patterns of behavior over time. Things as simple as the lunchroom, photocopier, or watercooler can act as attractors in the system—they can become places where people meet, talk, and make decisions.

Of course, like our model landscape, attractor valleys differ enormously in shape. Some have steep walls and small floors, and it is easy to identify where the marble will come to rest. Others have shallow walls or wide flat areas that make it very hard to identify exactly where the marble will end up. You know it is likely to be somewhere in the valley, but just where is unpredictable. Still others, called strange attractors, cause behavior to move in one direction at some times and in other directions at other times. Small differences may be all it takes to cause a discontinuous shift in the attractor!

So, from a coaching perspective, being alive to the shape of the system landscape is important. Unfortunately, unlike marbles and model landscapes, real-life attractor landscapes are not always easily identifiable, particularly in advance. However, we often can begin to see the shape of attractors in patterns of behavior they elicit. For example, budget meetings in one organization may be a mad scramble to secure as many resources as possible from the collective pot. In another organization, budget meetings may revolve almost exclusively around cost cutting. In such situations, systemic coaching does not seek to ask who is to blame for undesired behaviors. Rather, it asks, “How useful are each of these behavioral patterns in terms of the client’s goals? What are the assumptions, rules, and feedback loops that elicit these behaviors? What changes could be made to elicit different, more useful behaviors?”

Noticing patterns is made even harder by another feature of complex adaptive systems, namely, sensitivity to initial conditions. Human systems often show huge sensitivity to very fine differences in the initial state of the system—so much so that it is often impossible to even identify precisely what is causing the final emergent behavior of the system.

This sensitivity was first shown by the meteorologist Lorenz (1963) in his work on weather systems. He found that even the most minute changes in the initial condition of a weather system tend to amplify over time to produce major changes in the outcome of the system. This is what is meant by the famous “butterfly flapping its wings in Brazil” analogy.

Some authors suggest that the complexity created by nonlinear feedback, amplification, and sensitivity to initial conditions makes all complex adaptive systems radically unpredictable (Stacey, 2000). To a large degree this is true. One can never predict with precision or certainty how systems will react to the things we do, particularly in the long term. They are too complex. Yet experience shows us that many systems respond in a broadly predictable way.

For example, if I forget a close friend’s birthday, I can predict the friend will be hurt and that tension may arise between us. It is likely that something will need to be done to repair the relationship before it returns to a relatively even keel. What I cannot predict is exactly what this process will look like. It may also be that my forgetting will prove to be the undoing of the friendship because it has come on the heels of several other slights (by me or others). Equally, it is possible that my friend might see that I was under unusual pressure at work, be understanding of my neglect, and let the incident pass without mention.

While precise prediction may be impossible, human behavior in systems is often predictable in a fuzzy way a good deal of the time. The further we move away from the present, the less predictable or more fuzzy the potential outcomes. At the same time, human behavior is typically not utterly random. When it appears so, it is seen as either an indicator of mental illness or a sign that we have missed some important feature of the event landscape. For example, I would normally have a high degree of confidence that my friend will not respond to my failure to remember his birthday by becoming vegetarian, joining the French Foreign Legion, or attacking me with a cauliflower.

Yet, as coaches we are often asked to treat system members as if they are isolated units. This is particularly true in situations involving remedial coaching. It is not unusual for individuals to be referred for coaching in the hope that they will be “fixed” and that their negative impact on team or group performance might be alleviated. This is analogous to something that often happens in family therapy. Parents will bring a child to therapy for behavioral problems, when in fact the child’s behavior is actually a symptom of the distress evident in the wider family system. In recognition of this, family therapists typically say that the child is the “identified patient,” and then proceed to work with the whole family system (Parry, 1996).

One might argue that in executive coaching we are not dealing with helpless and hapless children. We deal with adults, and adults are accountable for their behavior. Furthermore, this accountability is important because an individual’s actions within a system can have a significant impact, for good or ill, on the system as a whole. One has only to think of Sir Bob Geldof and Band Aid, or Nick Leeson and the failure of Barings Bank to see the impact an individual can have. However, as unique as these individuals might appear, treating them as entities isolated from the wider systems in which their actions occurred is problematic. Band Aid’s achievements would have been impossible but for the cooperation of a host of people. Leeson’s activities were enabled by a system that lacked appropriate feedback mechanisms or the will to implement them. Leeson’s fraud was enacted within a culture that placed a premium on success, regardless of how it is achieved. According to a colleague at Barings Bank, “The management trusted him too much. . . . Nick was the star trader at Barings because his trades contributed a huge amount of money. Because of this, a lot of the checks and balances that should have been noted were ignored” (Borthwick, 2003).

The idea that the individual is the only agent contributing to an organizational outcome can be reassuring for the organization. It enables the members to have heroes and scapegoats, and to explain outcomes in ways that do not call for significant personal change. While locating causality in the individual may protect system members from having to face their complicity in the outcome, in the long run it fails to address problematic systemic patterns. Colluding with this avoidance serves to undermine the coaching engagement and weaken both the client and the organization.

An example of unhelpful collusion occurs when coaches act as gobetweens, delivering management feedback to the coachee or advocating on behalf of the coachee. Such behavior has the potential to subvert more mature channels of communication in the organization, weaken relationships, foster dependence, and maintain the very patterns of poor communication the coaching was meant to address.

The failure to acknowledge the wider system can be just as damaging in times of success. A common example of this is the impact on team members when a narcissistic leader fails to properly acknowledge the contributions of the team in times of success. Such behavior weakens team morale and commitment, and reduces the leader’s influence at the very time when success should be enhancing both! If done often enough, the legacy can be a cancerous resentment that team members feel unable to openly express, but which undermines every project undertaken.

