Capella University Global Leadership Lens & the Multidisciplinary Lens Discussion
For this question, select an organization or industry with which you are familiar and, referring to scholarly literature, provide 2 additional references, and respond to the following 3 points below:
1. Examine one multidisciplinary environmental context (social, political, cultural, legal, or economic) that is influencing global leaders today
2. Using the Hernandez et al. 2011 article (attached), define the loci and mechanisms that might be considered in selecting appropriate leadership systems.
3. Discuss the relevance of theoretical models in aligning leadership systems with context in helping to maximize an organization’s potential and contribute to its viability.
ORDER NOW FOR AN ORIGINAL PAPER ASSIGNMENT: Capella University Global Leadership Lens & the Multidisciplinary Lens Discussion
The loci and mechanisms of leadership: Exploring a more comprehensive
view of leadership theory
Morela Hernandez a,⁎, Marion B. Eberly b,1, Bruce J. Avolio a,2, Michael D. Johnson a,3
a Michael G. Foster School of Business, University of Washington, Seattle, Paccar Hall, Box 353226, Seattle, WA 98195-3200, USA
b Milgard School of Business, University of Washington, Tacoma, 1900 Commerce Street, Box 358420, Tacoma, WA 98402-3100, USA
article info abstract
Available online 15 October 2011 Through a qualitative review of the leadership literature, we derive two fundamental principles for codifying the last century of leadership theory and research: the locus and mechanism
of leadership. Our systematic review and categorization of past theories suggests that further
development of the follower, collective and context loci, and the affect mechanism is needed.
Building on these insights, we propose that by simultaneously considering all five loci and four
mechanisms in their theories, leadership scholars can advance a more comprehensive and integrative understanding of the leadership phenomenon. We then demonstrate the potential of
using the locus and mechanism framework for examining and expanding current and future
leadership theory including work on shared and strategic leadership.
Published by Elsevier Inc.
It is not uncommon for both leadership practitioners and academics to lament the range of definitions that are typically used
in the literature to describe leadership. The differences in how leadership has been defined have resulted in disparate approaches
to conceptualizing, measuring, investigating, and critiquing leadership. For example, some authors have focused solely on the
leader to explain leadership, whereas others have examined leadership from a relational, group, or follower-centered perspective.
To add to the differentiation that has emerged in the leadership literature, other authors have focused on examining leader traits
versus behaviors, while still others have drawn from the cognition and affect literatures to explicate leadership and its effects.
In this article, we propose some fundamental principles that can be used to integrate the numerous theories of leadership generated over the past 100 years. We suggest that the diverse range of perspectives developed within the leadership literature all
represent legitimate ways of conceptualizing leadership; with each having contributed to furthering our understanding of
what constitutes leadership in terms of its breadth and depth. At the same time, such breadth makes it extremely difficult not
only to create a coherent picture of the state of leadership theory and research, but also to guide what is needed in terms of
the next stage of theory development.
More than a decade ago Graen and Uhl-Bien (1995: 220) expressed this same concern, suggesting that, “as new leadership
theories emerge, attempts to classify them into general categories of approaches are becoming more difficult.” Their taxonomy
for classifying leadership research encompassed three domains: the leader, the follower, and the leader–follower relationship,
where each could be analyzed from different levels (e.g., the individual, dyad, group, organization level). Notably absent from
their classification scheme was the context in which leaders, followers, and their relationship are embedded (Avolio, 2007). Moreover, while their initial taxonomy provided a starting point, Graen and Uhl-Bien (1995) have acknowledged it was not complete,
in part because it excluded important domains (e.g., groups and teams, the context).
The Leadership Quarterly 22 (2011) 1165–1185
⁎ Corresponding author. Tel.: +1 206 221 5140.
E-mail addresses: email@example.com (M. Hernandez), firstname.lastname@example.org (M.B. Eberly), email@example.com (B.J. Avolio), firstname.lastname@example.org (M.D. Johnson). 1 Tel.: +1 253 692 5630. 2 Tel.: +1 206 543 7908. 3 Tel.: +1 206 616 2756.
1048-9843/$ – see front matter. Published by Elsevier Inc.
Contents lists available at SciVerse ScienceDirect
The Leadership Quarterly
journal homepage: www.elsevier.com/locate/leaqua
We propose that to formulate what might be a more comprehensive leadership system, in which leaders, followers, collectives,
and context co-exist, integration across diverse theoretical perspectives is necessary. Our purpose here is thus twofold. First, we
set out to formulate a common language with which leadership researchers may explain, argue, and defend their theoretical
views without restricting new developments or expansions to existing theories. Accordingly, we begin by conducting a qualitative
review of the core leadership theories—a process that is analogous to conducting an empirical meta-analysis, although our focus
here was on analyzing and aggregating theories as opposed to empirical findings associated with those theories, in order to derive
an integrative view of leadership. Based on the results of this qualitative review, we have concluded that each leadership theory
attempts to answer two fundamental questions, which we used to devise our basic rules or principles for codification: “Where
does leadership come from?” and “How is leadership transmitted?” Accordingly, we propose two dimensions to classify leadership theories in a meaningful and useful way: the locus and the mechanism of leadership (see Fig. 1).
Second, through a systematic review and categorization of the leadership literature utilizing our two-dimensional framework,
we demonstrate that considerable gaps exist in the development of the follower, collective and context loci, and the affective
mechanism of leadership. We propose that to fully understand the complex intricacies of the leadership process, a simultaneous
consideration of all five loci and four mechanisms will be needed. To demonstrate the potential of our proposed locus and mechanism framework in advancing current and future leadership theory, we use it to examine two promising approaches that have
received growing attention in the leadership literature: shared leadership and strategic leadership. Both approaches currently
conceptualize leadership via a limited set of loci and mechanisms thus underestimating the complexity of these different leadership constructs. Using our proposed framework, we believe we are better able to examine and clarify the theoretical limitations
and opportunities for each theory and target future research directions.
1. A common language: the loci and mechanisms of leadership
1.1. Where leadership comes from: the loci of leadership
The first dimension for categorizing leadership theory is the locus of leadership. We define locus of leadership as the source
from which leadership arises. When considering the locus of leadership, we ask whether the leader is the sole initiator of leadership or if other loci initiate leadership. Various leadership theories conceptualize the locus as an individual (e.g., the leader), multiple people (e.g., a group of followers), and/or the context within which people interact. Supporting our focus on the origins
Fig. 1. Placing leadership theories within the two-dimensional framework. (The order of the loci and mechanisms and the size and distribution of the theories’
boxes were chosen to maximize graphical clarity. Please note that some theories are represented by two separate boxes.)
1166 M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185
of leadership or locus, Hollander and Julian (1969: 395) suggested that the sources of leadership require “greater attention in future investigations,” and Conger and Kanungo (1987: 639) called attention to examining what constituted the “locus of charismatic leadership.” Based on our qualitative review, we categorize existing leadership theories into five loci: leader, follower,
leader–follower dyad, collective, and context. We note that these loci do not lie on a continuum, but rather represent reasonably
independent categories. The following description of the five loci guided our categorization of each theory.
The most straightforward and traditional approach is to view the leader as the locus of leadership. Leadership theories classified into this category, state that leadership either totally or partially arises from the leader. Theories that posit the followers as
the locus of leadership identify unique aspects of followers which by themselves (i.e., independent of the leader) would make
leadership possible. Conceptualizing followers as a locus of leadership is therefore distinct from the concept of followership,
which examines “what followers do when they follow”(Rost, 2008: 54).
Theories that discuss leadership origins within the leader–follower relationship may be placed in the dyad category, with
the emphasis on leadership arising from specific features of the relationship rather than unique partners in the relationship.
The collective locus includes those theories where leadership is presumed to arise from the interconnected relationships of people
within a specific group of individuals (e.g., work teams). It is important to note that in this latter category, leadership is conceptualized as a group-level phenomenon (and thus is distinct from the individual leader or follower loci), arising from the set of
relationships among group members, not just a single relationship (and thus is distinct from the dyad locus).
Finally, theories that describe features of the environment as giving rise to leadership are assigned into the context category.
This category includes leadership theories that take into account relationships beyond a predefined group and/or recognize the
power of the context such as team members’ social networks, cultures, or norms within an organization in explaining the source
of leadership. Other aspects of the organizational environment such as uncertainty and crisis, for example, can also influence the
formation of leadership.
We define context following Johns (2006: 386) as being “situational opportunities and constraints that affect the occurrence
and meaning of organizational behavior”. The context category is necessarily broad and includes multiple levels. For example,
the context can exist at the individual level (e.g., an employee’s job design), the group level (e.g., the reward system of the
group), the organizational level (e.g., organizational culture), and the societal level (e.g., national cultural norms), as well as
across these respective levels.
Interestingly, context can both exert influence on, and receive influence from, organizational members. Mowday and Sutton
(1993) discussed this two-way influence, noting that both powerful members, and less powerful members acting in concert,
can alter the organizational context. For example, a strong new organizational leader can present a compelling vision or introduce
new policies that rapidly change the context in which the followers work. Alternatively, aggregating followers’ perceptions in
terms of what constitutes the “climate” can create a powerful context that influences organizational newcomers.
