Changing the rules: The implications of complexity science for leadership research and practice

Full text

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 1 of 64
Changing the rules: The implications of complexity science for leadership
research and practice
James K. Hazy
Adelphi University
Department of Management, Marketing and Decision Sciences
Mary Uhl-Bien
University of Nebraska
Professor and Howard Hawks Chair in Business Ethics and Leadership
Department of Management
January 30, 2012
Please do not quote prior to publication. Cite as:
Hazy, J. K. & Uhl-Bien, M. (in press). Changing the Rules: The implications of complexity
science for leadership research and practice. David Day (ed.) The Oxford Handbook of
Leadership and Organizations. Oxford: Oxford University press.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 2 of 64
Changing the rules: The implications of complexity science for leadership
research and practice
James K. Hazy
Adelphi University
Department of Management, Marketing and Decision Sciences
Mary Uhl-Bien
University of Nebraska
Professor and Howard Hawks Chair in Business Ethics and Leadership
Department of Management
ABSTRACT
The study of complexity has become an important lens through which to view and
understand the causes and potencies of individual action and interaction in organizations as well
as their meaning for leadership research and practice. This review of key complexity ideas and
their theoretical implications for leadership describes emerging theories in the field, highlights
the growing empirical support for these approaches, and sets an agenda for future research. The
thesis averred is this: Just as complexity has become an overarching theoretical paradigm in the
natural sciences, it is providing the basis for a paradigm shift in the social sciences, particularly
in leadership and organizational studies. Complex systems leadership theory describes the
process whereby the rules governing local interactions are changed in response to and
anticipation of changing circumstances. In shifting the focus from the individual to the
organizing process itself, the complexity leadership perspective has important implications for
both research and practice.
Keywords: Complexity leadership, complex systems leadership theory, human interaction
dynamics, generative leadership, adaptive leadership, unifying leadership, adaptation,
innovation, chaos theory

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 3 of 64
INTRODUCTION
Complex Systems Leadership Theory (CSLT) is about interactions and emergence. It is
about events and how these shape future action (Lichtenstein et al., 2006), and it is about how
human activity is organized into a system of choices and actions when organizations are
considered to be complex adaptive systems (Marion & Uhl-Bien, 2001). By describing an
overarching dynamic theory of human organizing, CSLT transcends traditional approaches to
leadership research by offering a theoretical framework within which prior results can be better
understood, evaluated, and integrated into a common view of how human agency drives
collective performance and adaptation.
In the complexity approach, “leadership” is not considered to be a person or persons.
Rather, it is the recognizable pattern of organizing activity among autonomous heterogeneous
individuals as they form into a system of action (Lichtenstein et al., 2006; Hazy, Goldstein &
Lichtenstein, 2007a; Uhl-Bien, Marion & McKelvey, 2007). At the same time, for organizing to
occur, leadership must perform certain functions, what Katz and Kahn (1966) called the
“influential increment.”
Interpreting Katz and Kahn (1966), Hazy (2011a) argues that when human interactions
are considered as complex systems, leadership performs three functions as it organizes human
activity. First, it influences human interactions in ways that unify individuals into organized
groups. This includes what might be called the strategic functions, such as setting vision and
strategy (Boal & Schultz, 2007) and establishing identities and ethics (Hazy, in press-a). Second,
leadership changes the rules so as to generate a variety of ideas and plans of action (Hazy, 2006)
as a mechanism for adapting to changing circumstances (Uhl-Bien et al., 2007). Creativity and
problem solving (Guastello, 2007) are important elements of this function, as are the constraints

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 4 of 64
on action that enable the emergence of novelty (Goldstein, Hazy & Lichtenstein, 2010). Third,
rules are changed in ways that enable the convergence of disparate, sometimes conflicting
individual perspectives, preferences and activities into effective and predictable collective action
(Phelps & Hubler, 2007; Dal Forno & Merlone, 2007; Hazy, 2008b). In sum, leadership is about
changing the rules that guide individual choices and interactions. In complex adaptive systems,
“changing the rules of interaction” locally can also change organizational outcomes more
globally. CSLT studies this process.
Individuals, of course, enact pieces of this functional puzzle, and as such, the complexity
paradigm implies certain things about individuals and their capacity to succeed as they engage in
leadership situations (Hannah, Eggers, & Jennings, 2008; Lord, Hannah, & Jennings, 2011). But
isolated individual behaviors are not leadership per se. Leadership is in the whole; it serves to
form, sustain and grow the system, just as product development or accountability processes are
parts of the whole system. No one person is “governance”; likewise, no one person is
“leadership.” At the same time, individuals must enact the leadership process just as they enact
other organizational processes. In this sense, CSLT offers a systems’ perspective within which
traditional views of leadership that include individual skills and actions can be integrated into a
process perspective (Hazy, 2011b).
For the purposes of the analysis herein, we define complex systems leadership as system
processes that change the rules of interaction and do so in specific ways that form human
interaction dynamics (HID) into a complex adaptive system (Hazy et al., 2007) in a manner
analogous to how physical and biological interactions are understood as systems. Core to
complexity is the realization that the rules governing the individual human interactions of day-to-
day experience are what determine the social structures that emerge (Goldstein, 1989, 2007,

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 5 of 64
2011; Holland, 1975; McKelvey, 2004). These emergent forms can sometimes be recognized as
stable properties, and when they are, they can be evaluated and managed within a particular
economic, political, social and technological context.
However, interventions cannot directly cause outcomes to change. Emergent forms can
only be affected by judiciously changing the local rules that govern the interactions of others and
from which the relevant outcomes emerge (Goldstein et al., 2010). It is within the nexus
connecting local rules to emergent forms that leadership gathers its potency. This is why
leadership is central to human experience. It is important for both organizing to succeed as a
group, and as a means to enable individual success through others. Both of these have been the
focus of leadership research. CSLT connects these complementary aspects of leadership within a
systems framework.
Leadership guides both performance (i.e., “exploitation,” March 1991) and survival of
the system in the short-term, and adaptation (i.e., “exploration,” March, 1991) and thus its
prospects for survival over the longer-term. It does this by processing and using information
gathered as feedback from the environment and from within the system itself (Gell-Mann, 2002).
It senses this feedback and channels it to individuals who are in a position to use the information
it contains to find new ways of organizing in an effort to acquire, store and allocate resources of
all types. As such, complex system leadership theory (CSLT) transcends and integrates prior
research and offers a platform for understanding leadership in fundamentally new ways.
Thesis and overview
Our thesis is summarized as follows: Just as complexity has become an overarching
theoretical paradigm in the natural sciences, it is serving as the basis for a paradigm shift (Kuhn,
1962) in the social sciences, particularly in the areas of leadership and organizational studies. By

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 6 of 64
shifting the focus from the individual to the organizing process itself, a key value of complexity
is its strong implications for practice: Individual action must be considered in nonlinear systems
terms. In the fast-changing global ecosystem, approaches to management grounded in linear
assumptions may overly emphasize applying controls on interactions, thus failing to stimulate
information flows, learning and growth. New techniques that exploit nonlinearities and embrace
fast-paced interaction are needed.
To explain this, we begin with some of the challenges a complexity paradigm brings to
the field. We then describe how complexity thinking is applied to theorizing about leadership,
the growing empirical support for this approach, and the new methods that might change the
research process going forward. Because one of the key ideas of complexity is the continual
unfolding of newness, we conclude by looking ahead.
COMPLEXITY BRINGS CHALLENGES
The paradigm shift (Kuhn, 1962) towards complexity in leadership research brings with it
certain challenges, particularly to those who have traditionally seen leadership as something to
be admired in, or executed by, especially gifted or specially trained individuals. In this mindset,
individual “leaders” cause things to happen. As these traditional observers might see it, when
organizing is needed, that is, when one observes that leadership is necessary, this “leadership
vacuum” is translated into the idea that someone, some person, should “step up” to “take charge”
causing something to happen (MacGillivray, 2010).
With complexity, however, the scenario is different. While the need for leadership
remains, causality—at least much of the causality that really matters—is assumed to be indirect
and diffuse (Streatfield, 2002). When circumstances require leadership, individual observations
and experiences, connections and shared values, relative status, and the interaction dynamics

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 7 of 64
themselves, force the issue until leadership structures that cross levels-of-analysis emerge
through the constructive actions of individuals (Goldstein et al., 2010). These leadership
structures provide downward influence on individual interactions even as they are themselves the
emergent outcome of those same interactions, in a dynamic process that complexity pioneer
Haken (2006) calls circular causality. Certainly people “step up,” but they are not seen as
providing leadership for others. Rather, they are seen as being drawn into the leadership process
along with others. The system properties that begin to emerge are then sustained, evolved or
replaced within the system of interaction, as leadership unfolds dynamically (Panzar, Hazy,
McKelvey & Schwandt, 2007).
One difference in this perspective stems from the fact that complexity science has found
that the order we observe and are able to create in the world manifests itself simultaneously at
multiple levels of scale. Not only do unique dynamics unfold within individual interactions, but
also at the group level, the department level, the firm level and the institutional level. Each of
these levels, provides feedback to all of the other levels, influencing the dynamics of the others.
These changes in turn feed back once again to the other levels, and so on in an ongoing adaptive
spiral.
In short, feedback is all over the place and in all kinds of directions, making it
exceedingly difficult to meaningfully control events or to cause specific outcomes in the
traditional sense (Tobin, 2009). This would seem to account for the common practice in
management to attempt to control events by containing the flow of information and thus limiting
all diverging nonlinear effects that often accompany reinforcing feedback in complex systems.
Because a lot is happening at once, this logic goes, if one doesn’t control events and information
flows there can be unintended consequences, and it is better to stop them before they happen

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 8 of 64
rather than risk that they might challenge one’s assumptions about strategy and direction. CSLT
suggests ways for thoughtfully relaxing control and letting constructive deviations build upon
their successes, while identifying and dampening destructive deviations before they threaten to
pull the organization apart.
A note on terminology: To help clarify for the reader where we are focusing discussion in
this paper, we will follow Gell-Mann (2002) and use the term “fine-grained” to refer to
individuals interacting with one another, and “coarse-grained” to refer to the higher scale
properties of organizations that are of interest to leadership researchers, such as profits,
employee turnover, regulatory regimes, or even the leadership capability itself. In truth,
however, the experience of leadership floats between the fine- and the coarse- grained levels,
crossing scale and effectively becoming scale-free (Boisot & McKelvey, 2007). Organizational
life thus challenges people to act both individually and collectively at the same time in the face
of this complexity. We describe some of these challenges next.
Stability and Attractors: Coarse-Grained Prediction with Fine-Grained Uncertainty
One of the key insights from complexity science is that organizing, and thus leadership,
drives change at the fine-grained level of individual human interaction. Leadership does this by
changing the rules that govern the nature of connections and exchanges between individuals.
From these fine-grained interactions, persistent patterns emerge as coarse-grained system
properties that are sometimes quite stable at a higher level of scale. For example, warehouses and
logistics processes that at one point emerged almost by chance may remain at the center of stable
activities for months, years or even decades, attracting behaviors to them (Allen, 2001). Stable
properties like these can be recognized and modeled as organizational capabilities (e.g., the
logistics example we mentioned, but also accounting or customer service) (Dosi, Nelson, &

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 9 of 64
Winter, 2000; Helfat et al., 2007). To change these coarse-grained properties of organizations,
however, one must first change the rules governing fine-grained interactions of the people who
are implementing them.
CSLT defines leadership in complex adaptive systems as the social process that changes
the rules of interaction across levels of analysis, i.e., among individuals, work groups,
departments, organizations, and institutions. The leadership meta-capability in particular, is
defined as the routines, knowledge management, and decision making processes (Helfat et al.,
2007) that serve the coarse-grained function of changing the rules of interaction inside the
organization or the broader system (Hazy, 2004, 2006; Hazy, Goldstein & Lichtenstein, 2007b).
Thus, coarse-grained properties are changed only indirectly: they change when the rules of
interaction are changed at a finer-grained scale. This is an intuitive result: one must change the
way that people do what they do in order to change the outcomes they produce. CSLT seeks to
discover and specify the mechanisms, both direct and indirect, that enable this top-down/bottom-
up iterative process in real organizations, as well as how individuals learn to recognize and
become proficient in enacting this capability.
As alluded to above, stable coarse-grained properties are often associated with a
dynamical attractor within the system of HID, like for example, the logistics capability
described previously (Allen, 2001; Hazy & Ashley, 2011). This complexity term means that
there is a subset of all possible coarse-grained states of the system such that the state is “sticky,”
meaning that these attractor states effectively “pull” the system back to their original state if
something disturbs its normal functioning. After a storm that destroys some logistics equipment,
for example, activities are “automatically” enacted to return to the “normal” state. As an even
more general example, firms that maintain their profitability can be stable for a time. When

