The negotiation of sharing leadership in the context of professional hierarchy: Interactions on interprofessional teams

Abstract

While there is growing recognition of leadership as a collective phenomenon, the question of how leadership is shared in the context of hierarchical asymmetry has been neglected in the collective leadership literature. Our article addresses this gap by examining how sharing leadership is negotiated in team interactions that are steeped in asymmetry deriving from the professional hierarchy. Adopting a leadership-in-interaction approach, we draw on fine-grained analysis of observed interactions on interprofessional teams from two health care organizations to compare the discursive strategies used by professionals in a superior hierarchical position to the ones used by those in inferior positions to share leadership. These strategies are organized into a matrix of interactional moves that resist or enact the professional hierarchy. Empirical vignettes are provided to demonstrate how sharing leadership and hierarchical leadership can be co-present and even intertwined in an interaction. We show that leadership is shared (or not) as a result of how the professional hierarchy gets negotiated in interactions. More specifically, we conclude that the sharing of leadership in this context tends to occur prior to decision making, especially around problem formulation, if the interactional climate allows. Furthermore, it requires concrete effort: Those in superior positions of influence mindfully relax the hierarchy whereas those in inferior positions create moments of sharing leadership through resistance and struggle.

Keywords

Asymmetry interprofessional teams leadership-in-interaction professional hierarchy sharing leadership

Interprofessional teams are often regarded as a salve in today’s complex and knowledge-intensive world. Many organizations espouse such teams, whose members hail from different professions, with the express hope to connect specialized workers to better resolve their collective problems and enhance organizational efficiency (D’Amour et al., 2005; Pinchot and Pinchot, 1993). This change has important implications for leadership. While teams have been regarded as being guided by one formal leader (e.g. Zaccaro et al., 2001), there is now growing recognition that leadership can be distributed, shared, and enacted collectively (Contractor et al., 2012; Denis et al., 2012; Yammarino et al., 2012), as different professionals exert influence and lead others to achieve collective goals (Pearce and Conger, 2003). Such a collective approach to sharing leadership emphasizes a decentralized, participative, and egalitarian vision (Quick, 2017).

However, the system of professions is traditionally very hierarchical (Abbott, 1988; Freidson, 2001). Thus, bringing together different professionals who belong to such a system can invite significant power issues (Brown et al., 2010). Therefore, we question the understanding that a collective approach to leadership promotes a vision of egalitarian, mutual, and non-hierarchical relations (Quick, 2017). How then can collective leadership be accomplished in everyday work practices (Boden, 1994) in such a hierarchical context? Our article examines how sharing leadership is negotiated in team interactions that are steeped in asymmetry deriving from the professional hierarchy. While some authors (e.g. Crevani et al., 2007; Pearce, 2004; Thylefors and Persson, 2014; Van De Mieroop et al., 2020) recognize the complementarity of collective leadership and vertical leaders, that is those who formally hold superior positions in the organizational hierarchy, we still know little about the interactional work required from actors to enact collective leadership in the context of hierarchical asymmetry between actors (Denis et al., 2012; Sveiby, 2011). In fact, Holm and Fairhurst (2018) recently called this a “fault line in the quest to illuminate the nature of leadership” (694). Thus, we ask: How do vertical leaders and other professionals interactionally negotiate and accomplish the sharing of leadership? Sharing happens in interactions, and thus taking an interactional approach, founded on observations, to empirically explore this question will give us a fuller and more nuanced understanding of what actually happens when leadership is shared, which can remain obscured in interview data.

We take inspiration from a discursive conception of leadership that defines it as influence through meaning management to advance a task or goal (Clifton, 2012; Fairhurst, 2007). This social constructivist view sees leadership as a co-constructed accomplishment that is relational and emergent in interactions wherein “locally produced understandings of reality are enacted through talk” (Clifton, 2006: 209). Thus, members’ (inter)actions are the appropriate locus of study (Fairhurst and Grant, 2010). More specifically, we align with scholars of leadership-in-interaction (e.g. Larsson, 2017) who examine how leadership is collectively accomplished as an ongoing process in the naturally occurring interactions of everyday organizational practice (Boden, 1994). As Larsson and Lundholm (2010, 2013) and Meschitti (2019) point out, there are few practice studies of leadership that rely on observations of mundane activities, which limits our understanding of how leadership is actually accomplished. Therefore, we rely on fine-grained analysis of observed interactions on interprofessional teams to compare the discursive strategies used by professionals in a superior hierarchical position to the discursive strategies used by those in inferior positions in order to share leadership. Thus, we show that leadership is shared (or not) as a result of how the professional hierarchy gets negotiated in interactions, especially with regard to the meaning of the situation being discussed and the situated interactional positions of the team members. We conclude that the sharing of leadership in this context tends to occur prior to decision making, especially around problem formulation, if the interactional climate allows. Furthermore, it requires concrete effort: Those in superior positions of influence mindfully relax the hierarchy to share leadership, whereas those in inferior positions create moments of sharing leadership through resistance and struggle.

We begin by defining the collective approach to leadership and reviewing scholarship on asymmetric relations in collaborative contexts, and then explain why interprofessional collaboration in health care is an ideal case for examining sharing leadership in the context of professional asymmetry.

A collective approach to leadership

The literature on leadership in general is effervescent. For years, scholars have endeavored to understand, name, and qualify this phenomenon that can emerge within groups and organizations and mobilize members to accomplish specific objectives. Leadership has been referred to in a variety of ways, such as democratic (Blake and Mouton, 1985), transformational (Bass, 2010), and authentic (Avolio and Gardner, 2005) to name just a few. Each of these conceptions distinguishes certain qualities of leaders, and they all agree on an underlying traditional vision that views leadership as being held by heroic individuals (Clifton, 2014), who distinguish themselves from their followers by distinctive traits and behaviors and by a superior position in the hierarchy.

Over the past decade, however, a new approach—collective leadership (Pearce and Conger, 2003; Pearce and Manz, 2005)—has taken root (Avolio et al., 2009). It contrasts sharply and fundamentally from its predecessors by suggesting we consider leadership not as the consequence of an individual’s qualities, but rather as a joint accomplishment (Raelin, 2016). It sees leadership as a relational process, a phenomenon that is shared or distributed across different hierarchical levels, contingent on social interactions and networks of influence (Fletcher and Kaufer, 2003).

This new proposition goes by plural nomenclatures, including team leadership (Day et al., 2004), shared leadership (Pearce, 2004), and distributed leadership (Gronn, 2002), each offering a particular nuance to the collective approach to leadership. Denis et al. (2012) organize this literature by identifying four forms of collective leadership: producing, spreading, pooling, and sharing leadership. First, producing leadership is constructed in an emergent fashion, over time and evolving social interactions (Crevani et al., 2007; Quick, 2017). Second, spreading leadership refers to individuals, or sometimes even to organizations, who take turns playing the role of leader (Davis and Eisenhardt, 2011). Third, pooling leadership represents structured coalitions, generally a dyad or triad of individuals who exert a common influence on a group (Currie and Lockett, 2011). Finally, sharing leadership corresponds primarily to work teams where decision making is shared among members whose actions alternate and who mutually influence one another (Pearce and Manz, 2005). While they evoke different collectives and different processes, all four forms insist on a decentralized, participative, and relatively egalitarian vision of collective leadership. Hence, organizational contexts characterized by rigid structures, such as a hierarchy, can seriously hinder collective leadership (Friedrich et al., 2011; McHugh and Bennett, 1999).

The particular conception of collective leadership on which we focus in this article has ties with sharing leadership, the mutual influence that can take place on work teams, particularly on interprofessional teams. We adopt Denis et al.’s (2012) insistence on the process of sharing leadership, as compared to shared leadership (Pearce, 2004) as a product, because this approach is commensurate with a leadership-in-interaction approach (Larsson, 2017). To this end, in the context of interprofessional teams, we designate “sharing leadership” as interaction episodes in which individuals mutually seek to influence one another by actively engaging in joint making meaning by sharing ideas, points of view, or knowledge, with the common objective of understanding a problem or a task that must be resolved (Van De Mieroop et al., 2020). However, as we shall see, sharing leadership in the interprofessional context also requires that individuals attend to the professional hierarchy in various ways during team interactions.