That is not to say that coaching individuals to more effectively participate in their organizations is pointless. Just as the team is a complex adaptive system, so, too, is the individual. The individuals’ potential to effectively participate in positive interaction, along with their ability to resist the pull of destructive attractors, is an emergent property of their complex interaction of the parts that make them up. These parts include their mental models, personalities, habitual behaviors, emotions, physiologies, and genetics. We will explore this area more fully later in the chapter.

Coaching from a complex adaptive systems perspective does not draw a dichotomy between the individual and the team. Rather, working with an individual is working with the team/wider organization. Similarly, one cannot work with a team without having an impact at the level of the individual. Nevertheless, it is important for the coach to be aware of, and intentional about, which level of the system is being engaged (individual, team, or organization) at any given time in the coaching process, and to consider, as best one can, the foreseeable impacts on the other levels of the system.

Almost all interventions in coaching are multilayered in that they have impacts on many levels. For example, one might be coaching an executive with a goal of developing a more family friendly work-life balance. A typical behavioral change might be to leave the office at a reasonable hour in the evening (say, 6:00 P.M.). It is important to consider how that change might be perceived. This will be very context dependent. For example, when other staff feel they must stay late because the company is struggling, seeing the boss leaving on time may lead to increased stress and tension in the workplace. In other contexts (such as in some law firms) this behavior may be interpreted as a lack of suitability for advancement or even a thinly veiled criticism of the dominant culture. That does not mean the goal of leaving by 6:00 P.M. must be abandoned. Rather, it indicates that the resultant stress and tension or misinterpretation may need to be addressed so as to keep it within reasonable bounds.

Unfortunately, sensitivity to initial conditions also carries with it a difficulty that is often downplayed or ignored by coaches and consultants working from a linear approach—the problem of long-term unpredictability. We can often present our solutions or develop plans with clients that have an aura of authority about them. We are easily seduced by the apparent certainty of linear thinking. The old image of the corporate leader standing at the helm of the organization, guiding it through the stormy waters of life, is a powerful one. It leads to the expectation, at all levels of management, that a skilled leader should know the right levers to pull in order to enact change, and what will happen when they are used. The most skilled leader is the one who can turn the ship around quickly and with the greatest grace! This is the fundamental assumption of command and control leadership.

The traditional or classical command and control leader, according to Crawford and Brungardt (2000), was well suited to the industrialized world, where work flow depended on stable, predictable structures. While the methods of industry resulted in relatively stable output, they paid scant attention to the human resources they used in the process. However, in today’s world where knowledge and information drive an increasingly service-oriented economy, success requires the adaptive and diverse contribution of all employees. Command and control models of leadership are not well suited to eliciting free-flowing contributions to innovation and knowledge creation (Crawford & Brungardt, 2000).

If we are to take the process of emergence in complex systems seriously, then we must acknowledge that the leader can never stand outside the system and operate on it in such a clinical way. Leaders are fundamentally part of the system. The very notion of leadership can be understood as an emergent property of the co-created relationship between leader and follower (Dimitrov & Lederer, 2003). Even the simple act of observation affects system behavior. We have all experienced how behavior changes in a meeting when the boss enters the room. We take these changes for granted, but they show powerfully the interdependence of systems agents, and demonstrate the principle that systems agents tend to act locally—that is, according to local goals.

This radical unpredictability and interdependence has far-reaching implications for leadership and for coaching (Stacey, 2000). If leaders cannot stand outside the organization in order to direct it in predictable ways, what is their role? If coaches cannot assist leaders to find predictable solutions, what value are they to the organization?

One of the reasons why complexity theory has been so slow to infiltrate leadership theory is that it raises these highly anxiety-provoking questions. It weakens our ability to see the world as predictable and controllable. This is an existentially terrifying position to take. It undermines the very reason for being so many of us have taken for granted in our roles as leaders and change agents.

At another level, though, complexity theory is enormously liberating. If control and predictability are illusions, and stable order is a sign of decay, then no longer do leaders have to struggle under so heavy a burden of responsibility for forces they cannot, as a matter of principle, predict or control. But this does not mean that leaders at all levels of the system have no role. Their role remains crucial to the organization—it is simply no longer one characterized by disengaged diagnosis, command, and control (Crawford & Brungardt, 2000). Rather, as we shall see, complexity theory shows us that the leader’s role is perhaps more usefully conceived in terms of engaging with the system in a way that is characterized by ongoing responsive dialogue.

Attempts have been made to maintain a traditional command and control approach to organizational leadership by suggesting that the role of the leader is to push the organizational systems toward the edge of chaos so as to capitalize on the creativity found there (for example, Brown and Eisenhardt, 1998). Some of the techniques associated with this include setting stretch goals, establishing cross-functional teams, flattening hierarchical management structures, downsizing, and so on. However, such an approach neglects that inherent unpredictability of complexity and also assumes the leader is able to stand outside the system and see where, and in what way to push the system into creativity. There is no guarantee that new order created at the edge of chaos will be a beneficial one for the organization (Stacey, 2000). It also neglects the fact that most human systems typically gravitate toward the edge of chaos, because this is where self-organization occurs.

Leaders (and coaches) cannot stand outside the system in order to manipulate it in predicted ways, because they are part of the system, and their actions are immediately swept up into the web of nonlinear feedback and information flow. Rather than intervening in the functioning of the organization from time to time, the leader needs to be engaged in an ongoing dialogical relationship with the system.

