Responsible Leadership for Performance: A Theoretical Model and Hypotheses

Companies spend large amounts of money in pursuit of recipes for leadership success. In 1998 it was estimated that 86% of companies offer some form of leadership training (Boyett & Boyett; Zhu, May & Avolio, 2004). Conversely, few companies can attest to the verifiable contribution of this investment to their business performance, although much of the research literature claims leadership as critical to organizational performance and profitability (Bass, 1985; Bass, 1990; Clark, Clark & Campbell, 1992; Kotter, 199Oa, 199Ob; Meindl & Ehrlich, 1987; Rottenberg & Saloner, 1993; Yukl & Van Fleet, 1992; Zhu, May & Avolio, 2004). Leadership and leadership development will increasingly need to show a direct link to business performance. Leadership is decreasingly about recipes of actions and characteristics that are transferable to infinite contexts, reflected in the likes of popular leadership models such as that of Covey's principle-centered leadership (1989, 1991).

Brungardt (1996) highlights that, in spite of the abundance of leadership research over the last several years, which has given us a ".. .much better understanding of leaders and the leadership process the field of leadership studies continues to be "...riddled with paradoxes, inconsistencies and contradictions" (p.82). This lack of theoretical and empirical rigor is underscored by Klenke: "...contributing to the messy state of the art [of leadership] are controversies about theoretical and methodological issues as well as tensions between the disciplines contributing to leadership studies" (1993, p.112). Swanson (1995) stressed that as the role of performance improvement in organizations increasingly takes on strategic and global importance, executives are, and should be, increasingly held accountable in this arena. Leadership and leadership development should be seen as "core organizational efforts at improving performance" and must, like other organizational efforts, "recognize the organization's major business processes and their connectedness to core inputs and outputs for the purpose of adding value"(p. ix). In short, the stresses and demands of the emerging global organization and accompanying chaos and complexity of these business realities will likely call for leadership that can think and act fundamentally differently in the future (Lynham, 1998, 2000c, 2000d; McLagan & Nel, 1995).

Grounding the Problem

A review of the literature points to a number of inadequacies. First, the direct link between leadership and business performance is implied rather than explicit, i.e., the majority of studies that examine leadership are not studies that tend to link leadership practices to objective outputs of the leadership system (Bass, 1990; Holton & Lynham, 2000). second, the impact of leadership on performance is not considered from multiple domains of performance (Bass, 1990; Holton & Lynham, 2000; Lynham, 1998, 200Od; Yukl & Van Fleet, 1992). By multiple domains of performance, we mean that leadership is not often studied in ways that document its effect on individual indicators of performance, group indicators of performance, process indicators of performance, and organizational indicators of performance and specifically all of these at the same time. Third, absent from the literature is the multidimensional notion of responsibility (Collins & Porras, 1994; Freudberg, 1986; White Newman, 1993). We suggest that not only should leadership be responsible, but that the notion of responsibility is related to other ethical and moral factors that are often created and agreed upon within the system itself. Although some notions of leadership include the importance of value-centeredness (Clark & Clark, 1996; Covey, 1991), and others allude to the importance of renewal in leadership (Gardner, 1990), being responsible in leadership is predominantly associated with effectiveness (Bennis, 1994; Bhatia, 1995; Tannenbaum & Schmidt, 1973)~with getting things done. Fourth, missing from this body of literature is agreement on the dependent variable of leadership.

That the phenomenon of leadership may be a system, with inputs, processes, outputs and feedback, and in service to a larger performance system, is not deeply considered in the literature and represents a notable void in this body of knowledge. Given the huge amounts of money being spent on the training, coaching and development of leadership capabilities and capacity, and the increasingly diverse environs in which leadership is both applied and judged, the above are troublesome knowledge gaps and inadequacies (Boyett & Boyett, 1998; Brungardt, 1996; Trevion, Brown, & Hartman, 2003; Yukl & Van Fleet, 1992).

The problem statement driving this study is thus: available leadership theories neither explicitly nor adequately address the nature and challenges of leadership that is both responsible and focused on performance. There is a need for a theory of leadership that satisfies these multiple domains of concern and that integrates the practical overarching concerns for people and performance (Lynham, 1998, 200Oc, 200Od; Melrose, 1995). It is therefore the purpose of this study to begin to address this inadequacy hi current leadership theory.