Sharing leadership, interprofessional teams, and asymmetry

Interprofessional teams are an ideal instance for exploring sharing leadership in the context of professional hierarchy. On the one hand, the goal for such teams is collaboration, cooperation, knowledge sharing, and expertise integration (D’Amour et al., 2005; Fiorelli, 1988; McCallin, 2003). The hope is that responsibilities are shared, that all members contribute freely to problem-solving (Jones and Jones, 2011), and that success stems from a joint rather than an individual performance. Thus, interprofessional teams are essentially driven by the principles of sharing leadership. Hiller et al. (2006) point out that members of such teams can share leadership processes, such as planning and organizing, problem-solving, support and consideration, and development and mentoring, thus contributing to the collective tasks as well as to the team climate (Contractor et al., 2012). Similarly, Schnurr (2018) indicates that members can enact relational or transactional behaviors: the former includes efforts that influence the climate and social relations whereas the latter represents efforts that are related to tasks.

On the other hand, sharing leadership is difficult to accomplish on interprofessional teams, in particular because of the hierarchy inscribed in the very genesis of the professions (Abbott, 1988). Professional hierarchy leading to asymmetric relations can adversely affect the mutual influence and synergy of a team (Dwyer, 1991; Fiorelli, 1988). Indeed, interprofessional team interactions are imbued with pressures exerted by the professional institutions through regulatory, normative, and cultural-cognitive forces (Finn et al., 2010). Barbour (2010) argues these forces constitute the “institutional moorings of talk” (450–451), through which the institutions themselves are reproduced and which afford established ways of communicating and managing disputes. We argue that one such mooring is the professional hierarchy.

This is particularly observable in health care, which is inhabited by both collaboration and hierarchy. Indeed, driven by spiralling costs spurred by growing rates of chronic disease and comorbidities, as well as by challenges to access (Bourgeault and Mulvale, 2006; Canadian Health Services Research Foundation, 2012) and the increasing fragmentation of expertise (D’Amour and Oandasan, 2005), many health care organizations adopt interprofessional and team-based approaches. However, their accomplishment at the local interactional level is considerably challenged by the entrenched professional hierarchy that characterizes health care organizations (Finn et al., 2010; Lammers et al., 2003). In other words, the institutional moorings of talk often serve to reproduce in team interactions the asymmetry inherent to professional hierarchy (Cott, 1997; Finn, 2008; Lingard et al., 2012; Long et al., 2006).

Such asymmetry can manifest interactionally through differences in politeness based on hierarchical status, where lower status members exert more effort to save the face of higher-ranking members, while the inverse is not necessarily the case (Liu et al., 2014). Other interactional studies have shown that lower status professionals become adept at posing questions and hedging their statements in order to navigate status differences in team interactions (Arber, 2008). Lingard et al. (2012) showed how physicians, who are traditionally seen at the apex of the professional hierarchy, can make efforts to encourage collective leadership by using inclusive language such as the pronoun “we,” by considering the pertinence of different expertise to case discussions, and by using team members’ first names; however, decision making ultimately remained in physicians’ hands. The use of these interactional strategies is influenced by the emergent and structural interactional resources (Meschitti, 2019) on which members may draw, for instance humor, one’s position in the hierarchy, or a claim to expert knowledge.

These studies of interprofessional interactions resonate with what some collective leadership scholars suggest: It would be utopic to believe that all members are completely equal in collective leadership (Salovarra and Bathurst, 2018; Sveiby, 2011). Others, such as Pearce (2004) and Crevani et al. (2007), suggest that collective leadership and more so-called hierarchical leadership are not completely incompatible. In fact, certain individuals can play an important role in the emergence of sharing leadership (Fletcher and Kaufer, 2003). While it may appear paradoxical, Pearce (2004) sees this possibility particularly among vertical leaders, that is individuals formally designated as the head or the leader of a team. These individuals can foster the emergence of shared leadership by appropriately designing the team (i.e. articulating a vision, clarifying roles and tasks, selecting members, etc.) and by managing boundaries (i.e. facilitating positive relations with those outside the team). In this vein, Chreim et al. (2013) discussed the boundary work practiced by clinical and managerial leaders (i.e. vertical leaders) who work to attenuate professional boundaries in order to allow open discussion and common interventions in the context of interprofessional collaboration. Thylefors and Persson (2014) also show the complementarity of collective leadership (which they refer to as horizontal leadership) and vertical leadership; put together, these two forms contribute to the effectiveness of interprofessional teams. However, this collective leadership literature does not specifically address the process of sharing leadership (Denis et al., 2012), nor interactions more specifically.

Some recent studies are starting to shed light on this process. For instance, Holm and Fairhurst (2018) explore shared and hierarchical leadership through the notions of authority and authoring, to show that hierarchy, expertise, and claims to advance the collective task serve as interactional resources that can be granted or resisted by others. Thus, this emphasizes the fluid, negotiated, and enacted nature of leadership, whether collective or not. Similarly, Van De Mieroop et al. (2020) show how formal and informal leaders can share the “doing” of leadership-in-interaction through their deontic stance (the public display of one’s power or authority in certain domains relative to that of one’s interlocutors, i.e. a public claim to the authority that underpins leadership) and deontic status (one’s relative position of authority and power, whether or not one makes a public claim to this position). However, the authors of both studies emphatically call for further empirical research that untangles the dynamics of leadership-in-interaction to enhance our understanding of how leadership is achieved.

Thus, there is still a need for more detailed and in-depth understanding of the relationship between sharing leadership and the asymmetry that derives from the formal or professional hierarchy (Morgeson et al., 2010; Yammarino et al., 2012), as negotiated in interprofessional team interactions. Therefore, we ask the following question: How do vertical leaders and other professionals interactionally negotiate and accomplish the sharing of leadership in the context of an entrenched professional hierarchy?

We draw on the theoretical assumption that team members are not all equal (Denis et al., 2012; Sveiby, 2011). We assume that various sources of influence, such as expertise and skill, are spread among the members of interprofessional teams, but not necessarily evenly. Furthermore, such asymmetry can derive from hierarchical differences in how professionals can lay claim to certain discursive resources by which meaning gets managed, such as openings and closings, which can control the topic of discussion, as well as formulations, which serve to fix the meaning of previous talk (Clifton, 2006). Drawing on the general principles of the negotiated order perspective (Strauss, 1978), we thus conceptualize sharing leadership as a dynamic and fragile activity that involves mobilizing resources, claiming rights and obligations. It may fluctuate over time; it can be strong at certain moments, and weaker at others, depending on interactional moves made by team members and tacit agreements constructed in social interactions. As Meschitti (2019) explained, the work of leadership “is a dynamic process, which is shaped in each turn [of talk]; it draws on both emergent and structural resources, and it is closely related to the task at hand” (629). Therefore, we aim to participate in the budding scholarly discussion about collective leadership and leadership-in-interaction by empirically exploring the fleetingness and complexity of sharing leadership in interprofessional team interactions.

Methods

Research context and data

We rely on data drawn from two previous ethnographically inspired studies of interprofessional team meetings in the health care sector in Canada. The data used for the current study come from observations of two teams located in different organizations and different clinical contexts. However, in both contexts, physician dominance in interactional dynamics was repeatedly observed, and in fact this was the impetus for the current study, as it contrasted with the ideals of “clinical democracy” (Long et al., 2006) and shared decision making that are associated with interprofessional collaboration in health care (Fox et al., 2019).

The authors of this article each followed one team, conducting naturalistic observations of interprofessional meetings, over a period of six months each. The first team, Alpha, is an outpatient clinic in pediatric psychiatry in Eastern Canada, composed of 25 mental health professionals, including a psychiatrist, psychologists, social workers, psycho-educators, and nurses. This team met on a weekly basis to discuss their clinical cases. They discussed an average of 15 patients per meeting, which typically lasted 90 minutes. The second team, Beta, is a short-stay internal medicine team in an acute care hospital in Western Canada, composed of 7 regular members from nursing and allied health who discussed all patients, and 10–15 rotating physicians, medical residents, and bedside nurses who dropped in to discuss only the patients in their care. This team met daily to discuss an average of 35 patients per meeting, which lasted 50 minutes on average. Data from Alpha were collected through non-participant observations and detailed fieldnotes (Emerson et al., 1995); data from Beta were collected through non-participant observations, detailed fieldnotes, and audio recordings of team interactions that were transcribed verbatim. (We also conducted interviews and documentary analysis; while we do not focus specifically on this here, we follow Larsson and Lundholm (2010) in relying on these data to better understand and contextualize the data analyzed here.) While approval for each site was obtained from the ethics review boards of each author’s relevant institutions, permission to audio record meetings was only granted for Beta. One implication of this difference in data sets is that some subtlety is necessarily missing from the fieldnotes, for instance overlapping turns of talk. However, other important interactional details were rigorously noted, such as frequency of interruptions, tone, pace of talk, silences, and so on. Indeed, similarities in our respective observations were the genesis of this study.

