The Leadership Quarterly 23 (2012) 1043–1062

Contents lists available at SciVerse ScienceDirect

The Leadership Quarterly

journal homepage: www.elsevier.com/locate/leaqua

Organizational discourse analysis (ODA): Examining leadership as a

relational process
Gail T. Fairhurst a,⁎, Mary Uhl-Bien b, 1
a
Department of Communication, University of Cincinnati, Cincinnati, OH 45221-0184, USA
b
Howard Hawks Chair in Business Ethics and Leadership, College of Business, University of Nebraska, Lincoln, NE 68588-0491, USA

a r t i c l e i n f o a b s t r a c t

Article history: There was a time when survey research was our only viable means of studying leadership
Received 17 May 2012 processes. That is no longer the case. In its many forms, ODA offers a fly-on-the-wall
Received in revised form 24 August 2012 methodology for researchers to see more of how leadership unfolds in a co-created process
Accepted 3 October 2012
among relational beings. We showcase a discursive perspective and ODA as a broad set of
Available online 8 November 2012
methods for adding relational views to leadership research—and moving beyond leadership as
a quality or characteristic of individuals. We begin with an overview of leadership as a
Keywords: relational process. From there we provide a primer on a discursive perspective and ODA,
Leadership
followed by a review of several types of ODA with examples. We conclude with a set of key
Organizational discourse
questions leadership scholars might pose with respect to this broad class of approaches.
Co-creation
Communication © 2012 Elsevier Inc. All rights reserved.
Relational leadership

1. Introduction

Leaders and followers live in a relational world—a world in which leadership is co-created in systems of interconnected
relationships and richly interactive contexts (IBM Global CEO Study, 2010). Despite this, our theories of leadership and
approaches to leadership study are heavily vested in assumptions of individuality, in which leadership is a top-down influence of
leaders while followers, process, and context appear secondary (Fairhurst, 2009; Marion & Uhl-Bien, 2001; Osborn, Hunt, & Jauch,
2002). If we are to advance leadership theories that have more relevance to the world of practice, we must more seriously
consider issues of relationality in leadership studies (Bradbury & Lichtenstein, 2000). One way to do this is to acknowledge in our
research that leaders and followers are “relational beings” who constitute each other as such in an unfolding, dynamic relational
context (Ospina & Uhl-Bien, 2012).
A relational view recognizes leadership not as a trait or behavior of an individual leader, but as a phenomenon generated in
the interactions among people acting in context (Fairhurst, 2007a; Hosking & Morley, 1988). At the core of this view is the
assumption that leadership is co-constructed in social interaction processes that Day (2000:582) suggests “generally enable
groups of people to work together in meaningful ways” to produce leadership outcomes (Day, 2000; Drath, 2001; Hosking &
Morley, 1988; Ospina & Foldy, 2010). Because of this, communication is a key element of relationally-oriented leadership
(Shamir, 2007).

⁎ Corresponding author. Tel.: +1 513 556 4460.

E-mail addresses: [email protected] (G.T. Fairhurst), [email protected] (M. Uhl-Bien).
1
Tel.: +1 402 472 2314.

1048-9843/$ – see front matter © 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.leaqua.2012.10.005
1044 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062

Studying relational processes in leadership, therefore, requires theory and methods that go beyond individualistic
“variable”-based theorizing and survey approaches to the interactional processes at the heart of leadership. A discourse
perspective and associated forms of organizational discourse analysis (ODA) (Grant, Hardy, Oswick, & Putnam, 2004; Grant,
Putnam, & Hardy, 2011; Phillips & Oswick, 2012; Putnam & Fairhurst, 2001) offer such a means through a focus on the formative
power of language and communication. With a well-established set of approaches, ODA provides a viable alternative to survey
research when the need is to address interactional, relational, and contextual issues. ODA is interactional because it can study
leadership-as-it-happens, a relationship made possible only through the sequential flow of social interaction. It is relational in that
it sees leadership not as a solitary activity, but as people co-creating a relationship as they interact. Finally, ODA is contextual in
that it has the capacity to incorporate social context into leadership research in various ways. For example, ODA methods:
(a) focus on the leadership relationship as deeply rooted in power and culture (du Gay, Salaman, & Rees, 1996; Foucault, 1995);
(b) demonstrate how leadership actors (i.e., designated or emergent leaders, followers, or other stakeholders) impact the context
as much as they are impacted by it (Fairhurst, 2007a); (c) accord power to leadership actors to name those aspects of the context
that are most relevant to defining their “situation here and now” (Goffman, 1959); and/or (d) bring the task back into the study of
leadership (Gronn, 2000, 2002; Robinson, 2001).
As a broad class of methods, ODA can map to the specific orientations of a wide range of leadership researchers—from the highly
quantitative to the highly qualitative. Its consideration of social interaction, power, and organizing (Fairhurst, 2007a; Grant, Hardy, et
al., 2011; Grant, Putnam, et al., 2011; Grant et al., 2004) allows us to address process issues often identified as missing in leadership
research (Alvesson & Sveningsson, 2012; Seers & Chopin, 2012; Shamir, 2012). Moreover, ODA is consistent with views that associate
leadership with organizing skills (Hosking & Morley, 1988) and as a special case of relational organizing (Seers & Chopin, 2012).
The purpose of our paper, then, is to introduce a discourse perspective, broadly defined, and to offer ODA as a wide-ranging set
of methods for studying leadership as a relational process. We begin with an overview of this relational process. From there we
provide a primer on organizational discourse and ODA, followed by a review of several types of ODA with examples. We conclude
with a set of key questions leadership scholars might pose with respect to this broad class of approaches.

2. Studying leadership as a relational process

Leadership research has long recognized that leadership involves a relational process between leaders and followers (Graen &
Uhl-Bien, 1995; Hosking, 1988; Katz & Kahn, 1978; Rost, 1991), a process that can occur inside or outside a formal leadership role (Yukl,
2013). Despite this, leadership is most often studied as a characteristic or behavioral style of an individual leader serving in a formal role
(Bass & Avolio, 1990; Fleishman et al., 1991). Even theories that see leadership as a relationship most often assess it individualistically
through followers' (subordinates') or leaders' (managers') perceptual or self-reported scale ratings (Graen & Uhl-Bien, 1995).
This approach has limitations. For example, Dansereau, Yammarino, and Markham (1995) point out problems in leader–
member exchange research, which provides “little guidance as to how to view leaders or followers from a levels-of-analysis
perspective” (p. 254) and uses one measure to reflect relationships at all levels of analysis. Schriesheim, Castro, Zhou, and
Yammarino (2001) identify the misalignment in leadership theory between how it is theorized, how it is measured, and how it is
tested: “Most leadership approaches may be characterized as theorizing one thing (‘A’), while tests have typically examined
something else (‘B’)” (p. 541). House and Aditya (1997) criticize a field that recognizes relationships but ignores context, quoting
Hunt and Dodge (2000): “it is almost as though leadership scholars…have believed that leader–follower relationships exist in a
vacuum…[the] context in which leadership is enacted has been almost completely ignored” (p. 435).
To address these criticisms, leadership scholars are beginning to recognize broader, multi-paradigmatic perspectives sharing a
view “that leadership is co-created in relational interactions between people, and that…leadership is dynamic, developing and
changing over time” (Uhl-Bien & Ospina, 2012: 541). According to Denis, Langley, and Sergi (2012), these views represent a
“growing body of organizational research and theorizing that examines leadership not as a property of individuals and their
behaviors, but as a collective phenomenon that is distributed or shared among different people, potentially fluid, and constructed
in interaction” (p. 2). They locate leadership in the ways actors engage, interact, and negotiate with each other to influence
organizational understandings and produce outcomes (Barge, 2004a; Barge & Fairhurst, 2008).
In this way, they view leadership as a relational process co-created by leaders and followers in context. For example, discursive
scholars study the micro-behaviors behind the scale ratings to observe such things as sequential patterns of control; influential
acts of organizing (collective behavior); or co-defined relationship definitions. Dyadic or collective behavior is sine qua non here,
although discourse analysts interested in the reflexivity of individual leadership actors will probe interview accounts in search of
deeper meanings for the relationship, what Bradbury and Lichtenstein (2000) refer to as the “space between.” This doesn't make
discursive approaches superior to survey methods necessarily, but it does argue against how dominant survey methods have
become in the field of leadership today—especially when discursive theories and methods are breaking new ground in the
leadership-related questions they pose (Fairhurst & Grant, 2010).
In addition, paradigms (e.g., postpositivism, social constructionism, critical management studies, and so on), disciplines
(e.g., psychology, management, communication, sociology, and so on), and methods (e.g., quantitative, qualitative) often
approach research on vastly different assumptive grounds, but the lack of communication between them limits our ability to see
leadership as complexly as we might (Uhl-Bien & Ospina, 2012). A discursive perspective offers a potential remedy, first, because
it involves a wide variety of theories and methods that cross paradigms and disciplines. Second, by introducing a variety of
constructionist approaches and considering their potentialities vis-à-vis more postpositivist approaches, it can help clarify those
leadership questions for which an individual, cognitive, quantitative, and postpositivist view of leadership is best suited for and
 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062 1045

those questions for which a social, cultural, qualitative, and more constructionist view of leadership is best suited. It can even
illuminate the epistemology behind research areas such as the “romance” of leadership (Meindl, 1995; Meindl, Ehrlich, &
Dukerich, 1985), which have elements of both postpositivism and social constructionism. Instead of viewing paradigmatic,
disciplinary, or methodological perspectives as competing (with one “right” perspective), a discourse perspective and ODA
methods provide a platform through which we can begin to ask: What are their uses? What are their limits?

3. A primer on organizational discourse and ODA methods

There is more than one way to theorize discourse and use ODA methods to study leadership. As we will show, certain forms of ODA
mirror survey research by studying appointed leaders and direct reports, although the emphasis here is on the reciprocal and sequential
flow of social interaction. Others suspend the assumption of assigned leader roles to look for influential acts of organizing in the sequential
flow of action by any leadership actor. Still others focus on the language games of the actors involved and study what actors say and do in
the name of leadership. Here we refer to “language games” not as Pondy (1978) used the term strictly as a linguistic expression, but as
Wittgenstein (1953) described them in a much richer sense as “forms of life” (Kelly, 2008). Finally, some chart the Discourses (the
capitalization of which is explained below) of leadership, which are systems of thought anchored socio-historically in time (Foucault,
1995)—much as we saw a neo-charisma Discourse of leadership form around (the commonalities among) charismatic, transformational
and visionary leadership (Bryman, 1992, 1996). A focus on “Discourse” shows us how leadership actors are not just transformative
agents, but disciplined products of culture (Fairhurst, 2007a); they draw from linguistic repertoires that Discourses supply to
use language in very specific ways (Potter & Wetherell, 1987). For example, drawing from a neo-charisma Discourse, “leaders” become
visionaries and change agents while “managers” implement change (Bennis & Nanus, 1985; Conger, 1989; Zaleznik, 1977).
Thus, it is important to begin a discussion of organizational discourse with a very general introduction and note that many
forms of ODA have been chronicled in literature reviews (Fairhurst & Putnam, in press; Grant & Iedema, 2005; Phillips &
Oswick, 2012; Putnam & Fairhurst, 2001), handbooks (Bargiela-Chiappini, 2009; Grant et al., 2004; van Dijk, 1985), major
works collections (Grant, Hardy, et al., 2011), not to mention countless journal articles and edited volumes (Aritz & Walker,
2012; Grant, Keenoy, & Oswick, 1998; Holman & Thorpe, 2003; van Dijk, 1997; Westwood & Linstead, 2001). In a recent review
of the organizational discourse research across several top management journals, Hardy and Grant (2012) found that ODA was
represented approximately 9% of the time (although less so in U.S. management journals). This growing body of work has been
extended to such topics as organizational change (Balogun & Johnson, 2005; Jian, 2007; Leonardi & Jackson, 2004; Marshak &
Grant, 2008) and strategy (Knights & Morgan, 1991; Laine & Vaara, 2007; Law, 1994; Lilley, 2001) as well as leadership (Barge &
Fairhurst, 2008; du Gay et al., 1996; Fairhurst, 2007a; Kelly, 2008; Wodak, Kwon, & Clarke, 2011).
In more specific terms, several ODA methods are based on a broad class of constructionist theories that may include critical theories
(Alvesson & Deetz, 1996; Fairclough, 2006; Reed, 2004), postmodern and poststructuralist theories (Foucault, 1990, 1995), rhetoric
(Bitzer, 1968; Burke, 1962; Perelman & Olbrechts-Tyteca, 1969), ethnomethodology (Garfinkel, 1967, 2002), posthumanist theories
(Barad, 2003; Suchman, 2007), and a host of related others. All tend to focus on how organizational members use language and other
material means to socially construct their relationships and other organizational realities (e.g., the leadership relationship, or individual
and collective identities). This class of approaches was spurred on by the linguistic turn in social philosophy (Alvesson & Kärreman,
2000a; Rorty, 1967), which focused on the powers of language to construct reality, not just represent it. Prior to the linguistic turn, the
positivist view of language was that of a mirror reflecting reality—language was a mere artifact or surface manifestation of an existing
organizational reality (e.g., see early organizational culture studies where language is a cultural artifact). In contrast, ODA focuses on the
discourse itself—how it is organized and what it is doing (Potter & Wetherell, 1987).
Focusing on what discourse is doing as opposed to merely representing involve very different questions (e.g., how are the terms
“leader” versus “manager” being used to construct differences in task assignments, compensation, space allocations, use of
honorifics, and so on versus merely asking, what are the labels for those in charge?). As Bateson (1972) observed, the only
thing human beings have between them is the exchange of messages. Humans must use language to position themselves with
respect to one another, while these “positionings” then form patterned redundancies in social interaction that eventually get
institutionalized into organizational rules, roles, and systems. By contrast, other positionings may sew the seed of change that
alters the trajectory of organizational life.
However, discursive scholars are quick to eschew linguistic reductionism of any kind (e.g., it is highly unlikely that Bernie
Madoff can find just-the-right language to socially construct himself out of prison any time soon), so they focus instead on the
significant role language plays in situations involving ambiguity or uncertainty. Language helps us to frame events in one way
versus another, with action following accordingly (Fairhurst, 2011; Weick, Sutcliffe, & Obstfeld, 2005). It also points us to the
operations of power and material realities that are often occurring at both surface- and deep-structure levels of organizations and
society (Foucault, 1979, 1995; Giddens, 1979, 1984).
Yet, as many writers have observed (Fairhurst & Putnam, in press; Grant, Iedema, & Oswick, 2009; Grant, Hardy, et al., 2011; Grant,
Putnam, et al., 2011; Phillips & Oswick, 2012; Putnam & Fairhurst, 2001), even within a broad social constructionist epistemology
discursive approaches can differ quite a bit from one another. These differences may be reflected in their treatment of power and
influence (e.g., Are the workings of power the starting point for analysis, or must they be made relevant by the leadership actors
involved? Is power a negative and repressive property, or does it have both positive and negative features? How are deep-structures of
power best captured—as ideologies, discourses, or background understandings?). They may also be reflected in their treatment of agency
(e.g., When should leadership actors be viewed as transforming agents, when should they be seen as passive receptors of cultural
influences, and when should it be both? How should agency be conceived amidst socio-historical influences, the material structures of
1046 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062

