Leadership

[Link]

The Practice of Leadership in the Messy World of Organizations

Jean-Louis Denis, Ann Langley and Linda Rouleau
Leadership 2010 6: 67
DOI: 10.1177/1742715009354233
The online version of this article can be found at:
[Link]

Published by:
[Link]

Additional services and information for Leadership can be found at:

Email Alerts: [Link]
Subscriptions: [Link]
Reprints: [Link]
Permissions: [Link]
Citations: [Link]

>> Version of Record - Feb 15, 2010

What is This?

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 67

Leadership

The Practice of Leadership in the

Messy World of Organizations
Jean-Louis Denis, Universit de Montral, Canada
Ann Langley, HEC Montral, Canada
Linda Rouleau, HEC Montral, Canada

Abstract This article examines the practice of leadership in organizations characterized by ambiguous authority relationships. Drawing on three empirical case
studies illustrative of a long-term research program on change in health care organizations, we examine leadership as a practical activity focusing particularly on its
dynamic, collective, situated, and dialectical nature. We invite researchers on
leadership to look carefully at the embeddedness of leadership roles in context and
at the type and consequences of practices that leaders develop in such contexts.
Implications of these ideas for further research and for would-be leaders are
discussed.

Introduction
Leadership studies have a long history in the administrative sciences and the field
has accumulated an impressive body of knowledge (Parry & Bryman, 2006).
Leadership research has focused for example on individual leadership traits (Judge
et al., 2002; Stogdill, 1948), on transformational and transactional forms of leadership (Bass & Avolio, 1994; Bennis & Nanus, 1985; Judge & Piccolo, 2004; Tichy
& Devanna, 1986), on leadership styles and behaviours appropriate to various
contexts (Fiedler, 1967; House & Aditya, 1997), on the role of demographic and
cognitive diversity in top management team decision making and performance
(Carpenter et al., 2004; Hambrick & Mason, 1984), and on the psychodynamic
dimensions that drive individual leaders (Kets de Vries, 2001). Each wave of studies
has enriched understanding of the phenomenon. Yet much of this research still does
not fully capture the experience of doing leadership as a practical activity in
complex organizations.
Nevertheless, there have been several calls to focus more intensively on the activities of leaders and to inject greater dynamism into leadership studies. For example,
Pettigrew (1992) argued for a more processual and contextualist view of leadership that pays greater attention to how leadership emerges and evolves in concrete
social or organizational settings and to interactions between organizational contexts
The Author(s), 2010 Reprints and permissions: [Link]/[Link]
[Link] Vol 6(1): 6788 DOI: 10.1177/1742715009354233

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 68

Leadership

6(1) Articles

and leaders capabilities. In their review of leadership theories, Parry & Bryman
(2006) refer to a number of emerging alternative conceptions of leadership including the processual view. They also use the term distributed leadership (see also
Gronn, 2002) to describe a perspective in which leadership is considered less as the
result of single individuals but more as a collective process where individuals
negotiate their position with respect to others in more unpredictable ways than a
rational view of organizations would suggest.
In a related trend, some researchers have proposed taking what is called in social
sciences the practice turn (Schatzki et al., 2001) for studying leadership (Alvesson
& Sveningsson, 2003a, 2003b; Carroll et al., 2008; Knights & Wilmott, 1992)
grounded in social theories of practice. A practice perspective focuses on human
action and praxis in order to understand how people participate in the production
and reproduction of organization and society and in this case of leadership. It is in
this sense that Alvesson & Sveningsson (2003a, 2003b) have suggested looking at
leadership as the extraordinarization of mundane activities, emphasizing for
example, the importance of listening and chatting as foundational practices of leadership exercised in the daily activities of managers. Another study in this vein by
Samra-Fredericks (2003) focuses on conversations among strategists enabling the
identification of certain rhetorical moves by which leadership becomes manifest. A
practice theory view thus tends to focus on micro-level activities, examining in a
very fine-grained manner how they achieve their effects. To date however, there have
been relatively few empirical studies that explicitly adopt a practice perspective on
leadership, and most of these studies have remained focused on relatively narrowly
defined situations.
In this article, we revisit data from three case studies of leadership experiences in
the context of change in the health care field, attempting to combine a focus on microlevel practices (Carroll et al., 2008) with an understanding of the broader overall
processes (Pettigrew, 1992) within which they are embedded. We use this analysis to
illustrate the dynamic, collective, situated and dialectical nature of leadership
practices, suggesting a need to more strongly incorporate these dimensions into
leadership research and training and to encourage an awareness of them among
practitioners.
The health care setting that forms the backdrop for this analysis is somewhat
extreme in terms of the ambiguity of its authority relationships it is a particularly
messy world in which multiple groups with different values, interests and expertise compete for influence (Mintzberg, 1997). Yet, to the extent that leadership in any
organization is considered as an interpersonal phenomenon that builds on informal
sources of influence as well as formal hierarchical position, we argue that the research
ideas derived from these studies may hold useful lessons for leaders and for the study
of leadership practices. The attributes of pluralism and complexity particularly associated with health care are found in other sectors as exemplified by contemporary
work on knowledge-based organizations (Alvesson & Sveningsson, 2003a).
Moreover, some might claim that almost all large organizations are at least to some
extent pluralistic in the sense of incorporating multiple foci of power, and diverse
interests, values and expertise (Denis et al., 2005, 2007).
We begin by briefly presenting the three case studies that form the basis of our
analysis, and then review how each study illustrates the four dimensions mentioned

68

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 69

Leadership

The Practice of Leadership in the Messy World of Organizations Denis et al.

above. Finally we consider the implications of this analysis for research and
practice.

Three illustrative case studies

The three case studies considered here were originally developed as part of a longterm program of research on organizational change in health care organizations that
has given rise to several previous publications (Denis et al., 1996, 2000, 2001, 2006,
2009). The research is based on qualitative and longitudinal research methods including multiple data sources: meeting observations, extensive hand-written field-notes,
taped and transcribed interviews with managers, board members, professionals and
employees, extensive internal documents and a voluminous database of press reports
and editorials. Our interest in the leadership phenomenon initially developed around
a preoccupation for understanding why some organizations seem able to initiate and
implement deliberate change while others are characterized by more inertia.
In order to develop the practice view of leadership presented here, we will draw
on three particular case studies from this larger corpus. The first study (Denis et al.,
1996) focused on the cyclical nature of organizational change processes in contexts
characterized by ambiguity. The second study (Denis et al., 2000) examined the
process of integration of a new leader into a health care organization. The third study
(Denis et al., 2006) focused on the role of management tools and in particular on
how a system of numerical indicators was used in implementing a major strategic
decision, that is, the closure of nine short-term care hospitals. The present article
integrates common insights across the three studies, while deepening the analysis of
the leadership practices in these cases. The Appendix summarizes the original case
studies and the data sources for each one.
In order to enrich understanding of the key points developed in the body of the
paper, we begin by providing a short narrative vignette for each case placed in parallel
with illustrative quotations from interviews showing perceptions of the particular
practices and approaches of a key leader playing an important role in the case history.
We focus more particularly here on three specific individuals who were at the centre
of the action: John for Case 1, Ivor for Case 2 and Martin for Case 3 (fictitious
names).