An ongoing responsive engagement within the system as it unfolds over time would seem to be the model of leadership most in keeping with the dynamics of complex adaptive systems. In this way the leader can participate in, rather than seek to control, the self-organization of the system. Four broad areas of activity have been held to be important in sustaining organizations in the creative tension on the edge of chaos. These are: setting vision, creating boundaries, ensuring adequate communication flow, and empowerment (Crawford & Brungardt, 2000; Fraser & Greenhalgh, 2001; Plsek & Wilson, 2001; Stacey, 2000; Wheately, 1999; Youngblood, 1997).

This ongoing responsive engagement is highlighted in the work of Ralph Stacey. Stacey (2000), building on the work of ethnologists such as Goffman and Boden, has developed a metaphor that I believe captures well the dynamic nature of complex systems and sheds light on the roles of both leaders and coaches in complex systems. Stacey (2000) suggests that human systems (and the human person) are constructed by the complex responsive process of conversation.

How does a marriage come about? It starts with a conversation between two people. How does it end? They stop talking (or talk about the wrong things!). The same is true for companies, clubs, and nations.

Communication is creative of organization at both the individual and the wider systems level. We become who we are via the complex process of communication with other human beings. Case studies of people who are raised in isolation show that normal human development is arrested when people are deprived of the opportunity to interact with other people (Pines, 1997). Interacting with our external conversations is the stream of internal dialogue that is an essential part of the creation of meaning. It is a conversation we have with ourselves in which we seek to form intentions and to elaborate ideas, models, explanations, and other narratives so as to resolve the tensions that arise when we are faced with challenges, or to discharge or avoid emotional states that arise within us, or to elicit other desired emotional states. It is out of this ongoing dialogue that our sense of self develops and is maintained (Stacey, 2000).

So what are the system elements in conversation within the person? Figure 11.2 shows that this internal dialogue involves the whole person. All aspects of the self both inform and are informed by each other in an ongoing inner dialogue. Our genetics determine the broad boundaries of the system. Our personalities, habits, and defenses act as attractors for particular types of conversations, behaviors, and conclusions. Our physiology is a powerful player in our internal dialogue. Physiology includes our emotions, our health, and our ever-changing physical states. Our changing goals and values inform our intentions, as do our mental models and the theories and metaphors that guide us. All of these factors interact together, and in that interaction emerges our experience of self, our intentions, and our knowledge.

That knowledge is an emergent property is an important point. We often think of knowledge as something we acquire (or lose in forgetfulness). Yet careful examination of our experience shows us otherwise. The story of Avril illustrates this.

I met Avril on a train going from London to Stanstead Airport on my way to run a communication skills workshop for a group of general practitioners. Avril, a bright, engaging, and intelligent woman in her 40s, was the conductor on the train. As she took my ticket, she asked where I was going, and I told her I was going to teach doctors how to communicate with patients.

“Well!” she said, planting her hands on her hips. “They can certainly use some of that! You know, whenever I go to the doctors, I know exactly what I want to say, but when I get in there he asks me all these questions, and in no time flat I am hopelessly confused and I don’t even know why I am there.”

We can all relate to Avril’s experience of being undone by the conversation. We have probably all been dumbstruck or discombobulated by others at one time or another. This occurs whenever the conversations we are engaged in (both inner and outer) do not allow our knowledge to emerge in consciousness.

Just as our sense of self and our knowledge emerges in a complex mix of inner and outer conversations, so, too, the organizations of which we are a part are defined in the complex conversations that emerge among systems members. Organizational conversations seek to make meaning, form intentions, discharge, elicit, and avoid emotional states via ongoing narrative. In our workplace conversations we seek to make sense of what we see happening. We discuss and form plans of action. We hold conversations that are designed, consciously and unconsciously, to reduce anxiety, discharge frustration, and respond to and exercise power. We do all of this while attempting to maintain a sense of belonging or connectedness to the system. These competing demands hold us on the edge of chaos—the place of emergence and creativity.

Our mental models are key elements in these conversations. They are constructed by and emerge from within the complex responsive dialogue that goes on between people and within the person. They both enable meaningful conversations to be held and constrain the limits of meaning. They help us to select certain information out of the miasma of stimuli that surrounds us, and reject other stimuli as mere noise (Parry, 1996).

When the conversation strays from the meaning enabled by our current mental models, we experience ourselves as moving further toward chaos, and the resultant tension is creative of innovation. If the conversation strays too far from our mental models, then it becomes meaningless and we are thrown into disarray until a radical new understanding can be discovered (or the system disintegrates). This is a paradigm shift. Interestingly, if our conversations become repetitious and add nothing new to our mental models, we experience boredom and restlessness (Parry, 1996). Notice that the problematic movement here is toward stability. The system becomes closed and vulnerable to entropy. Our boredom and restlessness serve to motivate us to seek new and stimulating conversations. It seems we are constantly drawn to that place between sameness and radical newness—the edge of chaos again.

At the simplest level of mental model making, we learn what behaviors are acceptable and not acceptable in both the subtle and covert conversations that go on in response to those behaviors. The system tells us, often in no uncertain terms, that different types of action, gestures, themes, and narratives are acceptable or not. Coming late for a meeting, telling the wrong joke for the situation, making a speech that captures the feeling of the audience, all elicit responses (feedback) from the system members with whom we are in relationship. The latecomer observes others as they glance at their watches; the inappropriate comic receives polite smiles, confused concern, or looks of disapproval. The speaker observes the audience’s attentive gaze, affirmative gestures, and heartfelt applause.

At more complex levels of model making, we elaborate theories and metaphors to explain the world of our experience. This is done, once again, both in conversations with others and in our internal dialogue. In this dialogue we test the theories and metaphors against our experience, seek for congruencies and areas of disjunction. Our mental models and the patterns of our conversations are attractors through which we reproduce themes and narratives (Stacey, 2000). We are often able to see the attractors at play in the recurrent patterns and themes that our conversations reproduce.