society [e.g., social class], and/or other materialities [e.g., technology] with which humans interact in leadership situations?). They also
differ in their treatment of context and the whole question of levels of analysis (e.g., who decides and gives primary meaning to
the context under consideration—actors and/or analysts? Are there separate “macro” and “micro” levels of [societal, institutional,
organizational, and interpersonal] context, or is there is just the ongoing stream of behavior in which the [researcher-imposed] macro
level of analysis locates itself within the [researcher-imposed] micro level when considering leadership?). Space limitations do not permit
a full treatment of this subject for the study of leadership, but can be found elsewhere (Boden, 1994; Collinson, 2005, 2006; Fairhurst,
2007a, 2009; Fairhurst & Grant, 2010; Grint, 2000, 2005, 2010; Kelly, 2008).
Differences among discursive perspectives notwithstanding, most embrace a meaning-centered view of human communication—a
view in which the processes by which meanings get created, contested, and negotiated are fundamental to understanding leadership. To
take a simple example, consider how a team leader who believes strongly in empowerment must orient to a team whose previous
leaders were only autocratic. How does that team leader challenge their view of leadership and get them to take responsibility for
leading themselves and not always wait to be told what to do? Views of leadership here are likely to clash, and the meaning of leadership
must be negotiated if they are to work well together. Is leadership about being autocratic? Or, is it about self-management? Their
language games are different, to borrow from Wittgenstein (1953), and must be negotiated. Discourse scholars are interested in those
negotiations, however subtle and dance-like they may be.
This view is consistent with many emerging approaches in leadership that recognize the role of sensemaking (Gioia &
Chittipeddi, 1991; Pye, 2005; Rouleau, 2005; Weick, Sutcliffe, & Obstfeld, 1999; Weick et al., 2005) and identity management
(Collinson, 2006; Haslam & Platow, 2001; Hogg, 2001; Watson & Bargiela-Chiappini, 1998). It contrasts, however, with
postpositivist approaches that rely on a transmissional view of communication (i.e., the familiar Sender–Message–Receiver
model) (Fairhurst & Connaughton, in press). Transmissional views are used when the focus is on relational outcomes
(e.g., leader–member relationship quality) or leadership effectiveness (e.g., performance, satisfaction, commitment) rather than
on what the relationship means to the parties involved (e.g., whether the nature of exchange feels balanced). This latter point
opens the door to contested meanings. Although one might argue that leadership as the management of meaning has been
around since the 1980s (Pfeffer, 1981; Smircich & Morgan, 1982), in fact, it was not well understood because it was thought to be
the province of senior leaders (Bryman, 1992), and (vision) communication was usually directed downward (Fairhurst, 2001). By
contrast, a meaning-centered view of communication opens the door to understanding sensemaking and the potential struggles
over meaning wrought by diverging relational, organizational, or socio-historical influences.
The strong social constructionist ties (Berger & Luckman, 1966; Foucault, 1983, 1995) that underlie many ODA methods also produce
different, but complementary definitions of the term “discourse.” For novices, it is helpful to think of little “d” discourse as focusing on
language-in-use in social interaction, while big “D” Discourse signals a more Foucauldian view of discourse as a system of thought
anchored in time socio-historically (Alvesson & Kärreman, 2000b). Importantly, this latter view of Discourse is not just a way of thinking,
but a way of speaking. As mentioned in the neo-charisma Discourse example above, this view of Discourse supplies communicating
actors with a linguistic repertoire of terminology, habitual forms of argument, story themes, and so on (Potter & Wetherell, 1987).
Thus, if a subordinate expresses the following to her manager, “I'd like to play a role in this project,” to which the manager
responds, “You have my permission,” we see a “D”iscourse of authority being enacted here (i.e., culturally familiar language like
“permission” associated with authority in the West), but the “d”iscourse reflects an indirect request (i.e., there is an assertion of
interest, not a request, and “play a role” doesn't specify what kind of a role), while “You have my permission” is a direct,
authority-granting speech act. In order to narrow the scope of this paper, we chose to focus only on little “d” discourse, that is,
language-in-use in social interaction either between leadership actors themselves or between a leadership actor and an
interviewer; however, big “D”iscourse implications will be identified where relevant.
Finally, the social interaction that discourse analysts study comes packaged in texts, which are the material representations of talk
or interaction in written or recorded forms. Texts have their own properties, which can come into play in the research process in at
least two ways (Cooren, 2004; Taylor & Van Every, 2000). The first way is theoretical, as when researchers look at how leadership
actors make use of various texts to analyze how they layer and interweave to establish agency, authority, or legitimacy. Former Arthur
Anderson CEO, John Berardino, for example, obfuscated his testimony before the U.S. Congress on the Enron debacle by flooding his
testimony with a glut of accounting details from numerous texts that made the assignment of blame difficult (Craig & Amernic, 2004).
The second way is methodological and involves asking questions like, what kind of text(s) was used or produced to code the
interactions (e.g., a transcript, a video, or real-time)? Who produced the text (e.g., was a typist hired, or did the analyst transcribe it)?
How detailed was the text in representing the talk (e.g., how were talk-overs and interruptions handled)? How many pages of
transcribed data were coded? How many messages were coded, and how were they unitized (e.g., turn-at-talk, thought units, and so
on)? These are just a few examples of the many types of methodological questions discourse analysts might pose about texts.
Without question, a discursive perspective and ODA have grown into a large and wide body of research in the organizational
sciences, and readers are urged to consult this literature for more in-depth information about the theories and methods available
(Alvesson & Kärreman, 2011; Aritz & Walker, 2012; Bargiela-Chiappini, 2011; Fairhurst & Putnam, in press; Grant & Iedema, 2005;
Grant et al., 2004; Hardy & Grant, 2012; Iedema, 2011; Mumby, 2011; Mumby & Clair, 1997; Phillips & Oswick, 2012; Putnam &
Fairhurst, 2001; van Dijk, 1997).

4. The relevance of ODA for leadership

To recap thus far, by focusing on leader–follower relational interactions, several ODA methods have the potential to go beyond reports
of relationship quality to study the micro-behaviors of relationships-as-they-happen (Boden, 1994). Many also recognize that
 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062 1047

relationships are co-defined; therefore, a reliance on self-reports, especially in dyads, risks assuming a single relational reality and a
consensus that may not exist (Rogers, Millar, & Bavelas, 1985). Although ODA involving interview data are also reports about relationship
quality, they are not the same as a survey asking for a summary judgment in a scaled rating. Done properly, interview data can tell the
story of a relationship, including how individuals make sense of surprises, breaches, or other twists and turns and how they define the
context to explain the relationship, its history, and possible futures. ODA methods thus study individuals, dyads, and collectives, and not
just in qualitative form. ODA methods include the use of a priori coding schemes that are highly quantitative and often variable analytic.
Increasingly, we also see inductively-derived coding schemes that may be both quantitative and qualitative (Fairhurst, 2004a; Parry,
1998; Strauss & Corbin, 1998).
What distinguishes ODA from other approaches and methods is a general inclination to ask how and not why questions (Barge &
Fairhurst, 2008; Fairhurst, 2007a). Why questions are the domain of (postpositivist) hypothesis testing, causality, and predictive
relationships; they are often more reductionist with respect to context in the quest for generalizability (Liden & Antonakis, 2009).
Whereas why questions help us to understand questions of leader agency and effectiveness, how questions take apart the process of
leadership, often in great detail, in the quest for local knowledge. They help us to understand just how leadership is “brought off” in terms
of strings of jointly-produced utterances or actions; patterns of coordination (or lack thereof); struggles over meaning and interpretation;
and, increasingly, attention to objects (e.g., technology), sites (e.g., the physical setting) and bodies (e.g., how the trappings of power
manifests itself on the human body) (Ashcraft, Kuhn, & Cooren, 2009). Examples of how questions include: How is leadership constituted in
this situation, and who is making this claim (leadership actors, analysts, or both)? What is the task at hand, and how would one lead at that
task? How is leadership being co-created in the sequential and temporal forms of social interaction? How do leadership actors engage in
individual and collective identity work? How are leadership actors pairing with non-human objects (e.g., communication technologies) in order
to be perceived as leader-like (Barge, 2007; Fairhurst, 2007a; Fairhurst & Cooren, 2009; Fletcher, 2010; Kelly, 2008; Sinclair, 2005)?
Importantly, why and how kinds of questions should not be seen as competing, but complementary (Fairhurst, 2007a). This only
benefits leadership studies because of new questions that can now be asked about the relationalities involved in co-creating relationships.
Table 1 thus demonstrates two kinds of “how” leadership questions divided by analytic units, which is simply who or what an analyst
observes (Gronn, 2002). In this case and throughout the rest of the paper, we contrast ODA methods used with interaction data versus
ODA methods used with interview data. Interaction data are those in which two or more leadership actors (i.e., designated or emergent
leaders, followers, or other stakeholders) are engaged in social interaction that is observed or recorded. Interview data are those in which
leadership actors report on the relationship in a recorded response to interviewer questions.
Following the analytic units, the sample theoretical questions we pose get instantiated in observational units, which are the leadership
phenomena that discourse analysts want to explain, measure or deconstruct (Gronn, 2002). In the table, we show three observational
units for interaction data (control, mobilizing acts, and relationship work) and two observational units for interview data (sensemaking
and identity work). The discourse units in Table 1 correspond numerically to their observational units. For example, we showcase the
study of control via relational control coding and sensemaking through reflexive narration (i.e., insights gained through storytelling). We
should mention that it is also possible to study control via interview data or identity in interaction data (but not always easily).

5. Using ODA to study leadership as a relational process

Consistent with Table 1, we posit five research exemplars below, three with interaction data and two with interview data. ODA
certainly enables analysts to pose other questions, but they are not the focus here. Our research exemplars were chosen for the
range of ODA they present (from quantitative to qualitative) and how relevant they are for understanding leadership as a
relational process. Table 2 provides a more complete definition of the discourse analyses undertaken in these examples along
with a list of resources for further reading.
Contrary to predominant methodological approaches based on perceptual self-reports (i.e., retrospective summaries of
behavior), the first group of studies showcases actual interaction. One of the contributions of this type of ODA is a microscopic
zeroing in on leadership as it coheres in sequenced action, such as through the “interact” or the episodic structure of a “language

Table 1
Key theoretical questions and analytic, observational and discourse units. a,b

Key theoretical questions How is the relational brought off in terms How do leadership actors' engage in (1 or 2)
of (1, 2, or 3)? with respect to the relational?

Analytic units: Interaction data (2 people or more) Interviews (Individuals)

(Who or what an analyst observes)

Observational units: (That which is to be explained, 1. Control 1. Sensemaking

measured or deconstructed about leadership.) 2. Mobilizing a collective to act 2. Identity work
3. Relationship work

Discourse units:(Focus of discourse analysis) 1. Relational control moves 1. Reflexive narration

2. Mobilizing moves 2. Category work
3. Language games, humor,
terse story-telling
a
This framework is loosely adapted from Gronn (2002).
b
The numbers correspond between observational and discursive units.
1048 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062

game.” According to Boden (1994: 206, emphasis original), “By directly observing the sequential context and consequences of
quick verbal exchanges, analysts can track quite distinct interactional processes that are simultaneously organizational.” Indeed,
such a view gives credence to the tight coupling between leadership and organizing (Hosking & Morley, 1988).
The second group of studies focuses on interview data and, in particular, the role of narrative in sensemaking and identity
work. Leadership researchers have long understood the role of narrative and interview methods (Boje, 1991; Gronn, 1993; Parry
& Hansen, 2007; Sparrowe, 2005; Watson, 2001); however, these studies demonstrate the ways that narratives can be used

Table 2
Types of organizational discourse analysis.a

Interaction Analysis: Definition: Interaction analysis (IA) is a type of discourse analysis that focuses
(e.g., Relational control coding) exclusively on language-in-use, interaction process, and the coding of behavior according to a
predefined set of codes. It includes a host of quantitative approaches that draw from studies of
message functions and language structures to assess the frequency and types of coded verbal
behavior in organizational interaction.

Focus: The sequences and stages of interaction, their redundancy and predictability, and the link
between interactional structures and the organizational context (Putnam & Fairhurst, 2001).

Theory base: No common intellectual tradition anchors interaction analysis (Fairhurst, 2004a, 2004b).
Various theories including systems theory, behaviorist theory, and structuration theory.

Resources: Bakeman and Gottman (1986), Fairhurst (2004a), Rogers and Escudero (2004a, 2004b)

Critical discourse analysis Definition: Critical discourse analysis (CDA) is a broad class of approaches that
examine the often opaque power relationships between discursive practices, events and texts, and
wider social and cultural structures, relations or processes; and how power, hegemony, and struggles
over power shape and are shaped by discursive practices (Fairclough & Wodak, 1997).

Focus: Emphasis is given to the way discourse simultaneously produces and hides deep structure
relations of power and inequality; how organizations are sites of struggle where different groups
compete to shape a reality vested with their own interests; and how conflicts are resolved through
the control of symbolic, discursive, or material resources (Mumby & Clair, 1997).

