Leadership has traditionally been conceived of as the exercise of influence (Bass & Bass, 2009; Yukl, 2011). Note that no statement is made in this regard as to whether influence is exercised for good or ill (Shamir, House, & Arthur, 1993). And, no statement is made as to the setting, or role, in which influence is to be exercised ā in a firm, in politics, in a non-profit (Middlehurst, 2004). Not only is no statement made about the setting in which leadership is exercised, but no statement is made concerning the level at which the leader is acting, a CEO versus a first-line supervisor, although we know these different levels of action impose very different requirements for leader emergence and performance (Caughron & Mumford, 2012).
These ambiguities with respect to the level, role, and the social impacts associated with the exercise of influence have given rise to proliferation of theories of leadership. Put differently, different models of leadership are proposed to account for the exercise of influence with respect to different impacts, roles, and levels of leadership. For example, leaderāmember exchange (LMX) focuses on the exercise of influence vis-Ć -vis dyadic, supervisory, relationships (e.g., Liden & Maslyn, 1998; Graen & Uhl-Bien, 1995). Other theoretical models, for example, Zaccaro, Rittman, and Marks (2001), focus on how influence should be exercised when people are asked to influence teams and team performance. Other models of leadership ask how leaders should exercise influence to ensure prosocial outcomes ā an approach illustrated in models of ethical leadership (e.g., Kalshoven, Den Hartog, & De Hoogh, 2011). Still other models focus on the requirements imposed on leaders occupying a specific role or working in a specific type of firm ā for example, leadership in industry (Fleishman, 1953), education (Briggs, Morrison, & Coleman, 2012), or politics (Burns, 1978).
Mumford and his colleagues (e.g., Mumford, 2006; Mumford, Antes, Caughron, & Friedrich, 2008; Mumford, Espejo, Hunter, Bedell-Avers, Eubanks, & Connelly, 2007) have proposed another model intended to account for the effective exercise of influence in a certain setting, historically notable leaders operating as directors of organizations, where influence may be exercised for good or ill. This model is referred to as the Charismatic, Ideological, and Pragmatic (CIP) model ā a model that also draws a distinction between socialized and personalized leadership to take into account the social impacts of a leaderās exercise of influence. We will, in the present effort, examine the origins of this model of leadership, the findings emerging from initial research seeking to validate this model, and potential directions for future research.
Origins
People interested in leadership often begin to pursue this interest by studying eminent, historically notable, leaders ā leaders such as Benjamin Franklin (Mumford & Van Doorn, 2001), Franklin Delano Roosevelt (Goodwin, 2013), or Steve Jobs (Isaacson, 2012). For those interested in leadership, this might be appropriate. Historically notable leaders have changed and reshaped our world. They leave behind an extensive record of their actions which also describes the reasons for their actions. Historians analyze these records to describe not only their actions and their impacts, but also the origins of these actions. Although one might assume historic biographies describing the careers of outstanding leaders provided the basis for development of the CIP model, this is not the case. Rather, the origins of this model can be traced to the work being done by the United States Army in the 1980s and 1990s.
No war is ever good. However, as wars go, the Vietnam war was considered a failure, and a significant failure, by the United States Army. Many factors contributed to the failure of the United States Army in Vietnam ā factors ranging from politics to troop rotations. The Army, however, was concerned with one potentially critical case of this failure ā the actions taken by those occupying leadership roles. This overarching concern led the Army to initiate two streams of research. One stream was concerned with how leaders interacted with, and motivated, troops. The key person leading this stream of research was Bernie Bass, and his colleagues at Binghamton University, who focused on the theory of transformational leadership (Bass, 1990; Bass & Avolio, 1993; Bass & Riggio, 2006). The other stream of research was concerned with how those occupying leadership roles resolved the types of problems confronting them or brought to them. This stream of research was directed by Edwin Fleishman, Steve Zaccaro, and Michael Mumford (Mumford, Zaccaro, Harding, Jacobs, & Fleishman, 2000) ā all working at George Mason University.
In our research on leader problem solving we focused on the complex, high-level, cognitive skills that make it possible for leaders to solve complex, ill-defined, or poorly structured, often novel problems (Mumford & Connelly, 1991). In this effort, we asked some 1,800 Army officers ranging in grade from second lieutenant to full colonel, and the occasional general, to work on a variety of cognitive measures. They were asked to complete Christensen, Merrifield, and Guilfordās (1953) consequences measure to assess divergent thinking. They were also asked to complete a modified think aloud protocol examining how officers worked through complex, novel, ill-defined military problems where probe questions focused on creative thinking skills such as problem definition, information gathering, conceptual combination, idea generation, and implementation planning (Mumford, Mobley, Reiter-Palmon, Uhlman, & Doares, 1991). They were asked to resolve social conflict problems to allow assessment of variables held to underlie wisdom (Zaccaro, Mumford, Connelly, Marks, & Gilbert, 2000). In addition, standard measures of basic cognitive abilities and personality were administered. These measures were used to account for a variety of criteria, including medals awarded, critical incident performance, and attained rank.