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 10 of 64
profits are threatened, they take action to return to profitability. At the same time, firms that lose
money will eventually implode or disband. They are not dynamically stable.
Patterns of activity that enable stable profitable operations (i.e., “exploitation,” March
1991) form an attractor of coarse-grained properties. These in turn drive what is happening at the
fine-grained interaction level. As individual behavior patterns and choices at the fine-grained
level converge towards a pattern of interaction they reinforce the coarse-grained organizing form
in what amounts to an iterative stabilizing feedback process that extends back and forth across
levels of scale (Hazy & Ashley, 2011). Convergent interactions that enable stability and thus
predictability are achieved through information feedback processes, wherein coarse-grained
structure provides information to individual agents, and their actions in turn influence the
specific characteristics of the coarse-grain structures in order maintain their relative stability
(Hazy, 2011b).
These feedback loops, both positive or reinforcing and negative or stabilizing, can shape
the emergent, dynamically stable coarse-grained state of the system. In turn, coarse-grain
stability can imply fine-grained choices and actions, such that their emergent outcomes become
to some degree predictable in the aggregate even though any particular event remains difficult to
predict, a circumstance that is called statistical complexity (Prokopenko, Boschetti, & Ryan,
2009). For example, one roughly knows what to expect when entering a retail store, a coarse-
grained structure. At Wal-Mart, a greeter will greet you somewhere with high probability. At the
same time, there is no way to predict exactly when or where such an event will occur. Thus,
although there is a level of stability and predictability at the coarse-grained level, there is always
unavoidable uncertainty at the fine-grained level. Individual interactions are neither random nor

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 11 of 64
completely predictable. They are constrained, but not determined, by attractors (Hazy & Ashley,
2011; Hazy, 2011b).
Emergence: Fine-Grained Prediction with Coarse-Grained Uncertainty
On the flip side of the comfortable stability and general predictability of familiar coarse-
grained aspects of organizational life is the reality that things sometimes change. When this
happens, old coarse-grained institutional structures must change as well. This implies a
conundrum: If coarse-grained properties are recognizable patterns that emerge from within
individual interactions, how does adaptive change actually happen at the coarse-grained level
when human interaction is experienced and predicted at the fine-grained level? In evolutionary
systems, this occurs through the process of variation, selection and retention among genetically
related but distinct entities over many generations. In cognitively-enabled systems, this occurs
through an intra-generational learning process whereby organisms learn to respond to stimuli in
their environment in a single lifetime. CSLT offers a framework that describes how
organizations both evolve through variation, selection and retention over many generations and
also learn to adapt within a single organization in a given generation. How well organizations
learn impacts their ability to survive and thus to contribute to the evolution of organizational
forms (Hannan & Freeman, 1989).
The process wherein organizing forms evolve and learn is the purview of a key area of
complexity research: emergence. Much has been written about emergence as a general matter in
complexity (Goldstein, 2007, 2010; Lichtenstein & McKelvey, 2007) as well as with regard to
leadership (Plowman et al. 2007a,b; Lichtenstein & Plowman, 2009). The idea is that under
certain exogenous constraints, a changing system of fine-grained interactions can cause the
emergent coarse-grained properties that are observed to undergo a qualitative transformation in

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 12 of 64
their coarse-grained patterns and structure. Examples of this phenomenon that are taken from the
natural sciences—such as the phase transition from liquid to gas—have provided metaphorical
insights for leadership researchers.
One important natural science example of emergence is the appearance of what are called
dissipative structures (Prigogine, 1995; Prigogine & Stengers, 1984; MacIntosh & MacLean,
1999). An occurrence of this phenomenon is described in some detail by Goldstein et al. (2010)
to illustrate how coarse-grained structures emerge in quasi-closed systems of fluids during what
are called far-from-equilibrium conditions. In the case described, this type of emergence happens
when heat is continually applied to the bottom of a closed container of liquid. As the intensity of
heat crosses a certain critical threshold, internal fluctuations interact with one another in the
presence of the exogenous force of gravity to cause an observable qualitative shift in the coarse-
grained behavior of the system.
The system’s dynamic behavior rather suddenly shifts from a relatively calm state where
heat is transferred by the mechanism of conduction to an orderly state of circular flow that
transfers heat through the mechanism of convection. As this occurs, the emergence of persistent,
coarse-grained hexagonal convection cells can be observed. Prigogine (1995) calls these
“dissipative structures” because convection dissipates heat more quickly than conduction does.
The onset or settling down of this qualitative change in structure can be “toggled” by the
experimenter by increasing or decreasing the heat applied to the bottom of the container. Thus,
changing patterns in fine-grained interaction behavior among the molecules can be seen to relate
to a qualitative change in the coarse-grained properties that emerge in the system, and these are
themselves shaped by external constraints on the system, the shape of the container for example.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 13 of 64
Another example of emergence relates to what Haken (2006) calls “order parameters”
that emerge through a process of circular causality (Haken, 2006). Using the example of a laser,
Haken shows how under the right conditions the light waves emitted by individual atoms
eventually synchronize their phases as represented by an order parameter. This occurs as
constructive interference reinforces a particular phase determined by interaction with external
constraints while destructive interference dampens others that are not reinforced by the
environment. This means than bottom up processes can lead to emerging order. And emergent
order places downward pressure on bottom up events, “enslaving” them to be in phase with the
order parameter, a process called “entrainment.” This is what is meant by circular causality.
Related to the above, is the subset of systems whose order parameters characterize phase
transitions. This well-studied phenomenon of circular causality in natural systems involves a
change in physical state as energy that is being added to the system modifies the internal
structure of the system to change how the system processes energy and information. Water
freezing, iron becoming magnetized, and the onset of superconductivity are examples of this.
Well-established mathematical models describe how external constrains—for example, ambient
temperature and pressure in the case of water freezing—interact to influence the “order” that
emerges within the system of interactions. The progress of these changing dynamics is described
by mathematical models that relate an order parameter to the change in state (Goldstein et al.,
2010).
Common to these examples is the idea that the breakdown of order within extant coarse-
grained properties is a prerequisite for emergence. This condition has been called the “edge of
chaos” (Kauffman, 1995; Mitchell, Hraber, & Crutchfield, 1993), “far-from equilibrium”
(Prigogine, 1995; Meyer, Gaba, & Colwell, 2005) and “criticality” (Bak & Paczuski,1995) in

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 14 of 64
varying complexity situations. To emphasize the dynamic nature of this condition and of its
potential to enable a qualitative transition from one stability regime to another, we prefer the
Goldstein et al. (2010) term for the onset of these conditions: criticalization.
The term criticalization highlights the potential for change when a certain parameter
crosses a critical threshold, what is called a bifurcation in mathematical modeling. In the case
where complex systems of human interaction dynamics (HID) experience adaptive tension that
pushes the system beyond a critical threshold, Uhl-Bien, Marion, and McKelvey (2007) describe
the requisite complexity of the system as a prerequisite for the onset of emergence. The
interaction between the complexity present in the environment and that which develops within
the system has been codified by Boisot & McKelvey (2010) in what they call “Ashby Space,” to
recognize Ashby’s law of requisite variety (1956).
The challenge when going beyond complexity metaphors to develop a theory of
emergence in human systems is to explicate the emergence phenomenon in the HID case while
taking into account the differences inherent in human interactions when compared to physical
systems. Prietula (2011) describes some of the practical differences in the context of agent-based
modeling. Hazy and Ashley (2011) explore the implications of these differences when
developing a theory of emergence in HID.
According to these researchers, the difference boils down to this: In contrast to physical
systems where many agents of a particular class (like water molecules) interact with one another
in the same way, human interactions are heterogeneous, each being determined by individual
preferences, personal histories, social connections and perceived difference in power and status,
all of which are stored in the individuals’ memories. Prediction becomes problematic since these
memories are largely hidden from the observer.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 15 of 64
Further, human connections are interdependent rather than independent since individuals
incorporate into their choices not only information from their direct experiences, but also
information and knowledge that is received through communication with others. This
interdependent heterogeneity implies that many of the statistical methods used in the natural
sciences are not appropriate in the human case since traditional methods assume independence
and consistency across time. In many cases, neither of these assumptions is valid with human
beings (Hazy & Ashley, 2011). As a result, the development of a well-specified and robust
model describing the mechanisms of emergence in HID remains an ongoing challenge for the
field (Hazy, 2011b).
Complexity: A Journey from Novelty to New Paradigm
In the late 1980s and 1990s, complexity thinking for organizations had not yet come into
sharp focus. Still the application of complexity was all the rage within the social sciences
(Anderson, 1999; Cilliers, 1998; Dooley, 1997; Goldstein, 1989; Levinthal, 1997; McKelvey,
1997; Thietart & Forgues, 1995). Exotic concepts that were discovered in the natural sciences,
like chaos, strange attractors and the possibilities implied by emergence - where entirely new
order springs forth seemingly in whole cloth - led researchers to look for complexity applications
in management and human organizations (Wheatley, 1999). Initial interpretations of complexity
often led to an unfortunate tendency to recommend to practitioners a version of laissez-faire
leadership - arguing that a hands-off style was all that was needed and that employees would
simply “self-organize” to solve business problems.
In contrast, Marion and Uhl-Bien (2001) saw the promise of complexity as beyond
metaphoric and suggested specific areas for theoretical exploration and empirical research. This
article follows in their direction and that of others (Schneider & Somers, 2006) and reviews what

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 16 of 64
has happened since these early days, and it explores what these activities might mean for the
future of leadership research and practice.
Ontological and epistemological issues
Relevant to the complexity framing of business and organization has been the distinction
between ontology and epistemology. The philosophic challenge is to determine the extent to
which the complexity mindset is a reflection of what is real in the world (that is identifiable
through observation - ontology) versus the extent to which complexity is just a new or different
way of knowing or understanding what is happening or perceived to be happening in the world
(Boisot & McKelvey, 2010). In this latter way of thinking, complexity is just an analytical “tool-
kit.”
The distinction is not a trivial one. It exposes the question of how human beings deal with
experienced complexity. As this paper explores, some interesting questions this implies include:
Is the world a complex system? Is it essentially a computation that is unfolding (Dooley, 2007;
Richardson, 2010)? If so, what are its algorithms? Are there better, more complex ways to model
and predict the world? How does complexity science inform the study of cognitive
neuroscience? What about the study and practice of psychology? And of course, what can it tell
us about study of leaders and leadership? These concerns can be considered from various
philosophical perspectives, for example, as constructionist, constructivist, objective realism,
critical realism, etc.
From the practical perspective of epistemology, the increased use of nonlinear techniques
and probabilistic prediction models in management are aspects of the complexity revolution in
the social sciences that have already established their value in practice. Monte Carlo analyses,
information theoretic approaches, game theory modeling, and system dynamics approaches are

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 17 of 64
all now in the mainstream decision support tool-kit in business. All of these are complexity-
informed tools that support managers in their decision-making. The adoption of these tools,
presumably because of their usefulness, supports the notion that ontologically, human
organizations appear to act like complex systems. What is missing is a holistic modeling
approach - something analogous to a statistical mechanics of human interaction dynamics (HID)
- that would represent and study organizations using complexity theories framed in a useful
epistemology. Such an approach would be invaluable to practice as it would allow managers to
forgo their need to apply controls, which can dampen both learning and growth in the service of
a perceived sense of stability and predictability, a practice that can be counterproductive.
One additional point is relevant here. There are particular ontological implications of the
complexity notion of emergence. Does an organization exist ontologically, in other words, does
it have agency? There is an argument that insect colonies do in fact exhibit ontological agency.
Tens of thousands of honeybees swarm as a collective to find a new nest without central control.
Choosing a proper nest is critical for survival of the swarm, and thus for evolutionary adaptation.
This is also true for the individual bees, all of which carry the DNA of the queen (Seeley, 2010).
Can the same argument be made for a firm or a nation? Does the diversity of DNA in human
organizations make the super-organism argument untenable (Nowak, Tarnita, & Wilson, 2010)?
This is a fundamental question for leadership researchers. Stated differently, is
anthropomorphizing the organization wrongheaded, or forward thinking? Do individuals actually
“lead” organizations, or is something else going on, something akin to the emergence of
collective agency?