We suggest, then, that context is not simply the situation in which leadership is enacted. Rather, context can act as a direct
determinant of the nature of leadership; hence its inclusion as a locus of leadership. Accordingly, it is important to note that
the theories placed in this category should not merely categorize context as an antecedent, mediator or moderator variable,
but as an actual “spring” that generates leadership (Uhl-Bien, Marion, & McKelvey, 2007).
The concept of leadership locus is similar to what Graen and Uhl-Bien’s (1995) taxonomy calls domains of leadership. We
drew on this conceptualization in developing the locus dimension, but noted some key differences. First, the earlier concept of
domain focuses only on leader, follower, and leader–follower dyad; based on more recent theories, we added group and context
to this typology. Second, and more importantly, their description does not state that leadership originates from the domains,
but rather that the focus of the research is on the leader, follower, or their relationship. A focus on followers, for example,
could include followership behaviors such as obedience to an authoritarian leader, whereas the locus of followers demonstrates
how followers enact leadership behaviors such as convincing peers to embrace the organization’s vision.
Some may argue that our conceptualization of locus references levels of analysis—a concept upon which leadership theories
and measures have already been categorized and organized (Klein, Dansereau, & Hall, 1994). Levels of analysis are commonly
defined as “entities or objects of study” (Yammarino, Dionne, Chun, & Dansereau, 2005: 882), and are assigned to individual
variables as opposed to entire theories. Yammarino et al. (2005) reviewed a wide range of leadership articles to determine at
which level of analysis each paper’s variables were conceptualized and/or measured. Their coding revealed that most leadership
theories were examined at multiple levels of analysis, either separately or simultaneously (e.g., multilevel models).
The work of Yammarino and his colleagues illustrates a major distinction between levels of analysis and our conceptualization
of loci. The locus of leadership is a core component for each theory; it does not change as researchers empirically test the theory’s
propositions at different levels of analysis. One can examine, for example, ethical leadership at the individual level (e.g., how does
an ethical leader make decisions?) or across multiple levels (e.g., how does an ethical leader influence a group’s ethical norms?).
Whatever level at which ethical leadership is studied, it is the ethical leader who remains the locus from which leadership
1.2. How leadership is transmitted: the mechanisms of leadership
The second dimension discovered in our qualitative review of leadership theories is the mechanism of leadership. This is
the actual process through which the locus of leadership exercises influence; we thus define mechanism as the means by which
leadership is enacted. Leadership scholars have long acknowledged the importance of studying various mechanisms to advance
our understanding of how leaders exert influence. In his critical review of the charismatic and transformational leadership
M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185 1167
literature, Yukl (1999) asserted there is much to be discovered about the underlying processes through which leaders influence
follower attitudes, behaviors, and motivation. Kark & Van Dijk (2007: 511) explicitly discussed this idea when they suggested
that “one of the mechanisms by which transformational and/or charismatic (leaders) exert their influence on followers … is by
eliciting a promotion focus among their followers.”
As with the locus dimension, leadership scholars have described various approaches to how leadership is enacted. Our review
suggests that leadership theories can be categorized by four mechanisms: traits, behaviors, cognition, and affect. Again, it is important to recognize that they do not form a continuum, and therefore represent relatively independent categories.
Modern leadership theory originated in the view that leaders exhibit certain personality characteristics that help to differentiate them from other individuals (e.g., Galton, 1869). The term personality traits perhaps best captures the stable and enduring
qualities and patterns of individuals’ emotions, thoughts, and behaviors (Mischel & Shoda, 1995). Traits can exist not only at the
individual level but at the group/unit level as well, as illustrated by the Big Five factor model, which has been argued to be functionally isomorphic at the individual and collective level (Hofmann & Jones, 2005). Although traits can produce stable behavioral
patterns (Hogan, 1991), the mechanism of behaviors encompasses the types of behaviors that make leadership possible. Consequently, one can examine such behaviors independent of whether they are consistent with any specific traits (Bass & Bass, 2008).
Approaches to leadership that emphasize cognition focus on the thoughts and sense-making processes related to leadership.
These theories arose out of the “cognitive revolution” in psychology that emerged in the 1950s, and then imported into the
leadership literature in the early 1970s (Lord & Brown, 2004). Instead of simply focusing on traits embodied within the leader,
or behaviors carried out by the leader, the cognitive approach to leadership emphasizes how scripts and schemas influence the
ways leadership is conducted and ultimately perceived and interpreted. Cognitive scripts and schemas can directly influence
leaders and their decision-making processes, choices, and behaviors. Alternatively, leaders can activate their followers’ scripts
and schemas and therefore influence their thinking and attitudes, while followers’ cognitive elements can shape who is or is
not perceived as a leader.
Finally, affect captures the emotions and moods involved in leadership. Leadership is an emotional process for a variety of reasons.
Leaders’ emotions can influence their decision-making and ability to connect to followers. Leaders can transmit their emotions to followers, thus influencing how leadership develops and is perceived (Erez, Misangyi, Johnson, LePine, & Halverson, 2008). Followers’
own affect can bleed into their perceptions of leadership, and the emotional connection between leaders and followers then shapes
their relationship (Ashkanasy & Tse, 2000).While expressing a certain emotion might be considered a behavior, affect as a mechanism
of leadership captures the emotions and moods as feeling states that subsequently influence a variety of processes including behaviors
and cognitions (Brief & Weiss, 2002; Forgas, 1995; Forgas & George, 2001).
At this time, we would like to add a couple of important caveats. First, although our framework depicted in Fig. 1 appears
to make clear distinctions between the twenty cells, the differences between them are in fact fairly fluid. For example, although
we separate cognition and affect into two different mechanisms, this line must be considered as being blurry at best as has been
suggested in previous literature (e.g., Clore, Schwarz, & Conway, 1994; Zajonc, 1998); for the sake of conceptual clarity, however,
we treat them as distinct. Correspondingly, we categorized many of the leadership theories into more than one cell, emphasizing
the increased complexity of today’s theories, while at the same time attempting to present the most parsimonious classification. If
a theory was not placed into a particular cell, it implies the theory’s emphasis does not lie in the dimensions of the cell.
The original theorists may have acknowledged the existence of different loci or mechanisms of leadership, but did not identify
the theoretical rationale for them. Often, such acknowledgements have been framed as directions for future research. For example, although other leadership reviews and many textbook discussions make clear distinctions between the trait, behavioral and
contingency approaches to leadership, our reading of this literature indicates that early trait theorists actually reflected on the importance of behaviors and situations, and that leadership behaviorists also considered situational contingencies. Nevertheless, such
an incidental discussion of loci and mechanisms, in our opinion, did not warrant the inclusion of the theory in a particular cell.
Second, due to the fluidity of the cells, it was not always entirely unambiguous which loci and mechanisms were integral parts
of specific leadership theories; therefore, our placements of the theories may be in disagreement with others’ categorizations of
the theories. Fig. 1 should therefore not be regarded as a steadfast map, but rather as a framing of the leadership field in the spirit
of generating conversations about loci and mechanisms. We believe that the discussions that may arise from the use of loci and
mechanisms to categorize theories of leadership will help leadership scholars to better organize the myriad of approaches to examining leadership in future research. As in most emerging fields of inquiry, there has been a growing diversification of theories
pertaining to explaining what constitutes leadership over the last 100 years. Now, the time seems right for consolidating and
integrating what we know about leadership as the basis for the next phase in the field’s development.
1.4. The language of leadership: a starting point
Overall, our qualitative review of the leadership literature suggests that leadership theories can be meaningfully categorized
according to the loci and mechanisms of leadership they advocate, and that leadership theory has evolved to more complex conceptualizations of leadership by including additional loci and mechanisms. By way of analogy and for ease of use, the two dimensions can be compared to grammar: the locus of leadership is the subject of the sentence (that which acts) and the mechanism
is the verb (the action). It follows that combining the two will result in a complete sentence such as “The leader behaves” or
1168 M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185
“The follower feels.” To push the analogy further, a substantive analysis of leadership is not possible unless both dimensions are
considered, and we posit that the loci and mechanisms of leadership can be used as fundamental building blocks to understanding
what constitutes leadership. A list of nouns and verbs, however, is of little use unless one combines them into a coherent story.
Hence, while we started with a simple sentence employing only a noun and verb, we tell the story of leadership theory through
a combination of multiple loci and mechanisms that we suggest form a more comprehensive leadership system.
2. Leadership theories: a review and categorization
In the following sections, we review major theories that have significantly contributed to the leadership literature over the
past 100 years with the purpose of identifying trends and gaps in the existing literature. For each theory, we briefly discuss its
major predictions and theoretical rationale and then classify it according to our two-dimensional framework. Fig. 1 illustrates
our classification of the theories into the two-dimensional framework.
2.1. Trait theories
Evolving from the “great man” theories of the 19th century, the trait approach to leadership in the early 20th century argued
that certain personality characteristics distinguish leaders from non-leaders (Bowden, 1926; Galton, 1869; Gibb, 1947; Jenkins,
1947; Kohs & Irle, 1920; Terman, 1904). The initial hope was that the identification of such traits (such as intelligence, sociality,
and dependability) would enhance personnel selection. Reviews of the literature in the late 1940s, however, revealed, “no single
trait or group of characteristics has been isolated which sets off the leader from the members of his group” (Jenkins, 1947: 74–75).
