Tensions in Distributed Leadership

Abstract

Purpose: This article proposes the utility of using activity theory as an analytical lens to examine the theoretical construct of distributed leadership, specifically to illuminate tensions encountered by leaders and how they resolved these tensions. Research Method: The study adopted the naturalistic inquiry approach of a case study of an elementary school in Singapore in the process of implementing a project using information and communications technology for instruction. The leaders identified included the principal, vice principals, and middle managers, including a senior teacher. Activity theory was used as the lens for data analysis. Findings: The article makes three contributions to distributed leadership. First, boundary-spanning aspects of work give rise to tensions in the distribution of leadership. Second, tensions provide a context for the exercise of influence by leaders—demonstrated through innovative solutions to counter the tensions. Third, the distribution of leadership is dynamic with the centrality of leadership shifting between multiple levels of leaders as they interact in intersecting activity systems. The case study also confirms the importance of studying the relation between structure and human agency in examining the distribution of leadership.

Keywords

distributed leadership tensions activity theory technology in education dualities multiple levels of leaders

Introduction

When Gronn and Spillane first proposed their conceptions of distributed leadership, what was revolutionary was a shift from focusing on the leadership actions of an individual as sole agent to analyzing the “concertive” or “conjoint” actions of multiple individuals interacting and leading within a specific social and cultural context (Bennett, Wise, Woods, & Harvey, 2003; Gronn, 2002b; Spillane & Orlina, 2005; Woods, Bennett, Harvey, & Wise, 2004). As explained by Gronn and Ribbins (1996), “context is the vehicle through which the agency of particular leaders may be empirically understood” (p. 454).

Gronn (2002a) argued that Engestrom’s activity theory offers advantages in the study of distributed leadership. He observed that activity theory recognizes that activity is jointly performed and involves a division of labor. In addition, activity theory incorporates a range of contextual dimensions that affect the activity, and which are simultaneously affected by the activity, contextual dimensions which Gronn (2002a) commented had been missing from much of the earlier literature on leadership. This interdependency between the context and the activity in focus is reflected in Spillane’s argument that the situation in which distributed leadership is practiced “both defines leadership practice and is defined through leadership practice” (Spillane, 2006, p. 4). Multiple leaders interact with one another as well as with specific contextual dimensions in the activity system, giving rise to particular leadership actions, which in turn affect the activity system. Furthermore, Gronn (2002a) noted that the activity system model enables researchers to track how the actions of leaders are enabled or constrained by the organization structures within which they work. This resonates with activity theory, which seeks to understand structural contradictions or tensions, with the purpose of generating innovative solutions to resolve these contradictions and enable the activity system to evolve (Barab, Barnett, Yamagata-Lynch, Squire, & Keating, 2002; Hartley, 2009).

Both Gronn and Spillane indicate that distributed leadership has its theoretical roots in activity theory (Gronn, 2000; Spillane, Halverson, & Diamond, 2001, 2004). However, in actual practice, empirical studies on distributed leadership have to date rarely used activity theory as an interpretive lens to analyze the concept of distributed leadership (Harris, 2005, 2010; Mayrowetz, 2008). We suggest that there is merit in using activity theory as an analytical lens to examine the construct of distributed leadership, in order to understand the interplay among the various contextual dimensions and leadership practice. In Ho, Chen, and Ng (2016), the authors illustrated the use of activity theory to examine the distinct but complementary leadership provided by senior and middle management, as well as the role played by social norms in mediating the leadership activity. In this article on the same case study, we suggest that the use of activity theory as an interpretive lens enables us to surface constraints on distributed leadership and how these are overcome, through the analysis of structural or systemic tensions within or between activity systems.

Literature Review

The following section presents a literature review of distributed leadership. In addition, since distributed leadership’s concept of leadership as a dynamic and collectively performed activity has its theoretical roots in activity theory (Gronn, 2002b; Spillane et al., 2004), the literature on activity theory is also examined, with the focus on understanding the concept of system tensions or contradictions in activity systems.

Distributed Leadership and Activity theory

According to Bennett, Newton, Wise, Woods, and Economou (2003), what differentiates distributed leadership from similar concepts of leadership, such as shared leadership, is that it emerges from the interactions of a network of individuals, from conjoint as opposed to individual agency. Therefore, what distinguishes distributed leadership is its concept of leadership as a dynamic and collectively performed activity (Gronn, 2002b; Spillane et al., 2004). In distributed leadership, the unit of analysis is the leadership activity and the interactions among the multiple leaders involved in leadership practices within a specific social and cultural context (Copland, 2003; Spillane, 2005; Spillane & Orlina, 2005; Timperley, 2005).

Distributed leadership is positioned as having its theoretical roots in activity theory. In activity theory, human actions are meaningfully understood only within the context of an activity system, which includes the subject whose motivation is to transform an object into an outcome, which can be either material or intangible in nature (Barab, Barnett, Yamagata-Lynch, et al., 2002; Engestrom, 1999). Activity theory, which takes as its unit of analysis a collectively performed activity system, incorporates the concept that labor is divided among various individuals, and takes into consideration varied contextual dimensions that dynamically affect leadership practices and the division of such leadership practices (Gronn, 2002a; Spillane et al., 2001).

In activity theory, these dimensions include the community, tools, and rules (Yamagata-Lynch, 2003). The community refers to the social group that the object identifies being a member of while participating in the activity. Rules, which can be explicit or implicit, include norms, policies, or conventions of the community which can constrain or enable actions and interactions within the activity system (Kelceoglu, 2006; Marken, 2006). Tools include physical and conceptual artifacts, such as sign and language systems, and structures such as a timetable schedule, or a scheme of work. Similar to rules, tools can simultaneously enable and constrain relations between subject and object, or between the other dimensions (Barab, Barnett, Yamagata-Lynch, et al., 2002). In addition, activity theory recognizes potential constraints on interactions in a collectively performed activity, which distributed leadership is defined to be. These are known as tensions, or contradictions, which are explained further in the subsequent section of the literature review. Figure 1 illustrates the structure of an activity system as presented by activity theory (Engestrom, 1999).

Figure 1.

Model in Engestrom (1999).

Tensions in Distributed Leadership and Activity Theory

Inherent in the notion that activity systems are dynamic systems, activity theory postulates the concept of structural or systemic tensions, which arise due to overlapping or conflicting needs that drive an activity system and which need to be balanced in order for the system to continue to develop and evolve (Barab, Barnett, & Squire, 2002; Barab, Barnett, Yamagata-Lynch, et al., 2002). Tensions, also known as contradictions, can occur within or between activity systems (Engestrom, 2001; Marken, 2006; Yamagata-Lynch, 2003).

As the dynamic components of an activity system (subject, object, tools, rules, division of labor, and community) continuously interact with the other components, tensions may occur between or within any of these components. In a rare study that used activity theory as a lens to examine distributed leadership, Vennebo (2016) observed that even when leaders work toward a common overall purpose, differences in motives can result in tensions, whether in the object under construction, the choice of tools, or in the existing division of labor. However, tensions are not necessarily negative in impact. Researchers propose that tensions play a critical role in that they can drive innovation and change within the system, as the system evolves to resolve or balance these tensions (Barab, Schatz, & Scheckler, 2004; Engestrom, 1999; Yamagata-Lynch, 2003). Vennebo (2016, p. 5), in discussing the role of tensions in distributed leadership, observed that tensions can be “springboards for the expansion of organizational routines and the transformation of . . . practices.” Gronn (2002a) also suggested the utility of identifying the tensions in workplace activity systems, which would assist in the development of intervention strategies to resolve or circumvent these tensions.

Although tensions were mentioned in the literature on distributed leadership by some researchers (Storey, 2004; Torrance, 2013), these tensions were not explicitly linked to the concept of system tensions in activity theory. Tension was used generically to refer to conflicts or friction between individuals. However, there is suggestion that certain tensions were due to the changing (or dual) role of the teacher who had to take on leadership roles, or due to struggles for power/control as a result of conflicting needs/objectives (Timperley, 2005). This perspective of tensions resonates with activity theory’s concept of system tensions as “dualities,” as arising due to conflicting needs within the system. Usually the conflicts/frictions identified in the literature on distributed leadership are between “competing leaders,” often between the principal and middle managers/teacher leaders (Storey, 2004, p. 250; Timperley, 2005; Torrance, 2013), generally because of competing visions or priorities.