According to Stacey (2000), the free flow of conversation enables the emergence of new forms, understandings, and action. When the conversation becomes repetitive, the behavior of the organization becomes stuck—the boundaries of the system become closed and the organization is unable to adapt to changes in the environment. Bureaucracies are typical examples of stereotyped repetitive conversation patterns. They are stultifying for both the bureaucracy members and its customers.

Stacey (2000) identifies three factors associated with the free flow of conversation: connectivity, diversity, and anxiety. By connectivity is meant the richness of the themes and associations within the conversation. If system members are connected by a few loose themes, their communication is apt to be stereotyped. We can often see this in the pattern of communication of old school friends who do not meet regularly. Their communication tends to revolve around reminiscing over their schoolday experiences. Repeated contact tends to allow more connected and richer themes to emerge in the conversation.

Diversity enables the free flow of conversation by supplying novelty, difference, and tension. People from diverse backgrounds and experience hold different perspectives on events. These differences provide the critical levels of misunderstanding and cross-fertilization needed to stimulate novel connections (Stacey, 2000). Too much diversity and the communication falls into chaos. This is consistent with communication theorists such as Shannon, who found that messages had to have the correct mix of redundancy (or repetition) and surprise (or novelty) in order to be effective. Too much of what is already known and the listener will switch off. Too much novelty and the message becomes chaotic (Parry, 1996).

Anxiety is the third critical factor in the free flow of communication. Too little anxiety and the conversation lacks energy. Too much and the free flow of conversation is destroyed. People who are overly anxious will tend to close down communication (destructive stability again), or open a flood of information that leads to chaos. The facilitatory and inhibitory effects of anxiety have been long known in the psychological literature (Druskat & Wolff, 2001; Harris & Scherblom, 1999; Yerkes & Dodson, 1908). Hence, for the free flow of communication there must be what Stacey calls the “good enough holding of anxiety” (Stacey, 2000, p. 391).

Allied to anxiety is the notion of the use of power. Power is an attractor that constrains (and enables) behavior. Used in a coercive fashion, power is likely to elicit passivity and compliance (the closing down of free conversational flow) or rebellion and disintegration of the conversation (Stacey, 2000). At the same time, when power is absent from the conversation, connectedness and cohesion suffer.

The more senior the leadership role, the more connectivity and power the leader is able to wield, and the more diversity the leader is able to introduce. This means that leaders are able to influence the pattern and quality of organizational conversations in a way many others cannot. An old saying, “The leader casts a long shadow, and small movements by the leader are translated into large movements at the end of the shadow,” speaks to the power leaders wield. The follow-up question, “What is the single most common activity in organizations? Shadow watching!” speaks to the connectivity of leaders!

The conversational metaphor, as discussed so far, is consistent with the work of Dimitrov (1997) on fuzzy logic in teams. Dimitrov suggests a remarkably similar set of preconditions for effective group dynamics. Using fuzzy logic rules, he sees “preparedness to act together” as a function of willingness to engage in dialogue, trustworthiness, and creativity. Both the correspondence with connectivity, anxiety, and diversity, as well as the conversational dynamic of the system are clear.

If organizations are emergent properties organised by the models, themes, and dynamics of the conversations engaged in by system members, then we can say, along with Parry, Dimitrov, and Stacey, that the quality of the conversation determines the quality of the organization.

What gets said, how it is said, by whom and where, whose voice is heard, what can and can’t be talked about openly, what values and emotions energize our conversations, what patterns our conversations take—they all point to the attractors that shape the behavior of system members, and indicate the health of the organization. The quality of the conversation determines the quality of the organization. One has only to look at the way our current discourse around the war on terror is shaping our societies in order to see this.

Figure 11.3 summarizes the implications of the preceding discussion for coaching leaders at every level in an organization.

Complexity theory not only shows us what may be profitably focused on in the coaching conversation, it also helps us to understand how we might approach the coaching conversation so as to foster change and the open flow of communication.

Similarly, the preceding discussion also calls into question overly simplistic client-centric approaches such as the common or received version of the solutions-focused approach often seen in coaching. In this version, the solution is claimed to lie within the client. The coach’s role is to merely facilitate the client’s discoveriy of what the client already has within himself or herself. “Ask, don’t tell” is the catch cry. Yet sometimes no matter how long we ask the solution does not emerge, because it is not “in” the client, nor are the raw materials available for it to emerge via a process of questioning.

(It should be noted that the more sophisticated solutions-focused approach, such as that articulated by Jackson and McKergow (2002) is consistent with a complex adaptive systems approach. It sees the solution to a client’s problem in interactional terms, and as an emergent property of the interaction between the client and the coach.)

My experience as a coach, and as an educator and a supervisor of coaches, suggests that both the simplistic client-centric approach and the expert-centric approach are often more about managing the coach’s anxiety than they are reflective of what actually happens in effective coaching sessions. Effective coaches often do tell. They educate their clients. They share their mental models, and tell them things when the answer eludes the client. They also spend a lot of time asking. But the questions they ask are not atheoretical, as stated in the solution-focused texts (Jackson & McKergow, 2002). Rather, they are informed by their hypotheses about what is going on for the client. These hypotheses are built on the foundation of the coach’s understandings about what it means to be human, or in relationship, or in business, or healthy, or whatever else the coaching is about. Their domain-specific knowledge is constantly in play, but never overpowering the client. The coach’s telling is timely, the questions genuinely curious.

A very wise clinical supervisor I had years ago used to say to me, “The art of good therapy is to wait for the learning moment, and notice it when it arrives.”