The findings emerging from this effort might be summarized as follows. First, measures of complex cognitive skills, creative thinking skills, expertise, wisdom, and problem-solving skills, were powerful predictors (r = .45) of all three measures of leader performance (Connelly, Gilbert, Zaccaro, Threlfall, Marks, & Mumford, 2000). Second, these skills mediated relationships between basic abilities, such as intelligence, and added to the prediction obtained from measures of basic abilities (Vincent, Decker, & Mumford, 2002). Third, these skills developed over time as a function of career experience (Mumford, Marks, Connelly, Zaccaro, & Reiter-Palmon, 2000). Fourth, the predictive power of these skills in accounting for leadership performance is maintained over a 20-year period, accounting for continuance in the āup or outā Army system (Zaccaro, Connelly, Repchick, Daza, Young, Kilcullen, et al., 2015).
We presented our initial findings in this regard at meetings of the Society for Industrial and Organizational Psychology. In this panel session, Bernie Bass also presented the initial findings bearing on his model of transformational leadership ā a model which holds that exceptional leader performance is a behaviorally based phenomenon requiring idealized influence, or charisma, inspirational motivation, intellectual stimulation, and individualized consideration (Bass & Bass, 2009). Although Mumford was not at all comfortable with this model ā it struck him as an authoritarian appeal to mass emotion (Yukl, 1999) ā the findings presented by Bass (1985), Yammarino and Bass (1988,1989), and Bass, Avolio, and Atwater (1996) indicated that transformational leadership behaviors, especially charisma or idealized influence, also predicted the performance of military officers. Although the prediction was not as good as that obtained from our cognitive model, subsequent research has shown the prediction obtained is stable and generalizes across incidents of leadership performance in many domains (Lowe, Kroeck, & Sivasubramaniam, 1996).
Discomfort, and substantive differences in theories, often leads one to think ā especially when divergent theoretical models both seem equally effective in accounting for performance. A potential solution to this quandary emerged from the findings of Weick (1995) examining incidents of good and poor leader performance under conditions of crisis ā noting crises inherently present leaders with novel, complex, and ill-defined problems. His findings indicated that team performance resulted from how leaders understood or made sense of the crisis. And, leadersā articulation of this understanding, in other words, articulation of a vision, was the basis for attributions of charisma (Conger & Kanungo, 1994) and provided the basis for subsequent team performance. Given that charisma, articulation of a viable vision, is a key, albeit not the only aspect of transformational leadership, a potential solution to the theoretical problem at hand emerged. Sensemaking is based on leadersā cognitive skills and effective sensemaking allows leaders to formulate viable visions which, with effective articulation, might cause followers to perceive the leader as transformational (Mumford, 2006).
How people understand a crisis, or novel, ill-defined, complex, high-risk problems, is commonly held to depend on the mental models, or cause/goal linkages (Goldvarg & Johnson-Laird, 2001), they use to understand the problem at hand. And, as Day, Gronn, and Salas (2004) have noted, the leaderās articulation of a viable mental model for understanding the task at hand is a critical mechanism for inducing shared mental models among team members ā shared mental models which appear to be a critical determinant of team performance. These observations led Strange and Mumford (2002) to propose a model for the vision-formation process. Although complex, this model held that leadersā descriptive mental models were used to organize cases of prior experience. When unexplained events arise, leaders search through activated cases vis-Ć -vis their descriptive mental model of the social system at hand to identify key causes of performance and key goals to be pursued. The identification of key causes and key goals relevant to the performance setting at hand allows leaders to formulate a prescriptive mental model which, with forecasting, allows leaders to construct a viable vision ā a vision that might lead followers to attribute charisma to the leader.
In an initial test of this model, Strange and Mumford (2005) asked undergraduates to assume the role of a principal of a new, experimental secondary school. They were asked to formulate a plan for leading the school and to craft a visionary speech to be given to students, parents, and teachers. Students, parents, and teachers ā āreal peopleā ā appraised the affective impact and perceived utility of the speeches. Doctoral students appraised the quality, originality, and elegance of the plans. Prior to starting work on the plans and speeches, participants were instructed to think about causes, think about goals, think about both causes and goals, or think about neither vis-Ć -vis a set of case models. These case models were drawn from the literature on cooperative leading techniques, where the case models presented were either successful or unsuccessful. It was found that the strongest vision statements, and strongest leadership plans, emerged when participants were asked to think about causes with respect to successful cases or think about goals with respect to unsuccessful cases. Thus, in keeping with Strange and Mumfordās (2002) model, cause and goal analysis does seem to underlie vision formation.
Other studies have provided support for other aspects of this model. For example, Partlow, Medeiros, and Mumford (2015) have shown that the viability of leadersā visions depends, in part, on the leaderās mental model. Byrne, Shipman, and Mumford (2010) and Shipman, Byrne, and Mumford (2010) have shown forecasting is also a critical influence on leadersā production of viable visions. Although these findings provided some rather compelling support for Strange and Mumfordās (2002) model of the vision-formation process, they beg a question. Do all leaders think about the same kind of causes and the same kind of goals in vision formation?
Two distinct lines of work allowed for a provisional answer to this question. One line of work arose as Mumford left the northeast and moved to the west. As an old Philadelphian, and one consigned to a government transit hotel, he decided to read Benjamin Franklinās writings on his many life achievements. What was clear in reading F...