Theory Base: Bakhtin (1981), Foucault (1972, 1983, 1995), Giddens (1979, 1984), Gramsci (1971),
Fairclough (1995, 2006), Habermas (1984)

Resources: Chouliaraki and Fairclough (2010), Fairclough (2006), Grant et al. (2009), Mumby (2004),
Cunliffe (2009)

Conversation analysis Definition: Conversation analysis (CA) is a type of discourse analysis that focuses on
the detailed organization of talk-in-interaction. Its main purpose is to discern how people use various
interactional methods and procedures to produce their activities and make sense of their worlds.

Focus: Turn taking, membership categorization, adjacency pairs, insertion sequences, accounting practices,
topic shifts, conversational openings and closings, agenda setting, decision making, etc.

Theory base: Ethnomethodology (Boden, 1990, 1994; Garfinkel, 1967)

Resources: Atkinson and Heritage (1984), Boden (1994), Heritage (1997), Pomerantz and Fehr (1997),
Psathas (1995), Jayyusi (1984)

Narrative analyses Definition: Narrative is a huge area of study that generally distinguishes between
narrative and story, although not consistently across authors. Gabriel (2004) sees narrative as a particular
type of text, involving temporal chains of interrelated events or actions undertaken by actors, wherein all
discourse is in some way narrative. Story requires ‘emplotment’ (characters, sequencing, predicaments,
action, and so on) or the weaving together of events in some meaningful way. However, characters and
events in stories can be real or imagined (Czarniawska-Joerges, 1997). Stories may be viewed as ‘poetic
elaborations’ of narrative material (Gabriel, 2004).

Focus: Narrative focus may be sensemaking, identity or relationship management, information sharing,
socialization, organizing processes, change, ideological control, and so on.

Theory base: Varies depending upon the type of narrative analysis. It may include Boje (2001), Bruner
(1986, 1990), Czarniawska-Joerges (1997), Fisher (1984), or Greimas (1987), among others.

Resources: Boje (1991, 2001), Bruner (1986, 1990), Czarniawska-Joerges (1997, 2004), Fisher (1984),
Gabriel (2004)

Membership categorization See conversation analysis.

a
This table was adapted from Fairhurst (2007a).
 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062 1049

reflexively (i.e., self-consciously) to better understand issues of ethics and moral responsibilities as well as mined for the ways in
which identities form, evolve, and/or get contested.
As we will see below, ODA is varied and wide-ranging. Nevertheless, it can add renewed vigor to leadership research by providing
complementary insights to extant leadership theories. The process orientation of ODA can better confirm existing theories, much the
way Courtright, Fairhurst, and Rogers (1989) demonstrated differences between Burns and Stalker's (1961) organic and mechanistic
systems in the control patterns of two manufacturing plants. ODA can point the way to new theorizing and testable propositions
about collective and distributed leadership, much as we see in the work of Ospina and colleagues (Foldy, Goldman, & Ospina,
2008; Ospina & Foldy, 2010; Ospina & Hittleman, 2011; Ospina & Sorensen, 2006). In these, and other ways, integrating ODA
with relational leadership theorizing (Uhl-Bien & Ospina, 2012) offers tremendous opportunities for addressing empirical
challenges that, heretofore, have hindered the ability to advance broader—and richer—theoretical and practical understandings
of leadership.

6. Studying relationality with interaction data

With interaction data, ODA focuses on sequenced behavior, similar to the way Weick (1979) described the “act,” “interact,”
and “double interact.” In this familiar scheme, an “act” is the behavior of one person, an “interact” is the response of another to
that behavior, and a “double interact” is the response to the response, as can be seen in the following example by Weick:
(S)uppose that a supervisor wants to get a worker to stop doing task A and start doing task B. The worker's action is the
doing of task A; the supervisor tries to influence the worker to do task B. Obviously, we must know how the worker
responds to this directive before we can make any statement about the complete influence attempt. But to determine
the worker's response, we need a specific description of the original activity as a basis for comparison. The worker's
typical response pattern will probably be altered in some way by the supervisor's directive, and before we can
understand the meaning of this alteration, we need to know the action that was already under way. (p. 89)

In interaction analysis, the sequentiality of social interaction (i.e., a series of acts, interacts, and double interacts) is of primary
concern. As Boden (1994) argues, the sequential context and quick verbal exchanges show interactional processes (e.g., seeking
and granting permission) to be simultaneously organizational (e.g., authority displays). The sequential nature of social interaction
is demonstrated in three research questions in Table 1 (column 1) addressing how relational aspects of leadership are brought off
in terms of: 1) control, 2) mobilizing a collective to act, and 3) relationship work. Although ODA involving interaction data is not
restricted to these three categories of questions, we use them to illustrate the kind of work that leadership scholars can engage in
to show how “talk is the work” (Gronn, 1983), or how the “interactional” creates the “organizational.”
Examining interaction data typically involves some form of interaction analysis. McDermott and Roth (1978) defined
interaction analysis relative to a person's interaction context: “a person's behavior is best described in terms of the behavior of those
immediately about that person, those with whom the person is doing interactional work in the construction of recognizable social
scenes or events (p. 321, emphasis original). McDermott and Roth were actually describing a number of linguistically-oriented
approaches in the 1970s with an emerging emphasis on sequence and temporal form in social interaction. More recently, Fairhurst
(2004a) more narrowly defined interaction analysis solely in terms of a priori coding schemes, like the relational control coding scheme
below. To distinguish this type of discourse analysis from others, we retain this definition (see Table 2).
Broadly speaking, however, interaction data can be analyzed in terms of a priori coding schemes (Fairhurst, 2004a), emergent
coding schemes such as those derived from grounded theory analyses (Charmaz, 2000; Parry, 1998), or close textual analysis that
do not require coding per se (Sacks, Schegloff, & Jefferson, 1974). Analyses can be performed on audio- or videotape transcriptions
of actual dialogue (e.g., routine work conversations, talk at meetings, etc.; Fairhurst, 2004a), on-line conversations (Carroll &
Simpson, 2012), or real-time observations of organizational practices (Komaki, Zlotnick, & Jensen, 1986). Oftentimes, multiple
methods such as those gained through ethnography (Van Maanen, 1998) are also present.
To illustrate this process, what follows are sample studies when the data and data analysis techniques are sensitive to
sequence and temporal form. These approaches can be used in leadership research to address questions about the sequential
timing or pacing of leadership actions or messages. Once again, the exemplars below mirror the key theoretical questions posed
in Table 1 (column 1).

6.1. Example 1: How are the relational aspects of leadership brought off in terms of control?

Issues of authority and control are routine concerns in the study of leadership (e.g., Who has authority over what kinds of
decisions? How much is control shared? What does that look like?), and there are of course many ways to study it. A group of
discourse analysts known as interaction analysts (see Table 2) code every turn-at-talk for its control implications. For example,
Fairhurst and colleagues (Courtright et al., 1989; Fairhurst, Green, & Courtright, 1995; Fairhurst, Rogers, & Sarr, 1987) used such a
system to study manager-subordinate control patterns in a multi-national manufacturing concern. Known as relational control
coding (Rogers & Escudero, 2004a, 2004b; Rogers & Farace, 1975), each turn-at-talk was coded as to whether managers and
subordinates asserted control (designated ‘one-up’), acquiesced or requested control (‘one-down’), or neutralized the control
move of the previous utterance (‘one-across’). Importantly, only the form of the utterance counts in this coding scheme, not the
1050 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062

content. Although content can vary widely across conversations, the form of a person's utterance—that is, how a message is
delivered—produces relatively fixed and reproducible patterns (Rogers et al., 1985; Watzlawick, Beavin, & Jackson, 1967).
In relational control analysis, an “act” is a single one-up (↑), one-down (↓), or one-across code (→) (Rogers & Farace, 1975).
However, an act alone tells us little about how it was received, i.e., its relational consequences. Therefore, manager-subordinate
interaction coding considers the “interact,” two contiguous control moves (by manager and subordinate), as the most basic unit of
analysis. Consistent with Weick's (1979) view of interacts, this approach allows us to see whether a subordinate consistently
complies with a manager's one-up orders (M↑S↓) or, alternatively, resists (M↑S↑). Relational control analyses also have the
capability to examine more extended relational sequences (that is, double interacts and beyond such as in a conflict sequence,
M↑S↑M↑S↑M↑) through Markov chain or lag sequential analyses or other stochastic methods.
Relational control patterns are identified and tested through observed frequencies in the data. Coding enables interaction
analysts to quantitatively analyze the sequences (interact, double interact and beyond) and stages of interaction, their
redundancy and predictability, and the link between interactional structures and aspects of the organizational context (Fairhurst,
2004a; Putnam & Fairhurst, 2001). For example, Courtright et al. (1989) coded over 11,000 messages from 45 manager–
subordinate pairs, each supplying 30 minutes of routine work conversation. They demonstrated relational control differences in
manager–subordinate interactions in organic versus mechanistic systems in two manufacturing plants. In the organic plant
organized around a self-managing teams' philosophy, communication was characterized by a variety of question-answer
combinations (M↓S↑; S↓M↑), reciprocated discussion (→→), and a lack of hierarchical control (M↑S↓). By contrast, in the
mechanistic plant, there were more competitive interchanges (↑↑), interruptions (→↑), and statements of nonsupport (S → M↑).
Fairhurst et al. (1995) coded almost 50,000 messages from 132 manager-subordinate pairs also supplying 30 minute
conversations and found a similar pattern among plants organized as self-managing from the start versus those attempting to
convert to self-management with a history of hierarchical control to overcome.
Control in the manager-subordinate relationship, shared or otherwise, emerges from co-created message patterns that evolve
into higher order systems (e.g., as dyads form work units, and work units form divisions, and so on). All are predicated on the
view that systems emerge over repeated interactions that evolve into multi-leveled orders of pattern (Bateson, 1972; Rogers &
Escudero, 2004a, 2004b). These interaction analysts assess the actual redundancy of relational message patterns that most other
discourse analysts (and Weick) just assume, and many other leadership researchers make inferences about through self-reports.
In other words, we know that relationships form behavioral patterns, but we don't study the behaviors per se.
An a priori coding scheme like relational control coding operates best when issues of organizational power and control are
relatively out in the open, such as in high reliability organizations in which the crisis mode requires clear command and control
communications (Fairhurst, 2004b; Fairhurst & Cooren, 2004). By contrast, relational control coding may not be sensitive enough
to such important relational phenomena as maintaining another person's face (Goffman, 1959) or when communication is
intentionally ambiguous (Eisenberg, 2007), all of which can obscure the operations of power and control in highly politicized
organizational environments (Fairhurst, 2007a). Leadership researchers can certainly avail themselves of other ways to study
control, both with and without coding schemes. The challenge is not just one of adopting a relational unit of analysis, but how
much the analysis is able to capture key aspects of leaders' and followers' relationality in the distribution of control.

6.2. Example 2: How are the relational aspects of leadership brought off in terms of mobilizing a collective to act?

Leadership is often about mobilizing a collective to act (Barker, 2002; Rost, 1991), and increasingly discourse analysts are
turning their attention to what this looks like in social interaction. In the example we provide, the emphasis is again on coding
sequences of manager–subordinate behavior; however, in this case, there is no a priori coding scheme or presumption about
organizational roles or positions. Analysts observe and code an interaction, or series of interactions, in terms of what Hosking
(1988) described as “influential acts of organizing.” These “influential acts” are key moves by designated or emergent leaders
where the primary emphasis is on the turning points in social interaction that mobilize a collective to act.
In order to do this type of analysis, which may be both qualitative and quantitative, analysts must derive an emergent coding
scheme because they are tracking the presence of “influential acts” of one kind or another that may be regular, but less frequent.
Thus, we can contrast this with the (a priori) relational control coding scheme shown above in which every turn-at-talk is coded
and observed redundancies noted. An example of an “influential acts” approach can be seen in Wodak et al.'s (2011) field study of
an Australian aerospace firm. Using a transcribed dataset over 150 hours long, including 45 h of individual interviews and 100 h
of meetings, they employed a form of critical discourse analysis (see Table 2) combined with ethnographic data that was sensitive
to both little “d”iscourse and big “D”Discourse (although they did not use this wording) (Reisigl & Wodak, 2009; Wodak, 2009).
They analyzed “the structural context of the organization and the respective history of specific communities of practice (such as
regular senior management team meetings or committees with the same participants)” (i.e., the big “D” Discourse) “with the
situational context of the meetings and the co-text of each utterance” (i.e., the little “d” discourse, as revealed through the texts
they studied) (Wodak et al., 2011: 597). This multilevel approach allowed them to understand not only the local context of the
utterances, but the broader socio-historical conditions the organization and its work communities faced.
Their close textual analysis identified five overall discursive strategies involving a number of micro-behaviors used by the company
CEO and COO to build consensus. These included: bonding through such things as pronominal use (e.g., the collective “we”); encouraging
in which leaders open a space for others to talk (e.g., soliciting opinions via open questions; request for expert contributions;
questioning/supporting existing propositions via repetition, back-channeling and explicit praise; frequent use of indirect instead direct
speech-acts; and silence); directing to bring a discussion toward closure and resolution through equivocality reduction (e.g., directed
 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062 1051

challenges; interrogative questioning; interruptions; direct speech-acts of request; disagreement; issue advocacy; and the preemptive
closing down of discussion); modulating through the framing of external environmental threats or institutional imperatives to act
vis-à-vis the issue at hand (e.g., argumentative appeals to common knowledge); and re/committing, which moves from the presumed
consensus at hand towards a commitment to act (e.g., speech-acts of promises, reminding others of role or personal obligations) with
an eye towards building a collective identity to internalize motivation (Wodak et al., 2011: 603–607).
From this analysis, Wodak et al. (2011) add to our understanding of relational leadership by discerning the actual talk patterns of
egalitarian versus authoritarian leadership styles and its impact on a durable consensus. While a balanced use of the five discursive
strategies named above produced such a consensus in key critical incidents, consensus quickly dissolved when the strategies were
imbalanced or authoritatively deployed in other such incidents. Their analysis also showed how an individual's current standing in
the team, and the context of the meeting itself, influenced leaders' participation and ability to control team interactions.
Another study of emergent interaction coding by Carroll and Simpson (2012) examined 93 online forum postings (paired with
ethnographic observations) by 20 participants conversing about the contents of a leadership development seminar in which they
participated. Framing episodes were constructed through “gesture and response” (i.e., interacts that are ordered, not necessarily
contiguous moves) patterns that simultaneously addressed each other and the ongoing conversation. Three, in particular, stood
out as moving the discussion forward from one person to the next: kindling, which furthered online discussion through testing
meanings, frames, and assumptions; stretching, which expanded the frames and meanings of key concepts and ideas; and
spanning, which linked key ideas and concepts together.
What is common to the above studies is that, set within the sequential flow of action, certain discursive moves have organizing
potential in their ability to mobilize a collective to act. Relationality is defined here in terms of turning points in collective action that
generate leadership attributions, in both studies by analysts, but they could easily have been made by actors as well (Fairhurst, 2007b).