Informed consent was obtained from participants. In all, this article relied on 33 observations that were conducted and transcribed as observation reports (10 for Alpha) or as observation reports and transcribed recordings (23 for Beta). All identifying details have been modified to protect the confidentiality of the participants and the patients they discussed.

Data analysis

For the entire data set, we conducted thematic analysis (Braun and Clarke, 2006; Saldana, 2009) through data-driven and theory-driven coding (Fereday and Muir-Cochrane, 2006) of different interactional moves. Our analysis followed four broad steps. First, taking as our starting point the definition that sharing leadership occurs when individuals share influence through meaning management to advance a task or goal (Fairhurst, 2007), we identified interaction sequences from our observation reports and transcriptions where different professionals, including physicians, mutually and actively engaged in the co-production of clinical problems and decision processes. These sequences typically fell within recognizable openings and closings (Sacks, 1995) of the patient case discussions.

Second, we examined physicians’ discursive strategies to influence meaning management and action planning versus those of the other professionals in team discussions. We made this comparison because we had already observed and wished to further examine how interprofessional interactions were marked by an asymmetry that gave supremacy to physicians. More precisely, we sought to understand how physicians (i.e. those attended to as institutional vertical leaders in team interactions) contribute to the sharing of leadership and how other professionals also engage in this process. Consequently, to examine influence, we characterized what the team members’ actions seemed to do or try to do in the interaction, loosely following the precepts of conversation analysis (Drew, 2005; Pomerantz and Fehr, 1997). It is important to note that we could not conduct “pure” conversation analysis, as we did not have recordings for Alpha, but still wanted to explore similar observations of interactions across the two broader studies. Therefore, we worked iteratively with the same sequences to identify actions that were recurrent across selected sequences (or that, alternatively, constituted outliers), which then constituted our process codes (Corbin and Strauss, 2015). Turning our analysis in this direction led us to identify different interactional moves undertaken by physicians and by other professionals during clinical discussions, for example, using inclusive pronouns or cutting the other out of the discussion.

As we scrupulously analyzed the data in our field reports and transcribed recordings to identify the different interactional moves that became emergent codes, we were guided by what participants displayed and oriented to as being relevant (i.e. an emic interpretation of the data). Furthermore, some of these emergent codes from our interaction analysis were corroborated by participants in our interview data. For instance, some participants reflexively remarked on their interactions, such as the use of the honorific “Doctor” in their interactions, and on requests for information that could be framed as either invitations or demands. It is interesting to note that certain interactional moves that we identified also coincide with those remarked on by others in their research (e.g. Lingard et al., 2012). In this sense, we must acknowledge that our reading of the data was also colored by our familiarity of research in this area, and is in some ways also an etic read.

Third, it was noted (by ourselves as researchers, but also in interview discussions with the participants) that some of these actions could influence the team climate, and others the very reason the teams held clinical meetings, that is arriving at a shared understanding of and an action plan for patients’ situations. Therefore, we grouped these process codes into two sets: Establishing an interactional team climate and Doing knowledge work. This is in line with Schnurr’s (2018) distinction between relational and transactional behaviors in leadership interactions. Establishing an interactional climate includes actions undertaken by team members to create and influence the situation of the team meeting itself, for instance by negotiating relative status and access to the conversational floor. Through these moves, participants negotiated the meaning of their interactional situation of collaboration and their relative statuses within it. Doing knowledge work refers to actions involved in problem formulation and decision making about patient care. In this way, moves made in doing knowledge work were directly related to the leadership task of defining what was happening in the patient's situation and coming up with action plans (see Larsson and Lundholm, 2013, for an interesting exploration of sensemaking/giving as an organizing property of leadership). However, it is important to point out that this was an analytical distinction; often the interactional climate that was established influenced the knowledge work in which various team members could engage. Finally, we connected these actions to the main theme of our research question, specifically asymmetry in sharing leadership, by analyzing the role of interactional moves to establish a certain interactional climate or to accomplish knowledge work with regard to the professional hierarchy and sharing leadership.

Findings

In this section, we present a matrix that emerged from our analyses of the 33 observations. It summarizes the different interactional moves undertaken by physicians (the vertical leaders) and by other professionals during clinical discussions. Through these moves, they negotiated the sharing of leadership in interprofessional collaboration, in a context of entrenched professional hierarchy, both resisting and enacting the professional asymmetry. We then present two empirical vignettes from our data to illustrate the interactional dynamics described in the matrix. These vignettes serve as exemplars; they reflect the main patterns we observed in our data set. After each vignette, we walk the reader through the interaction, reinterpreting various moves to highlight and explain this negotiation. The first vignette is from fieldnotes from an Alpha meeting and has been translated from the original French. The second vignette is from a transcribed recording of a Beta meeting.

Matrix of interactional moves

This matrix (see Table 1) depicts how possibilities for sharing leadership are created or shut down in interprofessional team interactions by examining the moves that either enact or resist the asymmetry inherent to interprofessional teamwork. Resisting the asymmetry refers to those moves to downplay and attenuate asymmetry stemming from the professional hierarchy, whereas Enacting refers to the work that does the inverse. When they seek to suppress asymmetric power relations, team members foster the emergence of sharing leadership, whereas when they work to maintain it, they impede the emergence of sharing leadership.

Table 1.

Matrix of interactional moves.

	Enacting professional hierarchy and not sharing leadership		Resisting professional hierarchy and sharing leadership
	Physician’s actions	Other team member actions	Physician’s actions	Other team member actions
Establishing an interactional climate
Marking relationships	Using condescending language (e.g. insulting others’ competence; insisting on one’s superior knowledge)	Using honorifics (e.g. “Dr.” or the formal pronoun vous in French)	Using inclusive pronouns and speaking for the team	Using first names for all team members, inclusive pronouns, and speaking for the team
	Expressing strong emotions (e.g. anger and frustration)	Accommodating physician’s outbursts	Using humor and social niceties (politeness)	Expressing strong emotions (e.g. anger and frustration); using sarcasm
	Positioning oneself as the one to whom others must report	Reporting to the physician	Positioning the team as the one to whom others must report	Requesting information; reporting to the team
Influencing possibilities for participation	Cutting the other out of discussion (e.g. interrupting, ignoring, and speaking over or for others)	Remaining silent	Recognizing other as pertinent contributor (e.g. inviting others to speak; building on other’s contributions)	Recognizing other or self as pertinent contributor (e.g. insisting on the right to speak; inviting others to speak)
	Granting permission to speak	Requesting permission to speak	Making eye contact and inclusive body language	Jumping in to contribute
	Dominating through monologue or rapid pace	Following the rapid pace	Pausing or asking for contributions	Interrupting
Doing knowledge work
Formulating the problem	Challenging the other’s interpretation	Justifying one’s interpretation	Accepting the other’s interpretation	Challenging the other’s interpretation
Formulating the problem	Requesting information	Responding to request by providing information	Requesting a professional opinion	Providing a professional opinion (whether requested or not)
Making action plans	Resolving a problematization	–	–	Resolving a problematization
Making action plans	Assigning a task	Accepting a task	Accepting a task	Assigning a task

The matrix is horizontally divided to describe the actions grouped according to our second-order codes. As mentioned in the “Methods” section, Establishing an interactional climate includes all actions undertaken by team members to create and influence the situation of the team meeting itself. Through their interactions, the team members negotiate, confirm, and contest their relative status with one another, creating an environment that is more or less open, encouraging more or less free-flowing participation by the different professionals. Hence, through different discursive strategies that mark relationships, team members can invoke status or rights and obligations that perpetuate the asymmetry or, conversely, nurture the sharing of leadership and an inclusive climate. Other discursive strategies influence possibilities for participation; team members can open up or close down access to the conversational floor, thereby facilitating or impeding the emergence of shared leadership. Taken together, the interactional climate that gets negotiated also influences how team members contribute to managing the meaning of the problem at hand.

In fact, meaning management constitutes the team’s collective task and its reason for being, namely the sharing and weaving together of the information and knowledge held by different professionals on the team as they collectively negotiate a shared understanding of the patient’s situation, formulating the problem(s) to address, and, when necessary, making an action plan for the team. We named this second-level code “Doing knowledge work” because it is through assertions about the patient’s situation that its (inter)subjective reality is negotiated and established by the team’s members in the meeting. Sharing leadership occurs in doing this knowledge work when each involved professional can speak freely from their professional position and experience with the patient and when their contributions are recognized by the other team members as valid in meaning management. Indeed, this is the premise and the gold standard of interprofessional collaboration in health care, especially for complex cases (CIHC, 2009), where it is the problem at hand, and not the professional hierarchy, that ought to dictate whose expertise should be involved in the case. However, given the entrenched professional hierarchy in health care organizations, it also depends on one’s status. Thus, it is often necessary to create an interactional climate that encourages participation in an inclusive dialogue, and very often, physicians are key to the interactional climate that is established.