Case 1 vignette: change leadership under ambiguity

The first case study focused on the leadership of major
change in a hospital characterized by diffuse authority
and power relationships. Several people played leadership roles in this context, but perhaps the most interesting figure in the process was John, a public health
physician who at the start of the study did not hold a
particularly central position in the organization.
However, when the CEO ran into difficulties with a
planning process, he called on John to help out. From
this point on, John began to play an increasingly
important role, promoting the designation of the

John as a leader:
He was a guy who shook
up a lot of things.
(Doctor-manager)
Hes a guy whos very
dynamic, he assembles
people, hes a doer, he can
find the glue to stick
rubber to plastic. (MD)

69

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 70

Leadership

6(1) Articles

hospital as a teaching institution and mobilizing

colleagues from the medical staff and the board around
his ideas. His approach was very entrepreneurial and
hands-on. When he felt something needed doing, he
tended to go for it, becoming successively involved in
managing financial downsizing, and negotiating with
the ministry of health and with the university. He was
extremely effective in most of these enterprises,
producing impressive substantive results in terms of
fiscal equilibrium and the teaching affiliation, but his
activities were not always appreciated by people who
felt pushed aside or ignored in his unrelenting drive to
move things forward. Eventually, he and his main
collaborators lost the support of key members of the
medical staff who felt that the organization was
moving too quickly. John left the hospital and moved
on to other things, while the change process he had
initiated slowed down.

He was a catalyst, he
seeded ideas. (MD)
Things were done in a
cavalier fashion . . .
people were profoundly
upset by the way things
were done. (Manager)
People need to slow
down. You cannot keep on
breaking the china all the
time. (Manager)

Case 2 vignette: the process of integration of a new leader

The second case study focused on the process by
which a new leader called here Ivor integrated into
the CEO position in a large and prestigious teaching
hospital. Ivor had acquired a reputation for excellent
communication skills in his former post as CEO of a
smaller hospital and this was one reason why the
medical establishment was particularly keen to hire
him they felt that their hospital needed to develop a
higher profile. However, they were also hiring
someone who had a strong taste for innovation something that members of this more conservative organization were not quite ready for.
The case shows how difficult it may be for a new
leader, even a CEO, to impose his way of thinking on
an organization with strong incumbent stakeholders.
While his integration was ultimately successful, Ivor
found himself forced to adopt many of the perspectives
preferred by the powerful medical staff and he went
through numerous difficult trial and error negotiations
with his administrative team concerning quality
management practices. Ivors open-door style and
practices, his skill in communications and his informality in relations with others were popular with
physicians. They enabled him to rapidly assimilate the
perspectives that the medical staff favoured. He
became adept at finding ways to develop initiatives

Ivor as a leader:
A man of imagination, of
creativity of projects, of
considerable ambition, of
unbounded energy.
(Manager)
He has an open-door
policy and receives all
sorts of people in his
office. But sometimes, it
looks like interference.
(Manager)
I think he is gradually
becoming imbued with our
culture. Like an angel cake
when you add the syrup,
it seeps in. (MD)
He took the culture,
adopted it and adapted it.
(Manager)
Ive learned some
fascinating things. You
have power by remaining
distant here. If you are too

70

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 71

Leadership

The Practice of Leadership in the Messy World of Organizations Denis et al.

that met their concerns and interests but that also

contributed positively to organizational performance.
On the other hand, the same practices caused difficulties with the administrative team who sometimes
found themselves shut out. Moreover, Ivor gradually
came to realize that his open-door policy might
undermine other managers and leave him with little
recourse when a change of direction was required.

involved from the

beginning, you lose your
impartiality. (Ivor)

Case 3 vignette: the role of tools in leadership

The third case study focused on the orchestration of a
major downsizing decision in a health region. Again
many people played leadership roles, but the new CEO
of the regional board (Martin) was a key player. Martin
saw his task as creating consensus around a difficult
decision that involved closing down nine acute care
hospitals and transferring resources to the community.
For him, this was a noble objective, contributing to the
sustainability of the health care system.
The approach used to achieve the objective
involved two sets of public hearings in which stakeholders were asked to comment on a preliminary plan.
The first set of hearings was organized to achieve
consensus around the need for closures rather than
cutting across the board, while the second set involved
naming specific institutions. Throughout the hearings,
Martin and his team adopted an apolitical and rational
stance, presenting themselves as sincerely searching
for the best solution, while showing sensitivity to
participants concerns. In order to determine which
hospitals were to be closed, a numerical scoring
system was used and publicly defended. The skillful
design of the system, its apparent objectivity, and its
embedding in a process that demonstrated competence
and transparency, combined with the compatibility of
the decision and the values and interests of a dominant
coalition, enabled the project to pass. However, the
objectivity of the numerical system made changes to
the list of closures difficult even though this might
have been desirable. Moreover, some observers found
the process manipulative. Finally, the plan was
complex to implement and required management
skills different from those that had been required to
produce it.

Martin as a leader:
He had a very clear
vision. (Manager)
Martin dared. The CEO
sets the tone. (Manager)
[They] are doing this with
a surprising degree of
energy, transparency and
determination . . . to the
point where one wonders:
who are these people,
Martin X, for example.
(media report)
It takes courage to face
decisions like that . . . to
be able to stand up in
public to explain things,
and to be rational. (MD)
I think that most people
realized that the dice were
loaded. (Manager)
They prepare wonderful
files. I admire that. But
how many of them really
know how to manage?
(Manager)

71

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 72

Leadership

6(1) Articles

The stories provided here are intended to offer some of the flavour of the cases
and of the particular leadership activities and events within them. As can be seen,
each of the leaders was successful, and yet certain aspects of their action had less
positive consequences. We will draw on these basic stories and on other complementary data in the following discussion to illuminate four features that we argue to be
important in developing a deeper understanding of doing leadership.

Doing leadership
The perspective presented in this article emphasizes the importance of looking at the
micro-practices through which leadership is constituted. Specifically, drawing on the
three cases sketched earlier, we highlight successively the dynamic, collective,
situated and dialectical character of doing leadership as a practical activity. Table 1
summarizes how these basic dimensions apply to the three representative studies of
health care organizational change.

Leadership as dynamic
Leadership studies that reduce leadership and its consequences to inter-related sets
of variables are clearly very limited in their capacity to grasp the temporally rich
experience of what it means to be a leader visible even in the very short vignettes
provided earlier. Our research program and the three cases in particular reveal the
nature of leadership as a dynamic phenomenon a process that evolves over time in
context. Indeed, a longitudinal research perspective seems essential to better understand how context and leadership interact through time and how changes take place.
Leadership emerges, shifts, changes and flows around organizations as leaders and
others engage in everyday activities, interpret the meaning and consequences of prior
actions and engage in further actions.
At the centre of the dynamics of leadership is the recursive relationship between
leadership practices and their consequences. One way of looking at these consequences is to consider them in three categories: substantive, symbolic and political
(see also Denis et al., 2001). Substantive consequences are those that concern
concrete structural change. Symbolic consequences concern the evolution of
meaning among relevant stakeholders. Political consequences refer to the evolution
in leadership roles themselves. Political consequences such as increased or decreased
credibility and changes in formal position are particularly important for determining
leaders future scope for action. The dynamics of leadership are strongly related to
these political issues in the first two cases.
For example, in Case 1, Johns ascension to a key leadership role and eventual
withdrawal constitute a dynamic process that can only be understood by looking at
his activities and their consequences over time. His initial interventions were rather
popular. He took over the strategic planning process, succeeded in generating
consensus among his medical colleagues (something that the CEO had not achieved),
and piloted a report that demonstrated that the hospital was underfinanced. The result
of these initiatives was to create a web of support across the organization that
encouraged him to pursue these proposals further and gained him a promotion to an
administrative position. In pursuing the teaching hospital affiliation, his role widened

72

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 73

Leadership

The Practice of Leadership in the Messy World of Organizations Denis et al.