6.3. Example 3: How are the relational aspects of leadership brought off in terms of relationship work?

When people communicate in high power and authority contexts, their communication is not just task-based, but very
relationship-oriented as well. Often, people are more polite, use titles, speak formally, or use ingratiating language (directed
upward), to name just a few examples (Goffman, 1959). ODA researchers can easily track this “relationship work,” and there are
many ways to do this (Fairhurst, 2007a; Tracy, 1991).
One way focuses on the co-produced language games of leadership actors, which are co-created routines (or sets of routines)
with logics of their own, such as ordering food at a restaurant or running a meeting (Wittgenstein, 1953). Again, they are not just
language based as Pondy (1978) used the term (Kelly, 2008). Instead, language games as used here mean words and/or actions
asserting a definition of the relationship that—whether they reflect relationship turning points or business as usual—require other
role set members to challenge or play along. In either case, the unit of analysis might be a simple “gesture and response” or an
“episode.” Close textual analyses and grounded theory case-comparison methods, rich in contextual information, are often
necessary to grasp the nature of these relational units.
For example, the stories that leaders or followers tell might be analyzed for how, in their telling, they garner favor, support, or
rapport—or, alternatively, how they control, compete, or sustain a conflict (Boje, 1991). According to Ochs (1997), “the interactional
production of narrative maintains and transforms persons and relationships. How we think about ourselves and others is influenced by
both the message content of jointly told narratives and the experience of working together to construct a coherent narrative” (p. 185).
This insight was used by Fairhurst (1993) to examine the interactional foundations of leader–member exchange with 6 women
leaders and 16 of their members in a multi-national manufacturing concern. She undertook a form of close textual analysis known as
conversation analysis (see Table 2) in which people use various interactional methods to produce their activities and make sense of their
worlds (Pomerantz & Fehr, 1997; Sacks, 1992; Sacks et al., 1974). Previous to this type of analysis, Fairhurst et al. (1987) tested and found
a weak relationship between LMX and relational control (using the previously mentioned relational control coding scheme), although
LMX theory suggests clear differences in control between high and low LMX members (Graen & Scandura, 1987; Graen & Uhl-Bien,
1995). However, Fairhurst (1993) found a much stronger set of findings with discursive methods more sensitive to the context and
micro-relationalities. For example, consider the storytelling of one low LMX member (M) who just happened to be male, white, older,
less educated, but more tenured than his supervisor (L) who is female, black, young, more educated, and less tenured. Following the
conventions of conversation analysis (Sacks et al., 1974), note how the marking in this excerpt display the micro-behaviors of this
interaction (e.g., italics = emphasis and stress; (2.3)=2.3 seconds of time; no::: = elongation of a syllable; [ = simultaneous speech).

1 M: I almost ended up on disability twice this weekend.

2 L: What happened?
3 (2.3)
4 M: Goin' down the stairs my leg gave out, and I went
5 flying down the stairs.
6 L: Oh no::: (.) here at work? (1.9) [let me see.
7 M: [INDISTINGUISABLE
8 L: You're [hurt?.
9 M: [INDISTINGUISHABLE
10 L: Oh oh you hurt yourself at home?
11 M: Yeah (.9) Just couldn't put any weight on the leg and
1052 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062

12 went down the stairs,

13 L: Oh go:::d
14 M: Grabbed onto the wall. Cut that.
15 L: Oh god, I see. Let me see let me see this hand (1.4)
16 You say you cut that one. I don't I don't see, oh on your
17 finger?
18 (.7)
19 M: I got a couple little [cuts]
20 L: [You] have one on this hand too? Okay.
21 You're gonna have to be careful. You're gonna hurt
22 yourself real bad.
23 (1.4)
24 M: Then I almost drowned.
25 L: You almost drowned? (.6) Uh, wh-what happened?
26 M: I (.5) went swimming (1.1), stayed in the pool until I gave out (1.0).
27 L: Till you gave out?
28 M: Yeah, I was swimmin' 80–80 foot length of the pool and gave
29 out when it was 9 feet deep.
30 (3.8)
31 L: What happened? (2.8) How did you save yourself? What
32 happened?
33 M: Just kept swimmin'
34 (.8)
35 L: I'm glad you're okay, Herb. God you you scare me
36 sometimes.
37 M: I almost drowned a lotta times (2.2). It doesn't stop me
38 me.
39 L: You better take care of yourself (2.1). You should do that
40 cause you scare me with some of those stories you
41 tell me, I don't know.
42 M: I got a pill from the V.A. now I could kill myself if I
43 wanted to
44 (1.0)
45 L: You got a pill to kill yourself?
46 M: Yeah, if I wanted to kill myself, I got a pill to do
47 it now.
48 (1.6)
49 L: Well, I hope you don't take the thing.
50 M: Great big (1.8) yellow pill. No markings on it
51 (3.4)
52 L: ((Sigh)) Mr. Cunningham, (1.1) I don't know about you
53 sometimes. I wonder (1.2) Anyhow (2.6) let's get back
54 to this. As far as (.7) the associates with mechanical skills
55 working the line (1.8) who else can fill in?

This low LMX member tells fantastical stories of falling down stairs (lines 4–5); bodily injury (lines 4, 11–12); near drowning
(lines 24–38); and potential suicide (lines 42–50). Yet, these stories are a joint performance between leader and member (Boje,
1991). The leader readily plays along with each expression of strong interest and/or concern (lines 2, 6, 8, 10, 13, 15–17, 21–22,
25, 27, and so on) until the subject of suicide emerges (line 42), at which point she takes the unusual step of switching the subject
back to more work related concerns. Displaying more concern for a cut on the hand than talk of suicide makes sense only if the
member is no longer credible in her eyes (Fairhurst, 1993), and/or the leader was made uncomfortable by such a disclosure.
Turn-by-turn, sequenced action coheres into a language game, in this case, a power game that the older white male member is
playing with his younger, black female leader. Together they co-construct an interactional pattern of deceit, distrust, and
resistance to authority (Fairhurst & Hamlett, 2003). In the process, we see how storytelling, as an assisted control move,
constitutes, not merely reflects, this low quality LMX.
Fairhurst (1993) also showed discourse from a high quality LMX relationship in which there was insider joking, also a feature of
LMX theorizing (Graen & Scandura, 1987). Note how the male member teases his female leader as she responds with a narrative.

1 M: ((papers shuffling)) I wanted to talk about (.4) feasibility

2 (1.4) uhm ((papers shuffling)) (2.9) update on uh (.) making
3 operation, my perspective on that
4 L: Oh good (2.6) Mm-hmm.
5 M: Is this how you're gonna do this while we're on tape
6 keep sayin' “Oh good” to everything I mention?
7 L: HAH-HEH-heh-[heh-heh-heh-huh
8 M: [Heh-heh-huh-huh
9 L: That's what I told Carol, I said, “Hey, I know you've
10 already had this uh (.4) session already so if you do anything
11 out of characteris- out of character, I'm going to call you
12 on it.”
 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062 1053

The leader's backchannel comments at line 4 (“Oh good. (2.6) Mm-hmm.”) prompt the member to humorously tease the
leader by reflecting on her behavior in story-like fashion (lines 5–6: “Is this how you're gonna do this while we're on tape keep
sayin' ‘Oh good’ to everything I mention?”). The humor is face-threatening, but playful and indicative of a high quality LMX as the
leader reacts favorably to the jab with a hearty laugh (line 7) and her own narrative that she aligns with the sentiments of the
member (line 9: “That's what I told Carol…”). Again we see sequenced action cohere into a language game, a co-produced
narrative by leader and member revealing the way humor can disarm potentially face-threatening challenges.
As the above examples demonstrate, the narrative construction of LMX and its story performances are co-creations of leaders'
and members' talk. Relational meanings are continuously negotiated and co-defined, especially in “terse tellings,” a phrase Boje
(1991) uses to describe the theatrical license narrators take to accentuate, abbreviate, or otherwise edit selected stories. These
findings thus demonstrate the importance of analyzing high versus low LMX stories-in-their-telling versus capturing them as
survey items that may trigger socially desirable answers. When surveys are the goal, stories may get turned into “does your leader
(or follower) play power games?” and “does your leader (or follower) engage in insider joking?”, while “terse tellings” are likely
discarded as just so much minutiae.
To recap, we have clearly gravitated from a priori to emergent coding schemes to close textual analyses, and we have sampled
both quantitative and qualitative discourse analyses. All of this has been in service to understanding relationships among
leadership actors-as-they-happen. We now turn to a discussion of qualitative methods with two more categories based solely on
storytelling in interview data.

7. Studying relationality with interview data

Interview data has come a long way since its days as an anecdotal backdrop to survey research. Interviews are now often a key
aspect of ethnographic data (Van Maanen, 1998), and in-depth interviews have become common research tools for exploring
issues of meaning, sensemaking, identity, and relational history (Johnson, 2002; Ochs, 1997; Tsoukas & Hatch, 2001; Weick et al.,
2005). They contribute to our understanding of leadership as relational because the relationship is narrated and explained by at
least one of the participants. Quite often, there is very little else but interview data because so many conversations between
leadership actors today are impossible to record and analyze for legal or proprietary reasons.
As interviews have grown in length and depth, discursive methods normally require that they be recorded and transcribed.
Although interview data are generally not subject to close textual analysis of interviewer–interviewee interaction (Sacks et al.,
1974), the thematic coding of interviewee responses based on grounded theory analyses is today quite common (Charmaz, 2000;
Strauss & Corbin, 1998). The question often becomes, is this discourse analysis? Technically, the coding of content themes is a
feature of the discourse. However, discourse analysts often go beyond thematic analyses (Hardy & Grant, 2012) to examine other
discourse features such as narrative (Boje, 2001; Holstein & Gubrium, 2000); categorization work in talk about individual or
collective identities (e.g., “I'm a union member, but in name only”) (Jayyusi, 1984; Watson, 1997); pronominal use (i.e., the use of
pronouns) as identity markers such as the collective “we” (Cheney, Christensen, Conrad, & Lair, 2004); or the use of terminology
and habitual forms of argument to signal the operation of broader Discourses (e.g., associating “leaders” with having an
organizational vision in neo-charisma Discourse) (Wetherell, 1998). Below we provide examples of relational studies when the
data and data analysis techniques involve interviews. These examples mirror the key theoretical questions posed in Table 1
(column 2).

7.1. Example 4: How do leadership actors' engage in sensemaking with respect to the relational?

This category considers how some of the stories that leaders and followers tell in interviews are cast as opportunities for
reflexivity (i.e., self-conscious thought) and sensemaking, the unpacking of which may lead to deeper relational meanings,
collaborative learning, or opportunities for morally responsive action. Key theoretical questions thus form around the experience
of leadership and followership in moments of leading or following. Narrative analysis is crucial here because the experience of
leading and following can only be accessed retrospectively through stories told to analysts (Cunliffe & Eriksen, 2011).
For example, working from the social constructionist theorizing of Shotter (2008), Cunliffe (2008, 2009) adopts a
“relationally-responsive orientation” to leadership study in which leadership becomes a way of being with others and a
self-conscious way of relating to them. Interviewers thus pay careful attention to leadership actors' narratives and language so as
to better grasp their sensemaking opportunities for morally responsive action. Cunliffe assumes that leadership actors already
know a lot about who they are and why they behave the way they do. She's looking to tap into this self-knowledge to drive more
ethical leadership.
Cunliffe and Eriksen (2011) did this in their 3-year ethnography of Federal Security Directors (FSDs) in a busy U.S. airport.
In-depth interviews with FSDs drew attention to issues of character, involvement, dialogue and multi-vocality, responsibility,
integrity, and the ethics of reciprocity. All point to (the presence of and need for) moral responsiveness within the leadership
relationship, but also a kind of a practical wisdom of knowing-from-within (again, self-conscious) interactional moments. Such
moments are usually the products of dialogue, shared sensemaking, collaborative learning, and local knowledge, much as we see
in the following example from one of Cunliffe and Eriksen's FSDs:

I will wear a uniform and I will go out there and work a shift at a position…because I want to understand what they (the
screeners) are dealing with day in and day out…I'll come in here, put my uniform on and go out to the checkpoints and
1054 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062

work the whole shift with them. It is a grueling task. It is lifting bags all day long and after a while, it takes its toll on you
and of course, the screeners get very artful. They learn how to do it. (p. 1442).