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 18 of 64
COMPLEXITY APPROACHES TO LEADERSHIP RESEARCH
With the above background, we now turn to the various applications of complexity to
leadership. Although there are copious studies that apply conceptual ideas to leadership
metaphorically, such as the idea of fractals (Levick & Kuhn, 2007), the onset of chaos (Brown &
Eisenhardt, 1998), emergence and attractors (Shoup & Studer, 2010), and sensitivity to initial
conditions (Peterson & Meckler, 2001), others have cautioned against such metaphorical
applications (Simpson, 2007; Goldstein et al., 2010). Moving beyond metaphor, we focus only
on the approaches that uniquely assume a complexity ontology or that apply one of the
complexity inspired epistemologies: agent-based modeling (ABM), dynamic network analysis,
nonlinear dynamical systems (NDS) modeling, or information theoretic framings such as game
theory, and non-Gaussian statistical methods. (For a description of these methods see, for
example, Hazy, Millhiser & Solow, 2007, Boisot & McKelvey, 2007, and Guastello, 2002).
Because we are for the most part assuming a complex systems perspective
i
, presumably the
studies of emergent properties within HID would unfold in a manner analogous to dynamical
systems models of weather patterns, ecological models or epidemiology studies, or in the way
that agent-based models represent conditions of criticalization or uncertainty that signal an
impending phase transition, for example when a system changes its properties qualitatively—as
when water freezes into ice.
As Goldstein et al. (2010) point out, in human systems, situations of “criticalization” occur
when conditions in the environment combine with the state of the organization to create
uncertainty and unpredictability about where things are going. Researchers studying knowledge
generation in organizations describe the unfolding of plans and strategies during everyday
practice (Tsoukas, 2005). Often different individuals are observed espousing multiple different

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 19 of 64
plausible futures, and individuals are left to decide which approach is likely to “win” in the end
(Hazy & Ashley, 2011). Other authors have referred to conditions like this as “disequilibrium”
(Lichtenstein & Plowman, 2009) and identified these conditions as a prerequisite for emergence.
Given how little is known about the relationship between fine-grained interactions and the
emergence of coarse-grained properties (Hazy, 2011b), we first explore how leadership can be
relevant at the coarse-grained level. We begin with an overarching model as context for the other
approaches we will describe.
The Leadership and Capabilities Model (LCM)
Although many articles begin with a statement that the authors are assuming organizations
are complex adaptive systems (Boal & Schultz, 2007; Brown & Eisenhardt, 1997; Levinthal,
1997), their use of complexity is usually limited to the assumption that semiautonomous agents
interact, and that organizations somehow emerge from this process. Little has been written about
the course-grained properties that emerge, how they emerge, and what leadership has to do with
this. The Leadership and Capabilities Model (LCM) developed by Hazy (2006; 2011a) addresses
this gap by explicitly describing human organizing as a complex adaptive system of interactions
that performs certain functions.
To address the cross-level definitional issue with regards to the term “leadership,” Hazy
(2006, 2011a) builds upon Katz and Kahn (1966) to define “complex systems leadership” to be a
special kind of organizational capability (Dosi et al., 2000; Helfat, et al. 2007) that performs
particular system functions wherein the human organization is formed and evolved as a complex
adaptive system. In particular, complex systems leadership is the organizational capability that
iterates changes to the system’s configuration (by changing local rules of interaction) to test its
performance in the environment. As such, complex systems leadership is not what individuals

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 20 of 64
do; it represents an organizational capability in the sense described in business strategy (Helfat et
al., 2007): it is a set of routines, knowledge management, and decision making processes that
perform leadership functions in the same way that marketing and accounting are capabilities that
perform more instrumental system functions. All of these capabilities, including leadership, are
enacted by individuals and collectives in furtherance of functional requirements as the system
gathers and processes information, resources and energy to create order. Because complex
systems leadership reconfigures other capabilities, Hazy (2006) calls it a meta-capability.
Organizations with well-developed complex systems leadership meta-capabilities are able
to iteratively act on the system itself, changing the rules governing fine-grained interactions
within the system in response to success or failure (Hazy et al., 2007; Hazy 2011b). The coarse-
grained properties that emerge enable subsequent success of the system in changing and adapting
as the situation develops. The complex systems leadership meta-capability guides this process.
The idea of complex systems leadership as a meta-capability extends the work on
organizational capabilities, both dynamic and operational (Barney, 1991; Helfat et al., 2007;
Nelson & Winters, 1982; Teece, Pisano, & Shuen, 1997). However, in contrast to the capabilities
literature, individuals in the organization aren’t actualizing these capabilities; rather they are
enacting the leadership meta-capability of the organization that is acting on the system to test
configurations of other capabilities, and to identify those that work better than others.
Using the complex systems agent-based epistemology of system dynamics, Hazy (2004,
2006) built a system dynamics model of the leadership meta-capability that is called the
leadership and capabilities model (LCM). When the model is run, the leadership meta-capability
performs an iterated operation on the coarse-grained properties within the system that: i)
implements the exploitation (March 1991) of current capabilities, ii) promotes the exploration

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 21 of 64
and generative process of new capability creation (March, 1991), and iii) unifies the system to
maintain it as an entity with regard to local and global criticalization conditions as required by
the environment. It changes the properties or capabilities that have previously emerged, and
presumably it does this by changing local rules of interaction among individuals. By changing
the rules, the properties characterizing the system, including its capabilities, also change.
Depending upon the context, the complex systems leadership operation acts on the system
to perform three functions: The convergent operation within the system adjusts the properties of
the system to make them more predictable (Hazy & Ashley, 2011) and thus improve
exploitation. Rules are changed to dampen deviations by increasing individual productivity and
leveraging cooperative activities with technology and other assets. This is called the exploitation
value-gathering loop (Hazy, 2011a).
The generative operation responds to changing constraints in the environment and promotes
exploration, collaboration, creativity and innovation in system properties (Hazy, 2004, 2006;
Surie & Hazy, 2006; Uhl-Bien et al., 2007). If changing constraints on the system implies that a
qualitative change in coarse-grained properties is needed, the system often passes through a
period of criticalization as requisite complexity is engendered. For this to occur, fine-grained
rules of interaction are changed to promote experimentation and to reduce or eliminate premature
convergence. Because a variety of possible futures coexist, this is a manifestation of Ashby’s
(1956) notion of requisite variety. The exploration value-identifying loop enables adaptation
(Hazy, 2011a).
The unifying operation uses communication and symbolic activities, such as policies and
boundary rules for the proprietary use of information, to more clearly specify acceptable and
expected rules for system properties by promoting locally stable collective identities and systems

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 22 of 64
of ethics (Hazy, in press-a). This tunes the level of criticalization and requisite complexity (Uhl-
Bien & Marion, 2009) both locally and globally within and across the organization.
One implication of the LCM is that, as is the case for other organizational capabilities,
individuals learn to implement aspects of the leadership meta-capability in the same way they
learn to implement manufacturing, customer service, business planning, or any other capability
(Hazy, 2007a, 2008b). The acquisition of leadership skills by individuals is the result of social
learning of the meta-capability within organizations, just as the acquisition of marketing skills
results from social learning in marketing organizations. Individuals are not born to leadership
competency; they learn how to exercise this capability by being involved in organizing efforts
that exhibit the coarse-grained property of an effective leadership meta-capability.
Leadership and Emergence
Given how little is known at present about the emergence of coarse-grained properties in
the unique manifestation of the distinctly human social context, it is at present difficult to offer
specific suggestions about how any given individual should behave. The process of emergence
has been explored, however. Lichtenstein and Plowman (2009) draw on case studies of
emergent leadership in organizational settings, the case of a mission church (Plowman et al.,
2007 a, b) for example, to argue that there are four phases in the emergence process. First there
must be disequilibrium, unstable conditions that were described in the earlier section as
criticalization. During these periods of uncertainty, reinforcing feedback is offered to certain
process fluctuations through amplifying actions that seem to offer promising new ways to bring
back stability. Hazy (2011a) calls these experiments in novelty “constructive deviations” because
they deviate from what had previously been the norm, but they do so intentionally and
constructively
ii
The constructive deviations that are working are then combined with other

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 23 of 64
fragments of collective activities that are at work in other parts of the organization in a
recombination process. In this way, new structures grow through the gradual accretion of
constructive deviations that work. Eventually, stabilizing feedback that operates on this new way
of doing things—for example, lack of funds or time limitations, or a saturated market—bring the
organization back into a stable, albeit qualitatively different, approach to organizing. In the end,
one can observe that distinctly new properties (at the coarse-grained level) have emerged.
As further support for this classification, Beck & Chong (2009) identified these stages in
social entrepreneurship ventures they studied in Indonesia. Aspects of these phases were also
apparent to Butterfield, Shepherd and Woods (2011) in a social enterprise that was studied in
New Zealand, and Baker, Onyx and Edwards (2011) found evidence of recombination of
network components in a developing community of social enterprises in Australia. The next
stage of CSLT research in this area is to uncover the specific mechanisms at work in each of
these stages, that is, the mechanisms of emergence at work in HID (Hazy, 2011b).
A series of research projects in Sweden have provided some hints about these mechanisms.
These studies explored how first-line managers provide the pre-conditions to influence both the
interactions between individuals and the collective organizing which emerges from them
(Backström, 2009; Backström, Hagström, & Göransson, 2011; Backström, Wilhelmson, Olsson,
Åteg, & Åberg, 2011). Backström and his colleagues refer to this aspect of management as the
“directing task” which focuses on enabling a dynamical balance between the autonomy of
individuals and their integration into the emerging organizing structures. The communicative
competence of the employees and the reasons, places and times for them to meet and to
communicate were found to be important conditions for the emergence of structures like

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 24 of 64
collective culture and identity, institutionalized collective behavior, and patterns of work
relations (Backström, Wilhelmson et al., 2011).
As the above discussion highlights, at present it may be theoretical overreach to posit that
any one person is in control of the emergence process whereby the coarse-grained patterns that
characterize organizational capabilities come into being, at least as regards what is currently
known about complexity (Hazy, 2008b). The assertion that complexity implies a good deal of
ambiguity when linking fine-grained action to coarse-grained properties has found support in the
literature (Siggelkow & Rivkin, 2005). Morrison (2010) showed how the process of gathering
information, analyzing it, recommending approaches, and implementing projects can be a very
challenging undertaking for managers. When this occurs within a network of positive and
negative feedback loops, as is often the case, the challenge is even more daunting.
There is a point where predicting the resulting outcomes in the face of this nonlinearity is
problematic (Siggelkow & Rivkin, 2005; Morrison, 2010). Interdependent, heterogeneous
connections between individuals as they interact seem to make each case unique. This is not to
say these dynamics are not understandable in a general way, only that they are not yet
understood. We next describe work that focuses on discovering additional coarse-grained and
fine-grained leadership activities needed in organizations when they are considered to be
complex adaptive systems.
Complexity Leadership Theory
An approach that is compatible with, but varies somewhat, from the LCM model just
described is Complexity Leadership Theory (CLT) (Uhl-Bien et al., 2007; Uhl-Bien & Marion,
2009, 2010). Complexity leadership theory draws from concepts of complex adaptive systems
(CAS) to offer an organizational theory (OT) framework of leadership.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 25 of 64
Consistent with OT (Barnard, 1938; March, 1991; Selznick, 1948), complexity leadership
theory identifies two primary functions in organizations: an administrative function and an
adaptive function. The administrative function is associated with the bureaucratic structures of
the organization, and focuses on productivity and efficiency through choice, execution and
variance reduction to enhance efficiency and effectiveness (i.e., “exploitation,” March, 1991).
The adaptive function is associated with CAS structures and dynamics in organization, and
engages individuals and organizations in search, experimentation, and variation to enhance
creativity and learning (i.e., “exploration,” March, 1991).
CLT also identifies leadership forms associated with these functions. Administrative
leadership manages in the formal systems and structures (i.e., the administrative function) to
produce results through alignment (e.g., standardization) and control (e.g., visioning, motivating,
inspiring, directing, controlling). Adaptive leadership is a distributed form of leadership
generated in the informal structures and systems of the organization (i.e., the adaptive function).
It is associated with the bottom-up emergent dynamics within organizations. Adaptive leadership
is defined as a dynamic, interactive influence process among individuals and collectives for
which the objective is to lead one another to adaptability, innovation and learning (Uhl-Bien &
Marion, 2009; 2011).
Recognizing that the administrative and adaptive functions often operate in tension with
one another (Boisot & McKelvey, 2010; Heifetz & Laurie, 2001), and that organizations operate
in states of dynamic equilibrium (Boisot & Child, 1999; Davis, Eisenhardt & Bingham, 2009;
Galunic & Eisenhardt, 2001; Meyer, Gaba & Colwell, 2005; Stacey, 1995), CLT adds a third
leadership function called complexity leadership. Complexity leadership (formerly referred to as
enabling leadership) works to achieve requisite complexity through a dynamic combination of

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 26 of 64
loosening-tightening the administrative function and enabling-inhibiting the adaptive function
(Havermans, Den Hartog, Keegan, & Uhl-Bien, 2010; cf. McKelvey’s 2010 discussion of
complexity leadership action disciplines).
Complexity leadership differs from administrative leadership in that administrative
leadership is only able to affect the former (loosening-tightening the administrative function)—it
is not associated with the adaptive function. It is because of this that current frameworks of
leadership often fall short for practitioners in complex environments (IBM CEO Report, 2010).
Traditional leadership theories are grounded in assumptions of hierarchy and authority (i.e., the
administrative function) that do not recognize complexity dynamics or adaptive leadership
associated with the adaptive function. Therefore they are inadequate with respect to how to
engage the adaptive function, manage in and for emergence, or enable the requisite complexity
required for firm fitness and adaptability (Uhl-Bien & Marion, 2008).
In including both organizational-level (administrative) and micro-level (adaptive)
leadership behaviors, CLT addresses both coarse-grained and fine-grained leadership activities.
Coarse-grained leadership is considered in the administrative function and administrative
leadership roles that occur in organizational structures and systems. Fine-grained is considered in
the adaptive function and adaptive leadership roles and dynamics that are associated with
leadership and emergence. Complexity leadership helps enable transition between fine-grained
and coarse-grained by both operating in the interface between the adaptive and administrative
functions (Uhl-Bien et al., 2007; Uhl-Bien & Marion, 2009).
Leadership and the Implementation of Strategy: The Unifying Process
One of the anomalies that confronts organization and leadership researchers is the
perception that individuals, and in particular executive managers, are in charge of