Even though Spillane and Gronn attribute the origins of distributed leadership to activity theory, Engestrom’s concept of system tensions/contradictions is missing from much of their work. Hartley (2009, p. 145) observed that though Spillane referred to Engestrom, there was “an absence of any consideration of contradictions.” Nevertheless, Spillane and Orlina (2005) did acknowledge that a distributed perspective includes situations where leaders have different or “contrary goals” (p. 152).

Similarly, although he did not refer specifically to activity theory, Gronn highlighted tensions between “central control” and “local discretion” (Gronn, 2008, p. 145), which resonates with activity theory’s concept of conflicting needs, or dualities, within the system. Gronn (2002a) also referred to tensions arising due to a duality of roles, tensions arising due to differences in status although there may be no difference in expertise, or tensions due to differences in objectives. He refers to the latter as a “pluralistic domain of multiple agents pursuing different objectives” (Gronn, 2002b, p. 441). In addition, both Gronn and Spillane view structure as both enabling and constraining human agency, suggesting a constant tension between structure and human agency (Gronn, 2002b; Spillane et al., 2004; Spillane & Orlina, 2005).

Purpose of Article

We propose that there is merit to identify and examine tensions in distributed leadership, which has its roots in activity theory and is a collectively performed activity. Tensions, or contradictions, are inherent to activity systems. Tensions are critical to understanding what motivates particular actions and to understanding the evolution of the activity system as innovative effort is made to resolve/negotiate the tensions (Engestrom, 1999, 2000, 2001). In this article, we argue that a challenge for researchers and principals who are interested in distributed leadership is to recognize or even anticipate the tensions, understand how tensions affect the distribution of leadership or leaders’ actions, balance their influence, and enable distributed leadership to evolve further. A similar argument was presented by Vennebo (2016) who suggested that in exploring how distributed leadership is enacted, the question of whether tensions occur, and how these tensions are worked on, is important. Therefore, this article adds value to the growing literature on distributed leadership, by illuminating how leaders overcome tensions arising from conflicting or overlapping goals.

Methodology

Distributed leadership is enacted through interactions among leaders, followers, and the situation (Spillane et al., 2004). In order to analyze in detail the leadership activity and interactions among the stakeholders, this study adopted the naturalistic inquiry approach of a case study (Creswell, 2005; Lincoln & Guba, 1985). A school was studied over a period of 16 months, in the process of preparing and implementing a project using information and communications technology (ICT) to support instruction in mathematics and science.

The main research question was “How was distributed leadership practiced in a school reform involving the use of ICT for instruction?” Activity theory was used as a lens to examine the activity systems involved, comprising the subjects, objects, tools, rules, division of labor, and communities, which enabled an understanding of the distinct leadership tasks performed by various leaders and their interactions (Ho et al., 2016). This article focuses on what tensions were encountered in the distribution of leadership and how these tensions were resolved. The research questions asked were as follow:

What were the tensions encountered in the distribution of leadership?

How were these tensions resolved?

Tension is defined as dualities which arise due to overlapping or conflicting needs that drive a system and which need to be balanced in order for the system to continue to develop and evolve (Barab, Barnett, & Squire, 2002; Barab, Barnett, Yamagata-Lynch, et al., 2002). These dualities could refer to a duality of roles, of objectives/goals/visions, or of boundaries of influence/responsibility. Tensions can occur within or between activity systems, within or between the components of the activity system: subject, object, tools, rules, division of labor, and the community.

Distributed leadership is defined as the distribution of tasks and the distribution of influence processes (Robinson, 2008). The first definition is based on the theorization of leadership as the performance of particular tasks, specifically the use of “social, material and cultural resources necessary to establish the conditions for teaching and learning” (Spillane et al., 2004, p. 11). The second concept is based on the view that leadership involves deliberately exerting some form of influence on people in an attempt to affect their beliefs, values, and actions, so as to achieve desired and presumed shared organizational outcomes (Pounder, Ogawa, & Adams, 1995; Yukl, 2001). In the context of schools, leadership often involves the effort to influence the motivation or practice of teachers to improve instructional practice (Gronn & Hamilton, 2004; Harris, 2003; Harrison, 2005; Spillane, Diamond, & Jita, 2003). In our specific case study, the identified leaders sought to influence the teachers concerned to use ICT to improve their teaching practice, through performing various tasks such as coaching, modelling the process of developing ICT lesson plans, providing professional development, and providing access to ICT resources (Ho et al., 2016).

The distributed leadership perspective proposes that leadership is best understood through examining the leadership actions enacted and the “theories-in-use” of the leaders (Argyris & Schön, 1974, as cited in Spillane et al., 2004, p. 15). From a methodology perspective, this translated into the researchers observing leaders in action and interviewing leaders to elicit their explanations and interpretations of the actions observed. To observe leadership in action, the case study involved observations of 49 meetings, during which many instructional and other decisions were made, and analysis of over 150 e-mail correspondences related to the ICT project. The 49 meetings were scheduled during curriculum time (referred to in Singapore schools as Timetabled Time) mainly to enable the teachers involved in the ICT project to discuss the ICT lessons they were developing, and to share how the ICT project was progressing. All the six teachers involved in the ICT project were interviewed at least once.

Identified leaders were interviewed at least twice during the course of the study. Leaders were identified in two ways. First, the formally appointed leaders were interviewed to understand the thinking behind their observed actions as well as behind actions which the researchers did not have the opportunity to observe (Seidman, 2006). Second, all the participants, including the teachers and formally appointed leaders, were asked who they felt had provided leadership for the ICT project, what leadership tasks were performed, and who had influenced the teachers in their use of ICT in the ICT project. Possibly due to the hierarchical nature of Singapore schools, in which authority and leadership is accepted and expected from officially appointed leaders (Dimmock & Tan, 2013; Goh, 2009; Hairon & Goh, 2015; Hallinger & Kantamara, 2000; Koh, Gurr, Drysdale, & Ang, 2011), the leaders identified by participants were the formal leaders. A total of seven formally appointed leaders were interviewed at least twice, including the principal, the two vice principals (VPs), the head of department (HOD)/mathematics, the HOD/ICT, the senior teacher (ST) in charge of the project, and the level manager (LM) of Grade 4.

All interviews were semistructured in nature (Merriam, 1998), in that the key questions were planned prior to the interviews but questions were crafted during the interviews to probe or build on the participants’ responses to the set questions (Seidman, 2006). In interviewing the identified leaders, besides asking leaders for their interpretations of their roles and their observed actions, leaders were asked about their working relationship with the other leaders, and about any challenges which they faced or support which they received, in an attempt to surface tensions and how such tensions were resolved. One question asked of all the leaders was “What are the challenges and issues that you have faced so far? How did you go about resolving them?” A question asked of both the principal and VPs was “What do you see as your role in the implementation of the ICT project?” This question was triangulated by asking middle managers what they perceived to be the roles played by the senior managers.

Data Analysis

An inductive approach (Patton, 2002) to coding tensions was adopted. With activity theory’s concept of system tensions as a “sensitising concept” (Charmaz, 2006), the researchers were alerted to contexts involving conflicting or overlapping needs, for example, where an individual played dual roles (e.g., the same person as subject teacher and as the person in charge of the ICT project), where different participants had different objectives (e.g., the heads of the subject departments as opposed to the heads in charge of the ICT project) or where different participants had overlapping roles or boundaries of responsibility (e.g., the ST in charge of the ICT project and the head of ICT who is in charge of ICT aspects in the entire school). Analysis of such contexts, including the challenges highlighted by the various leaders through interviews, helped surface the key tensions and attempts to work on these challenges. The tensions surfaced and strategies adopted to balance these tensions were then mapped against the activity theory components, to enable a visual analysis of where the tensions manifested within the activity system(s). Table 1 illustrates the study’s use of activity theory as an analytical lens, and how the key components were aligned to the distributed leadership perspective. Objects differ depending on the activity system.

Table 1.

Use of Activity Theory as a Lens to Analyze Distributed Leadership.