Research Questions & Methodology

Given the above problem statement, the following research questions were used to develop and guide this study: 1) Can "A Theory of Responsible Leadership for Performance" be developed?; and, 2) Can "A Theory of Responsible Leadership for Performance" be operationalized [for later verification]?

Due to the applied nature of leadership, and the preliminary nature of this study, Dubin's (1978) two-part, eight-step, theory-to-research methodology for applied theory building is well suited to address the research question and subquestions. This applied theory building methodology includes: (1) identification of the units or concepts of the theory, (2) determining the laws of interaction among the units, (3) specifying the boundaries in which the theory is expected to apply, (4) specifying the system states in which the theoretical system operates, (5) articulating the propositions, comprising the logical deductions or truth statements about the theory in operation, (6) determining the empirical indicators used to make the propositions and therefore the theory testable, (7) identifying the hypotheses, that is, the statements about the predicted values and relationships among the units, and (8) testing the predicted values and relationships.

Dubin's (1978) methodology for applied theory building is complex and extremely detailed - so much so that a full discussion of the philosophical reasoning underlying the method is not practical in the context of this article. This article proceeds in sections that detail each step of the theory building methodology in Figure 1 and overview descriptions of the steps are provided drawing from Dubin with as much detail as space permits. For a full discussion of the intricacies, of quantitative theory building in applied disciplines, please refer to Dubin (1978).

A theory of Responsible Leadership for Performance (RLP) is a general, integrative theoretical framework of leadership that addresses the nature and challenges of leadership that are both responsible and focused on performance. Two core premises govern the framework. The first-that leadership is itself a system consisting of purposeful, integrated inputs, processes, outputs, feedback and boundaries. The second-that leadership takes place within a performance system, that is, a system of joint, coordinated and purposeful action. Leadership can therefore be conceived of as a system of interacting inputs, processes, outputs, and feedback that derive meaning, direction and purpose from the larger performance system and environment within which it occurs. From this perspective, leadership is defined as: a focused system of interacting inputs, process, outputs and feedback wherein individuals and/or groups influence and/or act on behalf of specific individuals or groups of individuals to achieve shared goals and commonly desired performance outcomes, within a specific performance system and environment.

We also think it important to define the term "ethical" as it is used throughout this article as a further descriptor when we present the notion of responsible leadership. The American Heritage College Dictionary defines ethical as 1) "involving or expressing moral approval or disapproval" and 2) "conforming to accepted professional standards of conduct" (2002, p. 343). While we imply that both of these components are important to the discussion of "responsible leadership," we would like to stress importance on the former. That is, when we present leadership as a system involving the consideration of the people within that system, their interpretation of what is responsible and as focused on an agreed important output, we suggest that this group will define "responsible" behavior in a way that involves morality. Again, the constituency is believed to define what is moral according to the negotiations of the individuals that collectively form that constituency.

Units

The units of a theory are the concepts of the theory-the basic ideas that make up the theory (Cohen, 1991; Dubin, 1978; Reynolds, 1971). The units represent the things about which the researcher is trying to make sense and are informed by literature (knowledge of) and experience (knowledge about) (Lynham, 2002a). The units can be plainly defined as the building blocks of the theory or the elements that come together in the theory.

The three units of the theoretical framework of RLP are: considerations of constituency; a framework of responsibleness; and domains of performance (see Figure 2). The units interact to form the inputs, process and outputs to the leadership system that is the essence of the theoretical framework. Each unit is further distinguished by conceptual dimensions. Considerations of constituency (the input) include three conceptual dimensions: whether the constituency (a) resides inside or outside the performance system; (b) has high or low authority over the performance system; and (c) has the potential for high or low impact on the performance system. A framework of responsibleness (the process) has three conceptual dimensions: (a) effective leadership practices; (b) ethical leadership habits; and (c) enduring leadership resources. The third unit, domains of performance (the output), consists of four conceptual dimensions: (a) the system mission; (b) the work process/es; (c) the social sub-systems; and (d) the individual performer. A brief discussion of each unit follows.