Stogdill (1948: 66) summarized the trait findings in the leadership literature, concluding that “leadership is not a matter of
passive status, or of the mere possession of some combination of traits.”
In their review of the core leadership traits, Kirkpatrick and Locke (1991) initiated what could be labeled the neo-trait approach. Their approach was different from the traditional trait theories, because it recognized that traits by themselves are not
the key to leadership, but are merely a “precondition” (Kirkpatrick & Locke, 1991). Promising advances were made by systematically analyzing leadership traits through the application of the Big Five personality framework (Judge, Bono, Ilies, & Gerhardt,
2002). These advances went beyond the traditional view of traits by emphasizing the need to understand the psychological
mechanisms through which traits translate into leadership effectiveness.
The classical as well as the neo-trait approaches to examining leadership focus primarily on the leader as the locus of leadership. However, even early theorists had already acknowledged the importance of followers and the context. For example, Terman
(1904: 59) recognized the role of followers in his statement that “some credit only the leaders as a factor in race progress, others
only the masses” and Bowden (1926: 158–159) discussed the importance of the context by suggesting that the situation can determine what the best personality is to lead. Nevertheless, the primary goal of these theorists was to emphasize the personality
characteristics of the leader and how those characteristics differ from those of non-leaders. Thus, trait theories focused on the
leader locus and trait mechanism.
Trait theorists’ emphasis on traits has been addressed in recent calls for answering the infamous “why” question: Little is
known yet about why traits should influence leadership effectiveness, and answering this question requires an integration of
traits with other mechanisms of leadership such as behaviors, cognitions, and affect (Judge et al., 2002). Ultimately, however,
traits did not emerge as sufficient predictors of leadership effectiveness, and the field of leadership was left with a deficiency
of predictors. Stogdill’s (1948) now classic review of the leadership literature represented a shift in thinking about leadership,
emphasizing the importance of behaviors and the situational context in which leadership is embedded.
2.2. Behavioral theories
After the suspension of the trait approach, leadership scholars turned to identifying specific behaviors and behavioral dimensions that would distinguish effective leaders from ineffective ones. Early behavioral researchers differentiated between authoritarian, democratic, and laissez-faire leadership behaviors (Lewin, Lippitt, & White, 1939). The behavioral approach was
significantly advanced by two large-scale efforts: the Ohio State University and the University of Michigan studies. In 1945,
the Ohio State leadership studies began the exploration of descriptive dimensions of leadership behavior and developed the Leader Behavior Description Questionnaire (LBDQ). Factor analyses of the LBDQ consistently revealed that leadership behaviors could
be reliably categorized along the dimensions of consideration and initiating structure (Fleishman, 1953; Stogdill & Coons, 1957).
Similarly, the University of Michigan studies revealed a distinction between employee-centered and production-centered leadership behaviors (Kahn & Katz, 1953; Katz & Kahn, 1952; Katz, Maccoby, Gurin, & Floor, 1951; Katz, Maccoby, & Morse, 1950; Likert,
1961; Mann, 1965).
A third, less well-known stream of behavioral research was conducted at Harvard University through laboratory observations.
This research emphasized the dual role of leadership in terms of task and social leaders (Bales, 1954). In summarizing the Harvard
Laboratory Studies on leadership, Bales (1954) referred to the term “co-leadership,” suggesting that it might be beneficial for
groups to allocate the task and relational leadership roles to different individuals, moving the locus of leadership from the individual to the collective.
All three research centers emphasized the distinction between task- and people-oriented leadership behaviors, but they diverged on the relative emphasis of each dimension in predicting leadership effectiveness. Similarly, Blake and Mouton’s (1964)
M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185 1169
managerial grid training continued the focus on this two-dimensional view and categorized leadership behaviors along the dimensions of concern for production and concern for people.
Behavioral approaches to leadership, like the trait theories that preceded them, primarily emphasized the leader as the locus
of leadership. Behaviors are used as an indicator to differentiate effective from ineffective leaders; an approach particularly apparent in Kahn and Katz’s (1953) work on distinguishing between production-oriented and employee-oriented supervisors. It is
important to note, however, that behavioral scholars began to recognize the “situational nature of leadership” (Fleishman,
1953: 6) and the existence of behavior by situation interactions (e.g., Mann, 1965) relatively early into this line of inquiry.
Nevertheless, the primary mechanism of leadership in the behavioral theories was indeed behaviors.
2.3. Contingency theories
In an attempt to more precisely define the effects of leadership and account for more variance in empirical studies of leader
effectiveness, scholars in the 1960s and 1970s turned to contingency theories that took into account situational factors acting
as potential constraints or opportunities for leaders. Fiedler’s contingency theory was among the first to attempt to reconcile previous inconsistent findings regarding leader traits and behaviors (Fiedler, 1964, 1971, 1976). He posited that leadership effectiveness depends on the interaction of leadership style with features of the situation he referred to as “situational favorableness”.
Based on the least preferred co-worker (LPC) score, a person would either be categorized as a task- or relationship-oriented
Situational favorableness was considered high when leader–member relations, task structure and/or position power were
high. Under these circumstances, a leader presumably is supported by the situation because it provides some influence and potential power. Fiedler argued that task-oriented leaders would be more effective in highly favorable or highly unfavorable situations, while relationship-oriented leaders would be more effective in moderately favorable situations. Thus, rather than arguing
that a certain style (e.g., relationship orientation) is better at all times (as in the trait and behavioral theories), Fiedler acknowledged
that the effectiveness of certain styles can depend on the environment in which they are embedded.
Similar to Fiedler’s work, House’s initial statement of path-goal theory focused on situational moderators upon which leaders’
effectiveness is contingent (House, 1971; House & Mitchell, 1974). Path-goal theory is grounded in Vroom’s expectancy theory of
motivation, which at its core suggests that people are more likely to engage in a specific behavior if they perceive a high probability that the behavior will lead to a valued outcome (Vroom, 1964). An effective leader, according to path-goal theory, clarifies
employees’ paths to work goals and the link between work goals and valued personal outcomes, thus making it explicit what employees need to do. Having a path laid out in front of them that leads to valued personal outcomes (e.g., merit pay, promotion)
was posited as a motivator for employees to pursue those outcomes and engage in the desired behaviors.
According to path-goal theory, the extent to which certain leadership styles (e.g., directive, supportive, participative, and
achievement-oriented leadership) aided in clarifying employees’ paths depended on the employees’ personal characteristics
and on contextual demands with which employees may have to cope. For example, followers with an internal locus of control
are described as being more satisfied with a participative leadership style, because they believe in their own ability to make a difference; those with an external locus, however, tend to be more satisfied with a directive leadership style. As an example of an
environmental contingency variable, path-goal theory suggests that the degree of task structure moderates the relationship between directive leader behavior and employee job satisfaction, such that the relationship is positive for low task structure and
negative for high task structure. When tasks are already highly structured, the path to success should already be identified and
any additional directions may be perceived as excessively controlling. Indeed, the high level of structure may even serve as substitute for directive or even participative leadership (Avolio, Kahai, & Dodge, 2000).
Other environmental contingency factors include the formal authority system of the organization and the employees’ primary
work group. Like Fiedler’s contingency theory, then, path-goal theory emphasizes the context as a potential locus of leadership.
In addition, followers are clearly separated from environmental factors and are treated as active components of the leadership
process depending on their locus of control, authoritarianism and abilities. Thus, a leader’s effectiveness depends on follower
attributes, locating (at least partially) the locus of leadership in the followers. Finally, the leader is discussed as a third locus
because leadership effectiveness depends on the combination of leadership style with context and follower attributes. The
main mechanism of leadership within path-goal theory is behaviors; these include clarifying goals and paths, providing specific
guidance (directive), showing concern for followers’ well-being (supportive), soliciting employee suggestions (participative)
and setting challenging goals (achievement-oriented). However, with regards to followers, the theory mainly emphasizes traits
(e.g., locus of control) and therefore also recognizes traits as a mechanism.
Hersey and Blanchard (1969, 1982) proposed a practitioner-oriented situational leadership theory that placed leader effectiveness squarely in the interaction between leader behaviors and the followers’ level of maturity. Specifically, the theory suggests
that leaders should match their behaviors with the followers’ maturity level by moving through the phases of telling, selling, participating, and delegating to correspond to increased follower readiness. Each phase was categorized via the two dimensions of
task and relationship behavior. For example, when followers are not yet willing to assume any responsibility (low maturity), a
leader should “tell” them what to do and display high levels of task behaviors while limiting relationship behaviors. As followers
become more mature, it is important for a leader to provide additional socio-emotional support while finally, in the delegation
phase, task guidance does not need to be provided anymore, because followers are fully aware of their responsibilities and are
willing to assume the challenges. The locus of leadership within this situational theory arises out of the dyad, because leaders
are presumed to be effective only to the extent that they adjust their styles to followers’ attributes. Because Hersey and Blanchard
1170 M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185
drew from the Ohio State studies and Blake and Mouton’s managerial grid to develop their theory, behaviors are the primary
mechanism of leadership within their theory.