Here the supervisor speaks of the “artfulness” of screeners' learning, a pointed reference to the knowledge that screeners
collaboratively derive and form meaning around given the task of continuously lifting bags. Analysts interrogate such points for
deeper relational meanings and, perhaps, new commitments for understanding in the relationship.
Still other work by Cunliffe (2004, 2009; Cunliffe & Jun, 2005) in the area of relational responsiveness focuses on reflexivity
and sensemaking in the leadership relationship that is distinct from simple reflection. Cunliffe (2009: 45) elaborates, “reflexivity
goes deeper than reflection, because it means interrogating the taken-for-granted by questioning our relationship with our social
world and the ways in which we account for our experience.” Leaders and managers can be “reflexive practitioners” in two ways.
First is through self-reflexivity in which they reconsider their own experiences of being in conversation with others as a basis for
exploring how they can personally act in more morally responsible ways. The aforementioned quote from an FSD who sees the
“artfulness” of his screeners' actions would likely qualify in this regard. Second is through critical-reflexivity in which they
interrogate the operations of power, organizational contradictions and paradoxes, and that which remains unspoken to arrive at
better (i.e., more morally grounded) ways of leading and managing. For example, consider the display of critical reflexivity by a
manager cited in Cunliffe (2009):
…by caring, by empathizing, by questioning the assumption behind systems, and by keeping an eye on what is truly
important. That's a challenge for me…But there are two ends of the ‘manage’ spectrum…Managing and management.
Managing is a way of being and relating—a continually emerging, embodied practice. Management is a series of
disembodied activities. And, balancing the two, I believe, is critical: just as it is in the implementation of any philosophy,
system or habit. A leader can spend too much time on facts and details, concrete steps of implementation, HR policies and
procedures and the like. A leader can also overly focus on being philosophic—to the neglect of proven leadership tools,
principles and techniques as taught by the likes of Peter Drucker and Henry Mintzberg. So maintaining a balance is
important…(p. 48–49)

Note how the contrast between managing and management gives this manager license to question the system in which he
resides, but also the one he knowingly creates as he seeks a balance among his responsibilities. Cunliffe's approach is to look for
those sensemaking moments in interview discourse of self and critical reflexivity to simultaneously learn about and reinforce
more ethical management and leadership.
Barge(Barge 2004a, 2004b, 2004c; Barge & Craig 2009; Barge & Fairhurst, 2008; Barge & Little, 2002) draws from many of the same
sources as Cunliffe, but with the added emphasis on practical theory to aid actors' sensemaking. Reminiscent of the action science
tradition of Argyris and Schön (1996; Schön, 1983), practical theory is theory-as-instrumentality, in Craig's (1995) words, not just to
learn “what communication is, but also what it should be” (p. ix). As such, Barge (2007) interrogates leader and member storytelling
in interviews as an impetus to action and intervention. Storytelling becomes “systemic” because there may be multiple stories to tell
whose convergences and divergences must be explored to respect everyone's place in the system yet “create a coherent narrative
about the situation from the various stories…to take action” (Barge, 2007: 12). Likewise, Barge and Oliver (2003) draw from positive
psychology to urge leadership actors to pursue the positive stories of organizational life—to work with appreciation in pursuing that
which is life-generating and affirmative so as to bring about positive organizational change.
The logic behind Barge and Cunliffe's work is that leadership and followership are co-created in and through the relational nature
of social interaction. If leadership is a social construction (Fairhurst & Grant, 2010), it follows that relational identities are potentially
affirmed or reaffirmed with each communication opportunity. By necessity then, just who leaders and followers are and how they
relate to one another must be open to reinvention when their sensemaking is problematized in well-crafted, in-depth interviews.
Thus, we see how the interview genre is alive with the potential for reflexivity, moral accountability, and positive change.

7.2. Example 5: How do leadership actors' engage in identity work with respect to the relational?

As we can see from the previous section, a meaning-centered view of communication is a prerequisite to leadership actors'
sensemaking accounts. As Drazin, Glynn, and Kazanjian (1999) explain, “Meaning—or sense—develops about the situation,
which allows the individual to act in some rational fashion; thus meaning—or sensemaking—is a primary generator of
individual action” (p. 293). However, if leadership actors or collectives develop a cause-map of the world as a result of their
sensemaking efforts, inevitably, they must situate themselves in this map (Drazin et al., 1999). Thus, we are likely to find
individual and collective identity work in their sensemaking accounts in the form of categorizing and framing linguistic activity
in response to questions such as, “Who am I (in this context)?” and “Who are we?” Interview data are the likely repository of
such accounts.
We can see this demonstrated in a program of research on the Episcopalian ministry by Kreiner, Hollensbe and Sheep (Kreiner,
Hollensbe, & Sheep, 2006a, 2006b; Sheep, 2006a, 2006b; Sheep, Hollensbe, & Kreiner, 2009). For example, these authors report
two studies of Episcopalian parish priests who, as leaders of their congregations, experience a number of identity tensions and
conflict among their multiple roles (Kreiner et al., 2006a). Study 1 was an exploratory survey designed to set up Study 2, which
was in-depth interviews of 60 U.S. Episcopal priests. Using grounded theory methods, they generated several identity-related
themes associated with leading their congregations. They chronicled identity demands, such as how one priest viewed his social
 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062 1055

identity as a calling: “It's not a profession; it's a calling. It's not a job; it's a life.” They chronicled identity tensions, such as the
following:
In our diocese, they used to call them ‘the black shirts.’ That group, those are the guys who we think went to bed with their
clericals on. Being a priest was almost a different status. You have human beings, and you have priests. (p. 1039)

[My parishioners] have no access to the internal part of me…I think that there are huge parts of who I am as a person [that] they
just don't have access to unless I'm very explicit and make that personal, internal part explicitly out there for them. (p. 1039)

In the first example, we see identity tensions emerge from “the black shirts” or those who over-identify with being a priest
such that there is loss of self-identity that may be imperceptible to the person. In the second example, identity tensions emerge
from not being able to be the “real me” to one's constituency, suggesting a lack of identity transparency.
Kreiner et al. (2006a) chronicled the three kinds of identity work tactics that priests engaged in to gain some semblance of balance
in their lives. Three kinds of tactics emerged, including differentiation, which separated personal identity from social (leadership)
identities that served to “separate the ‘me’ from the ‘we’” (p. 1043); integration, which involved the blending of individual identities
with the occupational and/or organizational leadership identity, thus serving to “fuse the ‘me’ with the ‘we’ to varying degrees,
moving the individual closer to identity integration” (p. 1047); and, finally, neutral/dual-function tactics that were contingency
mechanisms facilitating either differentiation or integration. The in-depth interviews also documented the priests' journeys regarding
their leadership identities and the kinds of identity work they undertook to strike a balance in their lives among multiple work-family
roles. However, as rich as these analyses are, they focus heavily on a thematic content analysis of interview talk.
It is when Sheep and colleagues further explore identity tensions in the hierarchy of the Episcopalian church (as a result of the
election of its first openly gay minister) that they move into the more familiar territory covered by ODA. Sheep (2006a, 2006b)
and Sheep et al. (2009) examine the identity work vis-à-vis membership categorization (see Table 2), which refers to the
interactional work of actors who use categories (e.g., “priests”) to make claims and/or their actions accountable (Jayyusi, 1984;
Sacks, 1992; Watson, 1997). Oftentimes, language users move beyond the simple use of categories to characterize or reconstitute
them as the occasion warrants (e.g., the historical reference to “blackshirts,” which pejoratively distinguished overly pious priests
from mere humans) (Jayyusi, 1984). Consider another example below from Sheep et al. (2009):
Interviewee: What do you mean? ((laughter)) This is great! Do you mean how it purports to be? I mean this is- Or how it
really is? ((laughter))

Interviewer: Um, let's do both. Let's start with purports.

Interviewee: Well, I mean, it purports to be a bridge Church between the best of Protestantism and the best of Roman
Catholicism—that it's a thinking person's Church that allows a person to major on the majors and not major on the minors.
And to find a balanced form of community that has both spiritual depth and intellectual integrity in the modern world.
That's the ostensible presentation….The reality is that it is a deeply upper middle class, deeply white, deeply Western,
deeply increasingly secularized denomination that is elitist in its outlook and out of touch with not only its own grass roots
but with much of the rest of the world. …And the Episcopal Church is very, very much self-consciously, you know, focused
on itself and its own sense of importance. And this is in a period of decline. (p. 21–22)

This passage lists a number of categories of purported identities (e.g., “a bridge Church,” “a thinking person's Church”).
However, the identities that are realized tell a much different story through a different set of categories invoked (e.g., “deeply
upper middle class,” “deeply white,” “deeply Western,” “deeply…secularized,” “elitist”) out of touch with itself and the world,
self-important, and in decline. Sheep et al. go on to describe how the category work of these language users has the capability to
divide this ministry—through categories that split—and the capability to unite the ministry—through categories that stretch. An
example of “stretching” a membership category can be seen in the “circus” metaphor invoked below:

And this is kind of a crazy place; we have, yeah, a lot of African students and we had a lot of gay people and this that and the
other—street people coming in, and then, all kinds of folks. And he said, ―I really want my family to be here because, you
know, this is the kind of a world I want my family to treasure …Um, it's like, instead of going to a circus and seeing only
elephant acts, you get to see the whole circus. And so I think that's more and more what we're becoming, and being proud
of that. (p. 24)

Sheep et al. argue that the circus metaphor implies key features of the Church—its diversity and comprehensiveness—without
having to explicitly mention them. Moreover, the “whole circus” membership categorization suggests “multiple identities, though
diverse, are not viewed as conflictual but as compatible parts of a whole” (p. 25, emphasis original). As Martin Luther King taught
us in his famous “I Have a Dream” speech, leadership comes alive with metaphors that unite, rather than divide. It would appear
that this Episcopalian priest-leader has found one.
To recap, interview discourse can be a very rich source of data for ODA; however, many researchers do not go beyond a
thematic content analyses of their data. As we saw in the research presented above on narrative and membership categorization,
ODA methods can add further depth and dimensionality to one's research beyond the study of content themes.
1056 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062

8. Discussion

In our discussion of the relevance of ODA to leadership study, we address four primary questions: First, why should leadership
scholars care about ODA? Second, does using ODA mean abandoning traditional scientific methods for more interpretive ones?
Third, what are the criteria for evaluating the adequacy of ODA results? Fourth, how does one acquire an ODA skill set?

8.1. Why should leadership scholars care about ODA?

There was a time when survey research was our only viable means of studying leadership processes. That is no longer the case.
In its many forms, ODA offers that fly-on-the-wall methodology for researchers to see more of how leadership unfolds in a
co-created process among relational beings (i.e., people acting in relation to and with respect to one another in sequenced action).
This is an insight that management scholars who study bargaining and negotiation have been embracing for some time now
(Olekalns, Smith, & Walsh, 1996; Weingart, Bennett, & Brett, 1993).
Hinde (1979) offers a deceptively simple example of a compelling rationale for studying sequential processes: he argues that it
makes a relational difference whether a husband and wife consistently kiss after they quarrel or quarrel after they kiss, even the
though the total amount of quarreling and kissing may be the same. This, of course, is marital behavior, but we could easily
extrapolate to putatively “democratic” leaders who invite discussion, but always manage to decide in the end, or invite discussion
after setting the parameters (Stohl & Cheney, 2001). By contrast, consider how command presence at a police or fire scene is not
really about an overall style or the setting up of “command post,” but stepping into the stream of chaos at just-the-right-moment
to lend order and, perhaps, calm (Cooren & Fairhurst, 2004; Fairhurst & Cooren, 2004). Finally, consider how Komaki (1998) was
able to distinguish between effective and ineffective leaders from sailboat races to manufacturing environments based on the
sequential placement and pacing of performance monitoring messages. It is unfortunate that examples from the leadership
literature are few and far between—because whether it is dyadic or collective behavior, traditional or distributed leadership, the
study of sequential processes opens up new and important avenues for leadership research.
We are not suggesting that survey methodologies are not useful; on the contrary, they are ideally suited for studying
perceptions and summary judgments (e.g., implicit theories of leadership). More importantly, they partner well with interaction
and interview data in leadership studies (Fairhurst, 1993; Fairhurst et al., 1987). A key reason to begin to collect other kinds of
data besides surveys is that when leadership processes are studied only through the latter, communication is seen as little more
than acts of transmission (recall, the Sender-Message-Receiver model) and not as about the creation of meaning for the
relationship and its many aspects. The transmissional view of communication is not incorrect, but it is woefully narrow (Ashcraft
et al., 2009).
To conceive of communication properly, relational communication theorists Rogers and Escudero (2004a, 2004b) argue
that analysts must “say something about the social, systemic, temporal, circular, reflexive, multi-leveled complexities of
communication” (p. 19). Relational communication analyses—in whatever form of ODA they take—have the potential to provide
new information that survey research is unable to deliver on the constitutive and formative aspects of leadership relationships.
They do this for two reasons. First, whether quantitatively or qualitatively, they study process via the micro-sequences of behavior
—they recognize that leadership, like all organizational action (Boden, 1994), coheres as a sequence. We saw this most clearly in
the first three research exemplars, but in very different ways. The first form of ODA showed that leadership coheres as a sequence
by coding every turn at talk; the second showed that leadership involves shared influence processes by coding influential acts of
organizing; and the third cast leadership as co-produced language games. Of these methods, one is not better than the other;
instead, it is their variety that provides rich opportunities to open new pathways for investigating and better understanding
leadership processes.
Second, ODA allows us to consider a great deal more of the context than survey methods. In the first research example of
relational control coding, the context was considered in two ways: the immediate interactional context of the previous utterance,
and the plant context of the organic or mechanistic system. In the second example, context was also multi-leveled: it was the
nature of the task at hand for an Australian aerospace team seeking a durable consensus, but it was also the team's structural
context and the history of this community of practice. In the third example, the multi-leveled context included prior utterances
made sensible by the context of the language games of high versus low LMX relationships. Finally, in the fourth and fifth examples
involving interview data, the context was what the leadership actors made relevant in terms of what was significant to them.
Thus, many forms of ODA shy away from researcher-imposed levels of analysis (Boden, 1994), leaving the job to actors, not
analysts, to decide the contextual factors that are sine qua non to how sense is made of “the situation here and now.” Once again,
one handling of context is not necessarily better than another, but collectively, these examples demonstrate a range of options for
considering how to address context in ways that are not possible with survey methods alone.

8.2. Does using ODA mean abandoning traditional scientific methods for more interpretive methods?

The answer to this question is “no,” not unless you are already inclined to cross paradigms, or your work leads you in that
direction anyway. That was the reason we presented a variety of ODA methods and exemplars ranging from the highly
quantitative and postpositivist to highly qualitative and interpretivist or constructionist. One's epistemological orientation will
dictate how much one values (a) the search for a common language of leadership, (b) a “right” and universal definition
of leadership, and (c) a search for generalizable knowledge. Many forms of ODA challenge these assumptions. The more
 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062 1057

constructionist the ODA scholar, the more likely are they to believe that, following Wittgenstein (1953), leadership is one of those
“blurred concepts,” and following Gallie (1956), an essentially contested one because leadership will always be in the eye of the
beholder. The more constructionist scholar believes it is better to view leadership as a family resemblance among language games
(Kelly, 2008; Wittgenstein, 1953) where leadership becomes a “design problem” for the actors involved (Kelly, White, Martin, &
Rouncefield, 2006). That is, leaders must figure out what leadership is in the context of what they do and, through their framing
and actions, persuade themselves and other people that they are doing it (Fairhurst, 2011; Grint, 2000).
While constructionist scholars takes these views as a starting point for their research, the more postpositivist scholars will not
(Fairhurst, 2007a; Fairhurst & Connaughton, in press), and that is still acceptable in the context of ODA. For postpositivist scholars,
ODA may mean a heavy use of a priori coding schemes, the creation of variables and statistical testing or, alternatively, emergent
coding schemes derived from grounded theory analyses. Our view is to “let a thousand flowers bloom” and encourage the study of
leadership from many and varied perspectives.