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 27 of 64
organizations—they set the directions, they make things happen. As we can see from the above
discussion, this does not square with a complex system viewpoint that organizing patterns
emerge through interactions. One point of synthesis, however, is the realization that there are
variations in the roles and the impact that individuals have in the system—i.e., the influence of
all individuals is not equal. Status and reputation differences have significant impacts on
outcomes (Axelrod, 1984; Simon, 1997). For example, executives typically set the strategy and
direction for the organization; they define the organizations boundaries, its identity, and its
ethics, and they hold it together (Hazy, in press-a). In short, they drive the unifying process at the
highest relevant level of coarse-graining. Thus, contrary to the complex adaptive systems most
often studied in the natural sciences, it does appear that in the case of human interactions,
through intentional action that recognizes and seeks to exploit emergent patterns, individual
agents can make a difference even in properties that are coarse-grained.
Strategy, “Tags,” and the Unifying Process
For example, Boal and Shultz (2007) focus on the strategic aspects of leadership at the top
of the organization. Amid their discussion about how communication strategies and story-telling
are core to business strategy-making, they assert that individuals performing leadership functions
use the complexity idea of “tags,” as defined by Holland (1975), to compartmentalize large
organizations into differentiated sub-structures. They argue that this process is useful because, as
Holland shows, the use of tags in interactions enables individuals who share the same
assumptions and aspirations to identify one another. Tags thus promote higher levels of quality
interaction within same-tagged identity groups. The use of tags also results in lower levels of
interaction across groups with different tags, which can make communication more efficient.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 28 of 64
In a more general sense, the implications of Boal and Schultz (2007) suggest that tags are
useful in furthering the unifying process identified by Hazy (2011a). Tags can be thought of as a
unifying mechanism which clarifies and helps resolve identity tensions locally and globally, and
tunes criticalization conditions locally in the face of further tension between adaptive versus
performance concerns. In complexity leadership theory terms (Uhl-Bien et al., 2007),
administrative leadership exploits innovations generated by their adaptive leadership to benefit
the firm, which also seems to be relevant to the strategic unifying process Boal and Schultz
describe. Perhaps, the use of tags is a mechanism of administrative leadership that serves to
further the unifying process within the organization.
Unifying leadership across scale
One might note, however, that the unifying process requires that these activities occur
across the organization and are not limited to the actions of top management. They occur at all
levels of the organization. To further explore the mechanisms of individual influence within
HID, particularly when organizing large-scale efforts, Hunt, Osborn and Boal (2009) describe
the application of complexity to what they call Level VI leadership (Jaques, 1989), the level
below strategic leadership. For Jaques, Level VI implies individual cognitive capacity that
enables a task horizon and job focus of three to five years for mid-career executives.
Citing the Siggelkow and Rivkin (2005) results, Hazy (2007b) argues that at the senior
management level, successful leadership is more about picking strong subordinates and advising
and supporting them with resources than it is about choosing actual projects. The process
whereby these decisions are taken has recently been explored by Anneloes et al. (2011) in the
context of the nature and quality of the interactions between top managers and their middle
management subordinates who are responsible for directly delivering operating results. At this

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 29 of 64
level, it may be more appropriate to worry about picking the right individuals, i.e., the right
champions, who have proven they can pick and drive winners, than to try to identify winning
projects oneself.
Extending this line of thinking, Hazy (in press-b) argues that when reputation and status
accrue to individuals who are successful, and when these changes are publicized to the
organization’s members, the organization gets smarter as a system. In his model, relative status
and reputation determine relative influence among individuals in a manner analogous to the
synaptic weights connecting neurons in the brain. These weightings determine which neurons are
activated in response to stimuli in the environment. Similarly, relative status and reputations play
a role in determining who is involved and to what degree when an organization is forming a
response to events in the environment. In this way, the organization “learns” in a manner
analogous to the learning algorithm in the brain’s neural network.
As an example, a successful merchant bank in Sweden, with a record of success even
during the financial crises, illustrates a complexity-aware structure of this type. Each department
operates as its own company, and each employee has full responsibility for a group of customers.
The company’s culture unifies these activities as a major mechanism for integrating different
groups into the company. Activities such as appointing managers with values that closely match
those of the company, and ensuring high quality communication among individuals were shown
to be central for integration into a unified firm with a common identity (Backström, Hagström et
al., 2011). The bank uses a very simple feedback process at the departmental level and uses this
to clarify the relative success, and thus the relative status and reputations of departments (and
thus of individuals), by publishing a monthly ranking of all departments.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 30 of 64
Based upon these complexity ideas, Hazy (in press-b) argues that one of the most important
aspects of effective management is to focus on structuring projects so as to receive clear and
unbiased feedback about success or failure and to act on the feedback by adjusting the status and
reputations of those involved. Thus, an important imperative of leadership is to hold people to
account. This would be an element of administrative leadership (Uhl-Bien et al., 2007).
We turn next to the question of what it might take for an individual, a “leader,” to
differentiate his or her actions in order to successfully engage a system that has complex
characteristics.
The Individual in Complex Organizing
Core to the challenge of developing complex systems leadership theory is the unique role
played by individual human beings, particularly high status ones, or powerful ones, in complex
organizations (Denis, Langley & Pineault, 2000). Does complexity suggest what it takes to be a
successful individual “leader” in this context? Building on the work of Zaccaro (1999) and
Hooijberg (1996) this challenge has been taken up by Hannah, Lord, Jennings and others
(Hannah, Jennings & Nobel, 2010; Hannah, Woolfolk & Lord, 2009; Hannah, Eggers and
Jennings, 2008; Lord, Hannah, & Jennings, 2011) to explore the mental characteristics that
support success, as well as the potential for bias, for example, gender bias (Hogue & Lord,
2007). These issues become important to researchers for a number of reasons, one being their
implications to leadership development programs (Boyatzis, 2008).
As described earlier, Uhl-Bien et al. (2007) extend Ashby’s (1956) idea of requisite variety,
which says that in a cybernetic system the regulator must be at least as complex as the
environment it hopes to regulate, to argue that organizations must likewise match the complexity
of the environment, a condition called requisite complexity (Boisot & McKelvey, 2010). In

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 31 of 64
particular, they argue that complexity leaders must structure the organization to enable requisite
complexity. To further define this concept, Hannah, Lord, Jennings and others (Hannah et al.,
2010; Hannah et al., 2009; Hannah et al., 2008) have worked to define the characteristics of
individuals that enable requisite complexity in action. Most recently, Lord, Hannah, and Jennings
(2011) identified two dimensions of complexity, static and dynamic, the latter taking into
account the unfolding of complex conditions over time. By analogy, they extend the idea of
requisite complexity to groups (Lord, Hannah & Pearce, 2011). The processes of creativity and
innovation they describe require different skills, and support the system’s functional demand that
new information be gathered and synthesized in an ongoing generative process.
Innovation and Adaptation: The Generative Process
To sustain itself as an open system (von Bertalanffy, 1950), a complex system needs
continuing access to resources within its ecosystem (Hazy, Moskalev & Torras, 2009, 2010).
Since opportunities in an ecosystem ebb and flow, a complex system needs mechanisms to
explore the environment, identify resource-gathering opportunities and construct structures
within the system to begin to exploit these resources (Garud, Gehman, Kumaraswamy, 2011;
Garud, Kumaraswamy, & Sambamurthy, 2006). In short, the organization must balance
exploration and exploitation (March, 1991). Since ecologies change, this is not a once-and-done
process. This is why Hazy (2011a) calls the value-identifying loop a requisite function for
complex systems of human interaction dynamics (HID). This system requirement places a
functional demand upon the leadership meta-capability: to establish, evolve and regulate this
generative operation for the system by changing the interaction rules among individuals.
Garud et al. (2011) describe the generative environment at 3M as what they call
“complexity arrangements”:

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 32 of 64
We conceptualize the different combinations of practices – manifest structure (e.g.,
products, patents and platforms), relational processes (e.g., interactions between people
within and across platforms and businesses), temporal dynamics (e.g., moments of
serendipity enabled by the 15% option) and regulative guidelines (e.g., 30% stretch
objective) – that are activated at various stages of an innovation journey at 3M as
representing complexity arrangements (p.347).
These “complexity arrangements” enable a process of invention, but also innovation, and they
are enacted at the individual interaction level. CSLT explores the mechanisms that evolve the
rules that govern local interactions to enable the generative process of innovation and adaptation.
Divergence through discovery
A literal example of this might be the value-identifying loop associated with an as yet
undiscovered oil field in the Gulf of Mexico. Resources are allocated to exploration (March
1991), albeit with no a priori knowledge of the likelihood that a new field will be discovered at
any particular place. Once indications of potential are identified and the probability of success
increases (after a geological study, for example), locally specific capabilities must be constructed
to explore this possibility further. With additional positive indications, more resources are
allocated to take advantage of this potential, and so on, as long as the opportunity remains viable
(meaning the probability of success relative to risk remains high).
Complex system leadership evolves local rules of interaction to enact this process. As
experiments to acquire resources produce information, feedback (under promising conditions)
leads to significant expected value with regards the resources that could be discovered. This
positive feedback loop is generative of possible future ecological niches for the system.
However, if risks and rewards are not properly recognized and modeled, this feedback doesn’t

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 33 of 64
have to be beneficial for the organization. The unfolding impact of the BP Gulf Oil Spill in 2010
is a value-destroying example of the same positive feedback affect. Thus proper assessment of
information about risks and benefits is an important aspect of the interaction rules that are
evolved by complex systems leadership.
Events caught up in positive feedback loops can build upon themselves rapidly and their
effects can come to dominate the HID within an organization, as happened at Intel after the
discovery of the microprocessor (Hazy 2008a). Diverging value-identifying loops that build upon
themselves exhibit divergence through discovery because information about the opportunity
grows rapidly after discovery, often feeding upon itself. This information helps individuals to
organize the system to deal with rapid growth and expansion. In the above example, this means
that the petroleum industry ecosystem began as a dynamically stable system exploiting known
resources. Initially the system had no information about this particular oil field (or about the
implications of the Deep Water Horizon’s blowout in the negative case).
Upon discovery, and assuming continued investment of resources, change unfolds. For
example, researchers (Aasen & Berg, 2008;
Aasen & Johannessen, 2007; Johannessen & Aasen,
2007) used the complex responsive processes framework to explore the innovation process at a
subsea oil recovery case at Statoil ASA a Norwegian oil company. They observed that
“innovation emerges from the experiences of everyday social interaction, where patterns
gradually perceived as meaningful are created and adopted” (p. 44).
Using systems dynamic modeling, the divergence of discovery as a complexity
phenomenon was shown to describe organizational transformations at NCR (Hazy, 2006) and the
rise of the microprocessor at Intel (Hazy, 2008a). It was also described in for a technology
company (Hazy, 2008c), and in a case study reported by Surie and Hazy (2006). In the latter

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 34 of 64
case, an Indian manufacturing company was working with foreign partners who were seeking to
access production capacity to outsource their manufacturing. During the negotiation process, the
Indian company discovered and exploited entirely new global markets for its products. New
information about global markets gradually changed the Indian firm’s perception of their
potential customer sets. Their visible world diverged from a narrow, traditional, domestic-only
market, into the wider visibility of a truly world market.
Ecology of Innovation
What these examples have in common, according to Goldstein et al. (2010), is that they
occur in organizations wherein complex systems leadership has managed to position the
organization within an “ecology of innovation.” This means that the firm is well situated within
its network of suppliers, customers and partnerships and maintains excellent communication
across its connections so as to engender knowledge generation, discovery and thus a level of
divergence as new opportunities are discovered. Their point is that leadership occurs at the
nexus of interactions, where generative human dynamics lead to creativity and innovation
(Andriani, 2011; Beck & Chong, 2009; Garud et al. 2011; Garud et al. 2006; Tapsell & Woods,
2009). The key to this is in the quality of interactions, what Goldstein et al. (2010) call
interaction resonance, and what Garud et al. (2011) call relational processes.
In the Complex Responsive Processes (CRP) perspective, Stacey (1993, 1995) argues
that certain conditions are necessary for impactful interactions. These include: trust, the holding
of anxiety, power relationships that are both cooperative and competitive, and conversational
practices that don’t block explorations (Simpson, 2007, p.475). These interaction-level tensions
must be navigated to achieve a high level of interaction resonance. This in turn implies certain
leadership activities that are necessary for adaptation (cf. adaptive leadership in complexity

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 35 of 64
leadership theory, Uhl-Bien et al. 2007; Uhl-Bien & Marion, 2009). Complex systems leadership
evolves rules of interaction to enable interaction resonance.
Goldstein et al. (2010) go on to argue that conditions of highly resonant interactions,
bounded in the right way, can lead the system to criticalization in the complexity sense, where
the system is poised for a phase transition between two possible dynamic states: one less ordered
and thus more symmetric, and one more ordered or organized into specialties and thus less
symmetric. The transition from one state to the other can be modeled mathematically if an order
parameter can be identified to describe the change (Haken, 2006). Identifying the order
parameter for phase transitions in HID remains an open unanswered question in CSLT.
More specifically, on one side of the phase transition, the more ordered case, individuals
are oriented in certain roles and tasks, and are thus less interchangeable with others. Some are
marketers doing marketing things and are only interchangeable with other marketers; some are
accountants doing accounting things and can only be replaced by accountants, and so forth.
Because fine-grained elements of the system can be differentiated, there is ordering in the system
and thus there are fewer ways the system is self-similar. It is thus less symmetric. On the other
side of the phase transition, everyone does everything and thus, in a certain sense, everyone is
interchangeable with everyone else (ignoring for a moment individual heterogeneity). Such a
disordered system is thus more symmetric.
This is why qualitative change from one state to the other, called a “phase transition,”
represents a discontinuous change in the in the level of order in the system and this change is
often referred to as “symmetry breaking” (Haken, 2006; Guastello, 2002). Ecologies of
innovation enable symmetry-breaking qualitative change. Generative rules of interaction at the
fine-grained level are an enabling prerequisite to innovation reordering of this type (Hazy, 2009).