Activity Theory	Interpretation for the Study	Distributed Leadership Perspective
Subjects	The leaders involved	Multiple leaders
Outcome	The desired change in the follower or organization	Leadership as a deliberate attempt to influence people to achieve desired outcomes (Yukl, 2001)
Division of labor	Horizontal and vertical roles and relationships that affect how leadership actions are distributed	Multiple leaders in various positions performing various leadership tasks (Spillane, 2006)
Tools	Tools used by the leaders to achieve the desired changes	Included in the situation in which leadership is enacted, also referred to as artifacts (Spillane et al., 2001, 2004)
Rules/norms	The relevant rules/norms which enabled and or constrained the leadership actions	Social norms and organization routines are mentioned as shaping leadership practice and being reshaped by leadership practice (Spillane et al., 2001, 2004; Spillane, 2006)
Tensions/contradictions	Tensions between activity systems or between the components of an activity system which constrain the subject from accomplishing their intended outcomes	Consideration of how the cultural and social situation, including organizational structure, can enable or constrain leadership activity (Gronn, 2002b; Harris, 2005; Woods et al., 2004; Spillane & Orlina, 2005)

In the case study, the division of labor refers to horizontal and vertical roles and relationships within the community that affect the division of leadership tasks. The horizontal division of labor refers to actions spread across members of the community who have equivalent status, such as the various HODs, while the vertical division of labor refers to actions which are distributed up and down divisions of power, such as the division of labor between the HOD and the level head (LH) of the same subject. The tools, which could be material or psychological, would depend on the object and the desired outcome. For example, if the outcome is to influence teachers’ use of ICT, possible tools could include ICT lesson plans and ICT infrastructure. Rules include organization rules embedded in formal structures which govern the behavior of people within the community.

Selection of Case Study

Stake (1995) identified two main types of case studies: instrumental and intrinsic. The approach selected for this research was instrumental case study, as selection of the case was deliberate to answer specific research questions related to distributed leadership. The school, Greenville Elementary, was chosen from 66 schools which wrote in to the Ministry of Education (MOE), Singapore, in 2006, to propose an ICT project that they would like implement. In their proposals, schools outlined the specific ICT projects that they intended to implement, including subject area(s) and the intended roles of various school personnel.

Based on Spillane and colleagues’ argument that their theory of distributed leadership focuses on how leadership practice is distributed among both positional and informal leaders (Spillane et al., 2001), one main criterion for case selection was an indication that at least two people would be involved in the conceptualization and implementation of the ICT project. At the selection of case stage, our assumption was that the principal, HOD/ICT and the subject heads (SHs) of department would be potential leaders. This assumption was based on leadership sources identified in the literature on technology implementation in schools (Creighton, 2003; Divaharan, 2007; Langran, 2006). In Greenville Elementary, the person indicated as being in charge of the ICT project was an ST, and the team members, at the time of the proposal, included the principal, the two VPs, the HOD/mathematics and the HOD/ICT.

The other key assumption made was that in implementing a new ICT project, there would be a need to influence the teachers concerned to use ICT in their lessons, which suggested that leadership practices were likely to be observed. Although there are multiple interpretations of leadership in the literature, there appears to be concurrence that leadership involves deliberately exerting some form of influence on people in an attempt to affect their beliefs, values, and actions, so as to achieve desired and presumed shared organizational outcomes (Pounder et al., 1995). Thus, the case was selected on the basis that there was potential to study the leader-plus aspect (multiple individuals functioning as leaders) and the practice aspect (interactions among the leaders) of distributed leadership (Tian, Risku, & Collin, 2015).

Brief Description of School

In this section, a brief description of the school is provided, together with the official leadership hierarchy in Greenville Elementary as well as the main leaders involved in the ICT project. All names are pseudonyms to protect the identities of the participants and the school.

At the senior management level, Greenville Elementary had a principal and two VPs from 2005 to 2006. Middle management in the school comprised the HODs, SHs, LHs, and LMs. The title “level head (LH)” needs to be clarified as it is misleading. As opposed to the LM who is in charge of all the core subjects for one grade level, the LH is the second-in-command for a subject, just below the HOD, across all the grade levels. The LH involved in the ICT project was Sarah, LH/Science, who was simultaneously the LM for the Primary 4 level (grade 4), which was the level at which the ICT project was first implemented. As LM, Sarah was responsible for overseeing the three core subjects (English, mathematics, and science) at that level. She joined the ICT project midpoint as she was on course till June 2007. The main HODs involved in the ICT projects were the HOD/ICT, Ida, who is overall in charge of all ICT-related programs in the school and Ben Ling, the HOD/Mathematics, since the ICT project started with the use of ICT to teach mathematics. In addition, Ben Ling was deployed by the principal to teach Primary 4 the year the ICT project was first implemented. Supporting the HOD/ICT was a SH/ICT, Lai Ling, who was also deployed by the principal to teach Primary 4.

Besides the various heads, Greenville Elementary also had a number of STs, one of whom, Cassie, was appointed by the principal to be in charge of conceptualizing and leading the ICT project. The post of ST was relatively new and was created by the MOE as part of a parallel teaching track to the traditional leadership track of HOD, VP, and principal. The leadership track is the track for leadership positions in the schools while the teaching track provides advancement opportunities for teachers who make teaching excellence in the classroom the primary focus in their careers (MOE, 2016). Personnel on the leadership track, including the HOD, the LH and the SH, are the ones expected to provide “leadership.” The ST is described as a mentor and role model but the terms “leader” and “leadership” are not included in the role description (MOE, 2016).

Figure 2 illustrates the hierarchical structure of the school to show the official difference in rank of the various positions. The roles of individuals in the ICT project are indicated. The LM position is indicated in brackets as unlike the other roles, the LM position in 2008 was an internal school appointment, while the other roles were official appointments by the MOE. Although STs reported to the VPs, similar to the various heads, the position of ST was considered to be lower in the organizational hierarchy compared with that of a head. Figure 2 attempts to represent this difference in status by placing the ST below the heads, and above the teachers. In Singapore, the principal and VPs are generally addressed by their surnames as a sign of respect for their higher hierarchical position, while heads and STs are addressed by names or surnames, depending on who is addressing them, and the latter’s position in the organization hierarchy relative to the former.

Figure 2.

Hierarchical organizational structure in Greenville Elementary and staff involved in information and communications technology (ICT) project. HOD, head of department; SH, subject head; LH, level head; ST, senior teacher.

In 2006, the Lead ICT scheme was implemented by the MOE, Singapore. Lead stands for Leading Experimentation and Development in ICT. The scheme was introduced by MOE to support schools that were ready to move to a higher level of ICT use by providing funding and consultation services. The principal of Greenville Elementary applied for the scheme. At that time, the school decided to begin with the Primary 4 level and with the mathematics subject, with Cassie, ST/Mathematics, appointed to be in charge of the project. The Lead ICT project was implemented in 2007 and the researchers observed the project from preparations in August 2006 till the end of the academic year in 2007. Although the project started with the mathematics subject, it evolved to include science from June 2007 when Sarah, LH/Science, returned from a course and was tasked to assist in incorporating science into the ICT project.

Findings and Discussion

Tensions in the Distribution of Leadership

The use of system tensions as a sensitizing concept (Charmaz, 2006) revealed some tensions, particularly between the middle managers who were directly involved in the ICT project and middle managers who were less directly involved in the ICT project. These tensions were not simply personal in nature but arose due to existing norms or rules with respect to the perceived authority or spheres of influence of different roles within a school setting. Individuals, including middle managers, who simultaneously played multiple roles in different activity systems, felt obliged to abide by certain norms and rules.

Our case study surfaced four interrelated activity systems with regard to the use (or lack of use) of ICT in the school:

At the organization level (school level)

At the process level (ICT project)

At the structural level (ICT department) and

At the structural level (subject departments)

The four activity systems were interrelated in that the individuals involved had working relationships across the systems, and these relationships were influenced by the individuals’ multiple roles in the different activity systems.

The main tensions which were surfaced were (a) between the subject department activity and the ICT project activity and (b) between the ICT department activity and the ICT project activity. The tensions arose partly because members of one community were simultaneously members of another community, but the social norms or rules or the intended outcome of each community differed.