So how then does the coaching conversation proceed in a way that opens the client’s situation for change, while remaining respectful of the skills, experience, and knowledge of both the client and coach? How do we notice the learning moments, and what do we do with them? The process illustrated in Figure 11.4 is untested except in the laboratory of the mind. It comes out of my experience as a coach and supervisor, and as a client of coaching and supervision. While I have shared this model with many coaches and it seems to resonate with their experience, it does not yet qualify as an evidence-based model. I present it here because I find it a useful metaphor that helps me understand what I am doing when I coach.

It is named “The Three Reflective Spaces” because it is a system made up primarily of three spaces that contain three conversations that interact together to create the coaching conversation.

The first reflective space is the internal conversation within the client. It is here that the coaching conversation starts. For the sake of simplicity, the client is here defined as the person we are coaching. The second reflective space is the shared space created between the coach and client. It is the space in which the external coaching conversation is conducted. The third reflective space is the space within the coach. As with the client, this is the space where the coach’s internal dialogue happens. It is the space where meaning and understanding emerge for the individual.

The external coaching conversation begins as clients tell their story. In other words, they put data into the shared space between the coach and client. This data is the product of their internal conversation and the myriad of conversations they have outside the coaching relationship. It is informed by and expressive of their experience, mental models, personality, physiology, goals and values, emotions, and habits and defenses. Everything that the client puts into this shared coaching space is a potentially rich source of information about the client in his/her world.

The coach listens to this rich data as it enters the shared space—noticing and selecting through the filter of his/her own experience, mental models, and emotional responses. We try to notice what is said, both verbally and non-verbally. We notice the story and the way it constrains possibilities and enables them. We notice what is not said, and what is avoided. We notice the feeling tone in the client, and the feeling tone it elicits in us. In this way the client’s communication enters into the coach’s personal reflective space. Here it continues to interact with the coach’s experience, mental models, emotions, personality, history, and so on, and we begin to see patterns as the client’s data elicits ideas, images, metaphors, and theories. Meaning or knowledge begins to emerge for us in this process. This processing often continues postsession and during the coach’s supervision.

The knowledge that is elicited is new knowledge—we coaches see it in the connections between what the client is experiencing and our own experience and understandings. When we are truly engaged in the conversation, this emergent knowledge has the character of insight, rather than of the mechanical overlay of our preexisting models on the client’s situation. It is an “aha” experience (Lewin & Regine, 2001). Yet the insight is tentative until shared and agreed—“Aha, perhaps this is what it means for them.”

Our experience, mental models, values, and intentions help us as coaches to see the data in a way that raises the possibility of change. The coach’s experience, theories, values, and intentions inform the dialogue, rather than predetermine its outcome. It is through this internal dialogue that data first disclosed by the client is transformed in an attempt to make it understandable in a way that facilitates change and goal attainment.

The fourth step is to put this transformed data back into the shared space for ongoing consideration in the conversation. The coachee then picks it up and, all going well, takes it into the crucible of his own internal dialogue. The conversation continues in an iterative way until both coach and client have developed a shared understanding, or shared mental model, that opens the way forward to action. In effect, new knowledge has been created in the coaching encounter. For that knowledge to be useful, it must result in action in the client’s world, and feedback on that action needs to be returned to the client and the coaching system.

For the coaching process to remain an effective conversation, the coaching system must be open. The input-transformation-output process is required at every level of the system. Information can (and often should) flow in from many points in the wider environment. This introduces both connectivity and diversity into the conversation—two of the three critical variables for fostering the conversation. Figure 11.5 illustrates some of the many sources of new information that may nourish the coaching conversation. The cycle of information flow, conversation, action, and feedback continues in an iterative way throughout the coaching engagement.

Anxiety and trust must also be held at good enough levels in the conversations. Deep respect for the client, an empathic and curious attitude, rapport building and listening skills, sensitivity to the emotional impact of the conversation, clear confidentiality boundaries, and an understanding of personality are all critical coaching competencies here. Holding anxiety in a good enough way is an ongoing process in the conversation. As we have discussed, too much or too little anxiety is destructive of the type of open disclosure needed for effective coaching.

The exercise of expert power is also an important issue in the coaching space. The three reflective spaces model helps us to understand the noncoercive use of domain-specific knowledge and expertise in coaching. Our traditional understanding of an expert is someone who holds a high level of domain-specific knowledge and skills. Expertise is the effective application of that knowledge and skill. However, if complexity theory is correct, then true knowledge is an emergent property of the intrapersonal and interpersonal conversation process.

In the same way, expertise, or the effective application of expert knowledge, is an emergent property of the conversation. It emerges in the creation of a shared mental model that opens the way for change. This understanding of expertise seems to tally with experience. We have all had coaching sessions that went poorly—where we felt inexpert and acted inexpertly, even though we had the raw materials for expertise. Sometimes this is because the conversation is such that it does not allow knowledge to emerge in the individual (as in the case of Avril), or because conversation did not enable the production of a useful shared mental model. A multitude of factors can derail the emergence of expertise. These include: inattention, anxiety, lack of trust, avoidance, defensiveness, lack of expert information, failure to empathize, or other internal or external conversations overwhelming the coaching conversation.

There appear to be at least six common internal conversations that derail coaching. These all involve failure to engage with the conversation in the creative space at the edge of chaos. They can occur within either the client or the coach, and frequently both coach and client collude in them together. They are:

All of these involve conversations that block the free flow of information. They seek to resolve the tension of being on the edge of chaos by either prematurely imposing order on the conversation, constraining what can be talked about, or introducing a level of anxiety that closes down communication. They stop us from really seeing our clients and meeting them where they are.