Unit 1: Considerations of Constituency

Leadership does not exist on its own, but rather in reciprocity to constituency, sometimes referred to as fellowship. It is the constituency that gives voice and purpose to leadership-that allows the individual and/or group to stand up as one (Bass, 1985, 1990; Bennis, Parikh, & Lessem, 1994; Block, 1993; Corderio, 2003; Kelley, 1992; Kouzes & Posner, 1995; Lynham, 1998, 200Od; Odom & Green, 2003; Perreault, 1997; Terry, 1993).

Leadership does not exist in isolation. Without fellowship there is no leadership (Autry, 1991; Bass, 1990; Block, 1993; Gardner, 1990; Kelley, 1992; Kouzes & Posner, 1995; Northouse, 1997; Perreault, 1997). The phenomenon of leadership comes about as a result of a need to pursue desired outcomes, that is, to achieve goals desired by stakeholders located both inside and outside a specific performance system (Beauchamp & Bowie, 1997; Holton, 1999; Frooman, 1999; Trevion, Brown & Hartman, 2003).

Every performance system has a constituency that represents those whom leadership in the performance system serves, and for whom the leadership produces desired results (Freeman, 1997; Frooman, 1999; Gardner, 1990; Gibson, Ivancevich & Donnelly, 1994; Greenleaf, 1997; Jones & Wicks, 1999; Kelley, 1992; Milgrom & Roberts, 1992; Ulrich, Zenger, & Smallwood, 1999). As a result, considerations of constituency, forms the first, catalyzing, and input unit of the theoretical framework of RLP.

Unit 2: A Framework of Responsibleness

The word responsible is associated with words like answerable, liable, accountable, amenable, reliable, dependable, trustworthy, and care. These terms share the meaning of "being obliged to answer...as for one's actions...to an authority that may impose a penalty for failure" (American Heritage Dictionary, 1992, p. 1537). The notion of responsibleness is associated with professional action, that is, action that is "based on careful, reflective thought about which response is professionally right in a particular situation" (Tennyson & Strom, 1986, p. 298).

White Newman (1993, as cited in Lynham, 1998, p. 211 ) offers a workable and appealing framework for responsible leadership. "Too often", says White Newman, "leadership writings and practice have emphasized how to be effective. They need to embrace much more than this singular focus". White Newman aptly describes the logic of the framework of responsible leadership as follows:

Being solely concerned with effect seems inadequate, potentially even dangerous, since it is obvious that a person can be effective-that is, make a difference-yet also be unethical. Knowing the harm such people can cause, I believe most of us want leaders who, because they are ethical, will make beneficial differences to the world. Too often, in these stressful times, individuals who are effective and ethical survive as leaders for a brief time. They do not endure. Some are replaced by external factors. Others cave in under the pressure of leadership-they burn out. So endurance becomes essential to leadership. Endurance encompasses refreshment for leaders and renewal for their groups [and systems], (cited in Lynham, 1998, p.)

Leadership that is responsible is that which demonstrates, and is judged to demonstrate, effectiveness, ethics, and endurance (DePree 1989, 1997, Trevion, Brown & Hartman, 2003; White Newman, 1993), and are necessary components of responsible leadership. What constitutes these 3E's is determined by the constituency of the performance system in which the leadership occurs (Bass, 1990; Beauchamp & Bowie, 1997; Brady, 1985; Freudberg, 1986; Frooman, 1999; Jones & Wicks, 1999; Khuntia & Suar, 2004; Knapp & Olson, 1996; Stavrou, Kleanthous, & Anastasiou, 2005). It is this framework of responsibleness that constitutes the second unit of the theoretical framework of RLP.

Luthan's (2001; 2002; 2003) work on the role of hope and its impact on leaders is the closest that we have found to what we mean through the use of the term responsible. In several studies involving relatively large samples, Luthans has been able to suggest that high-hope leaders are generally able to foster more productive and more enjoyable work environments (2003). The significance of this work seems to suggest that what we might call responsible leaders (and what Luthans might call hopeful), are leaders that communicate frequently with their co-workers, and for whom those co-workers enjoy coming to work.

Unit 3: Domains of Performance

According to The American Heritage Dictionary (1992), performance is about carrying something through to completion, that "to perform is to carry out action, an undertaking, or a procedure" and that the word "often connotes observance of due form or the exercise of skill or care" (p. 1345). From these descriptions, performance must be seen to have two parts: "an activity and the outcome of that activity" (Dean, 1997, p. 72). Performance also occurs within a context of requirements, that is, according to the requirements of a particular performance system and audience.