Table 1 Illustrations of four features of the practice of leadership in the three case studies
John: change
leadership

Ivor: becoming a
leader

Martin: the role of

numbers in leadership

Leadership as
Dynamic: Leadership
actions at one time can
change the potential
for effective leadership
later

Johns initial successes

led to increased power,
but as his actions
threatened other
interests, they made it
impossible to continue.

Ivors earlier assimilation

to the values of the
medical staff enabled
him to convince them
to change later.

Martins initial work in

building a coalition
and obtaining consensus
on the need for closures
made the subsequent
naming of targeted
hospitals easier.

Leadership as
Collective: Leadership
depends critically on a
constellation of
co-leaders who play
complementary roles

John built effective

alliances with the
medical council
chairman Chris and
Board member Mitch,
each playing
complementary roles in
promoting change.

Ivor had to negotiate his

position with respect to
the very powerful
medical staff and board
as he entered the
organization.

Martins initial moves

involved building a team
internally and with
Russell on the Board,
then widening the
coalition to other board
members and outsiders.

Leadership as Situated:
Leadership is manifest
in the micro-activities of
leaders in interaction
with others in specific
contexts

Johns leadership was

manifest in the way he
was able to persuade a
team to adopt a
contrary orientation to
the one they had in
mind previously using
energetic and inspiring
arguments linked to
desirable strategic goals.

Ivors leadership was

manifest in the way he
connected warmly, and
naturally with other
people in interactions,
acquiring both visibility
externally and
recognition internally.

Martins leadership was

manifest in the way he
was able to present
himself in public and
defend the hospital
closures using rational
arguments and
demonstrations of
competence and
transparency.

Leadership as
Dialectic: Practices that
seem to be effective
often have a downside.
The strengths of leaders
can become weaknesses

In opposing a variety of
organization members
using practices such as
those in the box above,
John stepped on a lot of
toes. He succeeded in
the moment, but
created opposition that
would later have its day.

Ivors open and

accessible style described
in the box above
enabled him to gain
credibility, but
sometimes meant
decisions were taken
too quickly and he was
left without a final
recourse in case of
problems.

Martins rational
approach described in
the box above was
successful in achieving
the closure decision, but
the number system
limited discretion and
the public hearings
process made the
project very complex.

further. His negotiations with the government led him to propose a retrenchment plan,
something that previously had been considered inconceivable. As one board member
noted, It was a curious thing to see. The Board tipped completely from one side to
the other. Suddenly sentences that before made everyone jump we need a balanced
budget were uttered. Because now we were told that a balanced budget was needed
to get the university affiliation. This constituted the height of Johns influence in the
organization, but also a turning point as certain people began to see his contribution
in less positive terms. Concerns became more and more evident as certain key
members of the medical staff began to see other potentially negative consequences
of the university affiliation. As opposition to the proposal grew, Johns store of
credibility as a leader diminished and it became urgent to move on.

73

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 74

Leadership

6(1) Articles

The dynamic processes described here were also inherent to Ivors integration as
a new leader described in Case 2. Ivor successfully succeeded his predecessor but
had to go through an integration process that took considerable time (27 months) and
was differently achieved throughout the organization. During this time, he used
several mechanisms to build his leadership. As illustrated in the vignette, with the
medical staff, he found himself obliged to shift his perspective to meet their expectations before being able to convince them to consider moves in new directions. In
parallel with these processes, he engaged in a long period of trial and error with the
administrative team around the implementation of a quality-improvement project.
The extensive support he had acquired from the medical staff early in the process of
integration was however helpful in protecting him in his struggles with the administrative team. His credibility with physicians made it easier to survive these conflicts
and adjustments. A new leader entering in an organization needs to be aware that
leadership is a dynamic process in the sense that one action or one form of integration may be a precursor of another. The interactive and dynamic integration process
between leader and organization was summed up by one observer as follows: Each
side compromised in their approach and now we see that there is harmony.
In Case 3, there are several dynamic phenomena, but a particularly interesting one
turns around the way in which Martin and his leadership team orchestrated the
decision to close hospitals in two distinct phases. The number system they used to
choose hospitals for closure was not put on the table immediately. Rather, prior public
consultations had already established the benefits of closing down some hospital
structures instead of spreading budgetary cutbacks across the system. Martins
leadership in this early phase, manifested in strong solidarity among regional board
members and support in the press enabled the board to claim consensus around the
initial decision a symbolic triumph that clearly influenced the acceptability of
subsequent discussions around the choice of sites for closure. At the same time, the
dynamics of the hearings generated other consequences that might be more problematic for future leadership actions. For example, the teams attentiveness to participants in the earlier hearings created increased complexity in the final proposal,
rendering future management and leadership tasks more challenging.
Overall, our point here is that leadership is something that evolves and manifests
itself over time, interacting with its context. As circumstances change, leaders may
easily gain or lose influence. Moreover, leaders may contribute either wittingly or
unwittingly to their own leadership gains and losses through their activities and
practices at any particular point in time and the consequences these may bring. As
Case 1 illustrates, and as we shall develop further in a later point, activities that seem
to be effective in achieving substantive impact may sometimes become problematic
later. This emphasis on the dynamic nature of leadership strongly recalls Pettigrews
(1992) call for a processual perspective on the emergence, development and
evolution of leadership roles.

Leadership as collective
Our research program also drew attention to the collective nature of leadership. We
argue with others (e.g. Gronn, 2002) that there is value, even within more hierarchical organizational forms to considering leadership not simply as an individual

74

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 75

Leadership

The Practice of Leadership in the Messy World of Organizations Denis et al.