We turn now to the illustrative vignettes to show the analyses that resulted in the matrix presented above. It is important to note that not all interactional moves presented in the matrix appear in these two vignettes as the moves were drawn from the broader data corpus.

Vignette 1: Influencing action plans: Struggling against an enacted professional hierarchy

This first vignette presents the beginning of a team meeting involving a psychiatrist (MD-Psy, the team’s acknowledged vertical leader), a speech therapist, two psychologists (Psy-1 and Psy-2), a psycho-educator, and two social workers (SW-1 and SW-2). This particular vertical leader (MD-Psy) had been noted in interviews with other professionals as being unpredictable, which sometimes destabilized the interactional climate of team meetings. The first item of discussion is a request made by the speech therapist to include other mental health professionals in the treatment of a young girl who was receiving speech therapy for language difficulties. Realizing the girl’s issues extended beyond his scope of practice, the speech therapist calls for the expertise of his colleagues. We have numbered the different actions of interest (the second step in our data analysis) and use them to refer the reader during our illustration. In the interpretive analysis following each excerpt, we indicate in italics the actions that constituted our codes.

Before the meeting begins, MD-Psy (the psychiatrist) declares that she wants to add patient follow-ups to the meeting’s agenda, claiming that there are not many compared to other meetings, “My God, it’s stressful to not have any cases!” (1)

“Isn’t it more stressful when there are many?” the speech therapist responds (2). His comment seems to fall on deaf ears (3).

SW-1 announces that she has something to say, but I (researcher) note that she has trouble making herself heard (4). She eventually manages to take the conversational floor to say, “I’ve been assigned to targeted clients in the North, so if anyone has any files, any cases, they can let me know” (5).

The first case discussion begins. The speech therapist requests that other professionals get involved. He describes the case of a young girl in Grade 5 who has a learning disability and verbal problems (6). The speech therapist asks Psy-1 (school psychologist) to get involved in the case, saying, “For the past three weeks, this girl hasn’t been going to school. The reason I’m calling on you (Psy-1) is because we need to review her academic record.” He adds, “In fact, my requests for professionals have doubled, which means I think it would be worthwhile to get a social worker involved also” (7).

SW-1 tries to make a comment (8) but she is cut off by MD-Psy (9).

The psycho-educator asks SW-1, “You wanted to say something?” (10).

MD-Psy immediately speaks for SW-1, exclaiming that SW-1 can’t take the case because she’s in the North and the case is in the South (11).

Another social worker (SW-2) comes stiffly to SW-1’s defense, saying, “No, she doesn’t want that. She just wants to say something, so can she?” (12).

Somewhat flustered, MD-Psy responds, “Oh, okay…” (13). SW-1 then goes on to recall at length a scene of the patient in question at the hospital (14).

MD-Psy reacts, “Now, I think it is important, because in your naiveté and your youth, you might not have noticed, when the hospital is involved, it’s all fine and dandy, but when the girl gets discharged from the hospital, that’s when the problems hit. The hospital counts on us here (the outpatient clinic), but… It’s even harder when it’s not one of our professionals who is accompanying her” (15). SW-1’s non-verbal reaction is visible: She moves her body back while her face expresses shock, eyes wide at this insult. However, she remains silent (16).

Finally, MD-Psy closes the case by telling the speech therapist that she will pass on his request, “I am going to reiterate your request, I’m going to formalize your request for professionals” (17).

In this vignette, we see a struggle in the interactional climate for access to and control of the conversational floor, a struggle that hinders free-flowing participation in the collective task of doing knowledge work. More specifically, one member, SW-1, becomes the flashpoint around whom the tension roils in this charged interactional climate. The vertical leader, MD-Psy, tries to dominate the interactional climate, thereby enacting the professional hierarchy. First, she expresses her frustration (action 1). The speech therapist tries to positively recast her frustration (action 2), positioning himself as an equal team member but also accommodating her emotional venting. This could be aimed at attenuating her expressed frustration and limit its negative influence on the interactional climate, given the team’s experience with her. However, MD-Psy ignores the speech therapist, cutting him off (action 3).

SW-1 also has trouble jumping in and being heard (action 4), but eventually announces her availability to help others if needed (action 5). While it is not noted here how she finally managed to gain the conversational floor, her observed difficulty indicates her explicit negotiation of the interactional climate to contribute to doing knowledge work. By explaining her assignment to the North, she signals her availability as a team contributor to knowledge work, but also delimits her obligation to only particular cases from this region, and this becomes a resource for MD-Psy, as we shall see.

As the case discussion begins in earnest, the speech therapist describes the patient case under review and formulates the problem at hand (action 6), requesting help, which constitutes an implicit assignment of tasks (action 7). He speaks for the team as he does so (“I’m calling on you because we need to review”), requesting help from Psy-1 in his team role as the school psychologist and from the profession of social work. By formulating the problem (i.e. managing meaning) and making action plans, the speech therapist performs a leadership action (Clifton, 2012; Fairhurst, 2007), one that is typically category-bound to the vertical leader (Clifton, 2009). Furthermore, by using inclusive language (“we”), he proposes a collective—or interactional unit (Meschitti, 2019)—of which they are all members.

This call for SW involvement in doing knowledge work is followed by a series of push-and-pull negotiations for access to the conversational floor. First, although the called-upon professional, SW-1, tries to respond and thereby position herself as part of this “we” interactional unit (action 8), she is interrupted by MD-Psy (action 9). Thus, implicitly taking a deontic stance that insists on her deontic status, MD-Psy positions herself as vertical leader with the right to restrictively influence the interactional climate by cutting others out of the discussion. This move provokes resistance on the part of two other team members, who try to make the interactional climate more inclusive. First, the psycho-educator explicitly invites SW-1 to speak (action 10). However, MD-Psy tries to bat away this effort by speaking over SW-1, assuming SW-1’s turn at talk (i.e. speaks for her) and interpreting the situation (action 11). This again positions her as vertical leader having a right to speak for all. More specifically, by mobilizing as an emergent resource (Meschitti, 2019) SW-1’s previous utterance about her availability for cases in the North, MD-Psy claims that SW-1 does not have the right to participate in the knowledge work because it is not in her jurisdiction.

Once again, her exclusionary move is contested by SW-2, who aligns with the psycho-educator’s attempt to influence the interactional climate. She does so by challenging the physician’s interpretation of SW-1’s interactional rights and by insisting on SW-1’s right to speak, while also requesting permission for SW-1 to speak (action 12). In other words, in this negotiation, SW-2’s actions actively resist the asymmetric interactional climate sought by MD-Psy (resisting her deontic stance), while paradoxically also enacting it by seeking MD-Psy’s permission, thus orienting to her as the vertical leader (accepting her deontic status). Disgruntled, MD-Psy accepts this positioning and grants permission to speak (action 13) to SW-1.

With the interactional climate opened up sufficiently for her to participate in doing knowledge work with the team, SW-1 describes the patient’s situation and thus contributes to formulating the problem (action 14). However, the relevance of her contribution is immediately discredited by MD-Psy, who challenges SW-1’s interpretation of the situation (action 15) and adds an additional emotional charge to the interactional climate, insulting SW-1 by using condescending language (“in your naiveté and your youth…”). Furthermore, she mobilizes a specific structural resource (Meschitti, 2019)—her superior organizational experience with the hospital—to position herself as more knowledgeable and therefore as having greater epistemic entitlement (i.e. having the right to display a state of knowledge in the interaction, Clifton, 2012) to interpret the situation (i.e. do knowledge work). This double whammy (insult to SW-1 and self-claim to epistemic entitlement) excludes SW-1 from participating in knowledge work. SW-1’s non-verbal reaction (action 16) displays her displeasure at this relative positioning, but she ultimately accepts it.

Finally, MD-Psy reinterprets the speech therapist’s appeal for help as being directed to her as the vertical leader (i.e. positions herself as the one to whom he was reporting, action 17), and performs a leadership role by assigning tasks (i.e. making decisions by planning future actions, see Clifton, 2009) through the formalization of his request for professionals.