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 36 of 64
Performance and Efficiency
In addition to unifying HID into complex adaptive systems, and the exploration and the
generative activities that enable organizing in response to opportunities, the system must also
converge to stable operations that effectively and efficiently exploit the resources to which it has
access. Thus convergence is the third functional demand placed on the leadership meta-
capability as the complex adaptive system (as enacted by its agents) seeks to acquire its requisite
resources and to conserve the resources it has accumulated as slack (Cyert & March, 1963).
Convergence for preservation
This functional demand is needed to encourage efficiency and to preserve slack resources
(Cyert & March, 1963). In the process, however, potentially productive activities and
information about alternative approaches and opportunities are lost (Haken, 2006). The human
relations or “consideration” aspects of the process of convergence that brings people on-board,
and the “initiating structure” elements that enable action-in-concert have long been associated
with leadership (Fleishman, 1953; Stogdill, 1974). Individuals who exhibited either or both of
these behaviors are called “leaders” by others even though leadership is actually emerging from
within the interactions. The tendency to attribute leadership to individuals is a strong one that has
also been observed in the complex system leadership context (MacGillivray, 2010).
Complexity ideas and methods have added a perspective on the “how” behind the “what”
of these behaviors. Consideration and initiating structure by individual actions enable
convergence of a disparate group toward a single, common objective. Convergence of action
satisfies a functional demand of the complex system on the leadership meta-capability as defined
by Hazy (2011a). More specifically, Phelps and Hubler (2007) showed how groups could set and
choose direction when a single individual was sufficiently motivated to move in a particular

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 37 of 64
direction, essentially shutting out and forgoing other possibilities, including preferences of
others. In several experimental settings, the role of individuals in catalyzing convergence toward
a particular outcome has also been shown to follow distinctively nonlinear patterns of behavior
by Guastello and his colleagues (Guastello, 2002; Guastello, 2007; Guastello & Bond, 2004,
2007a; 2007b; Guastello, Craven, Zygowicz, & Bock, 2005; Guastello & Guastello, 1998).
Complexity researchers have also studied the impact of leadership on group member self-
selection across groups. Dal Forno and Merlone (2007) demonstrated both experimentally and
computationally that the complex dynamics associated with the distribution of rewards and
punishment by individuals on teams had a significant effect on which teams individuals chose to
join, and this in turn impacted the projects that were ultimately completed.
Information flow is a critical enabler of project selection and execution. Schreiber and
Carley (2006) used data from field observations and dynamic network analysis to study how the
flow of information within teams impacted performance. They found that multiple network hubs
in the flow of information, rather than a single one, that is, a single “leader,” led to better
performance when complex functioning was required. Similarly, Solow and Leenawong (2003)
used Kauffman’s (1995) NK model to show that too much complexity within teams can lead to
situations of overload that greatly reduced performance. Insight on the efficacy of a centralized
control model versus more distributed decision-making was also shown using complexity
modeling (Solow & Szmerekovsky, 2006).
Hazy (2007a; 2008b) summarized many of these studies into the theory of complex
systems leadership. Synthesizing the studies above as well as others completed at the time, he
observed that five distinct aspects of convergent leadership had been identified in these
complexity inspired studies. These five mechanisms comprise the leadership response to the

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 38 of 64
requisite system demand for collective convergence within HID toward action in concert. These
distinct mechanisms include actions or communications by individual agents which:
1. Espouse an approach or cooperation strategy for working towards a common
objective, a “program of action,” such that choosing to participate in the program is
an attractor for the individual choices of two or more agents (Phelps & Hubler, 2007).
2. Catalyze social influence conditions such that at least one other agent chooses to
participate in the program being espoused rather than continuing to act for its own
account or according to an alternative program (Phelps & Hubler, 2007).
3. Catalyze choices and action in other agents intended to navigate complexity and
overcome the limits to cognitive capacity in an effort to avoid an interaction
catastrophe. This is when there is too much interaction and confusion causing
performance to drop precipitously (Solow & Leenawong, 2003)
4. Form a distinct output layer (the Executive Office) that expresses learning and action
for the system as a unity in the environment; to do so, the agents disambiguate
learning and enable the unambiguous expression of action by the system in the
environment (Hazy, 2007b).
5. Process feedback information regarding success or failure of enactments in the
environment or internal to the system, and translate this information into structural
changes in the influence network among agents. This is done by changing the
reputation and status of participating individuals and thus changing their relative
influence (Hazy, 2007b).
Hazy argues that all of these aspects of leadership are facilitated either directly or
indirectly by changing the rules of interaction. A theoretical framework for how these leadership
mechanisms are actualized within a group of individuals in an organizational setting is discussed
next.
Microenactments within a CAS
By analogy with insect swarming behavior, an emergent process that has been studied
intensely (Seeley, 2010), Hazy and Silberstang (2009a) suggest that individual interaction events
in groups likewise can be understood as signaling behavior among individuals that culminates in
a specific collective action - programs of action (PoAs) - independent of the specific content of
the action being considered. Whereas bees swarm in search of a new nest (Seeley, 2010), humans
“swarm” in an effort to act in concert with regards to any particular project being considered.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 39 of 64
Examples of PoAs around which people might “swarm” include the decision to disband a
meeting or any social gathering, to initiate a project, or just to meet again the following week. Or
they might involve taking a single step in a larger program: to launch as new product
development effort, for example. When the relative status and thus influence of the various
individuals involved is taken into account the authors posit that this dance of interaction is
actually how decisions are taken when the authority to make a decision is “in the room.”
To explore this proposition, the authors go on to describe individual signaling behaviors,
called “microenactments” (Hazy & Silberstang, 2009a, 2009b; Silberstang & Hazy, 2008). These
include signals that initiate a PoA, reject it, accept/imitate it, negotiate to modify it, or synthesize
it with others to form a qualitatively new project. Together these microenactments constitute a
language of interaction that can enable action in concert. For efficiency, this language unfolds as
separate enactments made by individuals, but they are experienced by others in the context of a
shared grammar that enables action-in-concert. Aspects of the grammar might include such rules
as when and how to propose new ideas (Beck & Chong, 2009), what constitutes a quorum that
enables decisions that are binding on the group (Phelps & Hubler, 2006), and when to accept the
decision of a quorum (Guastello, 2007; Goldstein et al., 2010).
Detailed research on the precise nature of the microenactments that individuals display as
well as the grammars that enable shared meaning within this communication system might lead
to a deeper understanding of human collective behavior. Although the palette might be distinct
for each class of problem, there may also be commonalities, and these might shine additional
light on the precise nature of leadership in daily practice. In the spirit of the interdisciplinary
nature of complexity thinking, this same problem has also been addressed using an agent-based
modeling framework by Panzar, Hazy, McKelvey and Schwandt (2007).

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 40 of 64
Developing fields, like the leadership-as-practice research where the focus is on the
dynamic activities of leadership rather than heroic individuals (Raelin, 2011), are linked to the
complexity way of thinking. Here we see mutual influence among individuals with common
objectives forming into collective action. Research into the precise nature of this process for
different classes of problems might lead to a deeper understanding of human collective behavior.
Independently, Mangiofico and Feyerherm (2011) have qualitatively identified many of the
activities described by Hazy and Silberstang in a non-profit organization. Likewise, research in
Shared Leadership can be viewed in this way (Lord et al., 2011).
A Contrarian Challenge to Systems Thinking - Complex Responsive Processes
The focus of the complex responsive processes approach to complexity is in
understanding the interpretive experience of the individual as human beings interact within
complex human organizing activity (Stacey, Griffin & Shaw, 2000; Griffin, 2001). This
approach distinguishes human complexity from the “systems approach” (von Bertalanffy, 1950;
Katz & Kahn, 1966) arguing, in effect, that human organizing is not appropriately studied using
systems models (Stacey, 1995). Instead, as Johannessen (2009) states:
Taking a complexity approach means that the focus of research attention is drawn towards
the exploration of the phenomenon of human interaction and emergence. Human
interaction is the cause of emergence, and human interaction only creates further
interaction and emergence. This is what is meant by a radical process view of reality (p.
217).
The implication of this framing is that the focus of research is on the unfolding of experience, on
the narratives themselves, rather than on structures or artifacts that can be modeled and studied
using systems approaches.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 41 of 64
Although coming from a different theoretical framing, the implications of this research
stream are not that different from the system-based approaches described in previous sections.
For example, in a case study Mangiofico and Feyerherm (2011) have integrated the notions of
complex response processes with the systems perspective of Goldstein et al. (2010) to identify
the four leadership skills that are implied by complexity thinking: i) perpetually scanning of the
ecology to identify flows of information and resources, ii) weaving webs of interaction among
actors within and across boundaries, iii) creating coherence among the signals that flow through
these networks, what Goldstein, Hazy and Lichtenstein call “interaction resonance”, and iv)
support for expanding innovation by offering stabilizing feedback to converge activity toward a
kind of dynamic stability that represents a new way of doing things. In a certain sense, these four
principles frame the practical implications of the new perspective that is offered by complexity
science.
The Emerging Complexity Paradigm
Figure 1 shows a new paradigm that is emerging. Complexity science provides
conceptual tools for thinking about organizing (left box in the figure) and identifies key system
demands that must be met if a CAS is to be effective at both performance and adaptation (right
box). Three organizing mechanisms and their relationship to the leadership process were
described in this chapter (center box). These can be used by managers to drive organizational
outcomes. At the bottom of the figure we identify the challenges that all of this complexity
brings forward for individuals who are asked to manage or to lead. These include complexity
within a given situation such as during a large-scale project, as well as complexity that results
from changes over time, as might be the case as technology advances more rapidly than ever.
Each of these areas has implications for practice and begs further research.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 42 of 64
This new paradigm is a significant contribution to leadership research because it begins
to connect individual behaviors and actions to system processes, functions and outcomes. To do
so it explores the complexity dynamics within fine-grained HID in areas such as criticalization,
and through the study of emergence, it also sheds light on how these might result in beneficial
changes to the organization’s coarse-grained properties. This perspective is new to leadership
research, and it represents a significant advance in understanding which is only beginning to
show its promise. A summary of representative empirical studies is shown in Table 1.
Functional Demands
On System
Complexity
Tool-kit
Convergence
Stability & Predictability
Enables Effective
Performance
Generative
Innovation Enables
Variety & Provides
Future Options for
Adaptive Opportunities
Unifying
Common Models,
Ethics, & Identities
Engender Unity within
Boundaries & Enable
Collective Action
Coarse-Graining
Attractors, Statistical
Properties, Stabilizing
Feedback, Phase Transitions,
Order Parameters
Scale-Crossing
Criticalization, Fluctuation,
Amplifying Feedback,
Divergence, Recombination,
Self-Similarity/Fractals,
Networks, Dissipative
Structures
Fine-Graining
Autonomous agents
Heterogeneity
Interdependence
Boundaries & Tagging
Individual Requisite Complexity
Static Complexity Dynamic Complexity
Cognitive & emotional traits Situational awareness, action, & feedback
Changing Rules of Interaction
Enacting Complexity Mechanisms
Organizing
Mechanisms
Entrainment
Leadership Reinforces
Structures for Day-to-Day
Effectiveness
Emergence
Leadership Promotes
High-Bandwidth
Information Sharing,
Experimentation &
Synthesis
Identity Tension
Leadership Orchestrates
Individual, Group &
Intergroup Connections
& Synthesizes
Overlapping Models &
Identities
Figure 1 - A 2012 view of the emerging Complex Systems Leadership Paradigm includes a
complexity tool-kit of ideas and the organizing mechanisms that can be used to satisfy the
functional demands of a complex adaptive system.
THE ROAD AHEAD
In 2007, Jennings and Dooley used textual analysis to identify the emergence of a new
paradigm in complex systems leadership research. Five years later the synthetic analysis