attribute but as a coalitional phenomenon both throughout the organization and

among leaders at the top. While the upper echelons approach proposed by
Hambrick (2007) and others has moved the focus away from CEOs towards their top
management teams, more could be done to open up our understanding of how team
members interact and coalesce dynamically in the context of practical activity. In the
vignettes shown earlier, we emphasized the roles of three key individuals. However,
in each case, the activities that are central to their practices would have been impossible if these individuals had not succeeded in aligning themselves with others in
leadership positions with whom they then coordinated their actions. Indeed, part of
the dynamics of leadership described earlier involves the construction and evolution
of what we call leadership constellations following Hodgson et al. (1965). We now
briefly illustrate the importance of these constellations to the three cases.
Johns leadership activities in Case 1 would have been impossible if he had not
constructed strong alliances with a number of people who played co-leadership roles.
Central among these was Chris, a physician who had been elected to the hospitals
medical council and who had extensive political connections as well as a desire to
see the organization develop, and Mitch, a proactive board member rather tired of
what he saw as the defeatist tone dominating board meetings. The CEO also
contributed to the constellation by maintaining his support despite the somewhat
invasive initiatives of John and Chris. Various respondents described the relationships
among these protagonists: Chris had been looking for some time for a Director of
Professional Services he could work with (John); Johns arrival was a precipitating
factor for Chris. This is the guy we need (Manager); Chris was a good politician
. . . there was an alchemy between John and Chris . . . They had good relations with
Mitch on the board which helped move things along (Manager). The group created
a powerful constellation of people who mutually coordinated their actions, playing
differentiated but complementary roles (see also Denis et al., 1996, 2001). Without
this circle of collaborators, constructed over time, none of the protagonists would
have been able to achieve much in an organization where power and influence were
widely distributed.
Case 2 also clearly shows the collective nature of leadership. As a new CEO, Ivor
had to take into account the configuration of coalitions in place, and in particular the
three key poles of influence in this organization: the Board, the Medical Council
Executive and the administrative team. Indeed, his task as a new leader involved
inserting himself into an existing constellation, and gradually finding a way to inflect
its evolution over time. In this case, the prestigious medical staff exercised enormous
influence requiring Ivor to develop linkages with them before he could begin to
attempt moves in new directions. The process of leader integration seen in this case
study contrasts quite strongly with a more traditional perspective on leader integration as taking charge (Gabarro, 1987). Instead, the process can be viewed as one of
mutual accommodation and interdependence, in which the capacity to lead collectively emerged gradually over time. As one doctor put it towards the end of the study,
It is impossible that the medical council executive and the CEO not have the same
information on the institution. We have a common front.
In the case of the decision concerning the hospital closures (Case 3), we also
observed the collective character of leadership. It would not have been possible for
Martin to put forward this extraordinary proposal without having built up a strong

75

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 76

Leadership

6(1) Articles

constellation with his own staff and Board members. The internal cohesiveness of
the team was palpable and noted by several observers: Theres a lot of cohesiveness
in the team. Martins collaborative relationship with Russell, the Chairman of the
Board was particularly strong: The Board Chair and the CEO form a team . . .
exactly on the same wavelength and they have articulated their roles in a complementary way that is extremely mutually supportive. This was particularly critical
when the time came to defend the closures in public as the consistency with which
members of the team were able to articulate the reasons behind the decision added
to their credibility (Denis et al., 2006). At the same time, the coalition for change had
to extend outside the organization to be successful. Thus Martin and other members
of the team built alliances with other key health care leaders to ensure support of their
project.
Overall, building on the original work of Hodgson et al. (1965), we suggest that a
constellation of leaders will be more effective if it presents three characteristics:
specialization, differentiation and complementarity. Specialization refers to what each
member of the constellation brings in term of expertise and legitimacy (e.g. clinical
expertise, expertise in managing professionals, in setting up decision-making
processes). Differentiation refers to the division of labour among the coalition in order
to avoid excessive overlap and possible competition among leaders a non-negligible risk in contexts where individuals may have forceful personalities and strong
ambitions. Complementarity refers to the scope of the resources (expertise, legitimacy,
relationships) that a constellation has in regard to the challenges faced by the organization. The idea here is that the more a constellation is aligned with the issues faced
by an organization, the more its leaders will be in a position to have an impact.

Leadership as situated
Leadership is also contextually situated and practically enacted. Such a view implies
a need to look simultaneously at the properties of context and at the micro-level detail
of leadership practices in situ to understand how they achieve their results as
suggested by Alvesson & Sveningsson (2003a, 2003b) and Carroll et al. (2008). To
illustrate this idea, we zoom in on a some particular episodes in the three cases,
illustrating activities of the three leaders that appeared characteristic of how they
actually did leadership in the here and now. An optimal source of data for capturing
the micro-practices of leadership is the observations of leaders in action, something
that we were able to capture at least partly in Case 3 and will use for our illustration
here. Another source that offers a useful perspective on practice is leaders own
accounts or narratives of their practice (Rouleau, 2009). Although these accounts
may be less detailed and comprise elements of impression management, they provide
deeper insight into the thinking behind individual leaders behaviours, and the knowledge driving their practices. This is the source of data we use in our illustrations for
Cases 1 and 2.
To illustrate episodes of leadership practice for Case 1, we present two short
narratives or verbatim stories from an in-depth interview with John. These are shown
in Table 2.
These stories describe two micro-episodes that have striking similarities despite
their very different context. They both reveal Johns influence as a leader. In both

76

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 77

Leadership

The Practice of Leadership in the Messy World of Organizations Denis et al.

Table 2 Two micro-narratives of Johns leadership practice
Story 1: the big COUP

Story 2: tedirecting the fund-raising

campaign

People were saying we were underfinanced.

So the Finance director proposed a process
for demonstrating this with an accounting
firm . . . And I came right in and said, No
well take a different route and do a financial
analysis in the context of the current situation
and for our future strategy. And we called it
the Big COUP because jokingly I said we
needed to give a big kick [coup de pied in
French] and create the Committee for the
Organization, Unification and Promotion of
the hospital. And so we agreed to that, so it
was a very participative strategy that created
a wide movement to recognize the financial
difficulties of the hospital . . . It enabled
consideration of the implications of our
strategy that we hadnt done yet. I profited
from this to do that. This wasnt necessarily
appreciated by the Finance Director who had
his consulting firm lined up and they lost a
contract. But the CEO and the Board
approved it.

In the summer, the CEO with the foundation

decided to undertake a fund-raising drive
[for the anaesthesia department]. That isnt
easy to sell but it had been accepted by the
Board of the hospital and the foundation.
They asked me to come and talk to the
Board of the Foundation and to tell them
about anaesthesia . . . I went to the meeting,
the Board Chair was there, and I started
talking about traumatology. I said I thought
that anaesthesia was a terrible theme to sell
to the community, that the hospital couldnt
really identify with that, but with a more
strategic choice which was traumatology. In
the end, the Board decided to convene a
couple of days later and from one thing to
the next, they converged around
traumatology. And the only person who
changed that was me because the
anaesthetists were very happy to go ahead
with anaesthesia but the rest of the hospital
wanted traumatology. So we had a
fundraising campaign around that and I was
the spokesperson for the hospital.

cases he enters a situation arguing in opposition to plans promoted by people who

have apparent organizational authority and legitimacy, and he succeeds in driving
things in a completely different direction. His persuasiveness clearly reposes at least
in part on the strength of his arguments but also on the excitement and enthusiasm
he manages to create around his ideas. In each case, he moves people away from a
short-term operational choice that he presents as dull and bureaucratic towards a
more strategic and even glamorous decision compatible with the strategic plan he has
previously been involved in developing. His ideas are all driven by an articulate and
integrated vision that no one else seems to be providing, and moreover, it is one that
looks to the future optimistically. In the first story, the symbolic language and wide
participation of the big COUP is able to generate enthusiasm across the hospital
around the potential for development. In the second story, the idea of encouraging
investment in a prestigious traumatology program is clearly far more attractive and
strategic than the CEOs original plan to orient the campaign around anaesthesia.
Johns success in moving people around to his views is also related to his obvious
entrepreneurial drive and willingness to place his own energy in the service of these
initiatives. He became the coordinator of the big COUP and the spokesperson for
the traumatology fundraising drive. Thus the highly concrete and situated nature of
leadership practice is evident in these specific incidents. It is through such episodes
that leadership is enacted.
At the same time, these micro-level manifestations of leadership do not occur in