Each performance system therefore defines performance to fit and serve its unique needs (Collins & Porras, 1994; Dean, 1997; Gibson, Ivanevich & Donnelly, 1994; Holton, 1999; Kolvitz, 1997; Passmore, 1997; Rummler & Brach, 1995; Tosti & Jackson, 1992; Von Bertalanffy, 1968; West, 1997). The performance of a system is multidimensional. Four commonly identified and significant domains of performance include the system mission and purpose, the work process/es, the social sub-systems, and the individual performer (Cummings & Worley, 2001; Gibson, Ivancevich & Donnelley, 1994; Holton, 1999; Mintzberg, 1994a, 1994b; Rummler & Brache, 1995; Swanson, 1996; Swanson & Holton, 1999; Tichy, 1983, 1997; Wimbiscus, 1995). It is these multiple domains of performance that form the third unit of the theoretical framework of RLP.

It is a common misconception that a focus on performance implies a focus on financial measures. Given this assumption, a common issue with the premises we have argued relates to how one can be both responsible and focused on performance. To clarify this point, we stress that what makes for "performance" is defined by the constituency. For example, an educational institution or non-profit organization would define performance differently than that of a for profit organization. Habitat for Humanity might measure performance in terms of the number of houses built, while a fortune 500 might focus on stock performance and building shareholder value. The point to be made here is that our vision is for leadership is that it is done responsibly (drawing from moral norms agreed on by the constituency) and that it is performance-based (aims toward the achievement of some output agreed on by the constituency).

Laws of Interaction

The laws of interaction, of which there are a number of kinds/types, describe the interaction among the units of a theory (Dubin, 1978). The laws of interaction among the three units of the theoretical framework of RLP include four categoric laws and three sequential laws. "A categoric law of interaction is one that states that values of a unit are associated with values of another unit" (Dubin, 1978, p. 98), while a sequential law is defined as one that is "...always employing a time dimension. The time dimension is used to order the relationship among two or more units" (p. 101).

According to Dubin's methodology, "all units are linked with categoric laws, as a change in any unit will provoke a change in at least one other unit" (Chermack, 2005, p. 824). Furthermore, "all units are also linked with sequential laws to denote the importance of the time element" (p. 824). The theoretical framework of RLP does not include any determinate laws and appoints considerations of constituency as the catalyst unit, that is, the unit "...whose presence is required for other interaction in the theoretical [framework]" (Chermack, 2005, p. 824). The laws of interaction of the theoretical framework of RLP are illustrated in Figure 3.

Boundaries

Establishing the boundaries of the theoretical framework of RLP requires specification of the domain or domains in which the framework is expected to operate (Dubin, 1978). Boundaries help locate the theoretical framework in the environment with which the theory is concerned (Chermack, 2005), and require that the theorist makes the logic used to determine them explicit (see Figure 4). The boundaries of the theoretical framework of RLP are shown in Figure 4.

There are three boundaries for the theoretical framework of RLP: 1) a leadership sub-system boundary 2) a performance system boundary, and 3) the contextual environment boundary. All boundaries in the theoretical framework are open boundaries, denoted by the dashed lines in Figure 5, indicating that the leadership (sub)system continuously exchanges information and resources with the performance system domain in which it occurs and with the larger exterior environmental domain acting on the performance system. Informed by the logic of general systems theory (Senge, 1990) and existing work on leadership as process, responsible leadership for performance is positioned as a focused (sub)system within a larger performance system and context, as indicated in Figure 6.

A common issue in theory building is in differentiating the units of the theory from the boundaries. For example, in this theory building exercise, it may be tempting to inquire about the difference between the unit "domains of performance" and the boundary of performance systems. It is generally helpful to keep in mind that the unit refers to the item that is present in the theory, and the boundary can be considered the limiting element of that unit. For example, this framework of responsible leadership requires that individuals, groups, processes, and the organization meet some performance requirements (usually those specified in the system or organization), but the notion of a performance system boundary is meant to suggest the space in which this performance must take place.