Expert knowledge is critically important to the coaching conversation. Without it, we are simply well-meaning amateurs. Expert knowledge helps us, and our clients, understand experience in new ways. It helps open the system boundaries. While expert knowledge must be held, it must be held lightly! The danger with expert knowledge is that we can make it a priori to the lived experience of the client. In other words, we can give our “expert mental models” a privileged status and then try to fit the data to the theory. When our theories and models move from being perspectives that nourish the conversation to the necessary conclusion of that conversation, they have moved from being information to ideology. When this happens, we, as coaches, have moved from a stance of curiosity and service to one of coercive arrogance!

Carl Rogers’ often quoted exhortation “all facts are friendly” reflects this open, or nondefensive, reflective engagement. Learning to take this stance is the work of a lifetime. It is no less than the process of growth and maturation. It requires the individual to engage constantly with the tension found at the edge of chaos, so as to seek new understandings that enable ever more effective, value-based action in the world.

This process is inherently challenging. It calls us to examine our assumptions, mental models, habitual ways of responding, goals, and values. It calls us to engage in honest dialogue both within ourselves and with others When done well, this reflective engagement not only issues in double loop learning (Argyris, 2004) and effective action, it also nourishes the spirit; and by calling us into ever more authentic engagement with others, it nourishes the world in which we live.

This is no less true for the coach than it is for the client. As coaches, we have an ethical imperative to develop our fitness to engage in the complex responsive system that is the coaching conversation. This involves ensuring that we as coaches dance within the tension at the edge of chaos. At a practical level this means entering into supervision, ongoing professional development, a commitment to informing our skills through rigorous learning and evidence-based practice, and the hard work of dealing with one’s own personality issues and personal growth. These are the behaviors that keep us in the dance, and develop richness and depth to our inner and outer dialogue.

While it may sound a daunting task, as they say, “We aim for progress, not perfection.” It is an ongoing iterative process. But in learning to engage more fully and effectively with an ever more diverse range of clients, we ourselves are nourished. All the facts are indeed friendly!

Let us now look at how complexity theory may inform the coaching conversations with Bonita and Bob.

So, rather than invent a coaching dialogue, I have chosen to discuss some of the potential issues I found myself wondering about (my internal dialogue) as I read Bonita’s and Bob’s case studies, and point to some of the questions and techniques that might be useful to explore.

As is often the situation, both Bonita’s and Bob’s cases show that there is work needed at the level of the personal and the wider organizational conversations. In the interest of brevity I will focus more on the personal conversations in Bonita’s case, and more on the organizational conversation in Bob’s case.

There are a number of feedback loops potentially operating here. One (and only one of many) potential feedback diagrams is illustrated in Figure 11.6. We are told that Ken reluctantly agreed to Bonita’s promotion, and there are clear signs of tension in his internal dialogue around Bonita. It is possible that resentment, or a degree of dismissive contempt, is leaking into the external conversation with Bonita. This leakage can be conscious or unconscious, overt or extremely subtle. Nevertheless we are often very adept at feeling it, even when we find it hard to rationally identify what is wrong.

Something in the external conversation triggers an internal conversation for Bonita, which seems to create for her a sense of anxiety and powerlessness. This leads her to not speak her mind and/or to avoid difficult issues. We can see signs of this internal dynamic in a number of places in the case study. It is seen with Rick and with Rita, and in Bonita’s difficulty managing boundaries between work and home. It may also be seen in her need to syndicate ideas with her colleagues. There thus appears to be some sort of attractor at work here. Precisely what the boundaries of that attractor are, and the internal and external pattern of behavior that attractor elicits, we are yet to discover. At a theoretical level, we might guess that it is likely to involve Bonita’s personality and history.

One of the tasks of coaching, then, would be to inquire into the meaning of these events and relationships—to ask Bonita to talk about the stories she tells herself in these situations, and how this impacts on the unfolding conversation in the external world. Often one can discern a common pattern to these stories and the processes they set in motion. My initial hunch (informed by my mental models, experience, and emotional responses or empathy) is that it may have something to do with overvaluing others’ expectations, which generate feelings of responsibility, while at the same time leaving Bonita feeling like she can never really live up to those expectations. This set of assumptions, beliefs, rules, and expectations forms a mental model that causes Bonita to seek and notice particular types of feedback, and that functions to create an anticipatory or feed-forward loop in which Bonita does not speak her mind, and she avoids potential confrontations. This then serves to reinforce Ken’s perception of her as not up to the task. Three self-reinforcing vicious cycles are created. The quality of the conversations determines the quality of the system!

In coaching, we are not so much interested in unpacking the history of Bonita’s beliefs and feedback and feed-forward loops as we are interested in understanding how the story unfolds in the here and now, and what other stories might be told. In the coaching conversation I would want to explore the features of the situation that tend to elicit that story. I would ask questions about when Bonita is able to tell herself different stories with different outcomes; for example, what is different about the story she tells herself with Ken and the somewhat more empowering story she tells herself with Rita? What other stories could she tell? Elements of cognitive behavioral, narrative, and psychodynamic models are likely to be useful in assisting Bonita in developing a new story or mental model that is more helpful. Figure 11.7 illustrates what this might initially look like.

Rather than engage in a self-focused critical inner dialogue, Bonita has engaged in the hard work of challenging her mental model, and is now more able to dispassionately notice what is going on in her, and in the conversation. The conversation is no longer “about her.” Anxiety is held in check, and her emotional reactions can now be seen as information that helps her to understand the interaction and as signals to change the conversation. The boundaries of her system begin to open.

Understanding her own emotions, and the responses they elicit, is also a potent source of information in understanding others. What is most personal is often most common. By connecting with herself—her goals and her values—while at the same time holding a respectful and curious stance toward what is going on for Ken in this encounter, Bonita is now dancing with the tension on the edge of chaos. She is in the place where tension is most likely to stimulate the emergence of new more creative and adaptive responses.