2.4. Social exchange perspective
At approximately the same time that contingency theories were becoming prominent in the leadership field, a somewhat
separate line of research emerged that focused on the dyadic relationship between leaders and their followers. This research
questioned the assumption of an average leadership style that may be equally effective with all followers and advanced the
field by shifting attention towards the leader–follower relationship as the primary domain of analysis (Dansereau, 1995; Graen &
Uhl-Bien, 1995; Schriesheim, Castro, & Cogliser, 1999).
Jacobs (1970) was among the first to approach leadership from a social exchange perspective. He criticized the simplicity of
prior leadership approaches and suggested that leadership can only be understood when viewed with followers being proactive
partners in the interaction. Drawing from social exchange theory, Jacobs discussed the transactional nature of groups. Specifically,
any group member who uniquely contributes to his or her group’s goals is reciprocated by receiving higher status and esteem by
fellow group members.
While Jacobs’s approach focused mainly on leader emergence, leader–member exchange (LMX) researchers applied his work
to formal leader–follower relationships. Originating in the vertical dyad linkage model, LMX emphasized how leaders develop
two distinct types of relationships with their followers: one based on “influence without authority” and one based on “influence
with authority”(Dansereau, Graen, & Haga, 1975: 48). The ability of the leader to influence without authority was characterized
by higher levels of mutual support, trust, loyalty, and latitude given to their followers. Conversely, influence with authority
was based primarily on more formal supervisory roles and techniques. The development of such distinct relationships during
the initial role making processes was hypothesized to lead to the formation of in-groups and out-groups (Graen & Cashman,
1975; Graen & Scandura, 1987). Followers in in-groups enjoy higher job satisfaction and organizational commitment, display
higher levels of task and extra-role performance, and are less likely to turn over compared to followers in out-groups (Graen &
Uhl-Bien, 1995). Thus, being a member of the leader’s in-group has been shown to have a positive effect on many follower outcomes.
It is evident from the foregoing discussion that LMX theorists see leadership arising from the unique aspects of the leader–
follower relationship rather than attributes of the leader or the follower alone. Thus, the locus of leadership lies within the dyad.
The primary leadership mechanism within LMX theory is behaviors. Descriptions of the role making process focus on the leader’s
provision of negotiating latitude and members’ reciprocation through the expenditure of greater effort and time and the assumption of greater responsibility. In fact, role making itself is defined as a process by which partners “work through how each will
behave in certain situations” (Graen & Cashman, 1975: 143, emphasis added).
More recent advancements in LMX theory integrate it with social network perspectives, elevating the locus to the collective
level. For example, Sparrowe and Liden (1997, 2005) suggested that LMX may only be fully understood when going beyond
the leader–follower relationship to a consideration of the social structure within which such relationships are embedded. Drawing from Sahlins'(1972) work on the dimensions of reciprocity, the authors identified how a leader’s and a member’s network independently and interactively influence LMX quality through the three phases of initial relationship development, sponsorship
and assimilation. Leadership in this conceptualization is located within the collective, because it is both the leader’s and the members’ positions within their respective networks that, among other factors, determines the quality of the relationship and therefore gives rise to leadership. Like “traditional” LMX theory, we can infer from Sparrowe and Liden’s discussions that behaviors are
the main mechanism of leadership (e.g., sponsorship appears to be a behavior).
2.5. Venturing into new leadership perspectives
Although the contingency theories developed in the 1960s and ’70s contributed extensively to the literature, some researchers
started to question the validity of the elusive and lofty leadership construct, and provided some compelling arguments for looking
at leadership through an entirely different lens. Kerr and Jermier (1978) roiled the field by suggesting that certain variables can
act as substitutes or neutralizers, rendering leadership unnecessary or ineffective. Drawing from House and Mitchell’s path-goal
theory, these authors identified a set of characteristics of the follower, the task, and the organization that, if present, would simply
make leadership behaviors (task- and relationship-oriented) redundant. Kerr and Jermier’s work moved away from focusing on
the leader as the locus of leadership, but emphasized instead the followers and the context. In other words, the extent to which
leadership is effective could be based simply on certain follower and contextual attributes, almost independent of what the leader
actually does. With regards to the mechanisms of leadership, this theory appears to emphasize traits and behaviors, such that
certain follower personality characteristics (e.g., need for independence and ability) act as substitutes and certain contextual
variables (e.g., standardization of task) lead to the emergence of leadership behaviors (e.g., setting goals).
At about the same time that Kerr and Jermier published their work, Lord and colleagues introduced implicit leadership theory
(Lord, 1977; Lord, Binning, Rush, & Thomas, 1978; Lord, Foti, & De Vader, 1984; Lord, Foti, & Phillips, 1982). Drawing from categorization and person perception research, implicit leadership theorists posit that followers have preconceived notions (implicit
theories) about what a prototypical leader looks like, and when placed within an ambiguous situation, seek confirming evidence
of those notions. If cues are found that support the prototype, the person will be perceived as a leader. For example, leader prototypes often include the display of power; if a person does in fact display power, others around them are more likely to perceive
him or her as a leader. Unless followers perceive someone as a leader (because he or she provides prototype-confirming cues),
M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185 1171
leadership is not likely to occur. As such, Lord and associates were among the first to recognize the importance of follower perspectives in explaining leadership (see also Hollander & Julian, 1969; Pfeffer, 1977). Implicit leadership theory, then, locates
the locus of leadership within the followers and the mechanism of leadership within cognition.
Meindl, Ehrlich, and Dukerich (1985) also followed a nontraditional follower-centered approach to examining leadership, suggesting that the concept of leadership is likely romanticized by followers. Drawing from attribution theory, Meindl et al. (1985)
suggested that followers seek to make sense of inherently complex and ambiguous organizational activities and outcomes by
attributing them to leadership, because it is an accessible and comprehensible explanation. This romanticized view of leadership
is particularly strong when organizational performance is extreme—either very good or very poor. Either case is a salient event
that triggers sense-making and attribution processes in followers, which result in them portraying the leader as either the hero
or the scapegoat.
This romance theory of leadership primarily focuses on followers as the locus of leadership, but we acknowledge that the
situation is also considered in determining how followers construe leadership. Instead of focusing on traits and behaviors as
mechanisms, Meindl et al.’s focused on followers’ cognitions, which are biased towards seeing the leader as the primary
cause for extraordinary events.
2.6. Transformational and charismatic leadership theories
Based on Burns (1978) conceptualization of transactional and transformational leadership in the political arena and Weber’s
(1947) discussion of charisma in organizations, theories of transformational and charismatic leadership surfaced in the early
1980s. Similar to path-goal theory, transactional leadership was grounded in Vroom’s (1964) expectancy theory of motivation
as transactional leaders clarify how followers’ needs will be fulfilled in exchange for completing their job requirements. Thus,
transactional leadership is based on economic principles and implies that rational followers are aware of their needs. The transactional leader’s role is primarily seen in recognizing follower needs and monitoring follower role fulfillment (Bass, 1985; Bono &
Judge, 2004; Burns, 1978). Transactional leaders enact the following behaviors: contingent reward, management by exceptionactive, and management-by-exception-passive.
Although transactional leadership provides an important element of effective leadership (after all, followers need to know
how to get valued rewards), it has been overshadowed by the simultaneous emergence of the more revolutionary theories of
transformational and charismatic leadership. A thorough discussion of the potential similarities and differences of the two theories is beyond the scope of this article. For the purpose of this discussion, we will treat both theories as comparable (e.g., House &
Bass (1985) based his theory of transformational leadership on Burns’ earlier work and defined it primarily with regards to the
effects individual transformational leaders have on their followers, although both authors in later writings escalated the locus
of these theories to the collective level. According to transformational leadership theory, the followers have trust and respect
for the leader, and are motivated to go above and beyond what is normally expected of them. Transformational leaders enact
their leadership through four distinct types of behaviors: individualized consideration, intellectual stimulation, idealized influence, and inspirational motivation.
The concept of charismatic leadership originated from Weber’s (1947) early work on the potential implications of charismatic
leadership for organizations. Like transformational leadership theory, charismatic leadership theories emphasized the behaviors
that allow such leaders to have exceptional influence over their followers (Yukl, 1999). Based on characteristics of the situation,
leader, and follower, followers attribute extraordinary qualities and charisma to the leader. Although both theories emphasize certain behaviors as mechanisms of leadership, the theories also suggest that cognitions and affect help explain the leader’s influence.
The existence of multiple pathways to leadership becomes particularly apparent in House’s (1977: 191) theory of charismatic
leadership, in that he stated that followers of charismatic leaders “model their behavior, feelings, and cognitions after the leader.”
Conger and Kanungo (1987) suggested that perceived charisma was based on an attribution process, such that followers’ perceptions of charisma based on observations of the leader’s behavior determined whether the person was actually considered a charismatic leader. Weber (1947) similarly emphasized the importance ascribed to the beliefs that followers have in the charismatic
leader in determining attributions of charismatic leadership.