System States

Dubin (1978, 1981) defined a system state as a condition of the system being modeled in which all the units of the system take on characteristic values that have persistence through time, regardless of the length of the time interval. All units of the system being modeled have values that are determinant, meaning they are measurable and distinctive for that state of the system.

A system state that accurately represents a condition of the system of RLP being modeled has three important characteristics, namely: inclusiveness ("where all the units of the system are included in the system state"), persistence ("where the system state persists through some meaningful period of time"), and distinctiveness ("where all units take on unique values for that system state") (Dubin, 1978; Toracco, 2000, p.54).

The theoretical framework of RLP conceptualizes leadership as a (sub)system, and has four system states which it transitions between. Borrowing from Chermack (2005), and to illustrate the differing states of the RLP system, the theoretical framework uses a 0;1 coding, where O "represents none of the thing or characteristic under examination" (p. 825)-meaning there are no considerations of constituency and therefore there is no leadership system in operation. According to the time-sequence embedded in the laws of interaction, actions regarding specific units precede actions regarding others. "As the system transitions from state to state, the unit values shift from O to 1...demonstrating that as each unit value shifts, the [theoretical framework] transitions from one state to the next (Chermack 2005, p. 825). The four system states of the theoretical framework of RLP are: (1) non-operation-where all units have a value of 0; (2) identification of considerations of constituency-where unit 1=1, units 2 and 3= 0; (3) determining requirements of responsible leadership-where unit 1=1, unit 2=1, unit 3 =0; and (4) determining and assessing performance and implications-where all units = 1, and the leadership system is fully operational. When in a fully operational state, feedback based on performance becomes an input to the leadership units and begins the process again, from either point.

Having completed the first four steps of Dubin's theory building methodology, a complete and informed theoretical framework of RLP can now be made explicit (see Figure 5). This theoretical framework satisfies phase one of the General Method of theory building research and indicates that the conceptual development of the theoretical framework of RLP has been completed (Dubin, 1978; Lynham, 2002a, 200Ob; Torraco, 1974, 2000).

Another common issue in theory building concerns the tendency to question the difference between the laws of interaction and the system states. Those individuals intuitively attuned to the logic of theory building will see this connection, although it is not always clear. For certain, laws of interaction must link to system states, but these items are not the same. For some it is helpful to think of system states as "phases" through which the theory will transition. According to Dubin (1978) system states are often designated by examining the laws of interaction, but the system states should describe the nature of the relationship whereas the laws of interaction may simply state that a relationship exists. For example, in this framework of responsible leadership for performance, we have suggested that "considerations of constituency" are related to "a framework of responsibleness. System state 2 suggests that there is a process through which the theory moves that transitions from a focus on considerations of constituency to a focus on responsibleness.

Once conceptually developed, the theoretical framework is used to articulate knowledge claims or truth statements about how the framework can be expected to operate in the real world. These knowledge claims are also known as the propositions of the theoretical framework. Eight strategic propositions have been specified for the theoretical framework of Responsible Leadership for Performance:

1. RLP is a theoretical framework of leadership as a system-in-focus, in which leadership is conceived as a purposeful, focused system, not an individual or a process managed by an individual.

2. All systems have a purpose. The purpose of RLP is to serve the needs and desired outcomes of the constituency of a performance system by positively impacting multiple domains of performance in a responsible (effective, ethical and enduring) manner.

3. The content of RLP is derived from all three units of the theoretical framework-considerations of constituency, a framework of responsibleness, and domains of performance. If all three units are not present and interacting, then there is no system of RLP in action.

4. For leadership to be considered responsible, it must demonstrate, and be judged to demonstrate, effectiveness, ethics and endurance. If one of these three attribute properties is missing from leadership, then that leadership cannot be considered responsible.

5. The units of the theoretical framework-considerations of constituency, a framework of responsibleness, and domains of performance-are interdependent. A change in one unit can be expected to produce a change in the other two units.

6. As responsibleness (effectiveness, ethics and endurance) increases, performance of the whole performance system can be expected to increase.

7. Constituency is a necessary requirement for responsible leadership for performance. Without constituency there is no RLP.

8. Without guiding inputs from constituency, and outputs in the form of multi-domain performance, the phenomenon of RLP collapses.