Tension between subject department activity (structural level) and ICT project activity (process level)

The main tension surfaced was that between the HODs, who required their department teachers to carry on with the dominant department activity of teaching their subject area(s), with or without the use of ICT as a tool, and Cassie (ST) and Sarah (LM) who had the mandate of influencing the Primary 4 teachers to use ICT to teach mathematics and science. Using the language of activity theory, this meant that the object and outcome of the department activity differed from the object and outcome of the ICT project activity, and the value of ICT as a tool also differed in the two activity systems. These differences, together with entrenched norms in the subject department, including the norm of heads holding more authority than an ST or LM, resulted in tension between the subject department activity and the ICT project activity, which manifested itself in conflicts between the two sets of leaders.

At the process level, the Primary 4 teachers who were involved in the ICT project community, including Cassie and Sarah, were simultaneously members of the subject department community at the structural level, and thus had to abide by their subject department norms and rules. With respect to organization structure, the Primary 4 teachers reported to their various HODs although they were simultaneously involved in the ICT project. Among middle management, while Cassie’s foremost responsibility was the ICT project and to influence the Primary 4 teachers to teach using ICT, the HODs of the various subjects had other priorities, including ensuring that their teachers taught the subject effectively, with or without the use of ICT. This fundamental difference in the leaders’ intended outcomes and the sharing of the Primary 4 teachers as objects resulted in tensions between the ICT project activity (process level) and the subject department activity (structural level), manifested in tensions between the various middle managers. Both Cassie and Sarah mentioned that they faced constraints as the desired outcome of the ICT project activity system was perceived as clashing with the subject department’s priorities and alternative outcome.

Figure 3 illustrates the two activity systems and the key components which experienced tensions (underlined). Tension is represented by a double-sided arrow.

Figure 3.

The subject department and information and communications technology (ICT) project activities and the key tension points. HOD, head of department.

The subjects in each activity system were the leaders with their desired outcomes and the key tools they used to influence teachers to achieve their outcomes. An example where there was a clash between the two sets of leaders was in the choice of instructional strategy or tool to teach the subject areas. In Greenville Elementary, before the ICT project, the HODs had started to implement activity-based learning as the key pedagogical tool to influence and guide teachers in the design of lessons. According to Cassie, the activity-based lessons were the “brainchild” of all the HODs (Second interview). Thus, Cassie felt that if she wanted to replace an existing activity-based lesson with an ICT lesson, she needed the support of the VP, Mrs. Ang, to get the approval of the HODs:

because sometimes in our school, we implement activity-based lessons for the Primary 4, so certain things if you want to change, for example, this topic, . . . if you want to [incorporate] ICT, you must get the consensus of the HODs . . . so Mrs. Ang [VP] will sort of come in to help the HODs see certain aspects because certain things it’s not so nice for me to say. (Second interview)

Normally Lead ICT is led by an ICT HOD or a HOD so they may not have some problems which I face because I’m a senior teacher, all right? Because as a senior teacher, your empowerment sometimes is not there, right, so certain things I know is a bit touchy or sensitive, it’s easier to work through Mrs. Ang [VP]. (Second interview)

Although she was officially assigned by the principal to lead the ICT project, Cassie was simultaneously an ST and a mathematics teacher, whose status was lower in the school hierarchy structure compared to a HOD. In addition, while she did not ‘report’ to the heads in that they were not her reporting officers and were not involved in appraising her performance, she was nevertheless a subject discipline teacher who had to abide by subject department norms and rules. In addition, her sphere of influence was mainly confined to the use of ICT for mathematics. Mathematics-related curriculum decisions had to be made by the HOD/mathematics. For example, Cassie had to seek the HOD/mathematics’ approval for the mathematics topics to include in the ICT project:

I did talk to Ben Ling, I mean, Mathematics HOD, about … which topics [we can teach using ICT]. (Cassie, Second interview)

Cassie’s explicit reference to Ben Ling as HOD/mathematics was revealing because she knew that the researcher interviewing her was aware of Ben Ling’s official position, and it was not necessary to inform the researcher about Ben Ling’s position. There was another incident when Cassie, in referring to Ben Ling, implied that the latter’s higher status in the school’s hierarchy structure placed some constraint on Cassie’s leadership effort to influence Ben Ling, who was also deployed by the principal as a Primary 4 teacher, to use ICT. Cassie shared her reluctance to ask Ben Ling, who is her HOD/mathematics, to develop ICT lesson plans although she had asked the other Primary 4 teachers to do so:

You know the people I asked . . . I asked the beginning teachers or the very young teachers . . . I also feel bad to tell . . . middle management what you need to do, so you noticed I didn’t tell Ben Ling [to develop ICT lessons]. (Cassie, Third interview)

Thus, it was clear that Cassie felt her ability to influence the teachers was constrained by the higher authority of the HODs, and she also felt that she needed the assistance of the VP to influence the HODs to support the teachers’ use of ICT.

In the case of Sarah, although Sarah’s role as LH of science was officially regarded as being higher in the organization structure than Cassie’s role as an ST, Sarah also felt that “my role there are limitations because I’m still under the HOD” (First interview). Sarah was “under the HOD” in more sense than one. Like the other Primary 4 Science teachers, Sarah reported to the science HOD and her role as LH of science was lower in the organization hierarchy compared with a HOD:

It’s very hard, you know. Because their own department [referring to the HODs] they also protect their own, you know. You see, they have their own programmes [such as the activity-based lessons] . . . they don’t want you to upset their programmes. (Sarah, LH/science, First interview)

Sarah, like Cassie, felt that if the school wanted ICT to “move,” senior management had to get the “buy in from the [subject] HOD first” (First interview) since HODs report to senior management. The existing HODs were “not so advanced in thinking when you talk about ICT” (Sarah, First interview). They were acknowledged by both Mrs. Ang and the principal to be “very safe people” (Mrs. Ang, VP, Second interview) who were “conservative” (Ms. Wong, Principal, Second interview) and did not like to take risks, such as using ICT when other traditional instructional tools, such as activity-based learning, had served their needs well. Similar to Cassie, Sarah felt there was a limitation to her influence on both the teachers and on the HODs.

In activity theory, while tensions generate conflicts and disturbances, they also generate innovative attempts to balance these tensions (Engestrom, 2001). Although the findings suggest that Cassie’s and Sarah’s leadership actions and sphere of influence were to some extent constrained by the department structure and norms, there was evidence of senior management using the organization/department structure and norms to counter anticipated tensions and achieve the desired outcome of teachers using ICT. This is consistent with the understanding in activity theory that the relations between subject and object are not direct or static, but are continuously mediated by various components of the activity level (system), including rules/norms (Yamagata-Lynch, 2001).

Effort to resolve the tension between the outcomes

The key tension was between the outcomes of the two activity systems, which arose due to the conflicting needs/motives of the two systems. Cognizant that the HODs had other priorities, and that there was some resistance among the heads to the use of ICT, one critical role which senior management played was to work with or through the various HODs to provide teachers and the other middle managers with the necessary support. The principal agreed that it was the role of senior management to speak to the HODs if they posed obstacles to the efforts of the other middle managers in championing the use of ICT. She made it clear to the HODs that they “have to keep abreast,” “have to run faster,” and that despite their lack of ICT knowledge, they “mustn’t be seen as an obstacle” but could seek to “help in other ways,” such as applying for money and helping with procurement (Principal, Second interview).

At the September 15 meeting in 2006, Ms. Wong, the principal, assured the teachers that senior management would to the following:

keep on consistently reminding the [subject] HODs . . . there’s the Lead ICT project . . . please, you know, where Primary 4 is concerned, give priority [to the use of ICT]. . . . But we try not to add in any more new [initiatives] . . . at least for Primary 4. (September 15)

Similarly, the VP, Mrs. Ang, shared that she had spoken to the HODs of the need to moderate their expectations of teachers, to be willing to let go of what they normally expect the Primary 4 teachers to do, because of the extra time and effort which the Primary 4 teachers were spending on the Lead ICT project:

If you [the HODs] keep expecting the teachers to do the SAME things and yet carry out Lead [the ICT project], it is just impossible. We are just . . . setting the teachers up for failure. So I actually say to the HODs that you must empower, let the teachers try [using ICT] and let’s say if they don’t do certain things [such as the activity-based lessons], let them be. As long as they are responsible, the children are learning, let it be. (Mrs. Ang, VP, Second interview)

Mrs. Ang also reassured both the relevant heads and the teachers that it was acceptable if there was an initial drop in academic results as a result of the teachers spending more time and effort to use ICT to engage the students in learning:

Whether the results will drop or not, is really not a top priority. We want them [the teachers] to really use this Lead ICT [project] to promote greater engagement in the classroom. (Liz, VP, Second interview)

Thus, senior management played an important role in managing organization boundaries to support middle managers who were leading the use of ICT but were lower in the organization hierarchy, by reinforcing the school’s priority in pushing for the use of ICT. Senior management were able to play such a role by deliberately harnessing their higher authority in the organization hierarchy to influence the middle managers.