77

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 78

Leadership

6(1) Articles

a vacuum. It is here that the dynamic and collective dimensions of leadership as

processes are important to consider as a backdrop to situated practices. For
example, in order to fully appreciate Johns success in persuading others in the
above cases, it is important to understand what had happened before these incidents
that might make his message particularly credible. Johns success with the prior
strategic planning exercise as well as his recent very skilful handing of a local
environmental crisis in which the hospitals performance was seen as exemplary
were elements that almost certainly made him a person worth listening to. In
addition, the strong collaborative relationship he had already developed with Chris
ensured that when he spoke, other members of the leadership collective were
already with him. Thus situated practice is embedded in and indeed partly
constructed by these underlying dynamics.
To illustrate Ivors situated practices in Case 2, we look briefly at an account from
a middle manager towards the end of the integration period describing her reactions
to the new leader:
He has created breakfasts with the staff and two of my employees were part of
these breakfast talks. So . . . for them, I think what came out was that he was
someone who is accessible. Contrary to the image of a Director General that
everyone has you know a big man who walks around with a briefcase
(laughter . . .) who is just not accessible . . . He has demystified that image that
people had . . . You know, in meetings, he will say, We met with people from
the ministry, or were going to Quebec next week, and we will discuss this and
this. Before we had information, but it was always very general . . . When Ivor
X arrived, all of a sudden I realized the Director General exists (laughter) and he
sees us, he meets with us.
In this quotation, the very existence of a leader within the organization is literally
made manifest by his practices of accessibility. At the same time, accessibility to his
own employees also acquires value because of his visibility and prestige in the local
scene: He has wide visibility outside the hospital in the media and everything . . .
and at some point, people say thats our Director General. At this point, it is clear
that Ivor has been successfully integrated into a leadership role, he has been appropriated as such by his followers) and that the practices described in the two quotations appear to have contributed to that. Once again however, it would be simplistic
to consider the effects of these practices in isolation without also examining the
dynamic process of adjustment that led to this point, or without considering the strong
alliance that he had developed with the medical staff (i.e. the collective dimension of
leadership) that gave him the credibility to appear in public and with employees as
a leader.
Case 3 again illustrates the importance of situated practices, and in this particular
case of practices associated with managerial tools. In our study of closing hospitals,
we saw that the number system Martin and his colleagues used to orchestrate the
decision needed to be continually sustained along the way by the leaders in place.
Thus different members of the leadership team had to discursively support the value
of using rationality incarnated by numbers. Throughout the public hearings, they had
to constantly reaffirm the objectivity of their scoring system and to discredit any
counter propositions that would weaken their approach: In terms of criteria, those

78

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 79

Leadership

The Practice of Leadership in the Messy World of Organizations Denis et al.

that we used, the five of them, are the most objective criteria that exist. In order to
sustain this claim of rationality and objectivity, they relied only on numbers derived
from the hospitals themselves and propagated a uniform discourse that explained the
rationale behind the scoring system and its relative advantages compared to others
systems. They also demonstrated an attitude of transparency in using the numbers in
public, attempting to show that they were at the service of the people at large rather
than of particular interests: We wanted to be seen as an objective organizer who
defends the populations interests. The reaction of the media to their efforts were
remarkably favourable (see Vignette 3), suggesting that they had positioned themselves well as disinterested advocates for the collective good. In addition, through
the public consultation process, Martin and his colleagues symbolically disempowered their adversaries by making it very difficult to offer counterarguments, since any
questioning of the number system automatically redirected the threat of closure to
sister institutions (Denis et al., 2006).
It was through these situated practices during the hearings themselves that the
tools used by Martin and his team became relatively powerful and facilitated their
leadership in solving the problem of budgetary cutbacks. Yet once again these
practices were embedded in a dynamic process that contributed to their success.
Without the carefully managed sequential process described earlier that built up to
the hearings and without the solidarity among the collective leadership team (and
particular between Martin and Russell), these leaders could not have sustained the
integrity of these processes.
In summary, the three leaders studied showed different patterns in their situated
practices as revealed in these illustrations. Johns entrepreneurial moves succeeded
in deviating colleagues from preconceived paths by offering attractive alternatives
embedded in an overall strategic vision. Ivors skilful communication practices
brought him appreciation from the people who looked to him for leadership and
enabled him to become integrated. Martins rigorously rational and objective
approach accompanied with openness and transparency enabled him to achieve
consensus around a very difficult decision. In all three cases however, the micro-level
effects of their situated practices could not be understood without knowledge of the
context in which they occurred (i.e. the broader dynamics of the process, and the relationships developed with key co-leaders the collective dimension). It is important
to realize also that all of these situated practices embed contradictions. Their effects
are not so simple or one-sided as we may have implied in this section. This brings
us to the fourth dimension we wish to emphasize in this article.

Leadership as dialectic
Throughout the research program described here, we encountered contradictions in
the exercise of leadership in situations of change. Collinson (2005) recently drew
attention to three dialectic dimensions of leadership practices that he labels
control/resistance, consent/dissent and men/women. He suggests for example that
apparently successful leadership practices inherently generate their own resistance,
that consent may be manufactured and hide overt or passive practices of dissent
among followers and that gender relations may be embedded in these tensions. He
notes that,

79

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 80

Leadership

6(1) Articles

a dialectical approach suggests that studies need to acknowledge the deep-seated

asymmetrical power relations of leadership dynamics. It recognizes that leaders
exercise considerable control and that their power can also have contradictory
outcomes which leaders either do not always understand or of which they are
unaware. (p. 1435)
The three cases we studied reveal that leadership has dialectic qualities based on the
equivocal nature of many leadership actions and practices. Indeed each of the
vignettes suggests a cyclical process in which each leaders practices had a dark
side that ultimately came to the surface and that we now explore in more depth.
If we consider Case 1, a second look at the two practice narratives in Table 2
suggests that Johns practices of entrepreneurial leadership were not without risk. For
example, in the Grand COUP story, we see that the Finance Director has been largely
undermined by Johns proposed solutions. In the fund-raising story, there are also
losers, starting with the anaesthetists who had expected to be the main beneficiaries
of the hospitals campaign. One might also surmise that the CEO who had initially
supported the anaesthetists might find the situation somewhat uncomfortable.
Through practices like these in which he aggressively promoted his strategic ideas,
John created pockets of dissatisfaction and tension surrounding himself even as he
achieved immense strides in developing the hospital and in improving its performance in collaboration with Chris, Mitch and others. Ultimately, as one observer put
it, You cannot keep on breaking the china. After some time in their positions, John
and Chris had created too many enemies and it was time to move on. It is perhaps
no accident that the person who replaced Chris as the new President of the Medical
Council was the chief anaesthetist.
Ivors situated practices as described earlier also had contradictory effects. While
his open-door policy and accessibility were appreciated by the medical staff and
employees, they became a bone of contention for his executive team who felt that
this practice tended to bypass their authority. Indeed, one manager offers a somewhat
different perspective on this practice from that indicated earlier:
He says I will go and eat with you in the cafeteria, and we are there at the table
at lunchtime with the employees of the hospital and the Director General in
shirtsleeves who says, Lets talk. Well, people who have never seen that in their
lives think that at last they have access to the pipeline. They take anything that is
said for a decision. Participation is opened up everyone wants to participate,
but should it be around the Board table or at the proper level and then it goes up
through the organization? . . . So now he risks having everyones problems in his
office.
Ivor himself also gradually came to see the risks of excessive openness and found
himself modulating his approach as illustrated in the following narrative:
Im going up the stairs . . . and I meet a doctor, a service head, coming down. He
says, Hi, look, I have a great idea for my service . . . we need to do things this
and this way. I say, Thats very interesting, we should think about it. Two days
later, I have a department head come into my office and say, You authorized
$300,000 for the complete reorganization of the service. You have to be very