Notice in Figure 11.7 that Ken’s part of the conversation has not yet changed. There is most often a time lag between introducing change and the adaptive responses of others in the system. We cannot predict what Ken’s response will be. We might hope that Bonita’s respectful and courageous responding will elicit more respectful and positive responses from Ken—but this might not be the case! Ken’s responses are radically unpredictable. The attractors at work for Ken may mean that he becomes more entrenched in his negative view of Bonita, interpreting her new behavior as contemptuous and manipulative. Whatever Ken’s new response, Bonita is now in a better position to incorporate it as new data about which to be curious. As she slowly learns how to converse more effectively with Ken, she will become more and more adept at dancing in the tension, and more adept at holding good enough levels of anxiety and growing the degree of trust in the relationship.

Some of the coaching techniques that might be of use in developing more effective internal and external conversations are conversational mapping and role-play. Conversational mapping involves planning out a potential conversation and then examining it from as many perspectives as possible. What goals and values might others have in the situation? How might others feel, act, or respond to the conversation, and what does this mean for holding anxiety and trust at good enough levels? Role play is often a very useful way of both exploring these issues, practicing different conversations, and becoming adept at noticing what is going on in the inner dialogue.

Conversational mapping, role-play, and introspection are useful techniques because they acknowledge that different system members have different goals and adapt in complex ways as the conversation unfolds. This adaptation is informed by their goals, values, mental models, and emotional responses. This is what is meant by the phrase “System members act locally.” The process of conversational mapping and role-play is useful, but not because it allows the client to control or predict the way the conversation will unfold—interdependence and sensitivity to initial conditions means that the real conversation is often quite different to the one that was practiced. Rather, these techniques provide a forum in which to uncover the mental maps that are operating and develop new ones, and at the same time they give the client practice in engaging in the conversation in a more responsive and less reactive way. The exploration of which values, goals, and tone the client would like to have characterize the conversation also forms an attractor that helps the client maintain purposeful styles of communication behavior.

Bonita is likely to want to develop her ability to act assertively. This may have ramifications at a wider team or systems level. Behaving more assertively represents a change in the conversation between members in the system. Interdependence means that such changes typically raise anxiety in the system, and may elicit pushback. For example, as Bonita begins to be more open and forthright in her communication, some her direct reports may worry that she is unhappy with them. Still other system members may interpret Bonita’s new behavior as in some way subversive of the established “way we do things here.” Both interpretations may elicit defensive behaviors.

Exploration and role-play may be useful in developing strategies designed to introduce this new style of conversation into the workplace. Depending on the nature of the existing organizational conversation, there may be a need to gain buy-in from key team members, or even support for the change from senior leadership. The principle here is to assist Bonita in identifying some ways in which she and her team can develop a shared mental model of how they want to communicate together. Some potential strategies may include: just modeling the desired behavior, explicitly flagging the new communication style with individuals or the whole team, personal sharing of her experience, developing a new shared language through training, or facilitated examination of the team communication process.

Bonita’s goals are also likely to include improving her leadership and increasing team productivity. At this level it is useful for Bonita to have an overview of attractors in the system, including how power is distributed through the system. Points to focus on here might include: What sort of things are spoken about at a team level? What is the tone and nature of the conversations that take place? Are they solution focused or problem focused, optimistic or pessimistic? What is spoken about, where, and with whom? What is not spoken about? How is communication curtailed in the organization? Who gets to speak? What values are actually enacted, and which values are merely espoused? What processes, rules, norms, and assumptions enable and constrain the organizational conversation? How do people feel about themselves and their work?

The selection and timing of questions, and the selection, timing, and type of sharing /education done by the coach are important. The coach processes the data Bonita brings to the coaching conversation by allowing his/her experience, training, models, theories, and humanity to grapple with data. The questions emerge out of this process, and are not a priori to it. They are asked and considered in the light of Bonita’s intentions and aspirations for the organization and her role as a leader.

The task of coaching here is to help Bonita identify her goals as a leader and then to assist her in shaping ongoing, iterative, value-based conversations with the wider system and its members. In these ongoing conversations Bonita will be called on to manage boundaries, make decisions, implement actions, and have hard conversations while maintaining connectivity and diversity while keeping anxiety with reasonable bounds.

The coaching conversation is a developmental task whereby Bonita seeks to improve her ability to enter into these iterative, complex responsive exchanges. Coaching will cease to be needed as Bonita gains skill and confidence and fitness for sustaining this dance on the edge of chaos.

Bob appears to be an attractor for these types of conversations. One might expect that there will be a significant pull toward recreating this conversation in the coaching relationship. Hence the first question the coach must ask is whether he or she is able to enter into an effective coaching dialogue with Bob, or whether referral to someone better equipped to deal with challenging personalities is required (Berglas, 2002). Whether coaching or therapy is called for will depend on whether Bob is able to be open to the possibility that he will need to make significant changes in his own behavior (see Cavanagh, 2005, for a fuller discussion of working with challenging clients).

Let us assume the coach is appropriately qualified and Bob shows an ability to approach change. Given the long-term nature of Bob’s behavior, the coaching engagement may need to be relatively long-term (one to two years) if effective and sustained change is to be achieved. The first significant challenge faced by the coach will be to establish empathic attunement with Bob while also establishing effective alignment for creating the type of conversations needed to leave the organization in a healthier state.

Establishing alignment and rapport may be achieved by first exploring what the “just right” organization would look like, and what this would mean for Bob. I would also want to explore Bob’s understanding of the sort of relationships among the members that would qualify as “just right” functioning. Specifically, it will include exploring what this means in terms of the types of conversations Bob will need to shape with his people. Bob may well find it difficult to articulate the more empathic and integrity-based elements of that new conversation. For example, because Bob seems to lack empathy and sees the world through a mental model of winners and losers, his understanding of courageous honesty may be little more than apportioning blame.