Whereas the more traditional leadership theories emphasized rational decision-making, transformational and charismatic
leadership theories both emphasized the importance of affect and values (Yukl, 1999). For example, in his seminal work on charisma, Weber (1947: 360) emphasized the “emotional form of communal relationship” that followers of a charismatic leader
form with each other. Later, Bass (1985) designated an entire part of his book to the “emotional component” of transformational
leadership proposing that inspirational motivation leads to follower emotional arousal, and that leader behaviors such as
instilling enthusiasm through pep talks and making employees feel proud of their accomplishments are part of the emotional
Kark and Van Dijk (2007), among others, promoted the idea that followers are likely to adopt their transformational/charismatic
leader’s positive affect through emotional contagion effects (Hatfield, Cacioppo, & Rapson, 1994). Bono and Ilies (2006) also hypothesized (and found) that charismatic leaders are more likely to express positive emotions than non-charismatic leaders, which are then
“caught” by their employees.
Finally, traits also appear to play a role in the transformational and charismatic theories. Bass (1985) discussed the importance
of personality characteristics such as dominance, self-confidence and need for influence in predicting transformational leadership.
House (1977) hypothesized that extremely high levels of self-confidence, dominance, and conviction in one’s own beliefs are
1172 M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185
necessary for a person to be perceived as charismatic by others. Moreover, Kark and Van Dijk (2007) discussed the importance of
one’s self-regulatory focus to predicting whether one becomes a transformational or a transactional leader.
Unlike prior theories of leadership, we see that traits, behaviors, cognition, and affect are all dominant leadership mechanisms
within transformational and charismatic leadership theory. This might suggest that as the constructs of leadership have become
more complex, so too have the mechanisms needed to explain such leadership. With regards to the locus of leadership, these theories appear to suggest that leadership arises within the leader–follower relationship. Virtually all of the published papers
on transformational and charismatic leadership emphasize the importance of the leader–follower relationship (as opposed to
discussing leaders and followers as distinct partners). For example, followers observe their leader’s behaviors from which
they may or may not develop perceptions of charisma (Conger & Kanungo, 1987, 1998). Similarly, Shamir, House, and Arthur’s
(1993) discussion of the followers’ self-concept emphasized the importance of followers’ reactions to leader behaviors that determines leader influence. This is not to suggest that transformational or charismatic leaders cannot influence groups overall,
but rather, that the locus resides in the leader and follower relationship.
2.7. Contemporary leadership theories
Over the past two decades, more nuanced, complex, and distinct approaches to leadership have emerged in the literature.
Based on Quinn’s (1984, 1988) competing values framework, the theory of leadership complexity (Denison, Hooijberg, &
Quinn, 1995; Hooijberg, Hunt, & Dodge, 1997) suggested that behavioral complexity is an important concept to study within
the realm of managerial leadership. Specifically, they argued that more effective leaders display a variety of different and sometimes contradictory behaviors to react to the wide range of situations they are typically exposed to and required to address. They
proposed that although cognitive complexity is important to leader effectiveness, behavioral complexity warrants more consideration, because ultimately leadership is performed through actions within events. Hooijberg et al. (1997) took this idea a step
further and developed what they labeled a “leaderplex” model that incorporates not only cognitive and behavioral complexity,
but also social complexity. Together, cognitive and social integration and differentiation – and one’s behavioral repertoire – are
used as predictors of leadership effectiveness.
The locus of leadership within the “leaderplex” model lies in the leader and the context. The theory focuses on leaders’ ability
to integrate and differentiate socially, cognitively, and behaviorally taking into account the context, while adapting their behaviors accordingly. If leaders do not have a particular behavior in their repertoire that matches the demands of the context, then
leadership will not arise and the enacted behavior will be ineffective. The mechanisms characterizing the leadership complexity
theory are behaviors, cognitions, and affect. The centrality of cognitive and behavioral complexity in this theory emphasizes that
both are primary means through which leadership is enacted. The social complexity component of the model explicitly discusses
the importance of the leader’s ability to recognize and regulate emotions within self and others, and the potential for emotions to
influence the leader’s cognition and behaviors (e.g., affect infusion model, Forgas, 1995).
A second contemporary approach to leadership is the social identity model of leadership effectiveness (SIMOL) (Hogg,
2001; Hogg & Terry, 2000; Van Knippenberg & Hogg, 2003). Based on Hogg and Terry’s (2000) discussion of the influence of social
identity on a variety of different organizational phenomena, Hogg (2001) described how research on social identity and selfcategorization can help us understand who is being perceived as a leader within a group setting via which member is most prototypical of the group. SIMOL theory argues that the most prototypical members are more likely to gain influence as group
members are socially attracted to them and thus tend to comply with their suggestions and ideas. Hogg and colleagues further
posited that group members attribute the exemplary member’s behaviors to leadership ability and personality.
We view the locus of leadership within SIMOL as residing in the collective: who is perceived as a leader depends on the group
prototype, which is based on the unique identity of the group. The primary mechanisms of leadership in SIMOL are cognitions,
behaviors, and affect. The development of prototypes and the attribution processes that follow from them represent the cognitive
aspects that give rise to who is or is not attributed leadership. Behaviors are a key mechanism, because group members evaluate
their colleagues’ prototypicality by observing their behaviors and deciding whether they are prototypical or not. Although not
given strong consideration in its original writings, leader self-categorization has been shown to change “what people think,
feel, and do” as they attempt to conform to the in-group prototype (Hogg & Terry, 2000; p. 187, emphasis added). Thus, in addition to behaviors and cognitions, affect is a significant part of the prototype and should therefore be considered as a mechanism of
A third recent approach to leadership examines the phenomenon through social network theory. Although LMX was integrated
with social network theory (see foregoing discussion of Sparrowe and Liden’s (1997, 2005) work), other leadership scholars have
recently attempted to provide a more generalized social network perspective to leadership. For example, Balkundi and Kilduff
(2005) discussed the idea that a leader’s cognitive representation of the patterns of relationships within various networks (e.g.,
ego, organizational, and inter-organizational) is the starting point for the way that the leader initiates and maintains social ties,
which in turn influences their leadership effectiveness. Although this theory sees the leader’s network acuity as the starting
point, the locus of leadership lies within the network of relationships or the collective. These authors argued that it is only through
a full understanding of the social networks and one’s role within them that leadership can arise.
Contrasting the earliest trait theories, Balkundi and Kilduff (2005) stated that their “network approach locates leadership
not in the attributes of individuals but in the relationships connecting individuals” (p. 942). The main mechanism of leadership
is associated with cognitions since a leader’s cognitive representations of networks determine both the choices leaders make
and the leader’s effectiveness.
M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185 1173
2.8. Emerging leadership theories
2.8.1. Value-driven, uplifting leadership
Following recent corporate scandals and examples of improper behaviors by key leadership figures, increased attention by
the public as well as the leadership research community is placed on the questions of morality, rightness, values, and integrity.
Although ethics and values had been incorporated in past leadership writings, most of this work has been predominantly normative as opposed to descriptive or predictive (Brown & Treviño, 2006; Ciulla, 2004). However, there are a series of emerging leadership theories that attempt to explicitly incorporate ideas of principles, honesty, and spirituality, including ethical, spiritual, and
Ethical leadership is defined as “the demonstration of normatively appropriate conduct through personal actions and interpersonal relationships, and the promotion of such conduct to followers through two-way communication, reinforcement, and
decision-making” (Brown, Treviño, & Harrison, 2005: 120). To understand the effects of ethical leadership, Brown et al. used social learning theory (Bandura, 1977, 1986), positing that followers are more likely to behave in an ethical manner when their
leader demonstrates ethical behaviors, attitudes and values. In this case, followers are more likely to emulate their leaders’ ethical
decision-making, pro-social behaviors and general motivation and attitudes, which can decrease their counterproductive work
behaviors (Brown & Treviño, 2006; Brown et al., 2005). Although ethical leadership theory acknowledges the importance
of followers as emulators, the emphasis lies on the leader and how he or she provides ethical guidance to followers. The
locus of leadership is therefore solely within the leader.
With regards to the leadership mechanisms, ethical leadership theory mainly focuses on traits, behaviors, and cognition.
As the definition illustrates, behaving in an ethical manner is considered a key component of becoming an ethical leader.
Only through such behaviors and through the communication of high ethical standards will followers perceive a leader as ethical.
Although not immediately apparent in the definition, ethical leadership theory argues that certain personality traits (such as
trustworthiness and empathy) make it more likely that a leader is seen as a “moral person” (Brown & Treviño, 2006). Additionally, ethics necessarily involves the cognitive evaluation of values (e.g., see Ciulla, 2004), and ethical leadership emphasizes that
leaders act as role models for what is perceived as right or wrong. Thus, ethical leaders do not only influence followers’ behaviors
through role modeling, but also through their beliefs about morality (Brown & Treviño, 2006; Brown et al., 2005).
A rather drastic shift in thinking about leadership occurred with the introduction of the theory of spiritual leadership (Fry,
2003; Reave, 2005). Spirituality has long been considered outside the domain of scientific inquiry. However, accumulating research has shown that spiritual values such as integrity, honesty, and humility are positively related to leadership effectiveness,
and may therefore warrant additional consideration. Fry (2003) drew heavily from motivation theories to develop his causal
model that links a leader’s values, attitudes, and behaviors to organizational outcomes through the fulfillment of followers’
needs for spiritual survival. A key tenet of this theory is the idea that leaders and followers have a need for spiritual survival,
which captures the needs for a calling (making a difference) and membership (being understood and appreciated within a
group). Spiritual leaders create a vision to ensure that followers perceive a meaningful contribution of their work (and thus fulfill
the need for calling) and develop an organizational culture based on altruistic love where everyone respects and cares for each
other (and thus fulfill the need for membership). Providing the paths for spiritual survival in turn is predicted to increase followers’ organizational commitment and productivity.