Overall, by restricting access to the conversational floor, by marking her superior status on the team, and by insulting her team members, MD-Psy influences the interactional climate in a way that is clearly hostile to the emergence of sharing leadership. However, the other team members did not passively accept the climate MD-Psy sought to establish; indeed, they actively resisted it at times to ensure that members lower on the hierarchy could contribute to the tasks of problem formulation and action planning. Nor was leadership completely relinquished to MD-Psy: The speech therapist (momentarily) performed a leadership role as he did knowledge work in his professional capacity. Ultimately, however, only the speech therapist’s intervention was successful in sharing leadership; SW-1 was solidly excluded. As a whole, the conflict in this vignette makes explicit the contested gray zone of sharing leadership that may be inherent to interprofessional collaboration in the context of professional hierarchy. The tension between collaboration and asymmetric relations is fully evident here in that non-physician team members must redouble their efforts to influence the interactional climate and thus access the conversational floor in order to integrate their professional expertise into the team's collective knowledge work. It is as though they are swimming against the stream. The next vignette offers a counterbalancing view, with an interactional climate that is much more open and inviting.

Vignette 2: Shared meaning making: Vertical leadership and sharing leadership in equilibrium

In this vignette from Beta, the team is discussing one of the few patients under the supervision of the Hospitalist (a hospital physician and recognized vertical leader). On this team, hospitalists would drop into the team meetings on a voluntary basis to discuss the patients in their charge, which is in itself evidence of asymmetric power relations. This particular hospitalist was well-liked by the team members: Observation fieldnotes indicate that he frequently scanned the room to see if anyone wished to participate and that team members were visibly physically relaxed in his presence. Moreover, he was one of the only hospitalists who regularly attended these team meetings.

Typically, at the beginning of case discussions (both when hospitalists were present and when they were not), the PCC (patient care coordinator or charge nurse, a vertical leader in nursing on the unit) would provide an overview of the case based on the notes of the previous PCC on shift, a leadership act of problem definition, or formulation (Clifton, 2006). Just prior to the excerpted interaction below, the PCC read a medical update about a patient whose neurodegenerative disease affects his mobility (action 1). She recounted the difficulty the imaging department had that morning in conducting a magnetic resonance imaging test because the patient did not lay still, despite the fact that he was sedated. The Hospitalist asked if “they” (the imaging team) were looking for a CNS (central nervous system) infection (action 2). The PCC replied that she doesn’t know because the patient chart was with the patient in the imaging department (action 3). The interaction continues as follows. ([ indicates the beginning of overlapping talk.)

(4) Hospitalist: But they must be thinking a CNS infection, which is ((inaudible)), (5) anyways, I’ll have to read through this.

(6) Home care: When he came in, I’ll just put in, he came in because he was getting increasingly confused. I think that’s why all the investigation.

(7) Hospitalist: Oh, is that what’s happening? Oh, okay.

(8) Pharm: I don’t have anything on antibiotics at this point. I think they were looking at a right-leg cellulitis, query H1N1.

(9) Home care: I hope it’s a query ((inaudible))

(10) Pharm: It’s negative anyways.

(11) Hospitalist: So, from leg cellulitis to central nervous system infection to a respiratory…

(12) PCC: Somewhere in there is a problem!

(13) Hospitalist: Well, we’ve narrowed it down. Right. (14) I will look into him this morning.

(15) Diet: And over the last few days, looks like his swallowing has deteriorated as well.

(16) Hospitalist: Oh really.

(17) Diet: He’s now on thickened fluids

(18) Hospitalist: We better fix him.

(19) PCC: Uh-hum!

(20) Physio: Well, that could be because he was em sedated, wasn’t he? With the EEG, they sedated him, ‘cause when I went, I went to try and get him up and he was just, you know, do- dozy. And they said it was because he’d been sedated for the EEG ((electroencephalogram)).

(21) Hospitalist: So, when, when was that and when did they do that swallowing assessment?

(22) Diet: I don’t know. I just see ((from the diet notes)) that it changed.

(23) PCC: It was the end of last week.

(24) Physio: That was on Friday, was the EEG.

(25) Hospitalist: We should probably reassess his swallowing. Cause they don’t want him to dehydrate [because we’re doing ((a couple of?)) things to him.

(26) Diet: [No. ((noted in fieldnotes: nods in agreement))

(27) Hospitalist: Um, okay, ((reads next patient’s name)).

As we will argue, this vignette demonstrates the peaceful co-existence of sharing leadership and vertical leadership. The first thing that is noticeable is that the interactional climate is significantly less charged than that of the previous vignette. In fact, the interactional climate seems to recede into the background of what demands our analytical attention. More specifically, very little negotiation of relative interactional status through positioning is observable here; team members do not hesitate to access the conversational floor (jumping in) in when it seems professionally appropriate to them. As mentioned, fieldnotes indicate that this Hospitalist fostered an inclusive interactional climate.

Insofar as leadership is performed when team members circumscribe and formulate the problem being discussed by the team (thus influencing meaning management; Fairhurst, 2007), leadership on this team was almost always shared between the team’s two vertical leaders, the PCC and the Hospitalist, especially at the beginning of case discussions when the PCC defined the current situation (action 1). Furthermore, as we shall see, when other team members mobilized their professional knowledge as structural resources (Meschitti, 2019) for doing knowledge work, it allowed them to claim epistemic entitlement and hence to share the leadership task of problem formulation.

It is important to highlight that such leadership sharing occurred simultaneous to the performance of vertical leadership: Both the PCC and the Hospitalist perpetuated the professional hierarchy and asymmetry when they positioned the Hospitalist as the vertical leader to whom others should report. This is evident even before the transcribed sequence begins, with the Hospitalist requesting information about the diagnostic query, thereby influencing the management of the case’s meaning (action 2). PCC takes up this request and his self-positioning as vertical leader by justifying why she cannot report to him (action 3). As the sequence begins, the Hospitalist makes a deductive knowledge claim that formulates the problem (“they must be thinking CNS infection,” action 4) and assigns himself the task of figuring it out (action 5). In the medical context, making decisions about treatment and diagnosis are the ultimate leadership power of formulation (Clifton, 2006), which is institutionally regulated (Barbour, 2010). Thus, here the Hospitalist is performing his institutional and organizational role as the vertical leader and enacting the professional hierarchy.

Typically, in these fast-paced meetings, the team would rapidly talk about an issue until an action plan was produced, after which the case discussion would be closed and a new one opened, often by a vertical leader. Indeed, this seems to be what the Hospitalist tries to do in action 5. However, the other professionals do not take up his move as a closure of the case discussion. Instead, they jump in to contribute, marking the inclusive, more symmetric interactional climate. With each person providing professional opinions, they collectively work to reduce the initial uncertainty marked by the Hospitalist’s speculation (i.e. they try to resolve the problematization). More specifically, as each professional chimes in with information from their professional domain of expertise (actions 6, 8, 15, 20), they claim epistemic entitlement to make a pertinent contribution to knowledge work.

We first see this in action 6, when the Home Care nurse explains that the patient has increasing confusion, thus contributing to the progressive formulation of the problem. Importantly, the Hospitalist marks his acceptance of her interpretation (“Oh, is that what’s happening? Oh, okay,” action 7; see Heritage, 1984 on change-of-state tokens) and of her claim to epistemic entitlement, recognizing her as a pertinent contributor and underscoring the inclusive climate (action 7).

It occurs again in actions 8 and 10, when the Pharmacist works to resolve the problematization by providing his professional opinion, deducing that no diagnosis has been made (CNS or otherwise) because he has no information on antibiotics. He mentions two different suspected causes of infection, and, following Home Care’s utterance of dismay about H1N1 (action 9), eliminates one (H1N1, action 10), thus sharing in the leadership work of resolving a problem formulation. Again, in action 11, the Hospitalist marks his acceptance of these interpretations, laying them out as three equal possibilities and thus underscoring the collective nature of the problem formulation (see Golden, 2006, on “so” marking inferential connections and other-attentiveness in interaction). Importantly, this also recognizes the Pharmacist as a pertinent contributor and further enacts the inclusive interactional climate.

PCC completes the Hospitalist’s utterance (action 12), which is a formulation of talk so far (Clifton, 2006). Interestingly, this marks both the inclusiveness of the interactional climate (she can freely jump in to contribute) and also her own position as vertical leader possessing the interactional right to co-construct such formulations. The Hospitalist then jokingly evaluates their collective resolution of the problem, using the inclusive pronoun “we” (action 13). This move proposes an interactional unit—the team—to which they all belong (Larsson and Lundholm, 2010), while also marking his position as vertical leader entitled to speak for the team. This positioning is further reiterated as he assigns a task to himself (“looking into him”), formulated as an action plan (action 14) and which, again, could be understood as a move to close the case discussion.