80

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 81

Leadership

The Practice of Leadership in the Messy World of Organizations Denis et al.

careful in an organization like that to create an environment and structures that

allow formal, democratic and open decision making.
For Case 3, we also see a dialectical effect in the use of the number system fabricated by Martin and his team at the regional board. Part of the reason the number
system was successful was because of its constructed objectivity. This enabled the
team in some sense to distance themselves from a difficult decision the numbers,
not they, were responsible for the choice of hospitals to be closed. And yet, by implicitly assigning agency to numbers, they actually diluted their own discretion as
leaders. In fact, among the nine hospitals targeted for closure, there was one organization among them that was probably more effective and efficient than the others
and that had managed to mobilize considerable community support. Some Boards
members were conscious of this situation but there was no way of making adjustments to the decision without jeopardizing the legitimacy of the whole process. Had
the Board taken seriously any of the arguments put forward contesting individual
closure decisions, it would have thrown the whole project up in the air. Once the
number system was implemented, the regional Board and the targeted hospitals
appeared to be locked into the decision.
In summary, as shown in this section, the situated practices of leaders rarely have
unequivocally positive effects. Even apparently successful practices embed within
themselves contradictory effects a dark side that may and often does come back
to haunt leaders.
Indeed, leadership roles are by nature transient, and leaders are subject to unexpected forces for change including the consequences of their actions, practices, and
decisions. They evolve in complex systems of interactions including patterns of
power and interests that they cannot fully control. Nor can they perfectly anticipate
the context and the outcomes of their decisions. This brings us full circle to the
dynamic nature of leadership processes. Indeed, we would argue that the dialectic
nature of leadership practices is a major force behind the dynamics of leadership (see
also Denis et al., 2001). Specifically, Johns practices in Case 1 were often substantively effective, but as time went on politically problematic. Ivor learned to adapt his
practices over time to avoid their more problematic political effects and use them
increasingly to his advantage. And Martin was successful in the short term, but had
to deal with some of the unanticipated consequences of his practices once he had
achieved consensus around the hospital closure decision.

Implications for future research

We hope that the adoption by researchers of a dynamic, collective, situated and
dialectic approach to the study of leadership practices in a diversity of settings will
increase our grasp of the context and process of leadership and the dilemmas of
achieving influence beyond the myth of absolute power. Re-acknowledging these
attributes of leadership has implications for future work. There is clearly room for
more research that documents in much richer detail the actual doing of leadership in
complex settings (Alvesson & Sveningsson. 2003a, 2003b; Carroll et al., 2008;
Pettigrew, 1992). In the light of the ideas and findings presented earlier, we will now
discuss four distinct directions in which we believe this line of research could develop

81

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 82

Leadership

6(1) Articles

under the following theme headings: contextualizing micro-leadership practices;

collective leadership practices and dynamics; the role of practical knowledge in
leadership; and the materiality of leadership.

Contextualizing micro-leadership practices

As we indicated at the beginning of this article, proponents of research on the doing
of leadership have taken two rather different approaches to this topic. On the one
hand, Pettigrew (1992) and other process scholars have emphasized the need to look
at the evolution of leadership roles over time. On the other hand, scholars adopting
a social practice lens (Alvesson & Sveningsson, 2003ab; Carroll et al., 2008; SamraFredericks, 2003) have urged a more micro-level focus on situated activities and
interactions and their consequences in the here and now. The observations presented
in the current article suggest that neither of these views is entirely complete. The
broad process perspective neglects the highly specific skills and practices through
which leaders enact their roles and influence others as illustrated in the examples
above. However, an exclusive focus on situated micro-level activity ignores what is
brought into the situation from previous activities and interactions (i.e. the dynamic
and collective dimensions). In other words, Johns persuasive interventions in
meetings, Ivors skilful listening and chatting with employees (Alvesson &
Sveningsson, 2003a), and Martins brave performance in public hearings depend in
part for their success on everything that has gone before as well as the collaborators
and allies that are brought onto the scene to support them either physically or
virtually.
In addition, micro-level studies of leadership in action, such as for example
Samra-Fredericks (2003) impressive analysis of the rhetorical devices used by six
strategists in a management meeting, tend to look at the consequences of these
practices within a very short interval surrounding the events the emphasis is often
on what the practices appear to accomplish in the present. Yet, as we showed when
we discussed the dialectic dimension, the long-term consequences may be somewhat
different, but nevertheless important. Researchers (as well as leaders) need to
consider the potential benefits and costs incurred by leaders in the future while
achieving influence in the present.
In sum, while we advocate a research agenda that takes a practice perspective on
leadership, we believe that it will be most productive if accompanied by a focus on
all four of the dimensions that we have delineated in this article, embedding an
emphasis on micro-level detail within a broader processual and contextual analysis.

Collective leadership practices and dynamics

We found the notion of the leadership constellation derived from Hodgson et al.
(1965) to be useful to understand the dynamic and collective dimensions of leadership. The three properties of a leadership constellation (differentiation, specialization
and complementarity) can be a starting point to further study empirically how
leadership capacity develops in organizations. We need to open the black box of these
properties in order to reveal how they are enacted and performed by individual
leaders in everyday practices. More specifically, how do leaders come to define their

82

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 83

Leadership

The Practice of Leadership in the Messy World of Organizations Denis et al.

roles within a constellation and adapt them to the behaviours of others members in
action and interaction? In addition, the process by which a constellation comes to
incorporate new leaders to respond to emerging challenges while maintaining its
functionality appears to be fundamental, and potentially problematic. We also have
relatively limited knowledge of the process and practice that may favour the emergence of new leaders in organizations and ensure more deliberate transitions in
leadership constellations.
Studies should also pay more attention to the evolving dynamics and tensions
within leadership constellations. In our research, we saw that constellations may
evolve through periods of momentum gain and momentum loss in the process of
change (Denis et al., 1996, 2001) evidenced very much in Case 1 (John). Relationships among the members of a constellation contribute to enabling or limiting
change, but change in turn places stress on leadership constellations (Denis et al.,
2001). It is important to understand the type of practices used by leaders to deal with
such contradictions.
Overall, the study of the enactment, emergence, formation and dissolution of
leadership constellations from a practice perspective is a potentially rich research
opportunity, although one that poses significant challenges in terms of the degree of
access and intimacy that may be required to capture these dynamics.