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 43 of 64
described herein has brought this paradigm into clearer focus. A review of the literature shows
that a new theory has evolved in what can best be described as the abductive theorizing approach
advocated in recent years by Max Boisot (2010). The new paradigm that has emerged both
transcends and challenges prior research approaches. It transcends prior approaches in that it
represents a general theory of human interaction dynamics (HID) consistent with analytical
approaches used in the natural sciences. It challenges them in that its generality implies
additional criteria and analytical techniques through which prior results may be evaluated.
Implications for Research
As this theory comes into focus, additional empirical research is needed to inform future
iterations of CSLT. In this next phase of abductive theory development, it is increasingly
important that quantitative methods be added to the mix so that constructs can be validated and
relationships between them identified and tested with statistical methods. However, these
methods must be implemented while cognizant of the limitations inherent in traditional methods
when events are not independent and where individuals have different preferences, histories and
values.
To do this, a new way of thinking, new methods, and a new set of skills are needed. Tools
like agent-based modeling, the mathematical treatment of dynamical systems, dynamic network
analysis, and power-law analysis can all inform these new techniques. However, because the
HID approach is qualitatively different from the interaction dynamics typically modeled in the
natural sciences, many new technological advances are also needed (Prietula, 2011).
As a result, quantitative research in CSLT will require advances on many levels, and
many of these will need to be built fresh, from the ground up. It will involve changing the rules
of interaction among all of those who are engaged in leadership research and practice. This is

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 44 of 64
perhaps the greatest challenge of all: The new paradigm is not just another way of thinking about
leadership; it is also a new way of thinking about social science more generally. The magnitude
of this challenge is daunting, certainly. But the potential benefits of action—particularly given
the many challenges that face the world in the coming decades—make a focused effort toward a
new and better way forward not merely a choice to be taken. It is an ethical imperative.
Implications for Practice
These developments also have significant implications for practice and thus for
leadership development. One of these is the realization that leadership development must go far
beyond the current focus on individual self-understanding and communications skills. In the fast-
moving economy of the 21
st
century, individuals will need tools to help them better understand
the nonlinear effects in their ecosystems. Increasingly, these are being enabled by social media
and socio-technical networks, for example. It would be beneficial to be able to anticipate how
various possible interventions might influence emergent outcomes, and then learn to respond
quickly and thoughtfully to unfolding events in real-time.
We expect that leadership development programs will increasingly be characterized by
computer assisted simulations which express real life organizations as complex adaptive systems
in a manner analogous to flight simulators in the airline industry, or battlefield simulations in the
military. One of the advantages that CSLT provides is a theoretical framework within which one
could identify opportunities for management interventions. The potential exists that in the not too
distant future, computer simulations could be used to evaluate actions of individuals and groups
and then present to the user the likely outcomes at the coarse-grained levels under various
scenarios. Not only would this be useful in training, it could also be used to pretest actions to

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 45 of 64
explore possible unintended consequences. With better information, more informed action could
be contemplated, and possible contingencies could be better evaluated.
Table 1. Empirical studies using various methods
Author(s) Description Method Complexity
Paradigm
Aasen & Johannessen (2007)
Aasen,& Berg (2008).
Innovation in the subsea
technology development
Longitudinal industrial
case ethnographic
Complex Responsive
Processes
Backström, 2009 Resource generation in
pharmacies
Case study CSLT/Directing
leadership
Backström, Hagström, &
Göransson, 2011
Integration of employees in
the company culture of a bank
Network analysis &
survey
CSLT/Directing
leadership
Backström, Wilhelmson,
Olsson, Åteg, & Åberg, 2011
Training of first line managers Action learning
network, survey and
interview
CSLT/Directing
leadership
Baker, Onyx & Edwards (2011) Community service projects Network analysis &
qualitative case study
CSLT/Generative
Leadership
Beck & Chong (2009) Community groups solving
local problems
Action learning
&participant observer
CSLT/Leadership of
emergence
Buckle Henning & Dugan
(2007)
Self-organized patterns in the
workplace
Grounded Theory CSLT
Butterfield, Shepherd & Woods
(2011)
Development Entrepreneurial
business model
Case study CSLT/Generative
leadership
Dal Forno & Merlone (2007) Project tem formation and
success
Laboratory study &
Agent-model
CSLT/Convergent
Leadership
Garud, Gehman, &
Kumaraswamy, (2011).
Innovation process at NCR Case study Complexity and
innovation
Groot (2009) Performance improvement at
Dutch Railways
Case Study/ personal
reflection
Complex Responsive
Processes
Guastello et al. (2004, 2005,
2007a,b)
Emergence of individuals as
team leaders in simulation
games
Laboratory Study Nonlinear dynamical
systems/ Game theory
Havermans, Den Hartog,
Keegan & Uhl-Bien (2010)
Leadership practices in
project-based organizations
Qualitative/Interviews Complexity
Leadership Theory
Hazy (2008a) 18 month study of 50 person,
technology firm
Survey/case study Leadership Meta-
capability
Hazy (2004, 2008a) Organizational
Transformations:
NCR & Intel
System dynamics
modeling of case study
Leadership Meta-
capability
MacGillivray (2010) Leadership in community
group
Phenomenographic
study
Complexity
Leadership Theory
Moerschell (2010) Leadership emergence in
recreational groups
Grounded theory Complexity and
punctuated equilibria
Phelps & Hubler (2007) Emergent leadership in youth
groups
Multi-Case study and
agent-based model
Dynamical Systems &
Bifurcation Theory
Plowman et al (2007) Transformation of a mission
church
Case study Leadership of
emergence
Schreiber & Carley (2006) Workgroups at NASA Network Analysis Dynamic Network
Analysis
Surie & Hazy (2006) Indian Manufacturing
Companies
Case studies CSLT/Generative
Leadership
Tapsell & Woods, (2009) Maori of New Zealand Case Study CSLT/Generative
Leadership
Tobin, J. H. (2009) US Hospital merger Case Study/ personal
reflection
Complex esponsive
Processes

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 46 of 64
The CSLT approach is a useful and important advance in any case. By making managers
aware of the conditions that imply different types of leadership action, and by clarifying what
actions lead to which outcomes in a given situation, the process of enacting leadership becomes
more understandable as a capability to be learned. This is distinct from more cultish approaches
to leadership development that treat “leadership” as a “mysterious” or “authentic” attribute of
certain special individuals, one that is hidden deep within a person’s soul and can never be fully
understood. Rather than leaving us to trust that certain special people will reach inside
themselves to lead us where they believe we should go, CSLT exposes the human interaction
dynamics that each of us can influence to construct the world in which we choose to live.
FUTURE DIRECTIONS
The advancement of complex systems leadership theory has brought to the fore several
important research questions:
1. How should the ubiquity of interdependent heterogeneity in HID be treated rigorously
when modeling complex systems leadership events and when analyzing empirical data?
New statistical techniques are needed that take into account individual learning and
memory and path dependence with regards human interactions which limit the validity of
many traditional methods (Hazy & Ashley, 2010).
2. What are the precise mechanisms of emergence in HID? What is the role of leadership in
this process? Can the number and duration of experiments be quantified to determine if
potential success is sustained through reinforcing feedback?
3. By what mechanisms do constraints on human interaction dynamics imply the particulars
of emergent coarse-grained properties? How are constraints and their impacts measured?
For example, how are constraints to resources, such as financial, human, temporal, and

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 47 of 64
technological, linked to leadership actions and outcomes? How do changes to the
constraints translate into qualitatively different properties? What is the role of leadership
in this process?
4. How do individuals who are seeking to have an impact recognize coarse-grain properties,
determine the need to change them, and then translate this into complex systems
leadership actions that locally change the rules of interaction in ways that alter these
emergent properties as intended?
5. By what mechanisms do individual agents influence the local rules of interactions of
others? What is the role of identity? Of ethics?
6. By what mechanisms do fine-grained interactions imply specific coarse-grained
properties? How do changes in local rules become manifest in qualitative changes in the
emergent properties? Is the analogy with phase transitions informative for this?
7. By what mechanisms do coarse-grained properties entrain fine-grained rules of
interaction in HID? How is this related to leadership as well as to cultural norms,
institutional effects, etc.?
8. When focusing on the rules that govern local interactions, how are these rules recognized,
developed, shared, remembered, adapted, and replicated? What role do identities play in
storing, sharing and evolving rules of interaction? How do ethics play into this question?
9. Can examples of collective agency with regard coarse-grained organizing forms in HID
be identified and shown to be ontologically distinct from individual intention and action?
What would this mean for leadership research?
10. What are the implications of this new paradigm for leadership development programs
going forward?

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 48 of 64
REFERENCES
Aasen, T. & Berg, M. (2008). A complexity perspective on innovation processes for subsea
technology development International Journal of Learning and Change, 3(3), pp. 294-
307.
Aasen, T.M.B. & Johannessen, S. (2007). Exploring Innovation Processes from a Complexity
Perspective. Part II: Experiences from the SIOR case. International Journal of Learning
and Change, 2 (4), 434–446.
Allen, P. (2001). What is complexity science? Knowledge of the limits to knowledge.
Emergence: A Journal of Complexity Issues in Organizations and Management, 3(1), 24-
44.
Anderson, P. (1999). Complexity theory and organization science. Organization Science, 10(3),
216-232.
Andriani, P. (2011). Complexity and Innovation. In P. Allen. S. Maguire, & B McKelvey (Eds.),
The Sage Handbook of Complexity and Management, pp 454-470. Los Angeles: SAGE.
Anneloes, M.L. Raes, Heijltjes, M.G., Glunk, U, Roe, R.A. (2011). The interface of the top
management team and middle managers: A Process Model. Academy of Management
Review, 36(1), 102-126.
Ashby, W. R. (1956). An Introduction to Cybernetics. London: Chapman and Hall;
Axelrod, R. (1984) The evolution of cooperation. New York: Basic books
Backström, T. (2009). How to organize for local resource generation. The Learning
Organization, 16(3).