However, the dietician jumps in to contribute and provides her professional opinion to raise a concern about swallowing (actions 15 and 17), thus keeping the discussion open and returning team talk to the problem formulation stage of doing knowledge work. The Hospitalist marks his acceptance of this interpretation (action 16, “Oh, really”; see Heritage, 1984) and uses an inclusive pronoun to rally the team to action: a vertical leader interpreting the team’s mandate (action 18). PCC explicitly aligns with this positioning (action 19).

At this point, the physiotherapist jumps in to build on the dietician’s contribution to problem formulation. Specifically, she challenges the dietician’s problematization of the patient’s swallowing, providing a different interpretation from her professional experience that resolves the problematization (action 20). This free-flowing back and forth sharing information and potential interpretations is what interprofessional collaboration is ideally supposed to resemble (CIHC, 2009) and demonstrates sharing leadership in terms of problem formulation.

The Hospitalist recognizes the pertinence of the physiotherapist’s contribution by building on it (“So”; Golden, 2006) and requesting further information to resolve the problematization (action 21). This is a nuanced move that again positions him as the team’s vertical leader and the one to who others must report while simultaneously resisting the asymmetry by asking the dietician to report from her professional perspective (i.e. requesting a professional opinion). It also positions the dieticians who have done the swallowing assessment (“they”) as being outside the current interactional unit. Hence his move also recognizes the dual membership of his interlocutors: Each member of the interprofessional team is also a member of and spokesperson for their professional team (e.g. physiotherapy, nutrition, or pharmacy).

The dietician accepts this positioning and justifies her lack of knowledge and inability to report (action 22), a move that recognizes the professional hierarchy in the current interaction. Responding to her inability to report, both the PCC and the physiotherapist jump in to contribute information that helps to resolve the problematization (actions 23 and 24). The Hospitalist then uses another inclusive pronoun to implicitly assign a task to the dietician (action 25), an action that again both enacts his vertical leadership role and resists the asymmetry. The dietician orients to his positioning as vertical leader by implicitly accepting the assigned task (nodding, action 26). Then the Hospitalist closes the case discussion and opens the next, again performing a vertical leadership function.

Overall, this vignette illustrates a more peaceful co-existence of the pressures that characterize interprofessional teamwork than what we saw in the first vignette. The Hospitalist inhabits his medical role as vertical leader at the apex of the professional hierarchy; he positions himself as the one to whom others must report and influences the course of the team’s meaning management activity by requesting information and assigning a diagnostic task to the dietician. Indeed, the other team members accept this positioning and they report to him and accept assigned tasks. However, he also attenuates the professional hierarchy, working to foster an interactional climate that allows others to freely participate, specifically through his use of inclusive pronouns and speaking for the team, both of which underscore the importance of the team. Moreover, he accepts and builds on the others’ contributions to doing knowledge work, positioning them as valuable contributors. The others also inhabit their professional team roles and freely mobilize their relevant expertise to try to collectively resolve the problematization. In this sense, they accomplish the desired yet oft-illusive synergy of sharing leadership.

Discussion

In this paper, we analyzed observed interprofessional team interactions to question how leadership can be shared in the context of asymmetric relations stemming from the professional hierarchy. Underlying our analysis is a socioconstructivist approach that sees leadership as relational and emergent in interactions (Larsson, 2017; Quick, 2017; Sergi et al., 2012), understood as influence through discursive meaning management to advance a task or goal (Fairhurst, 2007). We contribute to the literature on leadership with a study based on observations of everyday work, something which many authors (e.g. Holm and Fairhurst, 2018; Larsson and Lundholm, 2010, 2013; Meschitti, 2019; Van De Mieroop et al., 2020) claim is needed to demystify (and perhaps deglorify) conceptualizations of leadership. With regard to collective leadership, we make a contribution in two main ways. First, we empirically explore sharing leadership in contexts characterized by inherent asymmetry, in this instance stemming from the professional hierarchy. Second and more specifically, we illustrate how the sharing of leadership is discursively negotiated in interactions, especially with regard to the interactional climate. In this sense, we use the leadership-in-interaction approach to explore collective leadership, thus not only demonstrating the usefulness of the former but showing the fruitfulness of putting these two emerging literatures into dialogue with one another.

Scholars of both these literatures may find our matrix of interactional moves useful for outlining the different discursive strategies used in the negotiation of sharing leadership. Conceptually, this matrix highlights the connection between sharing leadership and the professional hierarchy, and the issue of asymmetrical power relations that has so far been neglected in collective leadership scholarship (Denis et al., 2012; Holm and Fairhurst, 2018; Sveiby, 2011 are notable exceptions). In particular, it illustrates the discursive strategies team members adopt to enact and perpetuate or resist the asymmetry, and thus mute or foster the possibility of sharing leadership. We see that team members (both vertical leaders and lower status professionals) enact or resist the asymmetry that issues from the professional hierarchy by engaging in actions that can affect the team’s interactional climate and/or its knowledge work.

Certain discursive strategies described here are certainly closely linked to the observations made by others, such as Lingard et al. (2012)and Liu et al. (2014), who highlight interactional patterns on interprofessional teams. However, our matrix contributes to the literature a more comprehensive discursive analysis of both higher and lower status professionals on interprofessional teams to resist or enact the asymmetry. Relatedly, we note that moves by physicians and other team members become more convergent on the “resisting” side of the matrix. That is, when resisting the asymmetry, interactions became more of a dance between equals who can mirror the other’s moves rather than complementing them. This resonates with sharing leadership, as all professionals participate actively and influence each other in the meaning management and the accomplishment of their common goals.

Furthermore, the matrix extends the list of collective leadership processes to include establishing a climate and doing knowledge work. These two forms certainly overlap with more traditionally considered dimensions of leadership, such as those related to tasks or transactional behaviors and the (social) climate or relational behaviors (Contractor et al., 2012; Schnurr, 2018). Our study demonstrates that collective leadership can also happen at these two levels. However, again, we note that the establishment of a certain climate is necessary to accomplish the work related to a task, in this case knowledge work. In other words, it is not sufficient to bring together professionals in a collective so that each may share their knowledge; a climate must be established to facilitate this sharing. This finding resonates with other studies of interprofessional collaboration that emphasize the importance of creating participative safety for lower status team members (e.g. Jones and Jones, 2011; Nembhard and Edmondson, 2006). This can be created by the vertical leader, for instance by using inclusive pronouns or humor (Larsson and Lundholm, 2010; Lingard et al., 2012). Indeed, Edmonson et al. (2001) state that the doctor can act as a partner rather than as a dictator. Interestingly, our study shows that other professionals can also contribute to the establishment of this safe climate, for example by insisting on their right to speak.

Regarding leadership related to knowledge work, our data show, perhaps unsurprisingly, the persistence of a dominant position of the vertical leader in decision making. In contrast, our analysis allowed us to show that interprofessional knowledge work is not limited to decision making, but also includes all the meaning management (Clifton, 2012; Fairhurst, 2007) activities related to problem formulation. In particular, our analysis suggests the importance of taking close account of the leadership work that takes place prior to decision making as it can influence the decisions that are ultimately made. However, this work may not be accounted for in studies that rely on surveys and interviews (Meschitti, 2019).

Thus, our research shows how vertical leadership and sharing leadership can be co-present: In the same discussion, one can observe the enactment of vertical leadership and of sharing leadership, from both physicians and non-physicians. Sharing leadership appears as a complex and sometimes charged negotiation within interprofessional teams, especially regarding the interactional climate. In their daily interactions, team members must contend with the organizational expectation that they democratically participate in discussions by integrating their disciplinary expertise while also being yoked with institutional norms (Barbour, 2010) imposing distinct status differences between members. While Crevani et al. (2007) recognized the possible complementarity of collective leadership and vertical leaders, our empirically grounded study explains some of the ways in which an asymmetrical professional hierarchy can affect leadership interactions and especially the efforts required by vertical leaders and other team members to attenuate this asymmetry. This lends support to the view of leadership as performed and negotiated by multiple team members, rather than as appointed to a heroic individual in a given role (Berlin, 2015; Clifton, 2017; Salovarra and Bathurst, 2018). Likewise, such a view allows that leadership can “shift and distribute itself among several organizational members” (Fairhurst, 2007: 6). By examining what team members actually do in their interactions (i.e. their actions), we can tease out such shifting and distribution of leadership in the influence that different team members wield in both establishing the intersubjective situation of the meeting and the meaning management tasks that constitute the teamwork itself.