8.3. What are the criteria for evaluating the adequacy of ODA results?

Because ODA involves such a wide array of methods, there are a number of ways in which to answer this question. As
mentioned, ODA methods involving a priori coding schemes often operate from within a postpositive perspective where its
conventional standards for validity and reliability apply. However, there are often additional reliability burdens beyond coding
scheme reliability, such as in unitizing the talk samples (e.g., is the unit of analysis a sentence, thought-unit, or turn-at-talk?).
Therefore, unitizing reliability and other issues regarding the handling of sequential data must be addressed (Bakeman &
Gottman, 1986; Gottman, 1982). Similarly, ODA methods that make use of grounded theory also follow the validity and reliability
conventions in this area as well (Strauss & Corbin, 1998), although discourse analysts often adopt a more constructionist version
of grounded theory, such as we find in the work of Charmaz (2000, 2006). The same is true for ODA methods used in connection
with ethnography (Van Maanen, 1998) and more qualitative methods generally (Creswell, 2007; Patton, 2002).
Importantly, most constructionist forms of ODA are not subject to statistical testing. As Patton (2002) allows, “There is no
equivalent of a statistical significance test or factor score to tell the analyst when results are important or what quotations fit
together under the same theme” (pp. 57–58). The first thing that novices to more constructionist ODA methods must know is that
these discourse analysts are very concerned about the rigor of their findings (Boden, 1990, 1994; Jackson, 1986; Pomerantz, 1990;
Potter, 1997; Potter & Wetherell, 1987; Taylor, 2001; Wetherell, 2001; Wood & Kroger, 2000), but validity is not a simple issue.
Validation criteria turn on such things as logical coherence, the generation of novel findings with “punch” or insight, plausibility,
and grounding in previous research (Wetherell, 2001). For ODA methods that focus on social interaction, the participants'
responses are crucial to demonstrating how they are making sense of what has gone before. Because it is the participants who
must coordinate the management of meaning, the ways in which they show what they understand as the conversation unfolds is
crucial to the analysis (Boden, 1990, 1994; Molotch & Boden, 1985). Some discourse analysts distinguish between meanings that
could be conventionally assigned (i.e., as a member of the same language community) to a set of utterances versus those that
might be more private and idiosyncratic—and then restrict their observations only to conventionally assigned meanings
(Fairhurst, 1993). Other discourse analysts observe the interaction and then interrogate the participants afterwards to address
issues of meaning and interpretation (Tracy, 1991, 1995). For ODA methods involving interviews, analysts frequently build their
narratives by asking the host organization to comment upon their veridicality (Kreiner et al., 2006a; Sheep, 2006a, 2006b).
The very real challenge ahead is for increasing numbers of leadership researchers to understand postpositivism and social
constructionism as multiple epistemological orientations (read, Discourses), each of which are specifically suited to asking and
answering certain questions over others (Deetz, 1996), and each having issues of reliability and validity that follow accordingly.

8.4. How does one acquire an ODA skill set?

For leadership researchers trained in postpositivist methods, the simplest way to begin is with an a priori coding scheme
applied to transcribed interaction data, much as we demonstrated with the quantitative relational control coding scheme that
Fairhurst and colleagues used. While this often produces a larger volume of data relative to surveys, a coding scheme narrows the
field of interest considerably. The payoff for this kind of work is the experience of identifying relationality in the sequential and
temporal forms of social interaction. It is far and away a different visceral experience than static measures of relationship, which
are not able to achieve the kind of micro-behavioral precision accessible with interaction data.
One might then move on to inductively deriving a coding scheme, again based on transcribed interaction data. Without the
binds of an a priori scheme, there is more freedom to let the data “speak” for itself in critical conversations, turning points, or
altered trajectories based on one's theoretical interests to derive a workable coding scheme. We saw this with Wodak and
colleagues' (2011) interest in a durable consensus versus those that failed to hold, and Carroll and Simpson's (2012) framing
movements in on-line conversations.
Third, as we saw with the power games of the second Fairhurst study (i.e., involving LMX) , some kinds of relationship work by
leadership actors will be quite important, but relatively infrequent. This is the hardest form of ODA to apply, but having
experience with interaction data will help. So will reading the ODA literature, which will heighten one's sensitivity to various
language forms (e.g., category work, account-making, politeness forms, assertions, and so on).
Fourth, although it is not necessary to have had experience with interaction data to analyze interview data meaningfully, it
will help tremendously in understanding the sequential flow of narratives told in interviews and kinds of membership
1058 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062

categorization work that is done to study the making and re-making of identities by leadership actors. However, as mentioned in
the fifth example, ODA involves a great deal more than a (grounded theory) thematic analysis of content in one's data (Hardy &
Grant, 2012).

9. Conclusion

Without question, scholars recognize that leadership is a highly contextualized, interactive process, and we have rich theories
and strong conceptual definitions describing leadership as a relational process. The biggest challenge in advancing empirical
understandings of this view of leadership is: How can we study leadership as a relational process? In the current paper we
address this challenge by showing that a discursive perspective and ODA methods are an established and viable alternative or
complement to survey methodology when researchers need to investigate interactional, relational, and contextual processes
associated with leadership. In so doing, we show how ODA addresses the pervasive problem of misalignment between theory and
methods in leadership research, and offer a discursive perspective as a promising avenue for advancing rich new theoretical and
practical understandings of leadership.

References

Alvesson, M., & Deetz, S. A. (1996). Critical theory and postmodernism approaches to organizational studies. In S. R. Clegg, C. Hardy, & W. R. Nord (Eds.), Handbook
of organization studies (pp. 191–217). London: Sage.
Alvesson, M., & Kärreman, D. (2000a). Taking the linguistic turn in organizational research. The Journal of Applied Behavioral Science, 36, 1125–1149.
Alvesson, M., & Kärreman, D. (2000b). Varieties of discourse: On the study of organizations through discourse analysis. Human Relations, 53, 1125–1149.
Alvesson, M., & Kärreman, D. (2011). Decolonizing discourse: Critical reflections on organizational discourse analysis. Human Relations, 64, 1121–1146.
Alvesson, M., & Sveningsson, S. (2012). Un- and re-packing leadership: context, relations, constructions and politics. In M. Uhl-Bien, & S. Ospina (Eds.), Advancing
relational leadership research: A dialogue among perspectives (pp. 203–225). Charlotte, NC: Information Age Publishing.
Argyris, C., & Schön, D. A. (1996). Organizational learning II: Theory, method and practice. Madison: Fairleigh Dickinson University Press. Reading, MA: Addison-Wesley.
Aritz, J., & Walker, R. C. (Eds.). (2012). Discourse perspectives on organizational communication.
Ashcraft, K. L., Kuhn, T., & Cooren, F. (2009). Constitutional amendments: “Materializing” organizational communication. The Academy of Management Annals, 3,
1–64.
Atkinson, J. M., & Heritage, J. (Eds.). (1984). Structures of social action: Studies in conversation analysis. Cambridge, UK: Cambrdige University Press.
Bakeman, R., & Gottman, J. M. (1986). Observing interaction: An introduction to sequential analysis. Cambridge, UK: Cambridge University Press.
Bakhtin, M. (1981). The dialogical imagination. Austin, TX: University of Texas Press.
Balogun, J., & Johnson, G. (2005). From intended strategies to unintended outcomes: The impact of change in recipient sensemaking. Organization Studies, 26,
1573–1601.
Barad, K. (2003). Posthumanist performativity: Toward an understanding of how matter comes to matter. Signs, 28, 801–831.
Barge, J. K. (2004a). Reflexivity and managerial practice. Communication Monographs, 71, 70–96.
Barge, J. K. (2004b). Antenarrative and managerial practice. Communication Studies, 55, 106–127.
Barge, J. K. (2004c). Articulating CMM as a practical theory. Human Systems: The Journal of Systemic Consultation & Management, 15, 13–32.
Barge, J. K. (2007). The practice of systemic leadership. OD Practitioner, 39, 10–14.
Barge, J. K., & Craig, R. T. (2009). Practical theory. In L. R. Frey, & K. N. Cissna (Eds.), Handbook of applied communication (pp. 55–78). Mahwah, NJ: Lawrence
Erlbaum.
Barge, J. K., & Fairhurst, G. T. (2008). Living leadership: A systemic, constructionist approach. Leadership, 4, 227–251.
Barge, J. K., & Little, M. (2002). Dialogical wisdom, communicative practice, and organizational life. Communication Theory, 12, 375–397.
Barge, J. K., & Oliver, C. (2003). Working with appreciation in managerial practice. The Academy of Management Review, 28, 124–142.
Bargiela-Chiappini, F. (2009). The handbook of business discourse. Edinburgh: Edeinburgh University Press.
Bargiela-Chiappini, F. (2011). Discourse(s), social construction and language practices: In conversation with Alvesson and Kärreman. Human Relations, 64,
1177–1191.
Barker, R. A. (2002). On the nature of leadership. Lanham, Maryland: University Press of America.
Bass, B. M., & Avolio, B. (1990). Transformational leadership development: Manual for the multifactor leadership questionnaire. Palo Alto, CA: Consulting Psychologists
Press.
Bateson, G. (1972). Steps to an ecology of the mind. New York: Ballentine.
Bennis, W. G., & Nanus, B. (1985). Leaders: Strategies for taking charge. New York: Harper & Row.
Berger, P., & Luckman, T. L. (1966). The social construction of knowledge: A treatise on the sociology of knowledge. Garden City, NY: Doubleday.
Bitzer, L. F. (1968). The rhetorical situation. Philosophy and Rhetoric, 1, 1–14.
Boden, D. (1990). The world as it happens: Ethnomethodology and conversation analysis. In G. Ritzer (Ed.), Frontiers of social theory: The new synthesis
(pp. 185–213). New York: Columbia University Press.
Boden, D. (1994). The business of talk: Organizations in action. Cambridge, UK: Polity.
Boje, D. (1991). The storytelling organization: A study of story performance in an office supply firm. Administrative Science Quarterly, 36, 106–126.
Boje, D. (2001). Narrative methods for organizational and communication research. London: Sage.
Bradbury, H., & Lichtenstein, B. M. B. (2000). Relationality in organizational research: Exploring The Space Between. Organization Science, 11, 551–564.
Bruner, J. S. (1986). Actual minds, possible worlds. Cambridge, MA: Harvard University Press.
Bruner, J. S. (1990). Acts of meaning. Cambridge, MA: Harvard University Press.
Bryman, A. (1992). Charisma and leadership in organizations. London: Sage.
Bryman, A. (1996). Leadership in organizations. In S. R. Clegg, C. Hardy, & W. R. Nord (Eds.), Handbook of organization studies (pp. 276–292). London: Sage.
Burke, K. (1962). A rhetoric of motives. Berkeley: University of California Press.
Burns, T., & Stalker, G. M. (1961). The management of innovation. London: Tavistock.
Carroll, B., & Simpson, B. (2012). Capturing sociality in the movement between frames: An illustration from leadership development. Human Relations, 65,
1283–1309.
Charmaz, K. (2000). Grounded theory: Objectivist and constructivist methods. In N. K. Denzin, & Y. S. Lincoln (Eds.), Handbook of Qualitative Research
(pp. 509–535). (2nd ed.). Thousand Oaks: Sage.
Charmaz, K. (2006). Constructing grounded theory. London: Sage.
Cheney, G., Christensen, L. T., Conrad, C., & Lair, D. J. (2004). Corporate rhetoric as organizational discourse. In C. H. D. Grant, C. Oswick, & L. Putnam (Eds.), The Sage
handbook of organizational discourse (pp. 70–103). London: Sage.
Chouliaraki, L., & Fairclough, N. (2010). Critical discourse analysis in organizational studies: Towards an integrationist methodology. Journal of Management
Studies, 47, 1213–1218.
 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062 1059

Collinson, D. L. (2005). Dialectics of leadership. Human Relations, 58, 1419–1442.

