The goal of this paper is to describe a mind-set that is needed for those in leadership roles to succeed in today’s increasingly complex and fast-changing business environments. The predominant feeling in organizations today is of continuous motion rather than of the punctuated action of the last century. Last century’s model involved periods of careful planning followed by tightly controlled execution, followed by further planning, and so forth and so on. In contrast, today’s work environment is constantly in motion and often loosely controlled.

Over the course of several research projects in Sweden, my colleagues and I have identified what we believe is a distinct mode of thinking, communicating and interacting with others that enables success in today’s work environments. It is a manner of working that recognizes the turbulence and uncertainty in the flow of events and seeks to “lead” others through this complex flow of events in search of successful outcomes.

The Greek word for “stream”, or “flow” or “current” is rheo, and so I call this way of leading others: Rheo Leadership. Because the galloping motion of others along the flow of events can neither be completely controlled, as in a machine, nor allowed to run totally free, as with mustangs on the prairie, the methods of Rheo Leadership are described not as “steering” but as “rein control”.

In natural environments it is not uncommon to find this same kind of “rein control” in complex systems. This occurs when competing forces like the divergent effects of different enzymes influence the system’s dynamics in opposing directions. Balancing the dynamics at work in the system by toggling competing influences is what I mean by rein control. In the case of human systems, we are trying to find the same kind of competing forces, what we call “categorical dualities” that work in opposing directions within human interaction dynamics (HID). Once identified, these competing tendencies can be used by leaders in organizations to differentially exert rein control over others to guide their organizations to successful outcomes.

The Rheo leadership model is not meant to explain individual behavior in an organization. It is meant to simplify the understanding of one task that leaders must perform effectively to succeed in modern organizations. Leaders must recognize and influence the flow of events and do so through others. As such, they must know both what forces are influencing the activities of others and how as leaders they can influence these forces. To do this requires that the leader has both a rational and an intuitive understanding of two kinds of structures: those consciously formulated by managers and those that are tacit and emergent. Each of these types of structures operates on the same three sub-categories of activity: acting, thinking, and relating.

Beyond the recognition of these structures, managers must know how to influence them to exert rein control over others in fast moving environments. This involves understanding the lens through which individuals see and experience these structures and are influenced by them. Traditional management focuses on designing manager-formulated structures like incentive programs or project plans. This paper focuses primarily on tacit-emergent structures that characterize the informal structure of work activities with the intent of bringing influence of these structures within the leaders purview as well.

In the context of this structure, the lens of experience that can be used to facilitate rein control involves a keen understanding and appreciation three dualities and their implications for action: integrated-autonomy, convergent-divergence, and value creating-discovering. The first relates to the dynamic balance between the level of autonomy of the actor and the integration of his or her behaviors into the existing structures of the organization. The second duality relates to the nature of the change processes that emerge as a consequence of interactions in the organization. In particular, interactions in today’s work environments may lead to both the convergence of activities—bringing some people and activities together—and to their divergence—tending to tear some aspects of the work apart, moving certain activities in new directions. The third relates to the activities of the collective; balancing exploring and discovering new opportunities on one hand and creating value through exploiting these opportunities and resources on the other. Tugging on the reins of leadership to move an employee’s perception of these structures toward one side or the other for each duality can be used to influence the dynamics of the work system. Dynamically adjusting these categorical dualities is the essence of the metaphor “rein control.”

Ideally any particular work practice is partly controlled by manager-formulated structures like incentive plans, partly controlled by tacit-emergent structures like norms, conventions, work habits, etc., and partly uncontrolled, leaving varying amounts of room for autonomy and individual motion for employees. The integrated-autonomy reins influence employees’ propensity to be integrated into these structures (and thus be influenced by them) versus their desire to remain autonomous with respect to these types of informal structural constraints. The convergent-divergence reins influence the changing strength of the tacit emergent structures with respect to responding to new information and tolerating differences of perspective and opinion, and thus the changing degrees-of-freedom for individuals with respect to acting, thinking and relating about the information they hear about or learn about through experience.

The goal of this paper is to use these ideas, drawn from field research, to make a contribution to the development of HID as a tool for leaders in organizations. The paper is based on studies of first line managers in different kinds of organizations in Swedish work life. It focuses on the leadership task of enabling subordinates to self-organize into a competent collective. The first studies where exploratory, performed in organizations perceived to be models of effectiveness in self-managed teams and long-term economic sustainability. The latter studies were action research interventions where first line managers were trained to use what we now call the Rheo Leadership task. Experiences from these studies are presented in the final section of the paper.