The coach can assist Bob in discovering the meaning of empathy and honesty and courage in communication both by modeling this in his/her interactions with Bob and by creating good enough holding of anxiety to allow Bob to approach the blind spots in what appears to be a long-standing pattern of behavior. In order to do this the coach must be able to see the world from Bob’s perspective, and demonstrate an understanding of the value Bob sees in that perspective. At the same time he/she must find a way to help Bob engage with the inherent tensions and contradictions that inevitably lie at the heart of his experience.

As in the case of Bonita, the use of conversational mapping, role-play, introspection, and feedback will be important in developing greater empathy and attunement in Bob for his people.

At the wider systems level, it will be important to explore the key differences in approach to organizational conversations between the merging South Korean/European and U.S. elements of the system. As we know from complexity theory, diversity needs to be held within critical limits if it is to have a positive impact on opening the system to change. Bob may need to assess the level of connectivity and diversity operating in the merging between these parts of the business.

Dimitrov’s (1997) work on the use of fuzzy logic when dealing with social complexity may be very useful in helping Bob shape conversations that encourage participatory action. Dimitrov has found that preparedness to act together is a function of willingness to engage in dialogue, confidence in trustworthiness, and ability to create options (creativity). Willingness to engage in dialogue is fostered by effective listening and validation of differing perspectives. As leader of the organization, Bob can demonstrate this in his communication with the business. This will require Bob to begin to remember what is spoken about, follow up on communication, and respond to people’s concerns and feelings. In other words, Bob will need to genuinely engage with individuals and teams.

Trustworthiness is built via shared responsibility and accountability (Dimitrov 1997). Demonstrating unidirectional accountability is destructive of trust. Follow-through and accountability in Bob’s action are important conditions for increased trust. Similarly, Bob will need to entrust responsibility to his people. Learning to delegate power while holding in check his anxiety about not being in control may be important components of Bob’s change program. Bob will need to ensure there are appropriate resources, training, and communication avenues available to enable effective power sharing. Similarly, the establishment of timely and effective feedback mechanisms to support accountability will be important.

The final condition, creativity, needs to be fostered. Punitive action in the event of mistakes is inimical to creativity. Permission to make mistakes, as well as rewards for novel approaches and diversity, are essential in building an atmosphere conducive to creativity. In particular, reward for combined creative action would encourage a sense of moving forward together, increasing cohesion and maximizing the organization’s ability to adapt to its new circumstances (Dimitrov, 1997; Harris & Scherblom, 1999). Once again, Bob’s ability to demonstrate this type of connectivity in his own relationships with organization members will be an important part of embedding this behavior in the organization.

These changes will be difficult for Bob. The quality of engagement Bob has with the organization is a critical part of modeling and supporting the emergence of a new organizational conversation. Equally so, the quality of the engagement the coach has with Bob will be critical in supporting the emergence of that new inner dialogue in Bob.

If the coach overidentifies with Bob, he/she will have no new perspective or information to bring to the conversation—no place from which to challenge Bob to growth. However, if the coach approaches Bob from outside—objectifying him and his behavior—then in order to protect his sense of self, Bob will almost certainly close his boundaries and the conversation will end up in silence or a useless cycle of accusation and denial. The coach must be able to maintain a paradoxical position of “attuned objectivity.” (Once again, we are at the edge of chaos.) It is in this place of creative tension that feedback can be most effective, provided that anxiety and trust are held within appropriate bounds.

Maintaining a position of attuned objectivity is a type of loving of our clients. It requires us to consider them as whole people, struggling like the rest of us to do their best in the world. But like the rest of us, struggling means struggling! It means grappling with the inner dragons that threaten to derail us. Attuned objectivity means taking the presence of the dragons seriously, while seeing the beauty inherent in the person and in his/her attempts to grow.

As mentioned earlier, coaching in the case of Bob is likely to be a longer-term engagement, with much less certain success, than in Bonita’s case. Relapse into old communication patterns is always a danger when more personality-based forces are at work. The impact of relapse in destroying trust and system openness is significant. At the same time, Bob’s impending life transition may provide the personal openness and motivation Bob needs to maintain himself at the edge of chaos.

There is potential for this to happen if events converge in such a way that the wider organization is destabilized and pushed into a highly reactive or defended state. Significant teething troubles with the merger, a major drop in share price, or the threat of downsizing may fit these conditions. A poorly conducted coaching-related feedback process at that point could elicit a level of blame and criticism toward Bob against which he is unable to defend. The more fragile Bob’s ego defenses, the less blame or criticism is needed to overwhelm him.

On the face of the case study supplied, Bob appears quite resilient, and such a worst-case scenario is unlikely to occur. Nevertheless, as the butterfly effect attests, a combination of sensitivity to initial conditions and amplifying feedback can lead to unpredictable outcomes. Hence it would be important for the coach to be particularly aware of Bob’s ability to engage with challenging feedback in a mature and resilient fashion, before initiating highly confronting feedback processes (e.g., public sharing of 360-degree feedback).

I believe that the metaphor of conversation offers us a rich image through which to understand the complexity of the human person, human systems, and the complex adaptive system that is the coaching engagement. Our role as coaches is to engage with our clients in the dialogue that carries us to the edge of chaos—to have the courage to approach the tensions that disturb us, yet paradoxically help us grow into more fully present and responsive human beings.

It is a great honor and responsibility with which we are entrusted. We honor that trust whenever we do the work of developing our fitness to enter into ever more effective conversations with our clients, so that they can enter into ever more life-giving and effective conversations with their families, colleagues, and the world.