Collinson, D. L. (2006). Rethinking followership: A post-structuralist analysis of follower identities. The Leadership Quarterly, 17, 179–189.
Conger, J. A. (1989). The charismatic leader. San Francisco: Jossey-Bass.
Cooren, F. (2004). Textual agency: How texts do things in organizational settings. Organization, 11, 373–393.
Cooren, F., & Fairhurst, G. (2004). Speech timing and spacing: The phenomenon of organizational closure. Organization, 11, 797–828.
Courtright, J. A., Fairhurst, G. T., & Rogers, L. E. (1989). Interaction patterns in organic and mechanistic systems. The Academy of Management Journal, 32, 773–802.
Craig, R. (1995). Forward. In W. Leeds-Hurwitz (Ed.), Social approaches to communication (pp. v–ix). New York: Guilford Press.
Craig, R. J., & Amernic, J. H. (2004). Enron discourse: The rhetoric of a resilient capitalism. Critical Perspective on Accounting, 15, 813–851.
Creswell, J. W. (2007). Qualitative inquiry & research design: Choosing among five approaches. Thousand Oaks: Sage.
Cunliffe, A. L. (2004). On becoming a critically reflexive practitioner. Journal of Management Education, 28, 407–426.
Cunliffe, A. L. (2008). Orientations to social constructionism: Relationally-responsive social constructionism and its implications for knowledge and learning.
Management Learning, 39, 123–139.
Cunliffe, A. L. (2009). A very short, fairly interesting and reasonably cheap book about management. Los Angeles: Sage.
Cunliffe, A. L., & Eriksen, M. (2011). Relational leadership. Human Relations, 64, 1425–1450.
Cunliffe, A. L., & Jun, J. S. (2005). The need for reflexivity in public administration. Administration & Society, 37, 225–242.
Czarniawska-Joerges, B. (1997). Narrating the organization: Dramas of institutional identity. Chicago: University of Chicago Press.
Dansereau, F., Yammarino, F. J., & Markham, S. E. (1995). Leadership: The multiple-level approaches. The Leadership Quarterly, 6, 97–109.
Day, D. V. (2000). Leadership development: A review in context. The Leadership Quarterly, 11, 581–613.
Deetz, S. A. (1996). Describing differences in approaches to organization science: Rethinking Burrell and Morgan and their legacy. Organization Science, 7, 190–207.
Denis, J. L., Langley, A., & Sergi, V. (2012). Leadership in the plural. The Academy of Management Annals, 6, 211–283.
Drath, W. (2001). The deep blue sea: Rethinking sources of leadership. San Francisco: Jossey-Bass & Center for Creative Leadership.
Drazin, R., Glynn, M. A., & Kazanjian, R. K. (1999). Multilevel theorizing about creativity in organizations: A sensemaking perspective. The Academy of Management
Review, 24, 286–307.
du Gay, P., Salaman, G., & Rees, B. (1996). The conduct of management and the management of conduct: Contemporary managerial discourse and the constitution
of the “competent manager”. Journal of Management Studies, 33, 263–282.
Eisenberg, E. (2007). Strategic ambiguities: Essays on communication, organization, and identity. Thousand Oaks: Sage.
Fairclough, N. (1995). Critical discourse analysis: The critical study of language. London: Longman.
Fairclough, N. (2006). Language and globalization. New York: Routledge.
Fairclough, N., & Wodak, R. (1997). Critical discourse analysis. In T. A. van Dijk (Ed.), Discourse as social interaction (pp. 258–284). London: Sage.
Fairhurst, G. T. (1993). The leader-member exchange patterns of women leaders in industry: A discourse analysis. Communication Monographs, 60, 321–351.
Fairhurst, G. (2001). Dualisms in leadership research. In F. M. Jablin, & L. L. Putnam (Eds.), The new handbook of organizational communication (pp. 379–439).
Newbury Park, CA: Sage.
Fairhurst, G. T. (2004a). Textuality and agency in interaction analysis. Organization, 11, 335–354.
Fairhurst, G. T. (2004b). Organizational relational control research: Problems and possibilities. In L. E. Rogers, & V.E. (Eds.), Relational communication: An
interactional perspective to the study of process and form (pp. 197–215). Mahwah, NJ: Lawrence Erlbaum.
Fairhurst, G. T. (2007a). Discursive leadership: In conversation with leadership psychology. Thousand Oaks: Sage.
Fairhurst, G. T. (2007b). Liberating leadership in Corporation After Mr. Sam: A response. In F. Cooren (Ed.), Interacting and organizing: Analyses of a board meeting
(pp. 53–71). Mahwah, NJ: Lawrence Erlbaum.
Fairhurst, G. T. (2009). Considering context in discursive leadership research. Human Relations, 62, 1607–1633.
Fairhurst, G. T. (2011). The power of framing: Creating the language of leadership. San Francisco: Jossey-Bass.
Fairhurst, G. T. & Connaughton, S. L. (in press). Leadership. In L. L. Putnam & D. Mumby (Eds.). The Sage handbook of organizational communication. Thousand Oaks:
Sage.
Fairhurst, G. T., & Cooren, F. (2004). Organizational language in use: Interaction analysis, conversation analysis, and speech act schematics. In D. Grant, C. Hardy, C.
Oswick, N. Phillips, & L. Putnam (Eds.), The Sage handbook of organizational discourse (pp. 131–152). London: Sage.
Fairhurst, G. T., & Cooren, F. (2009). Leadership as the hybrid production of presence(s). Leadership, 5, 1–22.
Fairhurst, G. T., & Grant, D. (2010). The social construction of leadership: A sailing guide. Management Communication Quarterly, 24, 171–210.
Fairhurst, G. T., Green, S. G., & Courtright, J. A. (1995). Inertial forces and the implementation of a socio-technical systems approach: A communication study.
Organization Science, 6, 168–185.
Fairhurst, G., & Hamlett, S. R. (2003). The narrative basis of leader–member exchange. In G. B. Graen (Ed.), Dealing with diversity (pp. 117–144). Greenwich, CT:
Information Age Publishing.
Fairhurst, G. T. & Putnam, L. L. (in press). Organizational discourse analysis. In L. L. Putnam & D. Mumby (Eds.). The Sage handbook of organizational communication.
Thousand Oaks: Sage.
Fairhurst, G. T., Rogers, L. E., & Sarr, R. (1987). Manager-subordinate control patterns and judgments about the relationship. In M. McLaughlin (Ed.),
Communication yearbook 10 (pp. 395–415). Beverly Hills, CA: Sage.
Fisher, W. R. (1984). Narration as a human communication paradigm: The case of public moral argument. Communication Monographs, 51, 1–22.
Fleishman, E. A., Mumford, M. D., Zaccaro, S. J., Levin, K. Y., Korotkin, A. L., & Hein, M. B. (1991). Taxonomic efforts in the description of leader behavior: A synthesis
and functional interpretation. The Leadership Quarterly, 2, 245–287.
Fletcher, J. K. (2010). Leadership as relational practice. In K. Bunker, K. Kram, & D. T. Hall (Eds.), Extraordinary leadership: Addressing the gaps in senior executive
development. San Francisco: Jossey-Bass.
Foldy, E. G., Goldman, L., & Ospina, S. (2008). Sensegiving and the role of cognitive shifts in the work of leadership. The Leadership Quarterly, 19, 514–529.
Foucault, M. (1972). The archeology of knowledge and the discourse on language. London, UK: Tavistock Publications.
Foucault, M. (1979). On governmentality. I & C, 6, 5–22.
Foucault, M. (1983). The subject and power. In H. L. Dreyfus, & P. Rabinow (Eds.), Michel Foucault: Beyond structuralism and hermeneutics (pp. 208–226). Chicago:
Chicago University Press.
Foucault, M. (1990). The history of sexuality, Vol. 1, New York: Vintage/Random House.
Foucault, M. (1995). Discipline and punish. New York: Vintage/Random House.
Gabriel, Y. (2004). Narratives, stories, and texts. In C. H. D. Grant, C. Oswick, N. Phillips, & L. Putnam (Eds.), Sage handbook of organizational discourse (pp. 61–77).
London: Sage.
Gallie, W. B. (1956). Essentially contested concepts. Proceedings of the Aristotelian Society, 56(1955–1956), 167–198 (New series).
Garfinkel, H. (1967). Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice Hall.
Garfinkel, H. (2002). Ethnomethodology's program: Working out Durkheim's aphorism. Lanham, MD: Rowman & Littlefield Publishers.
Giddens, A. (1979). Central problems in social theory. Berkeley: University of California Press.
Giddens, A. (1984). The constitution of society. Berkeley: University of California Press.
Gioia, D. A., & Chittipeddi, K. (1991). Sensemaking and sensegiving in strategic change initiation. Strategic Management Journal, 12, 433–448.
Goffman, E. (1959). The presentation of self in everyday life. New York: Doubleday.
Gottman, J. M. (1982). Temporal form: Toward a new languge for describing relationships. Journal of Marriage and Family, 44, 943–962.
Graen, G. B., & Scandura, T. A. (1987). Toward a psychology of dyadic organizing. In B. Staw, & L. L. Cummings (Eds.), Research in organizational behavior., 9. (pp. 175–208)
Greenwich, CT: JAI Press.
Graen, G. B., & Uhl-Bien, M. (1995). Relationship-based approach to leadership: Development of a leader-member exchange (LMX) theory of leadership over
25 years–Applying a multi-level multi-domain perspective. The Leadership Quarterly, 6, 219–247.
1060 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062

Gramsci, A. (1971). Selections from the prison notebooks. New York: International Publishers.
Grant, D., Hardy, C., Oswick, C., & Putnam, L. (2004). The Sage handbook of organizational discourse. London: Sage.
Grant, D., Hardy, C., & Putnam, L. L. (2011). Organizational discourse studies (3 volume set). Sage Major Works Series. Los Angeles, CA: Sage.
Grant, D., & Iedema, R. (2005). Discourse analysis and the study of organizations. Text: An Interdisciplinary Journal for the Study of Discourse, 25, 37–66.
Grant, D., Iedema, R., & Oswick, C. (2009). Discourse and critical management studies. In M. Alvesson, T. Bridgman, & H. Willmott (Eds.), The Oxford handbook of
critical management studies (pp. 213–231). London: Oxford University Press.
Grant, D., Keenoy, T., & Oswick, C. (Eds.). (1998). Discourse + Organization. London: Sage.
Grant, D., Putnam, L. L., & Hardy, C. (2011). History, key challenges, and contributions of organizational discourse studies. In D. Grant, C. Hardy, & L. L. Putnam
(Eds.), Organizational discourse studies (pp. xvii–xlii). London: Sage.
Greimas, A. J. (1987). On meaning: Selected writings in semiotic theory. London: Frances Pinter.
Grint, K. (2000). The arts of leadership. Oxford, UK: Oxford University Press.
Grint, K. (2005). Problems, problems, problems: The social construction of “leadership”. Human Relations, 58, 1467–1494.
Grint, K. (2010). Leadership: A very short introduction. Oxford: Oxford University Press.
Gronn, P. (1983). Talk as the work: The accomplishment of school administration. Administrative Science Quarterly, 28, 1–21.
Gronn, P. (1993). Psychobiography on the couch: Character, biography, and the comparative study of leaders. The Journal of Applied Behavioral Science, 29,
343–358.
Gronn, P. (2000). Distributed properties: A new architecture for leadership. Educational Management and Administration, 28, 317–338.
Gronn, P. (2002). Distributed leadership as a unit of analysis. The Leadership Quarterly, 13, 423–451.
Habermas, J. (1984). The theory of communicative action, Vol. 1: Reason and the rationalization of society. In (Eds.), Boston: Beacon Press.
Hardy, C., & Grant, D. (2012). Readers betware: Provocation, problematization and…problems. Human Relations, 65, 547–566.
Haslam, S. A., & Platow, M. J. (2001). Your wish is our command: The role of shared social identity in translating a leader's vision into followers' action. In M. A.
Hogg, & D. J. Terry (Eds.), Social identity processes in organizational contexts (pp. 213–228). Philadelphia, PA: Psychology Press.
Heritage, J. (1997). Conversation analysis and institutional talk. In D. Silverman (Ed.), Qualitative research: Theory, method and practice (pp. 161–182). London:
Sage.
Hinde, R. (1979). Toward understanding relationships. New York: Academic Press.
Hogg, M. (2001). Social identification, group prototypicality, and emergent leadership. In M. Hogg, & D. Terry (Eds.), Social identity processes in organizational
contexts (pp. 197–228). Philadelphia, PA: Psychology Press.
Holman, D., & Thorpe, R. (2003). Management and language: The manager as a practical author. London: Sage.
Holstein, J. A., & Gubrium, J. F. (2000). The self we live by: Narrative identity in a postmodern world. New York: Oxford University Press.
Hosking, D. M. (1988). Organizing, leadership and skilful process. Journal of Management Studies, 25, 147–166.
Hosking, D. M., & Morley, I. E. (1988). The skills of leadership. In J. G. Hunt, R. Baglia, & C. Schriesheim (Eds.), Emerging leadership vistas. Arlington Heights:
Lexington, Mass.
House, R. J., & Aditya, R. N. (1997). The social scientific study of leadership: Quo vadis? Journal of Management, 23, 409–473.
Hunt, J. G., & Dodge, G. E. (2000). Leadership deja vu all over again. The Leadership Quarterly Review of Literature, 11, 435–458.
IBM Global CEO Study (2010). Capitalizing on complexity. www-935.ibm.com/services/us/ceo/ceostudy2010/index.html
Iedema, R. (2011). Discourse studies in the 21st century: A response to Mats Alvesson and Dan Kärreman's ‘decolonializing discourse’. Human Relations, 64,
1163–1176.
Jackson, S. (1986). Building a case for claims about discourse structure. In D. G. Ellis, & W. A. Donohue (Eds.), Contemporary issues in language and discourse
processes (pp. 129–147). Hillsdale, NJ: Lawrence Erlbaum.
Jayyusi, L. (1984). Categorization and the moral order. Boston: Routledge & Kegan Paul.
Jian, G. (2007). Unpacking unintended consequences in planned organizational change: A process model. Management Communication Quarterly, 21, 5–28.
Johnson, J. M. (2002). In-depth interviewing. In J. F. Gubrium, & J. A. Holstein (Eds.), Handbook of interview research: Context and method (pp. 103–119). Thousand
Oaks: Sage.
Katz, D., & Kahn, D. L. (1978). The social psychology of organizations (2nd ed.). New York: John Wiley.
Kelly, S. (2008). Leadership: A categorical mistake? Human Relations, 61, 763–782.
Kelly, S., White, M. I., Martin, D., & Rouncefield, M. (2006). Leadership refrains: Patterns of leadership. Leadership, 2, 181–201.
Knights, D., & Morgan, G. (1991). Corporate strategy, organizations, and subjectivity: A critique. Organization Studies, 12, 251–273.
Komaki, J. L. (1998). Leadership from an operant perspective. London: Routledge.
Komaki, J. L., Zlotnick, S., & Jensen, M. (1986). Developing an operant-based taxonomy and observational index of supervisory behavior. The Journal of Applied
Psychology, 71, 260–269.
Kreiner, G. E., Hollensbe, E. C., & Sheep, M. L. (2006a). Where is the ‘me’ among the ‘we’? Identity work and the search for optimal balance. The Academy of
Management Journal, 49, 1031–1057.
Kreiner, G. E., Hollensbe, E. C., & Sheep, M. L. (2006b). On the edge of identity: Boundary dynamics at the interface of individual and organizational identities.
Human Relations, 59, 1315–1341.
Laine, P. -M., & Vaara, E. (2007). Struggling over subjectivity: A discursive analysis of strategic development in an engineering group. Human Relations, 60, 29–58.
Law, J. (1994). Organization, narrative and strategy. In J. Hassard, & M. Parker (Eds.), Towards a new theory of organizations (pp. 248–266). London: Routledge.
Leonardi, P. M., & Jackson, M. H. (2004). Technological determinism and discursive closure in organizational mergers. Journal of Organizational Change
Management, 17, 615–631.
Liden, R. C., & Antonakis, J. (2009). Considering context in psychological leadership research. Human Relations, 62, 1587–1606.
Lilley, S. (2001). The language of strategy. In R. Westwood, & S. Linstead (Eds.), The language of the organization (pp. 66–88). London: Sage.
Marion, R., & Uhl-Bien, M. (2001). Leadership in complex organizations. The Leadership Quarterly, 12, 389–418.
Marshak, R. J., & Grant, D. (2008). Organizational discourse and new organization development practices. British Journal of Management, 19, S7–S19.
McDermott, R. P., & Roth, D. R. (1978). The social organization of behavior: Interactional approaches. Annual Review of Anthropology, 7, 321–345.
Meindl, J. R. (1995). The romance of leadership as a follower-centric theory: A social constructionist approach. The Leadership Quarterly, 6, 329–341.
Meindl, J. R., Ehrlich, S. B., & Dukerich, J. M. (1985). The romance of leadership. Administrative Science Quarterly, 30, 78–102.
Molotch, H. L., & Boden, D. (1985). Talking social structure: Discourse, domination and the Watergate Hearings. American Sociological Review, 50, 273–288.
Mumby, D. K. (2004). Discourse, power and ideology: Unpacking the critical approach. In C. H. D. Grant, C. Oswick, & L. Putnam (Eds.), The Sage handbook of
organizational discourse (pp. 237–258). London: Sage.
Mumby, D. (2011). What's cooking in organizational discourse studies? A response to Alvesson and Kärreman. Human Relations, 64, 1147–1161.
Mumby, D. K., & Clair, R. P. (1997). Organizational discourse. In T. A. v. Dijk (Ed.), Discourse as social interaction, Vol. 2. (pp. 181–205)London: Sage.
Ochs, E. (1997). Narrative. In T. A. v. Dijk (Ed.), Discourse as structure and process, Vol. 1. (pp. 185–207)London: Sage.
Olekalns, M., Smith, P. L., & Walsh, T. (1996). The process of negotiating: Strategies, timing and outcomes. Organizational Behavior and Human Decision Processes, 68, 68–77.
Osborn, R. N., Hunt, J. G., & Jauch, L. R. (2002). Toward a contextual theory of leadership. The Leadership Quarterly, 13, 797–837.
Ospina, S., & Foldy, E. G. (2010). Building bridges from the margins: The work of leadership in social change organizations. The Leadership Quarterly, 21, 292–307.
Ospina, S., & Hittleman, M. (2011). Thinking sociologicallly about leadership. In M. Harvey, & R. Riggio (Eds.), Leadership studies: The dialogue of disciplines
(pp. 89–100). Cheltenham, UK: Edward Elgar.
Ospina, S., & Sorensen, G. (2006). A constructionist lens on leadership: Charting new territory. In G. Goethals, & G. Sorenson (Eds.), In quest of a general theory of
leadership (pp. 188–204). Cheltenham, UK: Edward Elgar.
Ospina, S., & Uhl-Bien, M. (2012). Mapping the terrain: Convergence and divergence around relational leadership. In M. Uhl-Bien, & S. Ospina (Eds.), Advancing
relational leadership research: A dialogue among perspectives. Charlotte, NC: Information Age Publishers.
 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062 1061