This leads us to remark on another, somewhat ironic paradox that stems from the co-presence of vertical and shared leadership: While we agree with Berlin (2015) that leadership, whether vertical or shared, is a collective performance, not all members of the interprofessional team have equal influence over the kind of leadership that is performed, as Fletcher and Kaufer (2003) and Sveiby (2011) noted. We conclude that, on the part of the higher-status professionals, sharing leadership requires a mindful and purposeful relaxing of the hierarchy, whereas for the lower-status professionals, it occurs more often through struggle and resistance, especially with regard to the interactional climate. This, in itself, is a manifestation of asymmetric power relations and of the “institutional moorings of talk” (Barbour, 2010). In other words, the professional hierarchy is both context generative, in the sense that it structures interactions, and context generated, getting negotiated in interactions on an ongoing basis.

What, then, does it mean for the emergence of shared leadership if it is dependent on the vertical leader’s “permission” and encouragement? In the health care sector, given the prevailing reality of the physician structurally inscribed at the apex of the professional hierarchy, it is conceivable that sharing leadership is only possible for the work of creating a shared understanding, what we call doing knowledge work. If so, then an especially important practical conclusion that our study suggests is that it is imperative that physicians as well as vertical leaders in other sectors be trained to find a balance between sharing leadership and enacting their vertical leadership role if they are to leave space for the others to participate fully in the interprofessional collaboration and provide their needed professional expertise. They can do so by fostering an inclusive interactional climate so that others feel safe to take the lead when the situation calls for it.

Conclusion

In this article, we delve into the in situ interactions from two interprofessional teams to better understand a neglected issue in the literature on sharing leadership: asymmetry that stems from the professional hierarchy. Interprofessional teams are an ideal context to analyze asymmetry and sharing leadership because by definition, they aim to foster knowledge sharing among members who belong to professions with different hierarchical positions. We contribute to the literature by showing that leadership is shared (or not) as a result of how the professional hierarchy gets negotiated in interactions. More specifically, we conclude that the sharing of leadership in this context tends to occur prior to decision making, especially around problem formulation, if the interactional climate allows. Furthermore, it requires concrete effort: Those in superior positions of influence must mindfully relax the hierarchy whereas those in inferior positions create moments of sharing leadership through resistance and struggle.

Our two vignettes of team discussions illustrate the interactional moves by which professionals resist or perpetuate an asymmetric professional hierarchy. The data analysis resulted in a comprehensive matrix that distinguishes interactional moves of both vertical leaders and other team members related to sharing leadership, according to their domain of action, namely the interactional climate or the knowledge work. Although sharing leadership and asymmetry appear as contradictory, our study suggests that both can be co-present and even intertwined. Thus, we empirically document the paradox that stems from the mutuality of vertical and shared leadership. Future research on sharing leadership could further employ a leadership-in-interaction approach to analyze the negotiation of interactional climate through conversational positioning understood as deontic stance and status (Van De Mieroop et al., 2020) and knowledge work through authority claims to “author” the knowledge work (Holm and Fairhurst, 2018). Another interesting path forward would be to examine the process of decision making and the contexts of interprofessional teamwork other than health care, such as government, industry, and education, to see if the same interactional dynamics prevail.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Stephanie Fox

Author biographies

Stephanie Fox earned her PhD in communication at Simon Fraser University and is assistant professor of communication at the Département de communication, Université de Montréal, Canada. Her research interests include interprofessional collaboration and conflict, sensemaking, and narrative theory and methods. She studies these issues in both primary care and hospital care contexts.

Mariline Comeau-Vallée earned her PhD in management at HEC Montréal, and is now professor of organizational development at École des Sciences de la Gestion, Université du Québec à Montréal, Canada. Her research interests include interprofessional dynamics, identity negotiation, and the management of paradoxical tensions. She has studied these topics in pluralistic settings, such as health care and the social economy.

References

Abbott

(1988) The System of Professions: An Essay on the Division of Expert Labour. Chicago, IL: University of Chicago Press.

Arber

(2008) Team meetings in specialist palliative care: Asking questions as a strategy within interprofessional interaction. Qualitative Health Research 18(10): 1323–1335.

Avolio

Walumbwa

Weber

(2009) Leadership: Current theories, research, and future directions. Annual Review of Psychology 60: 421–449.

Avolio

Gardner

(2005) Authentic leadership development: Getting to the root of positive forms of leadership. The Leadership Quarterly 16(3): 315–338.

Barbour

(2010) On the institutional moorings of talk in health care organizations. Management Communication Quarterly 24(3): 449–456.

Bass

(2010) The Bass Handbook of Leadership: Theory, Research, and Managerial Applications. New York: Simon & Schuster.

Berlin

(2015) Doctors’ functional leadership in psychiatric healthcare teams – A reversible leadership logic. Team Performance Management: An International Journal 21(3–4): 159–180.

Blake

Mouton

(1985) Presidential (grid) styles. Training and Development Journal 39: 30–34.

Boden

(1994) The Business of Talk: Organizations in Action. Cambridge: Polity Press.

10.

Bourgeault

Mulvale

(2006) Collaborative health care teams in Canada and the USA: Confronting the structural embeddedness of medical dominance. Health Sociology Review 15: 481–495.

11.

Braun

Clarke

(2006) Using thematic analysis in psychology. Qualitative Research in Psychology 3(2): 77–101.

12.

Brown

Kornberger

Clegg

, et al. (2010) “Invisible walls” and “silent hierarchies”: A case study of power relations in an architecture firm. Human Relations 63(4): 525–549.

13.

Canadian Health Services Research Foundation (2012) Interprofessional Collaborative Teams. Ottawa: Canadian Health Services Research Foundation. Available at: http://www.cfhi-fcass.ca/Libraries/Commissioned_Research_Reports/Virani-Interprofessional-EN.sflb.ashx (accessed 3 January 2019).

14.

Canadian Interprofessional Health Collaborative (CIHC) (2009) What Is…Collaborative Practice. Ottawa: Canadian Interprofessional Health Collaborative. Available at: https://www.cihc.ca/files/CIHC_Factsheets_CP_Feb09.pdf (accessed 14 October 2014).

15.

Chreim

Langley

Comeau-Vallee

, et al. (2013) Leadership as boundary work in healthcare teams. Leadership 9(2): 201–228.

16.

Clifton

(2006) A conversation analytical approach to business communication: The case of leadership. Journal of Business Communication 43(3): 202–219.

17.

Clifton

(2009) Beyond taxonomies of influence: “Doing” influence and making decisions in management team meetings. Journal of Business Communication 46(1): 57–79.

18.

Clifton

(2012) A discursive approach to leadership: Doing assessments and managing organizational meanings. Journal of Business Communication 49(2): 148–168.

19.

Clifton

(2014) Small stories, positioning, and the discursive construction of leader identity in business meetings. Leadership 10(1): 99–117.

20.

Clifton

(2017) Taking the (heroic) leader out of leadership. The in situ practice of distributed leadership in decision-making talk. In: Ilie C and Schnurr S (eds) Challenging Leadership Stereotypes Through Discourse: Power, Management and Gender, pp.1–24. Singapore: Springer Nature.

21.

Contractor

DeChurch

Carson

, et al. (2012) The topology of collective leadership. The Leadership Quarterly 23: 984–1011.

22.

Corbin

Strauss

(2015) Basis of Qualitative Research Techniques and Procedures. 4th ed. Thousand Oaks, CA: Sage.

23.

Cott

(1997) “We decide, you carry it out”: A social network analysis of multidisciplinary long-term care teams. Social Science Medicine 45(9): 1411–1421.

24.

Crevani

Lindgren

Packendorff

(2007) Shared leadership: A postheroic perspective on leadership as a collective construction. International Journal of Leadership Studies 3(1): 40–67.

25.

Currie

Lockett

(2011) Distributing leadership in health and social care: Concertive, conjoint or collective? International Journal of Management Reviews 13(3): 286–300.

26.

D’Amour

Ferrada-Videla

San Martín-Rodríguez

, et al. (2005) The conceptual basis for interprofessional collaboration: Core concepts and theoretical frameworks. Journal of Interprofessional Care 19(s1): 116–131.

27.

D’Amour

Oandasan

(2005) Interprofessionality as the field of interprofessional practice and interprofessional education: An emerging concept. Journal of Interprofessional Care 19(s1): 8–20.

28.

Davis

Eisenhardt

(2011) Rotating leadership and collaborative innovation: Recombination processes in symbiotic relationships. Administrative Science Quarterly 56(2): 159–201.