The role of practical knowledge in leadership

Looking at the three cases, it can be seen that John, Ivor and Martin rely on a set of
relational capabilities and competences deployed in their day-to-day activities to
achieve their leadership positions, although they each do this in very different ways.
Among other things, their capacity to practically learn from and deal with others is
at the centre of their organizational integration. We would argue that part of the
power of aspiring leaders to successfully act resides in their capacity to put to work
their practical knowledge of the context in positioning themselves with respect to
others. More research is needed to understand the practical knowledge that underlies
this positioning.
Moreover, effective leaders need to be able to move easily through a variety of
settings. We need to know more about the strategies used by leaders to acquire practical knowledge of a diversity of settings and to routinely manoeuvre among multiple
foci of decision making. Much of this knowledge is tacit, gained through conversations and observations acquired even as leaders pursue their efforts to shape their
context of action. However, by detailed observation and analysis of conversations and
interactions, it may possible to capture elements of this knowledge in action as
leaders intervene in the day to day. Rouleau & Baloguns (2007) recent work on the
knowledge resources of individual leaders begins to move in this direction.
Taking this view one step further, we would argue that for a relationally skilled
leader, appropriate emotions need to be rationally bounded and politically conveyed
through the materiality of the leaders body (Neumann, 2008). For example, an
open-door policy means nothing if the body seen does not show multiple signs of
openness. In the interview quotations offered above for Case 2, references to eating
in the cafeteria in shirtsleeves as opposed to big men who walk around with briefcases provide hints of the embodied nature of this practice. To be inspiring and to

83

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 84

Leadership

6(1) Articles

be convincing, a leader has also to give sense through the signs that his or her body
conveys when entering in relationships with others. This dimension has rarely been
explored in the literature but is an important hidden side of its situated and practical
nature.

The materiality of leadership

All leaders use tools of various kinds in their efforts to influence others. Case 3 in
particular invites leadership researchers to look more specifically at managerial tools
such as models, techniques, rules, standards, numbers systems, control system, documents, or contractual techniques that might be used to drive the organization towards
success. These tools are not just powerful devices for action. In following the controversies surrounding them, we suggest that it is possible to better understand how they
contribute or not to helping leaders secure compromises among multiple interests
often brought to bear in critical situations.
In this view, leadership is not just a matter of traits, or of personal, discursive and
symbolic characteristics; there is a materiality to leadership that intervenes in the web
of relationships that leaders develop within and around their organizations. Instead
of looking at leadership only through individual action, the collective and situated
character of leadership might also be revealed and understood by following the set
of managerial tools they successfully use to achieve their goals and to make relevant
decisions. Here we are suggesting that leadership can be studied through procedures
and tools rather than only focusing on intentions or on visible effects of power. Such
a position could allow researchers and practitioners to better understand the various
ways the organizational knowledge embedded in these tools is deployed and appropriated by multiple actors, potentially linking them together in a network as
suggested by Latour (1989).
Managerial tools are not only rational and technical. According to Latour (1989),
they convey an ostensive and a performative view of action. As institutionalists argue,
managerial tools are in some ways rational myths (Meyer & Rowan, 1977). In order
to be used by leaders in complex multiparty situations, their objectivity must be
socially constructed. However, leaders should never forget that behind their technical rationality, managerial tools hide metaphorical dimensions that convey a simplified version of organizational relations that they have participated in constructing and
in which they could become trapped as they use them to achieve their goals. The
study of the role of tools in leadership offers another fruitful avenue for research.

Lessons for leaders

The implications of seeing leadership as dynamic, collective, situated and dialectical for practitioners are numerous. An image of leadership as a collective enterprise
draws attention to the importance of creating a functional group of leaders. Leadership cannot be competently exercised in large, complex and messy organizations
without taking into account the multiple, dynamic and more or less fragile coalitions
that constitute them. Thus, leadership should not be viewed as an external authority
or symbol influencing others from outside. Rather, leaders need to see themselves as
embedded in networks that they do not fully control.

84

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 85

Leadership

The Practice of Leadership in the Messy World of Organizations Denis et al.

This perspective also demands more humility from people in leadership positions.
Leaders are transient in the sense that challenges and dilemmas change and a given
leader may not be the best person to contribute to organizational development in a
changing context. Individuals in a position of leadership need to accept the need to
leave space for others in an existing constellation something that seems hard to do
in many cases.
In addition, leaders need to pay attention to the fact that it is also through their
day-to-day interactions and actions that they are shaping their leadership capacity.
The knowledge they gain of the in-depth functioning of their organization through
daily interactions is an invaluable resource of learning. Leaders must also attempt to
practically tap into the value systems that reflect key aspects of organizational
identity and practices. This suggests, for example, that a leader able to incarnate
highly significant organizational values may have a better chance of gaining support
from others.
Finally, pressures are very high in organizations today to develop and implement
more sophisticated systems of management (information system, performance
management systems) and to rely more on formal agreements such as contracts and
management tools especially in the public sector. While, managerial tools are an asset
in the hands of aspiring leaders, their utilization is not without risk as we observed.
Such tools may increase the influence of leaders while constraining their agency at
the same time. Leaders need to pay attention to the positive and negative consequences of these tools for their roles in decision-making processes. Indeed, as this
article suggested, almost any successful leadership intervention could have a potential downside. As Collinson (2005) indicated, power does not come free: leaders need
to understand that their activities create waves of substantive, symbolic and political
consequences that may not all be fully evident in the passing moment.

Acknowledgements
The authors thank the Social Sciences and Humanities Research Council of Canada,
the Fonds qubcois pour la recherche sur la socit et la culture, the Fonds de
recherche en sant du Qubec and the Canadian Institutes for Health Research for
their support of the research described in this article. They also thank two anonymous reviewers for their comments on an earlier version of the paper.

References
Alvesson, M., & Sveningsson, S. (2003a) Managers Doing Leadership: The
Extra-ordinarization of the Mundane, Human Relations 56(12): 143559.
Alvesson, M., & Sveningsson, S. (2003b) The Great Disappearance Act: Difficulties in
Doing Leadership, Leadership Quarterly 14: 35981.
Balogun, J., & Johnson, G. (2004) Organizational Restructuring and Middle Manager
Sensemaking, Academy of Management Journal 47(4): 52349.
Bass, B.M., & Avolio, B.J. (1994) Improving Organizational Effectiveness Through
Transformational Leadership. Thousand Oaks, CA: Sage Publications.
Bennis, W., & Nanus, B. (1985) Leaders: Strategies for Taking Charge. New York: Harper &
Row.
Carpenter, M.A., Geletkanycz, M.A., & Sanders W. (2004) Upper Echelons Research