In our research we have mostly studied managers, but the model is useful for all of us. Everyone who is part of an organization might take on the role of a leader for a while and would benefit from using the Rheo model. As a team member, you can take the reins to increase creativity when work has stalled and new ideas are needed, or to increase collaboration when it’s time to come together. As a Rheo leader, you will have the conceptual tools to become an important source of forward momentum for your colleagues. As a result, you will be seen by them as a leader.

A categorization of structures in work systems

This article assumes that leaders use and influence the social structures that organize much of the actual work that is being done in organizations. To clarify the notion of social structure, it is useful to elaborate on the ontology and epistemology used in this article. The term “duality”, a central concept in this paper, means that there are two different and seemingly incompatible descriptions of one phenomenon. It seems like a paradox that both can be valid, but in fact each focuses on different aspects of the same wholeness. One example from physics is the wave-particle duality of light, where the propagation of electromagnetic radiation is analyzed using wave theory but its interaction with matter is described in terms of photons. Another example of duality is given by Giddens (1986, see also Schwandt & Szabla, this issue), who talks about the duality of structure in society. He argues that it is a mistake to reduce structure as functionalists do, where structure has primacy over action, influencing agents that are ignorant of them. But it is equally wrong to reduce structure as hermeneutic approaches do and “regard society as plastic creation of human subjects” (Giddens, 1986: 26). Both of them are valid at the same time in each human action. Structure is at the same time created by human action and is influencing it.

Inspired by critical realism (Bhaskar, 1978) we also use a duality of ontology, including both a social determined understanding of reality—a inference model if you will (Hazy & Backström, this issue) on the one hand, and the real phenomenon, an unaffected-by-the-observer reality, on the other. We believe that there are actual events “out there” about which we can collect empirical data. But our data are always affected by the filters we have when we collect them: our senses, language, pre-understanding, intentions and theories—our inference models (Hazy & Backström, this issue). Further, our ambitions are normally higher than just to describe the events out there. We are usually also trying to understand the mechanisms behind these events; for example the social structure that influences observable human actions. By studying actions of humans in a system it is possible to draw conclusions about and describe constraints, like social structures, that decide the dynamics of the system.

Fig. 1: Different categories of social structures in an organization

The social structures studied in this article are structures in organizations. For analytical reasons we divide the structures of an organization into different categories, see Figure 1. Some of them are formulated by managers or commissioned by them. For example, they might be developed in the technostructure of staff units and specialists (Mintzberg, 1983). Tacit-emergent structures, on the other hand, are emerging in social interaction with other people and are constrained by the manager-formulated structures. These informal structures are tacit since in everyday life most actors are neither aware of these structures nor able to choose between them or change them purposely. They are stored as different patterns of thinking (which we will call culture in this article), acting (which we will call an institution), and relating (which we will call relatonics). These have been studied and are described herein, but other emergent structures may be equally important, for example collective identity (Hazy & Silberstang, 2009). The remainder of this article focuses on tacit-emergent structures because they are more difficult for a leader to recognize, use and influence.

The emergence of tacit social structures

Central for the understanding of emergence as used in this article is that recurrent events at the agent level may form habits or path-dependent trails that constitute informal structure. This tacit-emergent structure persists for a time, and while it does it governs interactions among a subset of individuals or agents to a degree, thus forming what is called a multi-agent. This is a collection of agents that act as an entity. Tacit-emergent structure—for example, a dominant interpretation within an informal group that is addressing an issue (See Edwards & Baker, this issue)—makes it more probable that events like the prior event will again unfold in a similar way.

From an HID perspective, tacit-emergent structure means that path-dependence information has been stored in this informal structure enabling greater predictability (Hazy & Backström, this issue). The very notion of emergence, therefore, contains within it the presence of a categorical duality between acts/events at the actor or fine-grain level and the very same act/events as they unfold at the meta-actor, multi-agent or coarse-grain level (Hazy & Backström, this issue). This duality includes within it a deep necessity for circular causality—both downward and upward influence simultaneously in force—between these levels as is described, for example, by the mathematician Hermann Haken (1996).

Also relevant to tacit-emergent structures is the notion of structural coupling among components of a system (Maturana & Varela, 1987). When considering the interactions between people within a workgroup, structural coupling provides a conceptual framework that makes it possible to use the work of Haken (1996) and the notion of the order parameter to model and potentially better understand the emergence of social structures. Although the mathematical derivations of Haken and others are valid only close to instability points or singularities, there is empirical support for their conceptual usefulness even far from such points at least with regards to physical systems (Haken, 1996).