Spiritual leadership theory focuses on the leader’s spirituality and his or her ability to fulfill the followers’ spiritual needs.
The two core mechanisms put forth in this theory are traits and behaviors. Correspondingly, Reave (2005: 663) suggested this
categorization when stating that the “theory incorporates both the trait approach and the behavioral approach” of leadership.
To explore spiritual leadership one needs to start “with the leader’s own ethics and integrity” (Reave, 2005: 663) and then identify whether those values are transformed into actions (Fry, 2003; Reave, 2005). Integrity, honesty, and humility are considered
essential personal qualities of spiritual leaders that allow them to build trust and credibility and to achieve a consistency between
who one is and what one does. These qualities are then translated into behaviors, such as showing respect to others, treating
others fairly, expressing concern and altruistic love, and appreciating others’ contributions. Nonetheless, beyond traits and behaviors, spiritual leadership also captures an affective mechanism of leadership through its emphases on the importance of selfmanagement and reflective practices. For instance, by engaging in reflective practices, the leader is able to reduce stress and anxiety to help individuals cope with negative emotions and gain control over their own emotions (Neal, 2000; Reave, 2005). Affect
as a mechanism of spiritual leadership therefore complements the trait and behavior mechanisms.
Finally, arguably the most developed theory within this category of “value-driven and uplifting” leadership is authentic
leadership theory, which partially arose from the positive psychology and positive organizational scholarship literatures
(Avolio, Gardner, Walumbwa, Luthans, & May, 2004; Ilies, Morgeson, & Nahrgang, 2005). Authentic leadership theory is
based on the premise that knowing who one is, what one believes and values, and then acting upon those beliefs and
values (being authentic) positively influences follower work attitudes and behaviors. Authentic leaders not only influence
intra-organizational outcomes, but also extend their influence to broader societal problems. This influence is suggested as
being possible because authentic leaders elicit followers’ personal identification with the leader and social identification
with the group, which in turn increases followers’ levels of hope, trust, positive emotions, and optimism (Avolio et al.,
2004). Being a positive role model is also one of the key means through which authentic leaders develop authentic followers (Gardner, Avolio, Luthans, May, & Walumbwa, 2005). Along these lines, Ilies et al. (2005) discussed the importance
of positive emotional contagion, positive social exchanges, and the support of follower self-determination as potential processes through which authentic leaders influence follower well-being.
1174 M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185
Although authentic leadership theory often discusses authenticity from a leader’s perspective only, at its core it emphasizes a dyadic approach to leadership. For example, Avolio and Gardner (2005) discussed the importance of an “authentic
relationship” between leaders and followers that supports the achievement of common goals and mutual development.
Avolio et al. (2004) posited that authentic leaders induce optimism in followers by first identifying with them, again suggesting that the interplay between leader and followers gives rise to leadership. Finally, authentic leadership theorists
argue that it is followers that “authenticate the leader” (Gardner et al., 2005: 348). Classifying this theory into the dyad
category therefore appears most appropriate.
The main mechanisms of leadership discussed in this theory are behaviors, cognitions and affect. Role modeling is one of
the key behaviors through which authentic leaders and their followers enact leadership. Eliciting trust, hope, optimism and
identification in followers all have a cognitive component to it in terms of being another mechanism of leadership. In addition, the idea that authentic leadership involves the “unbiased processing” of self-relevant information to generate an accurate representation of one’s self speaks to the importance of cognitions as a mechanism for both leaders and followers (e.g.,
Gardner et al., 2005; Ilies et al., 2005). Finally, the affective components of authentic leadership can be seen in the discussions
of emotional intelligence, emotional contagion and affective components of identification with one’s leader or collective (Gardner
et al., 2005; Ilies et al., 2005).
2.8.2. Leadership in a cross-cultural context
A majority of research in the leadership field has been conducted in Western contexts. It was not until recently that scholars
attempted to systematically explore the cross-cultural aspects of leadership across a wide spectrum of cultures and organizations.
One of the most extensive research projects of international nature is the Global Leadership and Organizational Behavior Effectiveness (GLOBE) initiative (House, Javidan, Hanges, & Dorfman, 2002). This multi-phase research project confirmed what most
scholars had intuitively hypothesized: that effective leadership can be culture-bound and needs to take into consideration the
unique cultural context within which it operates. For example, whether a leader is a risk taker can have a very different impact
on followers in countries with high as opposed to low uncertainty avoidance (Javidan, Dorfman, Sully de Luque, & House,
2006). Similarly, a leader with an individualist value orientation is perceived differently in collectivistic cultures than in more
Western-oriented individualistic cultures (House, Hanges, Javidan, Dorfman, & Gupta, 2004).
Although differences exist across cultures, the GLOBE project also identified many leader attributes that are universally endorsed or rejected. For example, leaders with a high level of integrity and inspiration are considered effective in most countries
across the world. Likewise, autocratic and non-cooperative leaders are considered largely ineffective worldwide (Den Hartog,
House, Hanges, Ruiz-Quintanilla, & Dorfman, 1999; Javidan et al., 2006).
The GLOBE initiative, however, is mainly empirically driven and has not (yet) developed a unique theory of global leadership.
The main leadership theory that has guided the foundational work in this project is Lord et al.’s implicit leadership theory (see the
discussion above), suggesting that the beliefs of what constitutes good and bad leadership are substantially different across cultures (House et al., 2004). When assessing which attributes and behaviors are universally endorsed and which ones are culturebound, GLOBE research has also drawn from charismatic and transformational leadership (Den Hartog et al., 1999). It is therefore
difficult to pinpoint the unique theoretical contribution of this research. Due to its focus on leader attributes and behaviors and
the culturally endorsed implicit leadership theories (beliefs), however, the main mechanisms of leadership within the GLOBE project can be identified as traits, behaviors, and cognition.
Besides looking at the leader as a unique locus of leadership, this initiative’s main contribution lies in the identification of culture as the context from which implicit leadership theories, and consequently leadership arises. Culture strongly influences the
implicit beliefs that the people (the system) within the culture develop, and it is only when the leaders’ attributes and behaviors
match these culturally endorsed leadership theories, that leaders are accepted (House et al., 2002). Thus, context via its impact on
the cognitive mechanism of leadership can also be considered a locus of leadership in this project.
An emerging research stream that moves away from a traditional Western-view of the world is referred to as paternalistic leadership theory (Chen, Eberly, Chiang, Farh, & Cheng, in press; Pellegrini & Scandura, 2008). Paternalistic leadership is defined as “a style
that combines strong discipline and authority with fatherly benevolence” (Farh & Cheng, 2000: 91), and is grounded in Confucian
values. Paternalistic leadership is the prevailing leadership style in Chinese business contexts and is also prominent in other collectivist and high power distance countries such as Turkey, Mexico, Malaysia, Pakistan and India (Cheng, Chou, Wu, Huang, & Farh, 2004).
Although the idea of paternalism within leadership has certainly been around for quite some time (e.g., Weber, 1947), it has
not been systematically investigated and conceptualized until recently. Farh and Cheng (2000) argued that paternalistic leadership can be described by using three distinct dimensions: authoritarianism, benevolence, and moral leadership. Authoritarianism
captures the leader’s behaviors that assert control and authority and demand absolute obedience from followers. Benevolence refers to leaders’ individualized concern for their followers’ general well-being, which also extends to their family and personal life.
Moral leadership represents leader behaviors that demonstrate “superior personal virtues, self-discipline, and unselfishness”
(Cheng et al., 2004: 91). Paternalistic leadership theory, then, focuses on behaviors as the mechanisms of leadership.
The leader is depicted as the main source of leadership, because it is the leaders’ behaviors that make paternalistic leadership
possible. However, scholars have also explicitly recognized that the acceptance and effectiveness of paternalistic leadership depends on the cultural values (e.g., beliefs, cognition) of the country and/or organization. Within a Western culture, being authoritative may actually impede a leader from being accepted as a leader, thus preventing leadership from arising. We therefore
suggest that paternalistic leadership theory also emphasizes the context in which it is exercised as a locus of leadership and
cognition as a mechanism.
M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185 1175
3. The need to include multiple loci and mechanisms of leadership
Our qualitative review of the leadership literature details how the prominent leadership theories over the past century have
grown to now encompass a broader range of loci and mechanisms. The field of leadership has generally progressed from a sole
focus on the individual leader as the locus of leadership to an increased emphasis on the followers, the leader–follower relationship, the context, and the collective. Similarly, the field has moved from traits to behaviors and then to cognitions and affect, to
include all of these mechanisms of leadership. Having continuously learned from past results and building on prior theory, leadership scholars have significantly advanced our knowledge of leadership by considering these new categories. Nevertheless,
prominent gaps in our understanding of what constitutes leadership remain from both the locus and mechanism perspectives;
in particular, further development of the follower, collective and context loci, and the affective mechanism is warranted.
