Recognizing the need for study on the influence of temporal dynamics (e.g., Hackman, 2012; Wageman, Gardner, & Mortensen, 2012), researchers in the field of leadership have sought the process in which people in a group exercise leadership along with the process (Zaccaro, & Klimoski, 2002). In order to capture the process or dynamics of leadership, previous studies have explored leadership dynamics in a various settings (cf. Hackman, 2002). For instance, some of situations are top management teams (e.g., Srivastava, Bartol, & Locke, 2006; Wageman, Nunes, Burruss, & Hackman, 2008), the entrepreneur teams (e.g., Ensley, Hmieleski, & Pearce, 2006; Hmieleski, Cole, & Baron, 2012), and virtual teams (Hambley, O'Neil, & Kline, 2007; Kirkman, Rosen, Gibson, Tesluk, & McPherson, 2002).
In order to capture dynamics of leadership, this study focuses on the concept of shared leadership (cf. Pearce, & Conger, 2003), because if the author focuses on one person who would exercise leadership in a group, it makes more difficult to discuss how leadership emerges and changes during the process of a group activities. Moreover, I show that because of changes of the nature of work (i.e., the trend to knowledge work; cf. Pearce, 2004), it is becoming much difficult for one person to exercise leadership alone. According to Pearce (2004), "it is even more difficult for any one person to have all of knowledge, skills and abilities for the tasks" (p. 47). In short, researcher has difficulty determining who has leadership role in advance (cf. functional approach: Morgeson, DeRue, & Karam, 2009; Zaccaro, Rittman, & Marks, 2001).
This paper addresses two issues. First, after reviewing the theoretical methods that the existing research employed and discussing what extant research has investigated, the study points out that previous research has been likely to overlook the process in which people share leadership roles and functions in a group. Secondly, I argue that due to the first point, the researchers need to explore the possibility that there are some opportunities where it is likely one person that exercises leadership.
The remaining sections of the paper are structured as follows. First, I outline how the concept of shared leadership has been discussed. Second, the present study reviews what the existing research has explored empirically by a theoretical approach. Third, drawing on the results of extant research, this article provides the neglected issue in shared leadership, and explores the neglected issue theoretically. Fourth, given the discussion, a series of potential theoretical and practical implications are discussed, as well as a series of limitations. Finally, advancing the research, this study suggests some points to be tackled in future research.
The Concept Of Shared Leadership
Some researchers have mentioned that ideas similar to the concept of shared leadership have been discussed before. For example, Pearce and Conger (2003) showed that there had had been some theoretical ideas underlying the notion of shared leadership. According to Pearce and Conger (2003, p.6), the earliest researcher who mentioned the similar idea to the concept of shared leadership was Mary Parker Follett (1924). She introduced the idea of the law of the situation, and showed under some situations; people follow the person who is not appointed leader.
Although this related idea was identified about 90 years ago, it was not until about the 1990s those researchers have been increasingly interested in the concept of shared leadership (cf. D'lnnocenzo, Mathieu & Kukenberger, 2014). Table 1 (adopted from D'lnnocenzo et al. (2014)) shows some of definitions to which previous researches have referred. Table 1 shows the current situation in which researchers have stated different dimensions of shared leadership. Researchers have mentioned the slight different definitions of shared leadership, however, previous research has proposed that there are some features underlying the different definitions of shared leadership among the previous studies. Although the interest has been increasingly on shared leadership, there exist no united definitions about shared leadership.
D'Innocenzo et al. (2014) identified five themes that previous research on shared leadership had had in common. The first theme is the locus of leadership. It states whether the locus of leadership is inside or outside teams, and especially the later is related to the study on team leadership (cf. Zaccaro et al., 2001) and self-managed team (cf. Manz, & Sims, 1993). The second theme is about the authority of leadership, that is, whether or not leadership is formal within organization. According to D'lnnocenzo et al. (2014), the first and the second theme are about the sources of shared leadership.
In addition, the third is "the extent to which team members participates in leadership" (pp. 4-5). For the third theme, previous studies have employed two approaches: the aggregation approach and the social network theory approach (cf. D'lnnocenzo et al., 2014). The aggregate approach of shared leadership assumes that shared leadership entails several leadership styles (cf. Pearce & Sims, 2002). Pearce and Sims (2002) suggested shared leadership entailed aversive leadership, directive leadership, transactional leadership, transformational leadership and empowering leadership. On the other hand, the social network theory approach is literally the methodology based on the social network theory. The approach attempts to capture the concept of shared leadership by using the concepts of density and centricity, and shows how people connect each other and who mainly exercises and shares leadership.
The fourth theme is the dynamic quality of shared leadership. This theme is related to the assumption that leadership is static. The concept of shared leadership struggles to capture dynamics in which different people perform the leadership role along with task-performing processes and during the team's life cycle (e.g., Erez, LePine, & Elms, 2002). Thus, the research on shared leadership needs to pay attention to the influence of time on dynamics and the temporal shift of leadership roles among team members. Thus, the concept of shared leadership is partially contrasted with the traditional leadership concept (i.e., vertical leadership) in that shared leadership assumes that all members, not only the person who is at the appointed position, can exercise leadership. The final theme is the involvement of multi roles and functions. It suggests that to facilitate the task completion, shared leadership consists of multiple leadership roles and responsibility. Thus, shared leadership assumes that whether same or different, members in a team perform multiple leadership roles for completing the task. The aforementioned themes are typical characteristics found in previous definitions, however, the salient difference between vertical leadership and shared leadership remains unclear.
In order to make the notable difference between vertical leadership and shared leadership clear, this article sheds light on the study that discussed the relationship between shared leadership and vertical leadership. As for vertical leadership, there are two salient characteristics. The first is that vertical leadership assumes the downward influence from the top or the person who is in the hierarchical position (i.e., managers). On the other hand, shared leadership considers the multidirectional influence. In other words, shared leadership literature captures three different influences; downward and upward, lateral, and diagonal (cf. Pearce, Conger, & Locke, 2008). The second is, as aforementioned, that shared leadership assumes that members in a group can exercise leadership. This point is related to the first in that members include not only managers, but also subordinates in a group. Hence, shared leadership supposes that there is a lateral influence between subordinates.
Given the characteristics mentioned above, in this article, the author proposes the following definition: Shared leadership is an emergent and dynamic phenomenon in which, regardless of the direction of influence, team members share leadership roles, functions, and influences.
The Result of Prior Empirical Studies
As the interest in temporal dynamics has been gaining more attention in recent years, researchers have acknowledged the importance of shared leadership and investigated the effectiveness of shared leadership. Moreover, the exploration of the influence of shared leadership, previous studies have sought to identify those situations in which the influence of shared leadership is effective or ineffective. The accumulated results related to shared leadership have shown the effectiveness of shared leadership, however, this article argues that the previous study has been likely to overlook the mechanism related to the centrobaric concept of shared leadership. In short, extant research has directed little attention to the issue of how members share leadership within a group or during task-performing processes. In this study, I review the previous empirical...