Parry, K. W. (1998). Grounded theory and social process: A new direction for leadership research. The Leadership Quarterly, 9, 85–105.
Parry, K. W., & Hansen, H. (2007). The organizational story as leadership. Leadership, 3, 281–300.
Patton, M. Q. (2002). Qualitative research & evaluation methods (3rd edition). Thousand Oaks: Sage.
Perelman, C., & Olbrechts-Tyteca, L. (1969). The new rhetoric: A treatise on argumentation. Notre Dame: University of Notre Dame Press.
Pfeffer, J. (1981). Management as symbolic action: The creation and maintenance of organizational paradigms. In L. L. Cummings, & B. M. Staw (Eds.), Research in
organizational behavior, 3. (pp. 1–52)Greenwich, CT: JAI Press.
Phillips, N., & Oswick, C. (2012). Organizational discourse: Domains, debates, and directions. The Academy of Management Annals, 1–47.
Pomerantz, A. (1990). Conversation analytic claims. Communication Monographs, 57, 231–235.
Pomerantz, A., & Fehr, B. J. (1997). Conversation analysis: An approach to the study of social action as sense making practices. In T. A. v. Dijk (Ed.), Discourse as
social interaction, Vol. 2. (pp. 64–91)London: Sage.
Pondy, L. R. (1978). Leadership is a language game. In J. M. W. McCall, & M. M. Lombardo (Eds.), Leadership: Where else can we go? (pp. 88–99). Durham, NC: Duke
University Press.
Potter, J. (1997). Discourse analysis as a way of analyzing naturally occurring talk. In D. Silverman (Ed.), Qualitative research: Theory, method and practice
(pp. 144–160). London: Sage.
Potter, J., & Wetherell, M. (1987). Discourse and social psychology. London: Sage.
Psathas, G. (1995). Conversation analysis: The study of talk-in-interaction. London: Sage.
Putnam, L. L., & Fairhurst, G. T. (2001). Discourse analysis in organizations. In F. M. Jablin, & L. L. Putnam (Eds.), The new handbook of organizational communication
(pp. 78–136). Thousand Oaks, CA: Sage.
Pye, A. (2005). Leadership and organizing: Sensemaking in action. Leadership, 1, 31–50.
Reed, M. (2004). Getting real about organizational discourse. In C. H. D. Grant, C. Oswick, N. Phillips, & L. Putnam (Eds.), The Sage handbook of organizational
discourse (pp. 413–420). London: Sage.
Reisigl, M., & Wodak, R. (2009). The discourse-historical approach. In R. Wodak, & M. Meyer (Eds.), Methods of CDA 2nd (pp. 87–121). London: Sage.
Robinson, V. M. J. (2001). Embedding leadership in task performance. In K. Wong, & C. W. Evers (Eds.), Leadership for quality schooling (pp. 90–102). London:
Routledge/Falmer.
Rogers, L. E., & Escudero, V. (2004a). Relational communication: An interactional perspective to the study of process and form. Mahwah, NJ: Erlbaum.
Rogers, L. E., & Escudero, V. (2004b). Theoretical foundations. In L. E. Rogers, & V. Escudero (Eds.), Relational communication: An interactional perspective to the
study of process and form (pp. 222–239). Mahwah, NJ: Lawrence Erlbaum.
Rogers, L. E., & Farace, R. V. (1975). Relational communication analysis: New measurement procedures. Human Communication Research, 1, 222–239.
Rogers, L. E., Millar, F. E., & Bavelas, J. B. (1985). Methods for analyzing marital conflict discourse. Family Process, 24, 175–187.
Rorty, R. (Ed.). (1967). The linguistic turn: Recent essays in philosophical method. Chicago: University of Chicago Press.
Rost, J. C. (1991). Leadership for the twenty-first century. New York: Praeger.
Rouleau, L. (2005). Micro-practices of strategic sensemaking and sensegiving: How middle managers interpret and sell change every day. Journal of Management
Studies, 42, 1413–1441.
Sacks, H. (1992). Lectures on conversation, vols. 1 and 2, Oxford: Blackwell.
Sacks, H., Schegloff, E. A., & Jefferson, G. (1974). A simplest systematics for the organization of turn-taking for conversation. Language, 50, 696–735.
Schön, D. A. (1983). The reflective practitioner: How professionals think in action. New York: Basic Books.
Schriesheim, C. A., Castro, S. L., Zhou, X. T., & Yammarino, F. J. (2001). The folly of theorizing “A” while testing “B”: A selective level-of-analysis review of the field
and a detailed Leader–Member-Exchange illustration. The Leadership Quarterly, 12, 515–551.
Seers, A., & Chopin, S. M. (2012). The social production of leadership: From supervisor-subordinate linkages to relational organizing. In M. Uhl-Bien, & S. Opsina
(Eds.), Advancing relational leadership research: A dialogue among perspectives (pp. 43–81). Charlotte, NC: Information Age Publishing.
Shamir, B. (2007). From passive recipients to active co-producers: Followers' roles in the leadership process. Follower –Centered Perspectives on Leadership: A
Tribute to the Memory of James R. Meindl. In R. P. B. Shamir, M. Bligh, & M. Uhl-Bien (Eds.), Follower-centered perspectives on leadership: A tribute to the
memory of James R. Meindl (pp. ix–xxxix). Greenwich, CT: Information Age Publishing.
Shamir, B. (2012). Leadership research or post-leadership research? Advancing leadership theory versus throwing the baby out with the bath water. In M.
Uhl-Bien, & S. Ospina (Eds.), Advancing relational leadership research: A dialogue among perspectives (pp. 477–500). Charlotte, NC: Information Age Publishing.
Sheep, M. L. (2006a). When categories collide: A discursive psychology approach to the elasticity of multiple identities. Cincinnati: Department of Management,
University of Cincinnati.
Sheep, M. L. (2006). When categories collide: A discursive psychology approach to the elasticity of multiple identities. © ProQuest Dissertations & Theses.
Cincinnati, Ohio, University of Cincinnati: 202 pages.
Sheep, M. L., Hollensbe, E. C., & Kreiner, G. E. (2009). “Tearing the fabric” or “weaving the tapestry”? A discursive psychology approach to the culture-identity
interface. Paper presented at the annual conference of the Academy of Management.
Shotter, J. (2008). Conversational realities revisited: Life, language, body and world. Chagrin Falls, OH: Taos Institute Publications (First published by London, Sage,
1993).
Sinclair, A. (2005). Body possibilities in leadership. Leadership, 1, 387–406.
Smircich, L., & Morgan, G. (1982). Leadership: The management of meaning. The Journal of Applied Behavioral Science, 18, 257–273.
Sparrowe, R. (2005). Authentic leadership and the narrative self. The Leadership Quarterly, 16, 419–439.
Stohl, C., & Cheney, G. (2001). Participatory processes/paradoxical practices. Management Communication Quarterly, 14, 349–407.
Strauss, A. L., & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory. Thousand Oaks, CA: Sage.
Suchman, L. A. (2007). Human-machine configurations: Plans and situated actions (2nd edition). Cambridge: Cambridge University Press.
Taylor, S. (2001). Evaluating and applying discourse analytic research. In M. Wetherell, S. Taylor, & S. J. Yates (Eds.), Discourse as data: A guide to analysis. London:
Sage in association with Open University.
Taylor, J. R., & Van Every, E. (2000). The emergent organization: Communication at its site and surface. Mahwah, NJ: Lawrence Erlbaum.
Tracy, K. (Ed.). (1991). Understanding face-to-face interaction: Issues linking goals and discourse. Hillsdale, NJ: Lawrence Erlbaum.
Tracy, K. (1995). Action-implicative discourse analysis. Journal of Language and Social Psychology, 14, 195–215.
Tsoukas, H., & Hatch, M. J. (2001). Complex thinking, complex practice: The case for a narrative approach to organizational complexity. Human Relations, 53,
979–1013.
Uhl-Bien, M., & Ospina, S. (Eds.). (2012). Advancing relational leadership research: A dialogue among perspectives. Charlotte, NC: Information Age.
van Dijk, T. A. (1985). Handbook of discourse analysis. New York: Academic Press.
van Dijk, T. A. (1997). The study of discourse. In T. A. van Dijk (Ed.), Discourse as structure and process, Vol. 1. (pp. 1–34)London: Sage.
Van Maanen, J. (1998). Tales of the field: On writing ethnography. Chicago: University of Chicago Press.
Watson, R. (1997). Some general reflections on ‘categorization’ and ‘sequence’ in the analysis of conversation. In S. Hester, & P. Eglin (Eds.), Culture in action:
Studies in membership categorization analysis (pp. 49–76). Washington, D.C.: International Institute for Ethnomethodology and Conversation Analysis &
University Press of America.
Watson, T. (2001). Beyond managism: Negotiated narratives and critical management education in practice. British Journal of Management, 12, 385–396.
Watson, T. J., & Bargiela-Chiappini, F. (1998). Managerial sensemaking and occupational identities in Britan and Italy: The role of management magazines in the
process of discursive construction. Journal of Management Studies, 35, 285–301.
Watzlawick, P., Beavin, J. H., & Jackson, D. D. (1967). Pragmatics of human communication. New York: Norton.
Weick, K. (1979). The social psychology of organizing (2nd edition). Reading, MA: Addison-Wesley.
Weick, K., Sutcliffe, K. M., & Obstfeld, D. (1999). Organizing for high reliability: Processes of collective mindfulness. In B. M. Staw, & L. L. Cummings (Eds.), Research
in organizational behavior., 21. (pp. 81–123)Stamford, CT: JAI Press.
1062 G.T. Fairhurst, M. Uhl-Bien / The Leadership Quarterly 23 (2012) 1043–1062

Weick, K. E., Sutcliffe, K. M., & Obstfeld, D. (2005). Organizing and the process of sensemaking. Organization Science, 16, 409–421.
Weingart, L. R., Bennett, R. J., & Brett, J. M. (1993). The impact of consideration of issues and motivational orientation on group negotiation process and outcome.
The Journal of Applied Psychology, 78, 504–517.
Westwood, R., & Linstead, S. (Eds.). (2001). The language of organization. London: Sage.
Wetherell, M. (1998). Positioning and interpretative repertoires: Conversation analysis and post-structuralism in dialogue. Discourse & Society, 9, 387–412.
Wetherell, M. (2001). Debates in discourse research. In M. Wetherell, S. Taylor, & S. J. Yates (Eds.), Discourse theory and practice: A reader (pp. 380–398). London:
Sage.
Wittgenstein, L. (1953). Philosophical investigations. Oxford, UK: Blackwell.
Wodak, R. (2009). The discourse of politics in action: ‘Politics as usual’. Basingstoke: Palgrave Macmillan.
Wodak, R., Kwon, W., & Clarke, I. (2011). ‘Getting people on board’: Discursive leadership for consensus builidng in team meetings. Discourse & Society, 22,
592–616.
Wood, L. A., & Kroger, R. O. (2000). Doing discourse analysis: Methods for studying talk and text. Thousand Oaks: Sage.
Yukl, G. (2013). Leadership in organizations (8th edition). Boston: Pearson.
Zaleznik, A. (May-June). Managers and leaders: Are they different? Harvard Business Review, 55, 67–78.