Models of feedback through the structural coupling connections among group members can enable the amplification of the newly identified order parameter, serving as one mechanism that describes emergent structural congruence (Maturana & Varela, 1987). Individual differences in understanding and ways to act are damped or amplified by pre-existing and evolving social structures. Global patterns or coarse grained structures develop—perhaps one is a new property that can be modeled by the order parameter—and certain of these patterns win in the competition with all other possible patterns.

As time passes, the winning (i.e., surviving) self-reinforcing patterns are recognized as properties described by order parameters. When these are recognized and when models of them and the order parameter are communicated to others, the models can become attractors that entrain the actions of an ever increasing proportion of the population. This dominant coarse grain structure (cf. Edwards & Baker, this issue) is thus constructed by individual changes in understanding and in the implicit pressure that these apply about ways to act; it is also constrained by already existing structures. At the same time, these preexisting structures continue to entrain each individual’s understanding and ways of acting in a spiral of convergent-divergent circular causality.

The consequence is “a collection of individual interactions (what we call fine-grained HID) exhibiting emergent regularities or patterns at a higher level of scale (what we call coarse-grained properties)” (Hazy & Ashley, 2011: 60). This article discusses the management of three tacit-emergent social structures: culture, institutions and relatonics.

Human beings are wholes and all kinds of activities of a human being are really indivisibly interwoven. However, for analytical reasons we divide these activities into three types. Some activities, for example thinking, occur inside within the internal structures of an individual which are impossible for others to have direct knowledge of, be influenced by, or to control. Behaviors like communication, on the other hand, are possible for others to register with their sensory organs. Thus, others may be influenced by our behavior and may also give us feedback.

We make the equivalent distinction between interior and exterior in the reference models we develop to describe social structures (Hazy & Backström, this issue). An organization’s culture (Shweder & Sullivan, 1993) is the level of congruence in the internal structures of those who are members of the organization. It is the common simplified model that the collective of people use to have a similar way to understand the world. The institution (Berger & Luckmann, 1966) is the common habits of behavior, the expectation that what another member of the collective will do is normally approximately determined and predictable.

Returning to the three types of social structures, we find it functional to add the third category: relatonics (Backström & Döös, 2008). This is the infrastructure of relations that are being used while performing a work task. This could very well be seen as part of the institution, but since the interaction is of such central importance, both for the emergence of tacit social structures and for the competence of the organizations we have decided to give it special focus.

Categorical dualities as the leader’s reins to influence performance

With the prior discussion as background, this article next discusses ways for managers and leaders to use and influence the tacit emergent social structures that organize much of the actual work being done at a work place. This is especially important in a post-industrial work system characterized by increased decentralization, distributed leadership and high degree of worker autonomy, all of which is loosely constrained by manager-formulated structures intended to control work behavior (Backström et al., 2011). The goal is to use manager-formulated structure to create a big container that does not constrain the informal ways that one may act within the system. This larger container must allow individuals and groups to have the requisite space to enable flexibility as well as effectiveness. Under these conditions, we explore the tools that are left for the manager to use to control this kind of work system: the dualities that define the work system and the reins they provide for leadership.

The thinking about rein control for leaders within HID has been tested in several research projects where we have developed what we now call the Rheo task of the leader. Two of the here presented dualities, integrated autonomy and convergent divergence has been part of these projects about leadership, and are further presented below. The third duality value creating and discovering has been added to the model as a result of a project about radical innovations, where the duality between the explorative and the exploitative were discussed and transcended (Backström, 2012). For each of the four concepts that represent the poles of the dualities integrated autonomy and convergent divergence we present some organizational or social aspects that we have seen leaders use during our research. They are meant as illustrative examples to make the concept more concrete and understandable. The research doesn’t say whether they are the only ones or even the most important ones.

Integrated autonomy

Integrated autonomy relates to the breadth of the degrees-of-freedom within which various individuals are able to act, and this range of freedom can be regulated locally through rein control. It has to do with the employees’ relation to the social structures. This duality is studied and used in other articles (see for example: Backström, 2009). Integration means that the workers are entrained by the tacit emergent structures while autonomy means that they are able to use their degrees of freedom for action.

Our first inspiration in the development of this duality comes from Arthur Koestler’s (1978) discussion about holons. A holon is a stable sub-unit in a hierarchy of holons, for example the employee is both a holon and a sub-unit of the holon work group, that in turn is a sub-unit of the holon organization. Each holon has strong “self-assertive” tendencies which helps define it as a unique actor; at the same time holons also have strong “integrative” tendencies which help to tie them to the meta-holon. Only when these opposing tendencies are kept in equilibrium can the holon function successfully, according to Koestler (1978). Thus under the premises of a post-industrial organization with a high degree of autonomy, the level of employee integration is also an important variable.