In addition, senior management strategically deployed the HOD/mathematics to teach Primary 4, thus assigning her the simultaneous role of a Primary 4 teacher (a subordinate member role) in the ICT Project community. This was an example where the duality of role was deliberate and was engineered by leaders higher up in the hierarchy—senior management. The mathematics head’s involvement in both the subject department (structural level) and the ICT project (process level) communities and activities resulted in her supporting the ICT project as a HOD at the department level, thus minimizing potential tensions between the mathematics department activity and the ICT project activity, enabling Cassie to lead the ICT project. That this was a deliberate decision was revealed by both Ben Ling, the HOD/mathematics, and senior management. Ben Ling revealed that “Actually in the first place, both of us [herself and the LH for mathematics] were not [deployed to teach] in Primary 4 at all, then Ms. Wong [the principal] insisted that one of us go there” (Second interview). When senior management were asked why Ben Ling was deployed to teach mathematics at the Primary 4 level, three reasons were given, one concerning her professional growth, with the other two reasons related to the need for the HOD/mathematics to provide support and monitor the ICT project, by being personally involved in the ICT project as a teacher:

So we told our HODs: “Look, you cannot be behind. You have to come along.” That’s why Ben Ling is pulled in.” (Ms. Wong, Principal, Second interview)

She [Ben Ling] needed to know what is happening. . . . Even [when deployed] to teach Primary 4, she just knows what is happening. She really doesn’t have a very active role in it. So can you imagine if she’s not even there? [as in not even teaching the Primary 4 pupils mathematic]. (Mrs. Ang, VP, Second interview)

By deploying Ben Ling to teach mathematics, she became a member of the ICT project community and her outcome as a head of mathematics was expanded to include getting her teachers to teach mathematics effectively using ICT, which was part of the outcome of the ICT project activity. That this became an important part of her outcome was clear when she made the decision to remove one topic, “tessellation,” from the mathematics examination at the end of the year, so that the teachers had more time to use ICT to teach the existing topics, without having to worry about the students’ performance at the examination.

Effort to resolve potential tension between rules/norms

In activity theory, it is suggested that when an activity system adopts a new element from the outside, it can lead to tensions and attempts to change the activity (Engestrom, 2001). When Greenville Elementary decided to participate in the Lead ICT project, the ICT project was the new element within the school activity system. At the school level, senior management used their institutional authority to create a new division of labor to cater to this new element. They assigned to Cassie, a senior teacher, the official title of being in charge of the ICT project. They also assigned to Sarah the role of the Primary 4 LM, in addition to her existing role as LH of science, to support their aim to develop more teachers to use ICT in different subject areas. This seemingly minor leadership action in terms of staff deployment was critical because of the social norm in Greenville Elementary, which is arguably the norm in many Singapore schools, in which teachers generally show respect for authority and rank and will follow instructions given by people with official roles and positional authority (Dimmock & Tan, 2013; Goh, 2009; Hallinger, 2010). Dimmock and Tan (2013) refer to this in their conceptual article on what contributes to educational success in Singapore as the leader-teacher social compact. Certainly, while the teachers acknowledged Cassie’s expertise in ICT, which is necessary for ICT leadership, they also referred to and deferred to her as being “in charge” of the ICT project (Azman, Interview; Dean, Second interview). Thus, while Cassie experienced some constraint in her exercise of leadership for the ICT project with regard to the HODs, she did not encounter much resistance from the Primary 4 teachers.

At the ICT project level, Sarah exerted her influence on the new role of LM, which senior management had assigned to her, by interpreting it in a way that broke away from the established norm/rule in Greenville Elementary of subject committee members developing lesson plans only for their respective subjects. A potential barrier to the Primary 4 teachers’ development of information and communications technology lessons lay in the norm/rule in Greenville Elementary that only teachers who belong to a subject committee would be involved in developing lesson plans for that subject. This was a potential tension between an accepted department structure and norm of distributing work, and the object of teachers’ increased competency in developing ICT lessons for all subjects.

This “practice” or “structure” (e-mail from Cassie to researcher on February 5, 2008) began way back in the 1990s. That this was an integral and unquestioned norm was evident in the discourse during the key planning meeting on August 19, 2006. Even Cassie did not question this territorial mentality:

Now you have to think in terms of the mathematics committee, you know, because crafting lessons is always done by the mathematics committee. The mathematics committee should be helping to craft [the ICT lessons for mathematics]. (Cassie, August 19, 2006)

There was no doubt expressed during this meeting, held the year before the ICT project was implemented, that it would be the mathematics committee teachers who would be crafting the ICT lesson plans. There is also evidence that Mrs. Ang, the VP, was aware and concerned that if the ICT Project were positioned as purely a mathematics project, it could limit the development of ICT lessons to mathematics lessons and by only the teachers who were members of the mathematics committee, since this was aligned to the established working norm:

Actually I think it started with mathematics and we wanted the ICT project to be spearheaded by the mathematics committee… but I feel that if we go in that direction, then the ownership by the Primary 4 teachers may not be so great . . . I told Cassie. I think we don’t go that way. Let’s take this whole project as the P4 teachers’ project. Whether it’s mathematics, English or science, never mind. It’s the P4 teachers’ projects, and not just the mathematics committee project. (Mrs. Ang, VP, Second interview)

Once the ICT project started, the researchers observed that all the Primary 4 teachers, regardless of the subject committees they belonged to, were involved in developing ICT lesson plans for both mathematics and science. In appointing Sarah as the Primary 4 LM whose ‘official’ responsibility was to “oversee the three subjects [English, mathematics, and science]” (E-mail from Sarah to researcher on February 4, 2008), senior management invested Sarah with the “power” and authority to work with the teachers who were from different subject committees. The role of Primary 4 LM is similar to a “restructured leadership role” (Little, 1995) which spans subject boundaries. This was another strategy by which senior management managed organization boundaries.

Although school leaders harnessed the hierarchical structure to legitimize Sarah’s work with the Primary 4 teachers, Cassie realized that it was Sarah’s agency in interpreting her role as a LM in such a way that it enabled her to involve the teachers in developing ICT lesson plans for different subjects, regardless of the subject department they belonged to. As Cassie pointed out, it was possible for Sarah, as the LM, to have simply continued with the traditional norm of working by saying “If English [committee member], you must develop English [resources].” Cassie realized that it was partly due to Sarah’s personal agency in being willing to break out of the traditional silo department mentality and norm that the position of the LM enabled her to provide leadership for the ICT project across subjects. Sarah’s interpretation of the role of LM illustrates how a leader could innovatively go against norms to impact his/her relationship with the followers. In fact, Sarah initiated a new norm when she involved all the Primary 4 teachers, regardless of their subject departments, in the development of ICT lessons across all subjects, although this was not part of the department activity system’s outcome. The teachers’ acceptance of this new norm could explain why, at the teachers’ level, although they played the dual roles of ICT project member and subject department teacher, the teachers did not express any feeling of tension between these two roles.

Tension between ICT department activity (structural level) and ICT project activity (process level)

Another tension which arose was due to the existence of two simultaneous activity systems which supported the use of ICT in the school, one at the process level and one at the structural level: the Primary 4 ICT project activity and the ICT department activity. In this situation, although both activity systems shared the same object and outcome, the intended complementary division of labor among the leaders failed to materialize as some leaders (subjects) lacked the relevant expertise to carry out the required leadership/managerial tasks. In addition, the leader (subject) of the ICT project activity was simultaneously a member in the ICT department community, who as a rule reports to the HOD/ICT.