The next step according to Dubin (1978) is to determine empirical indicators. These empirical indicators inform the formulation of hypotheses necessary to verify the theoretical framework through disciplined inquiry, application, and continuous refinement.

Empirical Indicators

Informed by the above propositions and existing literature, four empirical indicators have been identified for the theoretical framework of RLP. These empirical indicators are necessary to be able to identify values and measures for the interacting units of the theoretical framework, which must next be verified through empirical research. Four initial empirical indicators, derived from propositions 6, 7, and 1, respectively, have been determined, and are presented in the statements following (for further discussion refer Lynham, 200Oa, 2000d).

The empirical indicators are chosen by the theorist, and are targeted at those thought most necessary to confirm/disconfirm the operation of the theoretical framework in the real world (Dubin, 1978). These values and measures can also be used on the units and the RLP system employed in other propositions of the theoretical framework.

The theoretical framework, together with the propositions and empirical indicators, represent the development and initial operationalization of the theoretical framework of RLP. Informed by existing related literature, the theoretical framework can now be made ready for verification through research designed to prove the adequacy and improve the theoretical framework. The results of such research can be used to further refine and develop the theoretical framework.

Hypotheses

Hypotheses are the link between the theoretical model under construction and the natural and/or social world in which it applies. Hypotheses are thus "the predictions about values of units of a theory in which empirical indicators are employed for the named units in each proposition" (Dubin, 1978, p. 206). While it is common for each proposition to yield several testable hypotheses, Dubin (1978) advises that the minimum requirements call for only one testable hypothesis for each proposition. It is also common in developing theory building efforts to begin with a set of hypotheses that can always be expanded upon at a later time. Therefore, given the novelty of the theoretical model developed in this manuscript, the hypotheses identified for preliminary verification of the model include the following:

Hypothesis 1: There will be a positive relationship between participation in the RLP system and constituent perceptions of effectiveness.

Hypothesis 2: There will be a positive relationship between participation in the RLP system and constituent perceptions of ethical behavior.

Hypothesis 3: There will be a positive relationship between participation in the RLP system and constituent perceptions of endurance.

Hypothesis 4: There will be a positive relationship between participation in the RLP system and mission related performance in terms of quality, quantity and/or time.

Hypothesis 5: There will be a positive relationship between participation in the RLP system and work process performance in terms of quality, quantity and/or time.

Hypothesis 6: There will be a positive relationship between participation in the RLP system and social sub-system and individual units of performance in terms of quality, quantity and/or time.

Hypothesis 7: There will be a positive relationship between participation in the RLP system and the level of participation of the constituency.

Hypothesis 8: There will be a positive relationship between participation in the RLP system and the perceived value of that leadership system by the constituency.

These hypotheses are intended to provide a preliminary evaluation of the RLP system. That is, these are not a comprehensive list of hypotheses that would be required to verify the theoretical model as theory, however, they do provide a starting point. Upon verification that aspects of the theoretical model hold and are accurate under empirical testing, further hypotheses can be formulated to suggest the interaction of one or more units as predictors of another. Advanced statistical analyses can be formulated including ANOVA and MANOVA comparison studies to examine these more complex experimental designs. In truth, however, it is not useful to outline such studies until the general adequacy of the theoretical model can be established. The hypotheses that have been identified are a starting point at which to do so.

Research & Practice Implications

The theoretical framework of RLP presented is one of the phenomenon of leadership as a system-in-focus. Perceiving leadership as a system-in-focus is new to leadership theory. Perceiving leadership as a system-in-focus that is aimed at leadership that is both responsible and about performance is atypical of leadership practice. In conclusion, some implications of this theoretical framework for future research and practice are briefly considered.

Research

Related literature indicates an abundance of leadership theories. However, also shown is that general, integrating theories of leadership are in short supply. The theoretical framework of RLP is potentially useful to leadership theory in a number of ways, two of which might include those indicated below.

First, the theoretical framework (of RLP) can be used to promote connectivity among existing leadership theories. For example, in both behavioral and transformation leadership theory, the phenomenon of leadership is shaped by the situation, or context, in which leadership occurs. By rethinking the contextual situation as a leadership system-in-focus, within a specific performance system, RLP enables both groups of theory to inform leadership as a means to an end and in service to multiple-domain, whole system performance. Through this refraining these two theoretical perspectives cease their rivalry and can each be understood and appreciated for how they inform the nature and challenges of the performance system.