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 49 of 64
Backström, T., Hagström, T., & Göransson, S. (2011). Communication as a mechanism for
culture integration. Paper presented at the 55th annual meeting of the international
society for the system sciences, Hull.
Backström, T., Wilhelmson, L., Olsson, B. K., Åteg, M., & Åberg, M. M. (2011). The Role of
Manager in the Post-Industrial Work System. In E. Seglod, E. Berglund, E. Bjurström, E.
Dahlquist, L. Hallén & E. Johansson (Eds.), Studies in industrial renewal (pp. 215-227).
Västerås: Mälardalen University Press.
Bak, P. & Paczuski, M. (1995). Complexity, contingency, and criticality". Proceedings of the
National Academy of Sciences of the USA, 92 (15), 6689–6696.
Baker, E., Onyx, J., & Edwards, M. (2011). Emergence, social capital and entrepreneurship:
Understanding networks from the inside. Emergence: Complexity and Organization,
13(3), 31-45.
Barnard, C. I. (1938). The functions of the executive. Cambridge, MA: Harvard University Press.
Barney, J. B. (1991). Firm resources and sustained competitive advantage. Journal of
Management, 17(1), 99-120.
Baumann, O. & Siggelkow, N. (2011). Complexity and Competitive Advantage. In P. Allen, S.
Maguire, & B McKelvey (Eds.), The Sage Handbook of Complexity and Management (pp
494-505). Los Angeles: SAGE.
Beck, D. & Chong, L.C. (2009). Creative Interaction in culturally diverse groups. In The
emergence of Collective Identity as a Means for creating and Sustaining Social Value. In
J. Goldstein, J.K.
Hazy & j. Silberstang (Eds.), Complexity Science and Social
Entrepreneurship ) pp.487-506). Litchfield Park, AZ: ISCE Publishing.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 50 of 64
Boal, K.B. & Schultz, P.L. (2007). Storytelling, time and evolution: The role of strategic
leadership in complex adaptive systems. The Leadership Quarterly, 18, 411-428.
Boisot, M., & Child, J. (1999). Organizations as adaptive systems in complex environments: The
case of china. Organization Science, 10(3), 237-252.
Boisot, M. & McKelvey, B. (2007). Extreme events, power laws, and adaptation: Towards an
econphysics of organization. Academy of Management Conference Best Papers
Proceedings, OMT.
Boisot, M. & McKelvey, B. (2010). Integrating Modernist and Postmodernist Perspectives on
Organizations: A Complexity Science Bridge. Academy of Management Review, 35(3),
415-433.
Boisot, M. (2010). Connecting the dots before the world does. Paper presented at the annual
meeting of the Academy of Management, Montreal.
Boyatzis, R. E. (2008). Leadership development from complexity perspective. Consulting
Psychology Journal: Practice and Research 60(4), 298-313.
Brown, S., & Eisenhardt, K. (1997). The art of continuous change: Linking complexity theory
and time-based evolution in relentlessly shifting organizations. Administrative Science
Quarterly, 42, 1-34.
Brown, S. & Eisenhardt, K. (1998). Competing on the edge: Strategy as structured chaos.
Boston: Harvard Business School Press.
Buckle-Henning, P. & Dugan, S. (2007). Leader’s detection of Problematic Self-Organized
Patterns in the Workplace. In J. K. Hazy, J. Goldstein & B. B. Lichtenstein (Eds.),
Complex Systems Leadership Theory (pp. 386-412). Mansfield, MA: ISCE Publishing
Company.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 51 of 64
Butterfield, L., Shepherd, D. & Woods, C (2011). Developing digital citizenship for digital tots:
Hector’s World Limited. Emergence: Complexity and Organization 13(3), 16-30.
Cilliers, P. (1998). Complexity and postmodernism: Understanding complex systems. London:
Routledge.
Cyert, R., & March, J. (1963). A behavioral theory of the firm. Englewood Cliffs, NJ: Prentice-
Hall.
Dal Forno A. & Merlone, U. (2007). The emergence of effective leaders: and experimental and
computational approach. In J. K. Hazy, J. Goldstein & B. B. Lichtenstein (Eds.), Complex
Systems Leadership Theory (pp. 205-226). Mansfield, MA: ISCE Publishing Company.
Davis, J. P., Eisenhardt, K. M., & Bingham, C. (2009). Optimal structure, market dynamism, and
the stategy of simple rules. Administrative Science Quarterly, 54, 413-452.
Denis, J-L., Langley, A., & Pineault, M. (2000). Becoming a leader in complex organizations,
Journal of Management Studies, 37(8), 1063-1099.
Dooley, K. J. (1997). A complex adaptive systems model of organization change. Nonlinear
Dynamics, Psychology and Life Sciences, 1(1), 69-97.
Dooley, K. J. (2007). Leadership and a computational model of organizations.
In J. K. Hazy, J.
Goldstein & B. B. Lichtenstein (Eds.), Complex Systems Leadership Theory (pp. 327-
322). Mansfield, MA: ISCE Publishing Company.
Dosi, G., Nelson, R. R., & Winter, S. G. (2000). The nature and dynamics of organizational
capabilities. Oxford: Oxford University Press.
Fleishman, E.A. (1953). The description of supervisory behavior. Personnel Psychology, 37, 1-6.
Galunic, D. C., & Eisenhardt, K. M. (2001). Architectural innovation and modular corporate
forms. Academy of Management Journal, 44(6), 1229-1249.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 52 of 64
Garud, R., Kumaraswamy, A., & Sambamurthy, V. (2006). Emergent by design: Performance
and transformation at Infosys Technologies. Organization Science, 17, 277-286.
Garud, R., Gehman, J., & Kumaraswamy, A. (2011). Complexity Arrangements For Sustained
Innovation: Lessons From 3M Corporation. Organization Studies, 32, 737-767.
Gell-Mann, M. (2002). What is complexity? In A. Q. Curzio & M. Fortis (eds.), Complexity and
industrial clusters: Dynamics and models in theory and practice (pp.13-24). Berlin:
Physica-Verlag.
Goldstein, J. (1989). A Far-From-Equilibrium Systems Approach to Resistance to Change.
Organizational Dynamics, 17,16-26.
Goldstein, J. A. (2007). A new model of emergence and its leadership implication. In J. K.
Hazy, J. Goldstein & B. B. Lichtenstein (Eds.), Complex Systems Leadership Theory (pp.
61-92). Mansfield, MA: ISCE Publishing Company.
Goldstein, J. A. (2011). Emergence in Complex Systems. In P. Allen, S. Maguire, & B
McKelvey (Eds.), The Sage Handbook of Complexity and Management, (pp 65-78).
Thousand Oaks: SAGE.
Goldstein, J., Hazy, J. K., & Lichtenstein, B. (2010). Complexity and the Nexus of Leadership:
leveraging nonlinear science to create ecologies of innovation. Englewood Cliffs:
Palgrave Macmillan.
Griffin, D. (2001). The emergence of leadership: Linking self-organization and ethics. London:
Routledge.
Groot, N. (2009). Senior executives and the emergence of local responsibilities: A complexity
approach to identity development and performance improvement. International Journal
of Learning and Change, 3 (3), 264–280.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 53 of 64
Guastello, S. J. (2002). Managing emergent phenomena: Nonlinear dynamics in work
organizations. Mahwah, New Jersey: Lawrence Erlbaum Associates.
Guastello, S. J. (2007). Nonlinear dynamics and leadership emergence. The Leadership
Quarterly, 18(4), 357-369.
Guastello, S. J., & Bond, R. W. J. (2004). Coordination learning in Stag Hunt games with
application to emergency management. Nonlinear Dynamics, Psychology, and Life
Sciences, 8, 345-374.
Guastello, S. J., & Bond, R. W. J. (2007a). The emergence of leadership in coordination
intensive games. Nonlinear Dynamics, Psychology, and Life Sciences, 11, 91-117.
Guastello, S. J., & Bond, R. W. J. (2007b). A swallowtail catastrophe model of leadership in
coordination-intensive games. Nonlinear Dynamics, Psychology, and Life Sciences, 11,
235-351.
Guastello, S. J., Craven, J., Zygowicz, K. M., & Bock, B. R. (2005). A rugged landscape model
for self-organization and emergent leadership in creative problem solving and production
groups. Nonlinear Dynamics, Psychology and Life Sciences, 9(3), 297-233.
Guastello, S. J., & Guastello, D. D. (1998). Origins of coordination and team effectiveness: A
perspective from game theory and non-linear dynamics. Journal of applied psychology,
83(3), 423-437.
Haken, H. (2006). Information and self-organization: A macroscopic approach to complex
systems (3
rd
ed.). Berlin: Springer.
Hannan, M. T. & Freeman, J. (1989). Organizational Ecology. Cambridge, MA: Harvard
University Press.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 54 of 64
Hannah, S.T., Eggers, J.T., & Jennings, P. L. (2008). Complex adaptive leadership: Defining
what constitutes effective leadership for complex organizational contexts. In G. B. Graen
& J. A. Graen (Eds.), Knowledge-driven corporation: Complex creative destruction (pp.
79-124). Charlotte, NC: Information Age Publishing.
Hannah. S. T., Jennings, P. L., & Nobel, O. B-Y. (2010). Tactical Military Leader Requisite
Complexity: Toward a referent Structure. Military Psychology, 22(4), 412-449.
Hannah. S. T., Woolfolk, R. L., & Lord, R. G. (2009) Leader self-structure: A framework for
positive leadership. Journal of Organizational Behavior, 30, 269-290.
Havermans, L., Den Hartog, D., Keegan, A., & Uhl-Bien, M. Leadership in project-based
organizations: Applying and extending complexity leadership theory through qualitative
exploration. Working Paper.
Hazy, J. K. (2004). A Leadership and Capabilities Framework for Organizational Change:
Simulating the Emergence of Leadership as an Organizational Meta-Capability, doctoral
dissertation, The George Washington University, Washington, DC.
Hazy, J. K. (2006). Measuring leadership effectiveness in complex socio-technical systems.
Emergence: Complexity and Organization (E:CO), 8(3), 58-77.
Hazy, J. K. (2007a). Computer models of leadership: Foundation for a new discipline or
meaningless diversion? The Leadership Quarterly, 18(4), 391-410.
Hazy, J. K. (2007b). Leading Large: How disambiguation and changing reputations enable
back-propagation learning in complex organizations. Paper presented at the Leadership
Quarterly FestSchift in Honor of Dr. Jerry Hunt, Lubbock, TX )Oct 12-15, 2007).
Hazy, J. K. (2008a). Leadership or luck? The system dynamics of Intel's shift to microprocessors
in the 1970s and 1980s. In M. Uhl-Bien, R. Marion & P. Hanges (Eds.), Complexity