85

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 86

Leadership

6(1) Articles
Revisited: Antecedents, Elements, and Consequences of Top Management Team
Composition, Journal of Management 30(6): 749.
Carroll, B., Levy, L., & Richmond, D. (2008) Leadership as Practice: Challenging the
Competency Paradigm, Leadership 4(4): 36379.
Collinson, D. (2005) Dialectics of Leadership, Human Relations 58(11): 141942.
Denis, J.L., Lamothe, L., & Langley, A. (2001) The Dynamics of Collective Leadership and
Strategic Change in Pluralistic Organizations, Academy of Management Journal 44(4):
80937.
Denis, J.L., Langley, A., & Cazale, L. (1996) Leadership and Strategic Change Under
Ambiguity, Organization Studies 17(4): 67399.
Denis, J.L., Langley, A., & Pineault, M. (2000) Becoming a Leader in a Complex
Organization, Journal of Management Studies 37(8): 106399.
Denis, J.L., Langley, A., & Rouleau L. (2005) Rethinking Leadership in Public
Organizations, in E. Ferlie, L. Lynn & C. Pollitt (eds) The Oxford Handbook of Public
Management, pp. 44667. Oxford: Oxford University Press.
Denis, J.L., Langley, A., & Rouleau, L. (2006) The Power of Numbers in Strategizing,
Strategic Organization 4(4): 34977.
Denis, J.L., Langley, A., & Rouleau L. (2007) Strategizing in Pluralistic Contexts:
Rethinking Theoretical Frames, Human Relations 60: 179215.
Denis, J.L., Lamothe, L., Langley, A., Breton, M., Gervais, J., Trottier, L.-H.,
Contandriopoulos, D., & Dubois, C.-A. (2009) The Reciprocal Dynamics of Organizing
and Sensemaking in the Implementation of Major Public Sector Reforms, Canadian
Public Administration 52(2): 22548.
Fiedler, F.E. (1967) A Theory of Leadership Effectiveness. New York: McGraw-Hill.
Gabarro, J. (1987) The Dynamics of Taking Charge. Boston, MA: Harvard Business School Press.
Gronn, P. (2002) Distributed Leadership as a Unit of Analysis, Leadership Quarterly 13(4):
42351.
Hambrick, D.C. (2007) Upper Echelons Theory: An Update, Academy of Management
Review 32(2): 33443.
Hambrick, D.C., & Mason, P.A. (1984) Upper Echelons: The Organization as a Reflection
of Its Top Managers, Academy of Management Review 9: 193206.
Hodgson, R.C., Levinson, D.J., & Zaleznik, A. (1965) The Executive Role Constellation.
Boston, MA: Harvard Business School Press.
House, R.J., & Aditya, R. (1997) The Social Scientific Study of Leadership: Quo Vadis?,
Journal of Management 23: 40974.
Judge, T.A., & Piccolo, R.F. (2004) Transformational and transactional leadership: A
Meta-analytic Test of Their Relative Validity, Journal of Applied Psychology 89(5):
75568.
Judge, T.A., Bono, J.E., Ilies, R., & Gerhart, M.W. (2002) Personality and Leadership: A
Qualitative and Quantitative Review, Journal of Applied Psychology 87(4): 76580.
Kets de Vries, M.F.R. (2001) The Leadership Mystique: A Users Manual for the Human
Enterprise. London: Financial Times/Prentice Hall.
Knights, D., & Willmott, H. (1992) Conceptualizing Leadership Processes: A Study of
Senior Managers in a Financial Services Company, Journal of Management Studies 29:
76182.
Latour, B. (1989) La science en action. Paris: Dcouverte.
Mintzberg, H. (1997) Toward Healthier Hospitals, Health Care Management Review 22(4):
918,
Parry, K.W., & Bryman, A. (2006) Leadership in Organizations, in S.R. Clegg, C. Hardy,
T.B. Lawrence & W.R. Nord (eds) The Sage Handbook of Organization Studies,
pp. 44768. Thousand Oaks, CA: Sage.

86

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 87

Leadership

The Practice of Leadership in the Messy World of Organizations Denis et al.

Pettigrew, A.M. (1992) On Studying Managerial Elites, Strategic Management Journal 1:
16382.
Rouleau, L. (2005) Micro-practices of Strategic Sensemaking and Sensegiving: How
Middle Managers Interpret and Sell Change Every Day, Journal of Management Studies
42(7): 141343.
Rouleau L. (2009) Studying Strategizing Through Narratives of Practice, in D. Golsorkhi,
L. Rouleau, D. Seidl & T. Vaara (eds) Cambridge Handbook of Strategy as Practice.
Cambridge: Cambridge University Press.
Rouleau, L., & Balogun, J. (2007). Exploring Middle Managers Strategic Sensemaking Role
in Practice, Advanced Institute of Management Research, AIM Working Paper Series
055-March 2007.
Schatzki, T., Knorr-Cetina, K., & von Savigny, E. (2001) The Practice Turn in Contemporary
Theory. London: Routledge.
Stogdill, R.M. (1948) Personal Factors Associated with Leadership: A Survey of the
Literature, Journal of Psychology: 3571.
Tichy, N.M., & Devanna, M.A. (1986) The Transformational Leader, Training and
Development Journal 40(7): 2732.

Appendix: original case studies and data sources used

The current article draws on the data from three previously published case studies.
Here we analyse these data transversally and at an individual-leader level of analysis
to illustrate four analytical dimensions of leadership practice.

Case 1: leadership of change in a context of ambiguity (Denis et al.,

1996 )
This study examined the evolution of leadership roles during a period of radical
change in which a general hospital acquires a university affiliation and implemented
a more integrated form of management. It traces the tactics used by leaders to implement change and how these tactics enhance or decrease leaders credibility and the
progress of change itself.
Data sources: Minutes of meetings, correspondence over 14 years, 17 in-depth retrospective
interviews.

Case 2: the process of integration of a new leader (Denis et al., 2000)

The study examined the processes of entering and establishing a position of leadership in a complex health care organization. The study shows how accommodation
between the leader and the organization was differently achieved with medical staff
and administrative team.
Data sources: Minutes of meetings, documents, 29 in-depth interviews (4 with CEO at regular
intervals).

Case 3: the role of numerical tools in leadership (Denis et al., 2006)

The study traced how a new leadership team orchestrated a very difficult decision
involving the closure of nine hospitals. In particular, the role of a number system in

87

Downloaded from [Link] at University of Roehampton on July 16, 2012

067-088 LEA354233 Denis_170 x 242mm 03/02/2010 09:39 Page 88

Leadership

6(1) Articles

legitimating the decision was examined. The study shows how the team built consensus around the decision despite its difficult nature.
Data sources: Minutes of meetings, public hearings transcripts and video tapes, 25 in-depth
interviews, observation of 61 local board meetings within hospitals and regional agency.

Jean-Louis Denis is Professor of healthcare administration at the University of

Montral. He holds the Canadian Health Services Research Foundation/Canadian
Institutes for Health Research Chair in the Governance and Transformation of Health
Care Organizations. His current research is in the area of leadership and strategic
change in healthcare organizations, integrated health networks and the role of
research-based evidence in the adoption of innovations in the healthcare sector.
[email: [Link]@[Link]]
Ann Langley is Professor of management at HEC Montral and Canada research
chair in strategic management in pluralistic settings. Her research focuses on strategic change, leadership, innovation and the use of management tools in complex
organizations with an emphasis on processual research approaches. She has
published over 50 articles and two books, most recently Strategy as Practice:
Research Directions and Resources with Gerry Johnson, Leif Melin and Richard
Whittington (Cambridge University Press, 2007).
Linda Rouleau is Professor at the management department of HEC Montreal. Her
research work focuses on micro-strategy and strategizing and on the transformation
of control and identity of middle managers in a context of organizational restructuring. In the last few years, she has published in peer reviewed journals such as Journal
of Management Studies, Human Relations, Journal of Management Inquiry,
Strategic Organization, etc. She is co-responsible for the GPS (Study Group of
strategy-as-practice, HEC Montreal) and research member of the CRIMT (a
Canadian research centre on globalization and work).

88

Downloaded from [Link] at University of Roehampton on July 16, 2012