In our interpretation of autonomy the employee can, knows how, and wants to be autonomous along many degrees-of-freedom. It requires a distribution of decisions to the employee, paired with enough competence and other resources placed in the hands of the employee. The employee must also be motivated to use the possibility to act independently. Managers are accustomed to giving instructions, advice or at least opinions when asked, and employees are accustomed to listening to their managers and asking if something is out of the ordinary. It takes time and a conscious effort from the manager to change these habits.

We have seen managers answering employee queries by saying, “you know best” and “it is your decision”, refusing to take back autonomy once it has been given to employees. (This laissez-faire leadership style can be a destructive if it is not balanced with the integration rein). It is also necessary that one’s colleagues on the team allow for individual autonomy as well so as to avoid coercive social control (Barker, 1999) of which we have seen examples in our empirical research. We also believe that increased autonomy can be a source of workplace burnout if adequate resources are not given to the individual. This is another reason why we believe that autonomy should often be thoughtfully balanced by tugging the integration rein to provide social support within the collective of workers.

Integration means that the employees’ degrees-of-freedom are subordinated to the structures of the organization, that their level of autonomy is limited. The challenge of influencing employees to follow manager-formulated structures is often described as a key task of leadership. The argument goes like this: the structures must be possible to use; they must be communicated effectively to all employees; and one must motivate employees to use them and further the organizations objectives. When it comes to tacit-emergent structures, these are by definition followed by employees only if the employee is swept into the process of their emergence. In order for all employees to be part of the emergence process, we believe that it is central to have leadership that strives to include everyone in important processes at the work place and to provide an open social climate that accepts that people are different. This is accomplished by toggling the reins associated with the integrated autonomy categorical duality. For example, leadership should include employees in the planning process of the work place (Backström et al., 2013). When this is the case, there is also a good chance that there will be strong tacit-emergent structures and also reciprocal consistency between the plan and the institutions, the culture and relatonics. The manager is in a position, in part due to a longer-term or a more holistic understanding of the organization, to enable the emergence of efficacious tacit-emergent social structures. To achieve this, the manager should take part in communication events and take care to consciously introduce the manager’s perspective into interactions and discourse. This challenge is dealt with using the reins enabled by the convergent divergence categorical duality which is discussed next.

Convergent divergence

Convergent divergence has to do with information gathering and use in the organization and the consequences of the unfolding system dynamics (Gell-Mann, 2002; Hazy & Ashley, 2011). Here, convergence means that interactions lead to more and more structural congruence between workers. New structures emerge as work activities tend toward the properties that have been described by an order parameter using models that are widely shared. Once externalized, these models can begin to act as structural attractors (Allen, 2001) for behaviors within the system because actions are evaluated and then modified based upon their expected impact on the model. Once a model drives behavior in this way, other possibilities are abandoned or extinguished, and the information that is embedded in them is forgotten. As these dynamics unfold, tacit-emergent structures approach the attractor. More and more of the thinking, acting and relating at the workplace becomes influenced and entrained as attractors are recognized as the way that things are done. In contrast, divergence implies that new information that might not fit the model is recognized and retained. This simultaneously reduces the attracting influence of these prior structures leaving more degrees-of-freedom for individual to act within the system. Rein control involves toggling between convergence and divergence to identify new possibilities while simultaneously enabling coherent collective action.

New information in a system can manifest itself and then either be dismissed as noise or recognized as significant through the poles of convergence and divergence respectively. This categorical duality—convergence in the presence of divergence (or divergence in the present of convergence)—also provides reins of control. In an earlier section, we discuss the emergence of tacit social structures through structural coupling when actors interact. This has the effect of reducing the degrees-of-freedom for employees and as such leads to convergence. But we have also indicated that interactions that identify new information may lead to divergence as the relevance of the new information becomes more widely appreciated.

According to Haken (1996) increased flow of energy or information can cause an imbalance and enable changes to the structures of a system. This occurs because divergence enables different structures to exist in parallel within the organization for a period of time (Hazy & Ashley, 2011). Time allows convergence to operate locally on some of these nascent patterns to make them more efficacious and beneficial to the system when compared to prior structures. With demonstrated benefits, these new ways of doing things may then replace the old ways in a process of continuous renewal. An organization is in need of both convergence and divergence to be flexible and innovative.