Based on the school’s organizational structure and officially espoused roles, the HOD/ICT (Ida) and SH/ICT (Lai Ling), both leaders of the school’s ICT department, should have been taking the lead in planning and providing relevant ICT resources for the overall implementation of ICT in Greenville Elementary, including the Lead ICT project. Ideally, the division of labor should be complementary in nature, as described by Azman, a teacher:

Because Mrs. Sng [Cassie] is the one in charge of the ICT project with her knowledge of ICT, but in school . . . the one in the ICT department will be Mrs. Tan [Lai Ling]. So she [Cassie] has to be backed up by Mrs. Tan, whatever Mrs. Sng wants, Mrs. Tan can also help.

If we analyze the ICT department and the ICT project activity systems, from the perspective of the designed organization (Spillane, Camburn, & Pustejovsky, 2008), there should be no tension between the systems as they are intended to be complementary in nature, since they share the same intended outcome of teachers using ICT in their lessons. Figure 4 shows the two activity systems and the tensions (underlined) which surfaced in the division of labor despite the shared outcomes in this context.

Figure 4.

The information and communications technology (ICT) department and ICT project activities: officially shared outcomes and tools. HOD, head of department; SH, subject head

In reality, in what Spillane et al. (2008) refer to as the “lived” organization, perhaps “because the HOD/ICT is not entirely into the ICT part” and “may not be very familiar with ICT” (Ms, Lee, VP, First and second interviews), from Cassie’s perspective as the person leading the ICT project, the ICT department was slow in providing the ICT resources she required. That is, from her perspective, the official division of labor as implied by the designed organization was not working.

Effort to resolve the tension in the division of labor

As the leader in charge of the ICT project and since she possessed the relevant expertise, Cassie thus often stepped in to get things moving on the infrastructure side although her official role was to provide instructional leadership:

Frankly speaking, she [Mrs. Ang, the VP] wants me to focus more on the teachers . . . she has been harping on that, but I have told her if infrastructure is not in place, I also cannot move much. (Cassie, E-mail to Researcher 1, November 3)

In addition, Cassie resolved her problem by approaching the principal or VP for help when she had difficulty going through the “right channel” (Second interview). She thus harnessed the norm that the school leaders possess the most authority in a school’s context, to override the authority of the HOD/ICT and SH/ICT. Unfortunately, one result of Cassie intervening in the infrastructure aspect was the creation of tension between her and Lai Ling, SH/ICT, who felt that Cassie was “STILL the mathematics ST and not [overall in charge of] ICT although she is in the ICT department” (Lai Ling, Third interview). Lai Ling in a sense identified the conflict that Cassie experienced in her dual roles as a member in the ICT department community, who should defer to the HOD/ICT and the SH/ICT, and as the person in charge of the ICT project who reports directly to senior management. If one follows the norm of the school, HODs and SHs are higher in the hierarchical structure, and thus Cassie should not be overriding the former.

She’s [Cassie is] a member there [in the ICT department]. But she is somehow the mathematics senior teacher so to a certain extent if you would look at the hierarchy, it would be Ida [HOD/ICT] then me, then she will be below . . . but she is doing everything that we are, above us and certain things she doesn’t [get permission from the HOD/ICT], she gets permission from the top [senior management]. . . . So sometimes, we don’t know who is the HOD/ICT right? [laughed]. (Lai Leng, Third interview)

As one of the researchers observed in her analysis of the third interview of Lai Ling, there seemed to be a tension between two roles: “the leader of an initiative [the ICT project] vis-a-vis the hierarchical organizational role [HOD/ICT and SH/ICT]”. Cassie herself felt frustrated that she could not rely on the official division of labor; she felt the need to interfere with the infrastructural set up for the ICT project but she felt that this interference took time away from her role as an ST, which was to develop the teachers:

I must call myself a “kaypoh” [a colloquial term for someone who interferes when it is not his business to do so] in this area . . . perhaps it is wiser to leave it to the HOD to see to such matters. With Lead ICT, there is actually a lot of work to be done . . . better not to give myself additional stress. (Cassie’s e-mail dated July 27)

[Infrastructural issues] take up time and . . . that is also not my territory you see, so I have to go through so many channels to get things done. (Cassie, Second interview)

In summary, there was a tension between the two activity systems, when the ICT department activity system (structural level) failed to deliver the support and resources required for the ICT project activity (process level), resulting in Cassie feeling “compelled” to take on a role which was not within her role as the person in charge of the ICT project and in discord with her role as an ICT department member. The intended division of labor between the ICT department leaders and the ICT project leaders did not materialize. In addition, the tension was a result of the perception that the ICT department and the ICT project constituted different territories/boundaries of responsibility, which were not meant to overlap. Cassie’s interference in the management of ICT resources was perceived by the ICT department as an invasion of their “territory” (Cassie, Second interview). If the designed organization had worked as intended, as shown in Figure 3, the two territories would have had complementary functions. However, when the intended division of labor did not materialize, and Cassie had to take over tasks which did not belong to her boundary of responsibility and which went against the norm of her reporting to the HOD/ICT, this resulted in tensions between the two activity systems.

Conclusion

This article contributes to the literature on distributed leadership in three significant ways. First, boundary-spanning aspects of work give rise to tensions in the distribution of leadership. The existence of tensions in the case study is accentuated because a tool like ICT can be used in various subject areas, spanning subject boundaries. This boundary-spanning aspect of work specifically gave rise to structural and normative tensions. This structural-related tension is exacerbated when the officially appointed leaders lack the expertise to exercise their leadership roles. In the case study, the HOD/ICT and the HOD/mathematics and HOD/science lacked the expertise to promote the use of ICT, resulting in a situation where an ST and LH who were lower in the organizational hierarchy had to be assigned to lead the ICT project. This unusual division of labor created disturbances, and thus tensions, in the leadership activity system.

Normative tensions surfaced due to conflicting priorities, perceived distinct boundaries of responsibility and organizational norms. The first two tensions resonate with two of the three issues that Storey (2004) surfaced in a rare study focusing on the problem of distributed leadership in schools. These tensions also resonate with activity theory’s proposition that tensions arise due to conflicting or overlapping needs.

The tensions related to organizational norms are not new in that the literature has long acknowledged that organizational rules govern the behavior of actors by prescribing role relations (Ogawa, 2004). When there is an attempt to distribute leadership in a school based on expertise, beyond the traditional formal roles, it is not surprising that tensions arise (Oswald & Engelbrecht, 2013). The contribution of this article is in identifying the tensions encountered and resolved by multiple leaders in the distributed leadership process. This is different from the hierarchical leader (e.g., principal) –subordinate (e.g., SH) tension that the literature has mainly focused on (Brown, Rutherford, & Boyle, 2000; Storey, 2004; Torrance, 2013). Cassie and Sarah traded in their subordinate roles in their respective departments to become leaders in the ICT project, while the hierarchical positional leaders HOD/ICT, Ida, and HOD/Math, Ben Ling assumed subordinate roles in the ICT project. Although the HODs appeared to be taking on subordinate roles, the reality is that they still retained their positional roles as leaders. In this sense, the tensions encountered were between leaders, specifically between leaders appointed to lead a process (Cassie and Sarah appointed to lead the ICT project) and positional leaders who are part of the organization structure (Ida and Ben Ling).

Second, tensions provide a context for the exercise of influence by leaders—demonstrated through innovative solutions to counter the tensions. While tensions may constrain the distribution of leadership, the existence or possibility of tensions can also give rise to innovative solutions to balance such tensions, so that the activity system can continue to function and evolve. The findings from this study highlight the following ways in which the leaders attempted to resolve tensions in the distribution of leadership. Senior management used mainly the contextual influences of organization structure, cultural norms, and staff deployment (Gurr, Drysdale, & Mulford, 2005) to counter tensions which they had anticipated could arise due to the existing hierarchical structure and differences in power between the different middle managers. They did this partially by enabling a spanning of organization boundaries, for example, through creating the restructured role (Little, 1995) of the Primary 4 level manger, which enabled Sarah to provide leadership across the three core subjects. The LM position enabled Sarah to play the role of a “boundary spanner” to connect different communities (Coldren & Spillane, 2007) to support the use of ICT, with the different communities being the three subject departments.