As depicted by the gray areas in Fig. 1, our review revealed that the field of leadership research has largely focused on examining what constitutes leadership by examining the individuals who serve in formal leadership roles and these individuals’ dyadic
relationships with their followers. Accordingly, loci other than the leader and the leader–follower dyad have remained relatively
understudied, which is problematic considering that “leaders cannot be thought of apart from the historic context in which they
arise, the setting in which they function (e.g., elective political office), and the system over which they preside (e.g., a particular
city or state). They are an integral part of the system, subject to the forces that affect the system” (Gardner, 1990: 1).
First, although leadership has been conceptualized in many different ways, the myriad of factors explored by researchers has
focused heavily on how leaders exert influence on followers (Bass & Bass, 2008). Most theories do not explicitly examine follower
behavior, which suggests an opportunity for stronger emphasis on a follower-centered approach to leadership. After a review
of follower-centered approaches to leadership, Avolio (2007: 26) suggested that “most leadership research has considered the
follower a passive or nonexistent element when examining what constitutes leadership,” before recommending that followers
be viewed as active participants in the leadership process. We concur and posit that a focused, systematic investigation of
follower-centered perspectives to leadership will clarify the role of followers in the leadership process.
Second, the most commonly researched model of leadership over the last two decades has been transformational and charismatic leadership, with the locus of leadership being the leader–follower dyad. Yet, the originating author for this work suggested
that research should extend his work on transformational leadership to include a focus on collective leadership (Burns, 1996). He
suggested that “the initiator (referring to leader) may continue as a single dominating ‘leader’ a la Castro, but more typically he or
she will merge with others in a series of participant interactions that will constitute collective leadership…I see crucial leadership
acts in the collective process” (Burns, 1996: 2–3).
Similar to Burns’ extensions of transformational leadership, the main proponent of Burns’ work on transformational leadership
noted that “transformational leadership could be shared among the team members…Instead of motivation being supplied by
identification of members with an idealized, charismatic leader, similar motivation would be supplied by identification with
the team…Inspiration would come from a sharing of mutually articulated goals” (Bass, 1998: 157). As these authors have suggested, there is a need to include the collective locus of leadership in our theorizing.
Third, although context is pervasive in organizational phenomena, it has largely been neglected in leadership research. Indeed,
only 16% of the scholarly articles on leadership in a recent review emphasized context (Porter & McLaughlin, 2006). This is unfortunate in that the nature of leadership likely changes in different contexts (Osborn, Hunt, & Jauch, 2002). Context is not a static
entity in organizations; it constantly shifts and evolves due to factors both external and internal to the organization. Thus, the nature of leadership within a particular organization can change over time as the context changes. For example, leadership can take
a different form during an organizational crisis, such as that brought on by the 2008 global recession. Conceptualizing the locus of
context is therefore an important factor in fully capturing the dynamic nature of leadership.
Finally, the affective mechanism of leadership remains relatively underexplored. Although research on the relationship
between leadership and emotions has risen significantly in the past decade (Barsade, Brief, & Spataro, 2003; Gooty, Connelly,
Griffith, & Gupta, 2010), the affect mechanism of leadership is not widely included across multiple loci of leadership. Given
followers can feel certain emotions towards their leader which influence their attitudes and behaviors, and that such emotions can also transfer to other team members, ultimately shaping the group climate (e.g., Ashkanasy, Dasborough, Ashkanasy,
Tee, & Tse, 2009), it may be especially important to further our understanding of the affect mechanism across follower, collective and
context loci of leadership.
More broadly, we propose that by simultaneously considering all five loci and four mechanisms, leadership researchers can
advance a more comprehensive view of the leadership phenomenon. In the following sections, we demonstrate how the application of our locus and mechanism framework can aid in expanding the theories of shared leadership and strategic leadership.
We briefly introduce each theory and identify the loci and mechanisms on which each is currently focused. We then describe
how the remaining loci and mechanisms can help explain the emergence or effectiveness of shared and strategic leadership
above and beyond current conceptualizations. Our discussion is intended to be illustrative of what other areas of leadership theory may focus on in the future and is not all-inclusive or comprehensive (i.e., we will not describe every single cell in Fig. 1). Our
purpose is rather to inspire the use of our framework in similar ways to expand current and future leadership theories.
4. Expanding shared leadership theory
Shared leadership is defined as “a dynamic, interactive influence process among individuals in groups for which the objective
is to lead one another to the achievement of group or organizational goals or both” (Pearce & Conger, 2003: 1). Although Gibb
1176 M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185
(1947) suggested decades ago that leadership can be seen as either distributed or focused, the concept of shared leadership had
not been systematically integrated into leadership research until recently. Shared leadership is described as a group-level phenomenon that moves away from the traditional notion of top-down influence and argues that the leadership role can be shared by team
members either simultaneously or in a rotating fashion resulting in lateral and upward influence (Carson, Tesluk, & Marrone, 2007;
Pearce & Conger, 2003). Leadership can be assumed by any team member who chooses to step into the leadership role and feels qualified to do so, based on some level of competency or expertise (Bales, 1954; Bass & Bass, 2008; Burns, 1996); thus, it “originates with
individual members of a team” and results in a “leadership network” (Carson et al., 2007). From this perspective, leadership is seen as
something that occurs within networks and travels through the various relationships within the network (Fletcher & Kaeufer, 2003).
As such, shared leadership theory originates from both the followers and collective loci of leadership.
In the current conceptualization of shared leadership, it has been suggested that leadership is enacted through practices (behaviors) and is not limited to “a set of personal characteristics and attributes located in people at the top” (Fletcher & Kaeufer, 2003: 22).
Carson et al. (2007: 1219) specifically defined shared leadership as originating from “activities” that individual team members engage
in to influence the rest of the team. Similarly, Cox, Pearce, and Perry (2003: 53) posited that shared leadership might be illustrated “as
a sort of behavioral mechanism” through a series of lateral exchanges between team members. It appears, therefore, that behaviors
are the primary mechanism of shared leadership theory.
According to our two-dimensional framework, shared leadership currently covers only part of the entire leadership system
and we posit that incorporating the other loci and mechanisms can aid in elevating shared leadership theory to an even more
comprehensive theoretical treatment of the leadership phenomenon. A full account of each locus and mechanism provides a
richer explanation of what constitutes shared leadership. Thus, in order to push the field forward, we theorize how shared leadership can partially originate from leaders, leader–follower dyads and the context and be transmitted through trait, cognition and
affect mechanisms. These discussions are not meant to be all encompassing, but our hope is that these preliminary ideas will
spark the beginning of a coherent, interesting research agenda to investigate shared leadership processes in teams or even
broader collective levels.
4.1. Loci of shared leadership beyond followers and collectives
We indicated earlier that the leader as the locus of leadership has been at the forefront of research over the past decades.
While we believe, as others have stated, that as a field we need to increasingly focus on the other loci of leadership, we cannot
lose sight of what we know. By exploring the leader locus extensively, the leadership field has developed a broad and deep
base of knowledge that encompasses how various leaders behave, think and feel in organizations, which can be applied to understanding multiple leaders in a shared leadership framework. Consequently, we suggest continuing to draw from and develop this
knowledge—the leader remains a critical component in the leadership process.
With respect to shared leadership, the question arises how leaders can encourage a culture where shared leadership can develop and succeed and what types of leaders and followers are better equipped to do so. Thus, the leader locus can help explain
when and how shared leadership can emerge successfully. A leader external to the collective can, for example, engage in sense
giving to influence team members’ sense making processes in a desired direction (Gioia & Chittipeddi, 1991). Sense giving allows
leaders to interpret environmental cues and place them into a framework that team members then can use to engage in their own
meaning interpretation. Having leaders who impose a common frame of reference on the team can aid in influencing the accuracy
and similarity of team member’s interpretative processes (Randall, Resick, & DeChurch, 2011) and change the team’s collective
capacity for sense making (Zohar & Luria, 2004), which in turn can facilitate the shared leadership process.
Similarly, leaders who enact participative leadership behaviors and provide equal emphasis to all team members can provide
the platform from which shared leadership can flourish (Dionne, Sayama, Hao, & Bush, 2010). Engaging all team members similarly and recognizing everyone’s contributions provides a level of integration and communication among the team that allows
it to begin to engage in a reciprocal influence process where no single team member is presumed to be more powerful, knowledgeable or influential than the others. A leader may then continuously deemphasize his or her positional power and begin to
share leadership across multiple persons and roles such that ultimately, the collective becomes the locus of leadership.
An important contributor to shared leadership emergence may be the development of leader–follower relationships where
followers take charge, challenge their leaders’ viewpoints and engage in leadership behaviors themselves. The transformational
leadership literature can likely provide some answers to the question of how this can be accomplished, because it emphasizes
the reciprocal dyadic process of turning followers into leaders. Transformational leaders appeal to followers’ higher-order values
and activate their followers’ collective identity (Howell & Shamir, 2005). They positively influence followers’ social identification
with the work group (Ashforth & Mael, 1989; Kark, Shamir, & Chen, 2003) and therefore switch followers’ focus from themselves
to the collective. Thus, a critical antecedent for shared leadership emergence seems to be followers’ ability to focus on the collective as an agent of change.