The basic idea in convergent divergence is that if a system has strong processes of convergence it will not be overwhelmed by divergent information as that leadership rein is loosened. Conversely, strongly divergent systems should not allow their participating individuals to be flagged by convergent tendencies as these reins are loosened. It is important to have a high degree of both convergence and divergence among actors in a system in a change process (Lubatkin et al., 2001).

Convergent divergence is also one of the most central aspects of group creativity. The diversity quality is important to have a broad range of knowledge and ideas to build on. The creative group is thus often composed of individuals with different backgrounds. To be able to use all relevant individual competencies in the group in the creative process each group member has to be autonomous: willing and able to develop, express and argue for his/her ideas in the interaction. Convergence is needed to be able to talk about the same thing, to develop a group idea (Olsson, 2008). The group members have to focus and relate their actions to an emerging group idea to be able to cooperate. The group idea is the vehicle that the group uses to travel together to be able to scout places of knowledge and ideas between their existing areas of experience (Backström & Olsson, 2010).

Divergence can be increased by introducing into interaction both motives for change and information that deviates from the normal. Organizations with the post-industrial work system often use ideas like customer focus, meetings with customers, a high degree of feed-back from customers, employee experiences, and reported quality problems to introduce opportunities for divergence into interactions. Other sources offering potential motives for change include continuous feedback about how one’s unit is performing in comparison to other units and reward-systems like profit sharing for workers. It is also important that the workers are confident that they will never be sacked because the work has become so effective that fewer workers are needed. Encouraging relatonics including persons working in other departments and in other organizations and contacts with persons from inspiring milieus also provide a way to increase divergence. These practices all serve to make external intelligence and different perspectives available for the workers.

Convergence can be enabled by giving employees space for, and reasons to communicate, as well as developing a competence in communication. For example, the managers with whom we have worked have changed the lay-out of the room and developed meeting structures in ways intended to increase possibilities of interaction. Also bonding activities outside of work are used; for example a company might offer a lunch program at a restaurant with good and cheap food and where the CEO eats with the workers and encourages other managers to do this as well. Another normative approach is to distribute responsibilities and tasks to small groups of employees and give group members reasons to interact with one another under common goals that take away competitive pressures that might otherwise inhibit cooperation. In the intervention study presented in the next section, we have developed a training program to help managers working with the employees’ competence in communication to enhance the structural couplings between them. It consisted of training in how to create a positive social climate, develop dialogue competence, and adopt an improvisational attitude.

Indications on the usefulness of the rheo leadership

In a project financed by Vinnova, Sweden’s innovation agency, my colleagues and I had the opportunity to use questionnaires before and after a training of managers in the Rheo task for leaders. The results give us indications on the usefulness of the rein control model (Backström et al., 2013). Nineteen managers from different kinds of organizations took part in training. The training was organized as an action learning network and consisted of eight workshops during one year (September 2009-June 2010), where the managers met with each other and the researchers once a month. The first and the last workshop were dedicated to start and end the process while the remaining six workshops had different themes. Two of them dealt with convergence (Dialogue and Balanced communication), two with divergence (Group creativity and External relations), one with integrated autonomy, and one workshop with the motivation of employees (Attractive work). The researchers facilitating the workshops put great effort in being Rheo role models focusing on integrated autonomy and convergent divergence while leading the managers in the training.

Questionnaires with 247 questions were given to all employees in the managers’ workgroups before (N=266) and after (N=215) the training. Seventeen scales were used to measure the effect of the training. The subordinates’ answers show a positive change in all of the seventeen scales (Backström et al., Forthcoming). Eleven of them can be interpreted as measuring the reins of Rheo leadership. One conclusion from these results is that the managers are able to strengthen both sides of the two categorical dualities, a potential indication of increased rein control. Their employees demonstrate higher levels of both autonomy and integration, and there are indications of moving communication and dialogue toward both divergence and convergence after the intervention. Our main conclusion is that it is possible to change subordinates behavior toward a normative ideal of the Rheo leadership model by training their managers in the model. Notably, the strength of the integration and convergence poles has increased more than the autonomy and divergence poles. We do not know if this was because this change was more needed, or if the structure of the intervention caused an imbalance, where the reins on one side became much stronger than the ones on the other side. Further research is needed to develop the training program in ways which further strengthen autonomy and divergence.


This paper is an effort to integrate thinking from different disciplines into one model of how HID can be used by leaders in organizations. The model is abductively developed and is therefore neither connected to the theories in an explicit enough way, nor fully tested empirically. Considerable work is still needed to develop a model that is fully valid and consistently successful, if that is even possible. The paper is a step forward in formulating the idea of a new Rheo task of the leader.