Senior management themselves also played the role of boundary spanners, between HODs and lower ranking middle leaders, as well as between heads and teachers. In the literature on middle management in schools, it is usually the HODs who act as the bridge between top-down initiatives and implementation by teachers in the classrooms (Busher & Harris, 1999; Gunter, 2001; Timperley, 2005). However, such a relationship is based on the assumption that initiatives are driven by senior management, usually the principal. When it comes to instructional reforms which require specific expertise, particularly ICT innovations, it is very likely that the initiative is driven by a middle manager, or even by a teacher (Collis & Moonen, 1994). In such a scenario, coupled with a hierarchical society in which positional authority is important (Ng et al., 2005) and in which zones of indifference are wide (Blackbourn & Wilkes, 1987; Henderson, 1990; Morris, 1990; Wilkes & Blackbourn, 1981; Wilkes & Love-Wilkes, 1989), it makes sense that senior management has to play an active role in supporting leadership by those who are lower in the hierarchy, by playing the role of mediator between the latter and HODs, who currently occupy the highest position among middle management in a school setting.

There was evidence that the agency of individual leaders could be used to challenge existing norms and create their own norms, enabling them to lead. While senior management created new roles to distribute leadership to those with expertise and interest in ICT, human agency was observed in Sarah’s use of her role as LM to involve all the Primary 4 teachers in developing ICT lessons, contrary to the established norm of teachers developing lessons only for their main subject areas. This case study confirms the importance of studying the relation between structure and human agency in examining the distribution of leadership (Gronn & Ribbins, 1996; Spillane et al., 2004). The study showed how structure, mainly organization structure and its accompanying norms and rules, constrained human agency. At the same time, the case study showed how agency could act through structure (Spillane et al., 2001). Spillane et al. (2001) note that while organization structures and routines constitute school leaders’ activities, these structures are also created/recreated by the actions of leaders and others who work in schools. Similarly, Harris (2013) shared research by Day et al. (2009) that heads in schools orchestrated the structural and cultural conditions which supported the distribution of leadership. The idea of heads orchestrating conditions further supports the importance of agency.

Third, the distribution of leadership is dynamic with the centrality of leadership shifting between multiple levels of leaders as they interact in intersecting activity systems. In the traditional hierarchical chain of command, a designated leader is assumed to be the locus or central authority with regard to the designated area of work. In Cassie’s case, it was her appointed role as the person in charge of the Lead ICT reform, where she was expected to have the authority to make decisions and intervene to resolve problems related to the project. However, Cassie encountered difficulty exercising her authority over the HOD/ICT and SH/ICT as these leaders were higher in the hierarchical chain of command. This resulted in Cassie shifting the centrality of authority back to the principal and VP who stepped in as mediator between Cassie and the HOD/ICT and SH/ICT. The principal and VP provided the voice or “right channel” which endorsed Cassie’s authority. This resulted in the dynamic shifting back of the centrality of leadership back to Cassie.

Finally, in this article, the sensitizing concept of system tensions in activity theory was used to analyze a particular case of distributed leadership to shed some light on the kinds of tensions which arose and how such tensions were balanced. One concern in the use of activity theory as an analytical lens in the study of leadership is the danger of researchers getting lost in the complexity and interactions of the various system components which mediate the relationship between the subject and the object. In using activity theory to study distributed leadership, it is critical to keep in focus the actions by leaders (the subject) as they seek to influence their followers (the object), and to understand the thinking behind these actions (Spillane et al., 2001). If this requirement is met, activity theory has value in enabling researchers to move beyond linear causation models to depict social phenomena, including distributed leadership, as “multiple . . . interacting elements” (Minnis & John-Steiner, 2001, p. 304), with the centrality of leadership continually shifting between multiple levels of leaders as they interact in intersecting activity systems.

Footnotes

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Jeanne Ho, previously a vice-principal, is currently on secondment to the National Institute of Education, Singapore. Her research interests include computer-mediated learning, teacher professional development, and school leadership.

David Ng is associate professor for policy and leadership studies at the National Institute of Education, Singapore. His areas of specialization include school leadership, principalship, and learning technology.

References

Barab

S. A.

Barnett

Squire

(2002). Developing an empirical account of a community of practice: Characterizing the essential tensions. Journal of the Learning Sciences, 11, 489-542.

Barab

S. A.

Barnett

Yamagata-Lynch

Squire

Keating

(2002). Using activity theory to understand the systemic tensions characterizing a technology-rich introductory astronomy course. Mind, Culture, and Activity, 9, 76-107.

Barab

S. A.

Schatz

Scheckler

(2004). Using activity theory to conceptualize online community and using online community to conceptualize activity theory. Mind, Culture, and Activity, 11, 25-47.

Bennett

Newton

Wise

Woods

Economou

(2003). The role and purpose of middle leaders in schools. Nottingham, England: National College for School Leadership.

Bennett

Wise

Woods

Harvey

J. A.

(2003). Distributed leadership. Retrieved from http://www.ncsl.org.uk/media/3C4/A2/distributed-leadership-literature-review.pdf

Blackbourn

J. M.

Wilkes

S. T.

(1987, November). The prediction of teacher morale using the supervisory conference rating, the zones of indifference instrument and selected personal variables. Paper presented at the annual meeting of the Mid-South Educational Research Association, Mobile, AL.

Brown

Rutherford

Boyle

(2000). Leadership for school improvement: The role of the head of department in UK secondary schools. School Effectiveness and School Improvement, 11, 237-258.

Busher

Harris

(1999). Leadership of school subject areas: Tensions and dimensions of managing in the middle. School Leadership & Management, 19, 305-317.

Charmaz

(2006). Constructing grounded theory: A practical guide through qualitative analysis. London, England: Sage.

10.

Coldren

A. F.

Spillane

J. P.

(2007). Making connections to teaching practice: The role of boundary practices in instructional leadership. Educational Policy, 21, 369-396.

11.

Collis

Moonen

(1994). Leadership for transition: Moving from the special project to systemwide integration. In Kearsley

Lynch

(Eds.), Educational technology leadership perspectives (pp. 113-135). Englewood Cliffs, NJ: Educational Technology.

12.

Copland

M. A.

(2003). Leadership of inquiry: Building and sustaining capacity for school improvement. Educational Evaluation and Policy Analysis, 25, 375-395.

13.

Creighton

(2003). The principal as technology leader. Thousand Oaks, CA: Corwin Press.

14.

Creswell

J. W.

(2005). Educational research: Planning, conducting, and evaluating quantitative and qualitative research (2nd ed.). Upper Saddle River, NJ: Merrill.

15.

Dimmock

Tan

C. Y.

(2013). Educational leadership in Singapore. Journal of Educational Administration, 51, 320-340.

16.

Divaharan

(2007). Secondary school socio-cultural context influencing teachers’ type of ICT use: A case study approach (Unpublished doctoral dissertation). Nanyang Technological University, Singapore.

17.

Engestrom

(1999). Expansive visibilization of work: An activity-theoretical perspective. Computer Supported Cooperative Work, 8, 63-93.

18.

Engestrom

(2000). Activity Theory as a framework for analyzing and redesigning work. Ergonomics, 43, 960-974.

19.

Engestrom

(2001). Expansive learning at work: Toward an activity theoretical reconceptualization. Journal of Education and Work, 14, 133-156.

20.

Goh

(2009). “Parallel leadership in an “unparallel” world:” Cultural constraints on the transferability of Western educational leadership theories across cultures. International Journal of Leadership in Education, 12, 319-345.

21.

Gronn

(2000). Distributed properties: A new architecture for leadership. Educational Management Administration & Leadership, 28, 317-338.

22.

Gronn

(2002a). Distributed leadership. In Leithwood

Hallinger

(Eds.), Second international handbook of educational leadership and adminstration (Vol. 2, pp. 653-693). Norwell, MA: Kluwer Academic.

23.

Gronn

(2002b). Distributed leadership as a unit of analysis. Leadership Quarterly, 13, 423-451.

24.

Gronn

(2008). The future of distributed leadership. Journal of Educational Administration, 46, 141-158.

25.

Gronn

Hamilton

(2004). “A bit more life in the leadership:” Co-principalship as distributed leadership practice. Leadership and Policy in Schools, 3, 3-35.

26.

Gronn

Ribbins

(1996). Leaders in context: Postpositivist approaches to understanding educational leadership. Educational Adminstration Quarterly, 32, 452-473.

27.

Gunter

(2001). Leaders and leadership in education. London, Englsnd: Paul Chapman.

28.