When followers define themselves via the collective they are a part of, they often also feel more empowered. Kark et al. (2003)
have shown that social identification is positively related to self-efficacy, organizational-based self-esteem, and collective efficacy,
which they used as proxies for empowerment. Transformational leaders empower their followers by delegating responsibilities
and encouraging them to think independently and challenge the leaders’ ideas (Kark et al., 2003). Followers enjoy more
M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185 1177
autonomy and independence in determining how to do their job and are confident that they have the skills and capabilities necessary to perform well (Spreitzer, 1995). From this dyadic influence process between transformational leaders and followers,
shared leadership emerges through reciprocation of leadership between fellow team members.
Yukl (2006) acknowledged the importance of the context with regards to shared leadership by defining it as a reciprocal, recursive influence process that takes place among multiple parties and is embedded in a particular social context. This suggests
that there may be contexts in which shared leadership is more or less likely to emerge and succeed. For example, in cultures
with high power distance orientations or individualist viewpoints, shared leadership may be less likely to emerge and be successful, because individuals’ socialization and cultural norms do not align with the basic assumptions underlying shared leadership
(i.e., power sharing, emphasizing group goals over individual goals).
Furthermore, shared leadership itself may become the context through which leadership acts are interpreted. Avolio and Bass
(1998) discussed how the construct of individualized consideration could be escalated to the group or even organizational level,
suggesting that shared leadership can become a norm and ultimately the context for new employees and top managers. As new
teams are formed or new team members come on board, they will be socialized according to the shared leadership norms and will
likely only succeed if they fit in by cultivating the same values.
4.2. Mechanisms of shared leadership beyond behaviors
As noted above, the earliest leadership theorists focused on traits as the mechanism of leadership, and although this emphasis
eventually gave way to more nuanced behavioral and contingent theories of leadership, the study of leadership traits has enjoyed
something of a renaissance of late. Meta-analytic evidence suggests that of the Big 5 personality factors (McCrae & Costa, 1987),
four of them—conscientiousness, extraversion, openness, and emotional stability—are significantly associated with leadership
(Judge et al., 2002). Extraversion in particular was a fairly strong predictor of leadership emergence; in the absence of formal
leadership, people who are highly extraverted tend to take charge. This trait is likely to have a similar effect on the emergence
of shared leadership, where teams composed of extraverts would offer more potential leaders willing to take charge when
Nevertheless, in order for shared leadership to emerge, extraversion alone may not be sufficient. One of the sub-dimensions of
the trait of extraversion is dominance, the extent to which people wish to wield power over others (McCrae & Costa, 1987). Highly extraverted people may be unwilling to cede leadership to others when appropriate, causing power struggles. Thus, extraversion may need to be tempered with the other trait on the “interpersonal plane of personality” (Costa & McCrae, 1992):
agreeableness. Agreeable people are helpful, considerate, and supportive of others; of the Big 5, this trait was found to have
the strongest relationship with team performance (Bell, 2007). Accordingly, although agreeableness was not meta-analytically associated with leadership emergence, because it may be associated with allowing others to take the lead, a configuration of high
extraversion and high agreeableness within a team may more likely result in shared leadership.
Taking this approach of looking at leadership profiles or composites, some authors have suggested that leadership represents
complex patterns of behavior, and therefore these patterns would be better explained by the leader profile versus single attribute
comparisons (Yukl, 2006). Yet, rarely have studies considered how the joint combinations of particular leader traits influence
leadership behavior (Zaccaro, Kemp, & Bader, 2004). We believe the focus on leader profiles is important for examining how
shared leadership emerges in that certain constellations of profiles representing the individuals in a team and the team itself
can affect how shared leadership is enacted.
At its most elemental level, shared leadership can simply involve role rotation or transference of roles among members in a
team, or sharing some roles while others are retained by specific individuals—suggesting that some peers could exhibit mutual
leadership (Benne & Sheats, 1948; Burns, 1996). This elemental view of leadership dates back to Mary Parker Follett’s (1924)
work on the ‘law of the situation,’ which put forth that individuals should not necessarily follow those with formal authority,
but rather those who have the most knowledge about the situation. This early view recognized that one must not only examine
the mechanism of behaviors but also the cognitive mechanisms, which explain how leadership is transferred among team
Shared leadership requires a shared understanding of the knowledge, skills, abilities and behaviors needed to complete the
tasks at hand and a shared judgment of which team members possess whatever is most conducive to taking over the leadership
role. Ideally, these judgments happen quite quickly and automatically so that the transference of the leadership role is fluid, dynamic, flexible and adaptable to new incoming information and contextual influences. This suggests that the successful enactment of shared leadership may be a function of the extent to which the collective has developed shared mental models. Team
mental models “capture the shared, organized understanding and mental representation of knowledge or beliefs relevant to
key elements of the team’s task environment” (Kozlowski & Ilgen, 2006: 83). They involve, for example, the shared organized understanding and mental representation of performance requirements, team member composition and resources, equipment
availability and effectiveness of team processes.
1178 M. Hernandez et al. / The Leadership Quarterly 22 (2011) 1165–1185
According to Mathieu, Heffner, Goodwin, Salas, and Cannon-Bowers (2000: 275), “highly similar mental models would suggest that teammates work toward common objectives and have a shared vision of how their team will function.” This occurs because the formation of shared mental models helps teams to similarly describe, explain, predict, and learn from events in their
environment; as such, they facilitate the transference of leadership roles or practices (Burke, Fiore, & Salas, 2003). In terms of
our model, this process involves examining not only individual cognition but also the collective cognition of the team. By creating
a basis of common beliefs and values among team members, shared leadership may not just shape how the team believes
it should behave, but how it actually behaves in addressing its tasks and challenges. Thus, examining the cognitive mechanisms
involved in shared leadership can raise our understanding of the exact processes through which team members engage with each
other in a reciprocal influence process.
The affective traits and states of team members can influence the emergence and successful enactment of shared leadership.
Pescosolido (2002) found that the leader’s emergence within a group could be determined by whoever is able to manage the
group’s emotions by, for example, modeling the appropriate emotional response. In highly emotional times, the leadership role
may therefore transfer to that team member who possesses the right skills and tools to manage their own and others’ emotions.
Affect researchers have also identified that groups whose members have similar affective reactions, and therefore are more
likely to develop a group affective tone, can function better than those groups where affective diversity is high (e.g., Barsade,
2002; Bartel & Saavedra, 2000). Tiedens, Sutton, and Fong (2004), for example, argued that emotional conformity increases
group members’ sense of interdependence, which subsequently facilitates task execution. Barsade (2002) demonstrated that
groups that converged on a positive emotional state experienced more cooperativeness, less conflict, and had higher perceptions
of task performance. Similarly, Fredrickson and Losada (2005) found that top management teams who displayed higher levels of
positivity when interacting with each other were more likely to promote deeper inquiry as opposed to simply advocating one position against another. The affective experience of the entire group can thus contribute to the shared leadership process.
Overall, our discussion shows that we can expand our understanding of shared leadership by considering the full spectrum of
loci and mechanisms put forth in this article. Viewing shared leadership through these different perspectives helps us explain
when shared leadership is more likely to emerge, how it can be sustained, who will assume the leadership role at any given
time, and myriad of other relevant factors. In addition, our conceptualizations illustrate how the locus and mechanism framework
can integrate existing leadership theories; for example, we suggest that transformational leadership may provide the foundation
at the leader–follower dyadic level from which shared leadership at the collective level may flourish.
5. Expanding strategic leadership theory
Strategic leadership has been an integral part of the strategic management literature since the early 1980s (Hambrick, 2007).
Strategic leadership theorists have proposed that organizational outcomes such as strategic choices and performance may be partially predicted by examining the background characteristics (e.g., values, personality, cognitions) and behaviors of those at the
top of an organization (Carpenter, Geletkanycz, & Sanders, 2004; Chatterjee & Hambrick, 2007; Finkelstein, Hambrick, & Cannella,
2008). The origins of strategic leadership theory can be traced to Hambrick and Mason’s (1984) upper echelons theory, in which
they posited that managers’ cognitions and values determine their cognitive information processing (selective perception, interpretation, etc.) which in turn drives their strategic decisions (Hambrick, 2007).
Boal and Hooijberg (2001) emphasized the change-oriented nature of strategic leadership and conceptualized it as firm’s managerial wisdom and capacity to learn and change. Hence, a key purpose for examining individuals who are at the top of an organization and have substantive responsibility for making strategic decisions is to investigate the creation of an overall purpose and
direction for the organization, which ultimately guide strategy formulation and implementation (Boal & Schultz, 2007; Makri &
Scandura, 2010). Hambrick and Mason (1984) recognized that individual executives do not work in isolation and that a consideration of the “dominant coalition” (e.g., top management team) can provide stronger explanations for organizational outcomes
than a focus on one or a few individuals only. Building on this past work, we propose that strategic leadership scholars have primarily examined the leader and a collective of leaders as the loci of strategic leadership. We can, however, increase the theory’s
explanatory power by also considering the leader–follower relationship, followers and the context as loci.
With regards to the mechanisms of leadership, strategic leadership theorists originally emphasized traits and cognitions as the
primary means by which leadership is enacted. The original premise of strategic leadership was that top managers’ background
characteristics including values and personalities determine what they pay attention to, how they interpret incoming information