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 55 of 64
Leadership, Part I: Conceptual foundations (pp. 347-378). Charlotte, NC: Information
Age Publishing.
Hazy, J. K. (2008b) Toward a theory of leadership in complex systems: Computational modeling
explorations. Nonlinear Dynamics, Psychology, and Life Sciences, 12(3) 281-310.
Hazy, J. K. (2008c). Patterns of Leadership. The System Dynamics of Intel's Shift to
Microprocessors in the 1970s and 1980s. In M. Uhl-Bien, R. Marion & P. Hanges (Eds.)
Complexity Leadership, Part I: Conceptual foundations (pp. 379-390). Charlotte, NC:
Information Age Publishing.
Hazy, J. K. (2009). Innovation Reordering: Five principles for leading continuous renewal. In
Schlomer, S. & Tomaschek, N. (Eds.), Leading in Complexity: New Ways of
Management (pp. 300). Seiten: Verlag fur Systemische Forschung.
Hazy, J. K. (2011a). Parsing the Influential Increment he Language of Complexity: Uncovering
the Systemic Mechanisms of Leadership Influence. International Journal of Complexity
in Leadership and Management, 1(2), 164-192.
Hazy, J. K. (2011b). Leadership as Process: A Theory of Formal and Informal Organizing in
Complex Adaptive Systems. Adelphi University School of Business Working Paper: SB-
WP-2011-02.
Hazy, J.K. (in press-a). Unifying leadership: Shaping identity, ethics and the rules of interaction.
International Journal of Society Systems and Science.
Hazy, J. K. (in press-b). Leading large organizations. International Journal of Complexity in
Leadership and Management.
Hazy, J.K. & Ashley, A. (2011). Unfolding the future: Bifurcation in Organizing Form and
Emergence in Social Systems. Emergence: Complexity and Organization 13(3), 77-91.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 56 of 64
Hazy, J. K., Goldstein, J. A., & Lichtenstein, B. B. (Eds.). (2007a) Complex Systems Leadership
Theory. Mansfield, MA: ISCE Publishers.
Hazy, J. K., Goldstein, J., & Lichtenstein, B. B. (2007b). Complex systems leadership theory:
An introduction. In J. K. Hazy, J. Goldstein & B. B. Lichtenstein (Eds.), Complex
Systems Leadership Theory (pp. 1-17). Mansfield, MA: ISCE Publishing Company.
Hazy, J. K., Millhiser, W. P., & Solow, D. (2007). Mathematical and computational models of
leadership: Past and Future. In J. K. Hazy, J. Goldstein & B. B. Lichtenstein (Eds.),
Complex Systems Leadership Theory (pp. 386-412). Mansfield, MA: ISCE Publishing
Company.
Hazy, J. K., Moskalev, S., & Torras, M. (2009). Toward a theory of social value creation:
Individual agency and the use of information within nested dynamical systems. In J. A.
Goldstein, Hazy, J. K. & Silberstang, J. (Eds.), Complexity Science and Social
Entrepreneurship (pp. 257-281). Litchfield Park, AZ: ISCE Publishing.
Hazy, J. K., Moskalev, S, & Torras, M. (2010). Mechanisms of Social Value Creation:
Extending Financial Modeling to Social Entrepreneurship and Social Innovation.
International Journal of Society Systems Science, 2(2), 134-157.
Hazy, J. K. & Silberstang, J. (2009a). Leadership within emergent events in complex systems:
micro-enactments and the mechanisms of organisational learning and change.
International Journal of learning and Change, 3(3), 230-247.
Hazy, J. K. & Silberstang, J. (2009b). The emergence of Collective Identity as a Means for
creating and Sustsaining Social Value. In J. A. Goldstein, J. K. Hazy and J. Silberstang
(Eds.), Complexity Science and Social Entrepreneurship,( pp.447-470). Litchfield Park,
AZ: ISCE Publishing.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 57 of 64
Heifetz, R. A., & Laurie, D. L. (2001). The work of leadership. Harvard Business Review,
79(11), 131-141.
Helfat, C. E., Finkelstein, S., Mitchell, W., Peteraf, M. A., Singh, H., Teece, D. J., & Winter, S.
G. (2007). Dynamic Capabilities: Understanding strategic change in organizations.
Malden, Mass: Blackwell Publishing.
Hogue, M., & Lord, R. G. (2007). A multilevel, complexity theory approch to understanding
gender bias in leadership. The Leadership Quarterly, 18 (4), 370-390.
Holland, J. H. (1975). Adaptation in natural and artificial systems. Cambridge, MA: The MIT
Press.
Hooijberg, R. (1996). A multidirectional approach to leadership: An extension of the concept of
behavioral complexity. Human Relations, 49, 917-946.
Hunt, J, G., Osborne, R. N., & Boal, K. .B (2009). The architecture of managerial leadership:
Stimulation and channeling of organizational emergence. The Leadership Quarterly, 20,
503-516.
Jaques, E. (1989). Requisite organization. Arlington, VA: Cason Hall.
Jennings, P. L. & Dooley, K. J. (2007). An emerging complexity paradigm in leadership
research. In J. K. Hazy, J. Goldstein & B. B. Lichtenstein (Eds.), Complex Systems
Leadership Theory (pp. 17-34). Mansfield, MA: ISCE Publishing Company.
Johannessen, S. (2009). The complexity turn in studies of organisations and leadership:
relevance and implications. International Journal of Learning and Change, 3(3), 214-
229.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 58 of 64
Johannessen, S. & Aasen, T.M. B. (2007). Exploring Innovation Processes from a Complexity
Perspective, Part I: Theoretical and Methodological Approach. International Journal of
Learning and Change, 2 (4), 420–433.
Katz, D., & Kahn, R. L. (1966). The social psychology of organizations (2nd ed.). New York:
John Wiley & Sons.
Kauffman, S. (1995). At home in the universe. Oxford: Oxford University Press.
Kuhn, T. S. (1962). The structure of scientific revolutions. Chicago: University of Chicago Press.
Levick, D. & Kuhn, L. (2007). Fractality, organizational management and creative change.
World Futures, 63(3-4), 265-274.
Levinthal, D. A. (1997). Adaptation on Rugged Landscapes. Management Science, 43 (7), 934-
950.
Lichtenstein, B. & McKelvey, B. (2007). Leadership in the four stages of emergence. In J. K.
Hazy, J. Goldstein & B. B. Lichtenstein (Eds.), Complex Systems Leadership Theory (pp.
93-108). Mansfield, MA: ISCE Publishing Company.
Lichtenstein, B. & Plowman, D.A., (2009). The Leadership of emergence: A complex systems
leadership theory of emergence at successive organizational levels. The Leadership
Quarterly, 20, 651–661.
Lichtenstein, B. B., Uhl-Bien, M., Marion, R., Seers, A., Orton, J. D., & Schreiber, C. (2006).
Complexity leadership theory: An interactive process on leading in complex adaptive
systems. Emergence: Complexity and Organization (E:CO), 8(4), 2-12.
Lord, R. G., Hannah, S. T., & Jennings, P.L. (2011). A framework for understanding leadership
and individual complexity. Organizational Psychology Review, 1(2), 104-127.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 59 of 64
Lord, R. G., Hannah. S. T., & Pearce, C. (2011). A framework for understanding leadership and
individual complexity. Organizational Psychology Review, 1(2), 104-127.
MacGillivray, A. (2010). Leadership in a network of communities: A phenomenographic study,
The Learning Organization, 17 (1), 24-40.
MacIntosh, R., & MacLean, D. (1999). Conditioned emergence: A dissipative structures
approach to transformation. Strategic Management Journal, 20, 297-316.
Mangiofico, G., & Feyerherm, A. E. (2011). The case for using a Complexity Perspective as
applied to Leadership. Working Paper.
March, J. G. (1991) Exploration and Exploitation in Organizational Learning. Organization
Science, 2, 71-87.
Marion, R., & Uhl-Bien, M. (2001). Leadership in complex organizations. Leadership Quarterly,
12(4), 389.
Marion, R., & Uhl-Bien, M. (2001). Leadership in complex organizations. The Leadership
Quarterly, 12(4), 389-418.
McKelvey, B. (2010). Complexity leadership: The secret of Jack Welch’s success. International
Journal of Complexity in Leadership and Management, 1(1), 4-36.
McKelvey, B. (2004). Toward a complexity science of entrepreneurship. Journal of Business
Venturing, 19, 313-342.
McKelvey, B. (1997). Quasi-natural organization science. Organization Science, 8, 352-380.
Meyer, A.D., Gaba, V., & Colwell, K.A. (2005). Organizing Far from Equilibrium: Nonlinear
Change in Organizational Fields. Organization Science, 16(5), 456-473.
Mitchell, M., Hraber, P., & Crutchfield, J.P. (1993). Revisiting the edge of chaos: Evolving
cellular automata to perform computations. Complex Systems, 7, 89-130.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 60 of 64
Moerschell, L. (2010). The intersection of punctuated equilibrium and leadership emergence
within the framework of Naturalistic Decision Making. Journal of Outdoor Recreation,
Education and Leadership, 2(2), 1-4.
Morrison, K. (2010). Complexity Theory, School Leadership and Management: Questions for
Theory and Practice. Educational Management Administration & Leadership, 38(3), 374-
393.
Nelson, R. R., & Winter, S. G. (1982). An evolutionary theory of economic change. Cambridge,
MA: The Belknap Press of Harvard University Press.
Nowak, M., Tarnita C.E., & Wilson, E.O. (2010). The evolution of eusociality. Nature 466,
August 26, 2010, p.1057. doi: 10.1038/nature0920.
Panzar, C., Hazy, J. K., McKelvey, B., & Schwandt, D. R. (2007). The paradox of complex
organizations: Leadership as integrative influence. In J. K. Hazy, J. Goldstein & B. B.
Lichtenstein (Eds.), Complex Systems Leadership Theory (pp. 305-326). Mansfield, MA:
ISCE Publishing Company.
Pascale, R., Sternin, J. & Sternin, M. (2010). Power of Positive Deviance: How Unlikely
Innovators Solve the World's Toughest Problems. Cambridge, MA: Harvard University
Press.
Peterson, M. F., & Meckler, M. R. (2001). Cuban-American entrepreneurs: Chance, complexity
and chaos. Organization Studies, 22(1), 31-57.
Phelps, K.C. & Hubler, A. (2007). Toward an Understanding on Membership and Leadership in
Youth Organizations: Sudden changes in average participation due to the behavior of one
individual. In J. K. Hazy, J. Goldstein & B. B. Lichtenstein (Eds.), Complex Systems
Leadership Theory (pp. 195-204). Mansfield, MA: ISCE Publishing Company.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 61 of 64
Plowman, D., Baker, L. T., Beck, T., Kulkarni, M., Solansky, S., & Travis, D. (2007a). Radical
change accidentally: The emergence and amplification of small change. Academy of
Management Journal, 50(3), 515-543.
Plowman, D., Baker, L. T., Beck, T., Kulkarni, M., Solansky, S., & Travis, D. (2007). The role
of leadership in emergent, self-organization. The Leadership Quarterly.
Prigogine, I. (1995). The end of certainty: Time, chaos, and the new laws of nature. New York:
The Free Press.
Prigogine, I. & Stengers, I. (1984). Order out of Chaos. New York: Heinemann.
Prietula, M. J. (2011) Thoughts on Complexity and Computational Models. In P. Allen, S.
Maguire, & B McKelvey (Eds.), The Sage Handbook of Complexity and Management (pp
93-100). Thousand Oaks: SAGE.
Prokopenko, M., Boschetti, F. & Ryan, A. J. (2009). An information-theoretic primer on
complexity, self-organization and emergence. Complexity, ISSN 1099-0526, 15(1), 11-
28.
Raelin, J. (2011). From leadership-as-practice to leaderful practice, Leadership, 7(2), 195-211
Richardson, K. (2010). Thinking about Complexity. Lichfield Park, AZ: Emergent Publishing.
Schneider, M., & Somers, M. (2006). Organizations as complex adaptive systems: Implications
of complexity theory for leadership research. The Leadership Quarterly, 17(4), 351-365.
Schreiber, C. & Carley, K. M. (2006). Leadership Style as an Enabler of Organizational
Complex Functioning. Emergence: Complexity and Organization, 8(4).
Seeley, T. (2010). Honeybee Democracy. Princeton: Princeton University press.
Selznick, P. (1948). Foundations of the theory of organizations. American Sociological Review,
13, 25−35.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 62 of 64
Shoup, J.R. & Studer, S. C. (2010). Leveraging Choas: The mysteries of leadership and policy
revealed. Lanham, MD: Rowman & Littlefield Education.
Siggelkow, N. & Rivkin, J. (2005). Speed and search: Designing organizations for turbulence
and complexity. Organization Science, 16, 101-122.
Silberstang, J. & Hazy, J. K. (2008). Toward a micro-enactment theory of leadership and the
emergence of innovation. The Innovation Journal: The Public Sector Innovation Journal,
13(3), Article 5.
Simpson, P. (2007). Organizing in the mist: A case study of leadership and complexity.
Leadership and Organization Development Journal, 5, 465-482.
Simon, H. A. (1997). Administrative behavior: A study of decision-making processes in
administrative organizations (4th ed.). New York: The Free Press.
Solow D., & Leenawong, C. (2003). Mathematical models for studying the value of cooperative
leadership in team replacement. Computational and Mathematical Organization Theory,
9(1), 61-81.
Solow, D., & Szmerekovsky, J. G. (2006). The role of leadership: What management science can
give back to the study of complex systems. Emergence: Complexity and Organization,
8(4), 52-60.
Stacey, R. D. (1993). Strategic Management and Organization Dynamics. London: Prentice
Hall.
Stacey, R. D. (1995). The science of complexity: An alternative perspective for strategic change
processes. Strategic Management Journal, 16, 477-495.
Stacey, R. D., Griffin, D., & Shaw, P. (2000). Complexity and management: fad or radical
challenge to system thinking? London: Routledge.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 63 of 64
Stogdill, R. M. (1974). Handbook of Leadership: A survey of the Literature. New York: Free
Press
Streatfield, P. J. (2001). The paradox of control in organizations. London: Routledge.
Surie, G., & Hazy, J. K. (2006). Generative leadership: Nurturing innovation in complex
systems. Emergence: Complexity and Organization (E:CO), 8(4), 13-26.
Tapsell, P. & Woods, C. (2009).A spiral of innovation framework for social entrepreneurship:
Social innovation at the generational dividein an indigenous context. In The emergence
of Collective Identity as a Means for creating and Sustaining Social Value. In J. A.
Goldstein, J. K. Hazy & J. Silberstang (Eds.), Complexity Science and Social
Entrepreneurship( pp.471-486). Litchfield Park, AZ: ISCE Publishing.
Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic management.
Strategic Management Journal, 18(7), 509-533.
Thietart, R-A. & Forgues, B. (1995). Chaos Theory and Organization. Organization Science, 6,
19-31.
Tobin, J.H. (2009). The myth of rational objectivity and leadership: The realities of a hospital
merger from a CEO’s perspective. International Journal of Learning and Change, 3 (3),
248–263.
Tsoukas, H. (2005). Complex Knowledge: Studies in Organizational Epistemology. Oxford:
Oxford University Press.
Uhl-Bien, M., & Marion, R. (2008). Complexity leadership, part 1: Conceptual foundations.
Charlotte: Information Age Publishing.
Uhl-Bien, M., & Marion, R. (2009). Complexity leadership in bureaucratic forms of organizing:
A meso model. The Leadership Quarterly, 20, 631–650.

Complex Systems Leadership Theory
Hazy & Uhl-Bien 2/11/2012 Page 64 of 64
Uhl-Bien, M., & Marion, R. (2011). Complexity leadership theory. In A. Bryman, D. Collinson,
K. Grint, B. Jackson, & M. Uhl-Bien (Eds.), The Sage Handbook of Leadership (pp. 468-
482). London: Sage.
Uhl-Bien, M., Marion, R., & McKelvey, B. (2007). Complexity leadership theory: Shifting
leadership from the industrial age to the knowledge era. The Leadership Quarterly, 18,
298-318.
von Bertalanffy, L. (1950). The theory of open systems in physics and biology. Science, 3, 23-
29.
Wheatley, M. J. (1999). Leadership and the new science: discovering order in a chaotic world.
San Francisco: Berrett-Koehler.
Zaccaro, S.J. (1999). Social complexity and the competencies required for effective military
leadership. In J. G. Hunt, G. E. Dodge, & L. Wong (Eds.), Out-of-the-box leadership:
Transforming the twenty-first century Army and other top performing organizations (pp.
131-151). Stamford, CT: JAI Press.
i
In a later section we describe a competing approach, Complex Responsive Processes (CRP) put forth by Stacey
(1993), which challenges the systems approach entirely.
ii
Constructive deviations are ad hoc experiments performed with the intention of achieving some purpose; this idea
is distinct from the notion of “positive deviance” which is a post hoc analysis and intervention technique used to
identify positively performing subgroups (outliers) in populations facing many of the same challenges. The
positively deviant solutions are then analyzed and understood before being replicated more broadly across the
population (Pascale, Sternin, & Sternin, 2010). This technique could presumably be applied with positive affect to
evaluate the relative success of the constructive deviations described herein as they are occurring within
organizations that are experiencing criticalization.