Gurr

Drysdale

Mulford

(2005). Successful principal leadership: Australian case studies. Journal of Educational Administration, 43, 539-551.

29.

Hairon

Goh

(2015). Pursuing the elusive construct of distributed leadership: Is the search over? Educational Management Administration & Leadership, 43, 693-718.

30.

Hallinger

(2010). Making education reform happen: Is there an “Asian” way? School Leadership & Management, 30, 401-418.

31.

Hallinger

Kantamara

(2000). Educational change in thailand: Opening a window onto leadership as a cultural process. School Leadership & Management, 20, 189-205.

32.

Harris

(2003). Teacher leadership as distributed leadership: Heresy, fantasy or possibility? School Leadership & Management, 23, 313-324.

33.

Harris

(2005). Distributed leadership. In Davies

(Ed.), The essentials of school leadership (pp. 160-172). Thousand Oaks, CA: Sage.

34.

Harris

(2013). Distributed leadership: Friend or foe? Educational Management Administration & Leadership, 41, 545-554.

35.

Harrison

N. L. T.

(2005). The impact of distributed leadership on teachers (Unpublished doctoral dissertation). Indiana University, Bloomington.

36.

Hartley

(2009). Education policy, distributed leadershipand socio-cultural theory. Educational Review, 61, 139-150.

37.

Hartley

(2010). Paradigms: How far does research in distributed leadership “stretch”? Educational Management Administration & Leadership, 38, 271-285.

38.

Henderson

J. C.

(1990, November). The relationship between teachers’ personality factors and Their compliance with administrative directives. Paper presented at the meeting of the Mid-South Educational Research Association, New Orleans, LA.

39.

J. P. Y.

Chen

V. D.-T.

(2016). Distributed leadership through the lens of activity theory. Educational Management Administration & Leadership, 44, 814-836.

40.

Kelceoglu

(2006). An exploratory study of first year elementary teachers’ utilization of technology (Unpublished doctoral dissertation). Ohio State University, Columbus.

41.

Koh

H. H.

Gurr

Drysdale

Ang

L. L.

(2011). How school leaders perceive the leadership role of middle leaders in Singapore primary schools? Asia Pacific Education Review, 12, 609.

42.

Langran

(2006, March). Technology leadership: Principals and technology coordinators working together. Paper presented at the Society for Information Technology and Teacher Education International Conference, Orlando, FL.

43.

Lincoln

Y. S.

Guba

E. G.

(1985). Naturalistic Inquiry. Beverly Hills, CA: Sage.

44.

Little

J. W.

(1995). Contested ground: The basis of school leadership in two restructuring high schools. Elementary School Journal, 96, 47-63.

45.

Marken

J. A.

(2006). Activity Theory as a lens for considering culture: A descriptive case study of a multinational company developing and supporting training around the world. Bloomington: Indiana University.

46.

Mayrowetz

(2008). Making sense of distributed leadership: Exploring the multiple usages of the concept in the field. Educational Administration Quarterly, 44, 424-435.

47.

Merriam

S. B.

(1998). Qualitative research and case study applications in education: Revised and expanded from case study research in education. San Francisco, CA: Jossey-Bass.

48.

Ministry of Education. (2016). Enhancing your strengths — career tracks. Retrieved from https://www.moe.gov.sg/careers/teach/career-information

49.

Minnis

John-Steiner

V. P.

(2001). Are we ready for a single, integrated theory? Human Development, 44, 296-310.

50.

Morris

H. J.

(1990, November). Administrative directives: A measure of the degree of acceptance as perceived by classroom teachers in public community colleges in Mississippi. Paper presented at the Annual Conference of the Southern Regional Council on Educational Administration Atlanta, GA.

51.

Jeyaraj

Lim

S. P.

Lee

Goh

Chew

(2005, May). Parallel leadership for school improvement in Singapore: A case study on the perceived roles of school principals. Paper presented at the International Conference on Education: Redesigning Pedagogy: Research, Policy, Practice, Singapore.

52.

Ogawa

R. T.

(2004). Leadership as a social construct: The expression of human agency within organizational constraint. In English

F. W.

(Ed.), The Sage handbook of educational leadership (pp. 89-108). Thousand Oaks, CA: Sage.

53.

Oswald

Engelbrecht

(2013). Leadership in disadvantaged primary schools: Two narratives of contrasting schools. Educational Management Administration & Leadership, 41, 620-639.

54.

Patton

M. Q.

(2002). Qualitative research and evaluation methods (3rd ed.). Thousand Oaks, CA: Sage.

55.

Pounder

D. G.

Ogawa

R. T.

Adams

E. A.

(1995). Leadership as an organisation-wide phenomena: Its impact on school performance. Educational Adminstration Quarterly, 31, 564-588.

56.

Robinson

V. M.

(2008). Forging the links between distributed leadership and educational outcomes. Journal of Educational Administration, 46, 241-256.

57.

Seidman

(2006). Interviewing as qualitative research (3rd ed.). New York, NY: Teachers College Press.

58.

Spillane

J. P.

(2005). Primary school leadership practice: How the subject matters. School Leadership & Management, 25, 383-397.

59.

Spillane

J. P.

(2006). Distributed leadership. San Francisco, CA: Jossey-Bass.

60.

Spillane

J. P.

Camburn

Pustejovsky

(2008). Taking a distributed perspective: Epistemological and methodological tradeoffs in operationalizing the leader plus aspect. Retrieved from http://www.sesp.northwestern.edu/docs/FINAL_Operationalizing_DL-4_Mar0620_08_1_.pdf

61.

Spillane

J. P.

Diamond

J. B.

Jita

(2003). Leading instruction: The distribution of leadership for instruction. Journal of Curriculum Studies, 35, 533-543.

62.

Spillane

J. P.

Halverson

Diamond

J. B.

(2001). Investigating school leadership practice: A distributed perspective. Educational Researcher, 30(3), 23-28.

63.

Spillane

J. P.

Halverson

Diamond

J. B.

(2004). Towards a theory of leadership practice: A distributed perspective. Journal of Curriculum Studies, 36, 3-34.

64.

Spillane

J. P.

Orlina

E. C.

(2005). Investigating leadership practice: Exploring the entailments of taking a distributed perspective. Leadership and Policy in Schools, 4, 157-176.

65.

Stake

R. E.

(1995). The art of case study research. Thousand Oaks, CA: Sage.

66.

Storey

(2004). The problem of distributed leadership in schools. School Leadership & Management, 24, 249-265.

67.

Tian

Risku

Collin

(2015). A meta-analysis of distributed leadership from 2002 to 2013: Theory devlopment, empirical evidence and future research focus. Educational Management Administration & Leadership, 44, 146-164. doi:10.1177/1741143214558576

68.

Timperley

H. S.

(2005). Distributed leadership: Developing theory from practice. Journal of Curriculum Studies, 37, 395-420.

69.

Torrance

(2013). Distributed leadership: Challenging five generally held assumptions. School Leadership & Management, 33, 354-372.

70.

Vennebo

K. F.

(2016). Innovative work in school development: Exploring leadership enactment. Educational Management Administration & Leadership. Advance online publication. doi:10.1177/1741143215617944

71.

Wilkes

S. T.

Blackbourn

J. M.

(1981). The design of an instrument to measure zones of indifference of teachers to directives issued by administrators. Starkville: Mississippi State University.

72.

Wilkes

S. T.

Love-Wilkes

(1989). The relationship of administrators’ experience and teachers’ acceptance of authority in Mississippi schools. Retrieved from http://search.ebscohost.com.libproxy.nie.edu.sg/login.aspx?direct=true&db=eric&AN=ED312759&site=ehost-live

73.

Woods

P. A.

Bennett

Harvey

J. A.

Wise

(2004). Variabilities and dualities in distributed leadership. Educational Management Administration & Leadership, 32, 439-457.

74.

Yamagata-Lynch

(2001). Using activity theory for the sociocultural case analysis of a teacher development programme involving technology integration (Unpublished doctoral dissertation). Indiana University, Bloomington.

75.

Yamagata-Lynch

(2003). Using activity theory as an analytic lens for examining technology professional development in schools. Mind, Culture, and Activity, 10, 100-119.

76.

Yukl

G. A.

(2001). Leadership in organisations (5th ed.). Upper Saddle River, NJ: Prentice Hall.