29.

Day

Gronn

Salas

(2004) Leadership capacity in teams. The Leadership Quarterly 15: 857–880.

30.

Denis

Langley

Sergi

(2012) Leadership in the plural. Academy of Management Annals 6(1): 211–283.

31.

Drew

(2005) Conversation analysis. In: Fitch C and Sanders RE (eds) Handbook of Language and Social Interaction. Mahwah, NJ: Lawrence Erlbaum Associates, Publishers, pp.71–102.

32.

Dwyer

(1991) Humor, power, and change in organizations. Human Relations 44(1): 1–19.

33.

Edmonson

Bohmer

Pisano

(2001) Disrupted routines: Team learning and new technology implementation in hospitals. Administrative Science Quarterly 46(4): 685–716.

34.

Emerson

Fretz

Shaw

(1995) Writing Ethnographic Field Notes. Chicago, IL: University of Chicago Press.

35.

Fairhurst

(2007) Discursive Leadership: In Conversation with Leadership Psychology. Los Angeles, CA: Sage.

36.

Fairhurst

Grant

(2010) The social construction of leadership: A sailing guide. Management Communication Quarterly 24(2): 171–210.

37.

Fereday

Muir-Cochrane

(2006) Demonstrating rigor using thematic analysis: A hybrid approach of inductive and deductive coding and theme development. International Journal of Qualitative Methods 5(1): 80–92.

38.

Finn

(2008) The language of teamwork: Reproducing professional divisions in the operating room. Human Relations 61(1): 103–130.

39.

Finn

Currie

Martin

(2010) Team work in context: Institutional mediation in the public-service professional bureaucracy. Organization Studies 31(8): 1069–1097.

40.

Fiorelli

(1988) Power in work groups: Team member’s perspectives. Human Relations 41(1): 1–12.

41.

Fletcher

Kaufer

(2003) Shared leadership: Paradox and possibility. In: Pearce CL and Conger JA (eds) Shared Leadership: Reframing the Hows and Whys of Leadership. Thousand Oaks, CA: Sage Publications, pp.21–47.

42.

Fox

Gaboury

Chiocchio

, et al. (2019) Communication and interprofessional collaboration in primary care: From ideal to reality in practice. Health Communication. DOI: 10.1080/10410236.2019.1666499.

43.

Freidson

(2001) Professionalism: The Third Logic. Chicago, IL: University of Chicago Press.

44.

Friedrich

Vessey

Schuelke

, et al. (2011) A Framework for Understanding Collective Leadership: The Selective Utilization of Leader and Team Expertise Within Networks (Project A790). Norman: Center for Applied Social Research, University of Oklahoma.

45.

Golden

(2006) Little words that matter: Discourse markers “So” and “Oh” and the doing of other-attentiveness in social interaction. Journal of Communication 56(4): 661–688.

46.

Gronn

(2002) Distributed leadership as a unit of analysis. The Leadership Quarterly 13: 423–451.

47.

Heritage

(1984) A change-of-stage token and aspects of its sequential placement. In: Atkinson M and Heritage J (eds) Structures of Social Action: Studies in Conversation Analysis. Cambridge: Cambridge University Press, pp.299–345.

48.

Hiller

Day

Vance

(2006) Collective enactment of leadership roles and team effectiveness: A field study. The Leadership Quarterly 17: 387–397.

49.

Holm

Fairhurst

(2018) Configuring shared and hierarchical leadership through authoring. Human Relations 71(5): 682–721.

50.

Jones

(2011) Improving teamwork, trust and safety: An ethnographic study of an interprofessional initiative. Journal of Interprofessional Care 25(3): 175–181.

51.

Lammers

Barbour

Duggan

(2003) Organizational forms and the provision of health care. In: Thompson TL, Dorsey AM, Miller KI, et al. (eds) Handbook of Health Communication. Mahwah, NJ: Lawrence Erlbaum Associates, pp.319–346.

52.

Larsson

(2017) Leadership in interaction. In: Storey J, Hartley J, Denis J-L, et al. (eds) The Routledge Companion to Leadership. New York: Routledge, pp.173–193.

53.

Larsson

Lundholm

(2010) Leadership as work-embedded influence: A micro-discursive analysis of an everyday interaction in a bank. Leadership 6(2): 159–184.

54.

Larsson

Lundholm

(2013) Talking work in a bank: A study of organizational properties of leadership in work interactions. Human Relations 66(8): 1101–1129.

55.

Lingard

Vanstone

Durrant

, et al. (2012) Conflicting messages: Examining the dynamics of leadership on interprofessional teams. Academic Medicine 87(12): 1762–1767.

56.

Liu

Manias

Gerdtz

(2014) Exploring power relations embedded in medication communication processes on general medical wards. Qualitative Research Journal 14(2): 161–178.

57.

Long

Forsyth

Iedema

RAM

, et al. (2006) The (im)possibilities of clinical democracy. Health Sociology Review 15(5): 506–519.

58.

McCallin

(2003) Interdisciplinary team leadership: A revisionist approach for an old problem? Journal of Nursing Management 11(6): 364–370.

59.

McHugh

Bennett

(1999) Introducing teamworking within a bureaucratic maze. Leadership & Organization Development Journal 20(2): 81–93.

60.

Meschitti

(2019) The power of positioning: How leadership work unfolds in team interactions. Leadership 15(5): 621–643.

61.

Morgeson

DeRue

Karam

(2010) Leadership in teams: A functional wapproach to understanding leadership structures and processes. Journal of Management 36(1): 5–39.

62.

Nembhard

Edmondson

(2006) Making it safe: The effects of leader inclusiveness and professional status on psychological safety and improvement efforts in health care teams. Journal of Organizational Behavior 27: 941–966.

63.

Pearce

(2004) The future of leadership: Combining vertical and shared leadership to transform knowledge work. Academy of Management Perspectives 18(1): 47–57.

64.

Pearce

Conger

(2003) All those years ago: The historical underpinning of shared leadership. In: Pearce CL and Conger JA (eds) Shared Leadership: Reframing the Hows and Whys of Leadership. Thousand Oaks, CA: Sage Publications, pp.1–13.

65.

Pearce

Manz

(2005) The new silver bullets of leadership: The importance of self- and shared leadership in knowledge work. Organizational Dynamics 34(2): 130–140.

66.

Pinchot

(1993) The End of Bureaucracy and the Rise of the Intelligent Organization. San Francisco, CA: Berrett-Koehler.

67.

Pomerantz

Fehr

(1997) Conversation analysis: An approach to the study of social action as sense making practices. In: van Dijk TA (ed) Discourse as Social Interaction: Discourse Studies: A Multidisciplinary Introduction. Vol. 2. London: Sage Publications, pp.64–91.

68.

Quick

(2017) Locating and building collective leadership and impact. Leadership 13(4): 445–471.

69.

Raelin

(2016) Imagine there are no leaders: Reframing leadership as collaborative agency. Leadership 12(2): 131–158.

70.

Sacks

(1995) Lectures on Conversation. Vol. I. Malden, MA: Blackwell Publishing.

71.

Saldana

(2009) The Coding Manual for Qualitative Researchers. London: Sage.

72.

Salovarra

Bathurst

(2018) Power-with leadership practices: An unfinished business. Leadership 14(2): 179–202.

73.

Schnurr

(2018) Leadership. In: Vine

(ed) The Routledge Handbook of Language in the Workplace. New York: Routledge, pp.323–334.

74.

Sergi

Denis

J-L

Langley

(2012) Opening up perspectives on plural leadership. Industrial and Organizational Psychology 5: 403–407.

75.

Strauss

(1978) Negotiations: Varieties, Contexts, Processes and Social Order. San Francisco, CA: Jossey-Bass Inc. Pub.

76.

Sveiby

(2011) Collective leadership with power symmetry: Lessons from aboriginal prehistory. Leadership 7(4): 385–414.

77.

Thylefors

IEC

Persson

(2014) The more, the better? Exploring vertical and horizontal leadership in cross-professional health care teams. Leadership in Health Services 27(2): 135–149.

78.

Van De Mieroop

Clifton

Verhelst

(2020) Investigating the interplay between formal and informal leaders in a shared leadership configuration: A multimodal conversation analytical study. Human Relations 73(4): 490-515.

79.

Yammarino

Salas

Serban

, et al. (2012) Collectivistic leadership approaches: Putting the “we” in leadership science and practice. Industrial and Organizational Psychology 5(4): 382–402.

80.

Zaccaro

Rittman

Marks

(2001) Team leadership. The Leadership Quarterly 12: 451–483.