Second, the theoretical framework can serve a transformative role with respect to existing and emerging theories of leadership. For example, trait theory is strongly informed by leadership in the military context. In this context the military makes for the particular performance system in which the leadership system-in-focus occurs. From the perspective of the theoretical framework of RLP, the military performance system must therefore be used to inform, shape and evaluate the required leadership system in terms of considerations of constituency, responsibleness and multi-domain performance. The nature and value of the three units of the framework are therefore changed by the purpose and nature of the performance system to which the leadership system-in-focus is in service. Considering the military as the performance boundary of the leadership systemin-focus significantly impacts who could, and should, be included in the considerations of constituency, what makes for responsible leadership in that performance system and context, and what constitutes acceptable performance in that performance system. This integrated perspective of leadership, in turn, informs the leadership traits best suited to the particular performance system, and highlights the need to reconsider these traits when the nature and purpose of the performance system itself changes. Thus, it is unlikely that the leadership traits best suited to a military performance will be the same as those best suited to a non-military one.

There are many diverse and competing theoretical perspectives on leadership, each explaining the phenomenon of leadership, each a way of both seeing and not seeing this phenomenon, and each with its own limitations. In a body of literature where theories of a conflicting and contradictory nature abound, this proliferation becomes fragmenting and confusing rather than unifying and clarifying (Yukl & van Fleet, 1992). More general theories of leadership, like that of RLP, need to be further developed, confirmed and refined to help to integrate and demystify this body of knowledge and understanding.

Practice

Go to the New Releases section at any modern-day bookstore, on-line or on-foot, and you will find at least a handful of books touting the latest answer to leadership. Included are marketing-driven recipes for quick success and results from consulting companies and so-called gurus, many of whom are best known for their management literature. Unfortunately, less seldom does one come across a scholarly, well-researched book on leadership that is written for the consumption of practice. The scholarly literature on leadership seldom resides on the bookshelves of popular bookstores.

The implications of using this theoretical framework in leadership practice are quite extensive. First, it would require that practitioners acknowledge that leadership is more a driver and less an outcome of performance. second, it means that leadership would have to be understood and pursued as a system-in-focus and thus in service to the performance system and context in which it occurs. Third, this theoretical framework requires that practitioners acknowledge the phenomenon of leadership as a system with inputs, processes, outputs and feedback, in the form of considerations of constituency, responsible leadership practices, habits and resources, and multiple performance outcomes. Fourth, the theory acknowledges and demands whole system performance as an indicator of leadership output, and fifth, are the implications of this theoretical framework to leadership development.

In brief, the implications of this theoretical framework for leadership development could indicate a drastic restructuring of current leadership development practices. For example, if this framework were to play out in an organization, we might find that the pathway to senior levels of management or to CEO might require a long tenure in the human resource, or organization development department. Activities like engaging in and directing training, organization development and team building are vital to understanding the relationships that make the organizations continued existence and growth possible. Suffice it to say that we assume leadership development practices would be required to change if the model we propose is investigated more fully and found to be a reasonable proxy for responsible leadership in general.

With growing demands for leadership that makes a difference, the increasing business realities of globalization and an increasing shortage of skilled, knowledge labor, the need to balance people and performance must become an important problem of leadership theory, research and practice. Leadership theories akin to that of RLP can be enormously helpful to this end. This theoretical framework could be used by practitioners to make sense of the multiple leadership theories at their disposal-that is, to frame and contextualize these theories in terms of the leadership system needs in their performance system. RLP could also be used to diagnose, develop, and evaluate current leadership and leadership capacity needs in various performance systems, ensuring that leadership is positioned as a key means to performance and not as an end in itself.

It is also our vision that current leadership practices must change. While we submit our model as currently founded only in the literature and how we conceive that responsible leadership MIGHT work, we hope to spend considerable efforts investigating the accuracy of our assessment. Should we find this theoretical model to be generally accurate (after considerable longitudinal investigation) we would also hope that leaders might see that responsible (effective, ethical, and enduring) leadership can be linked directly to the performance of the organization. Should this be the case, again founded on eventual years of cumulative research, the implications of our theoretical model could provoke changes in the way leadership is conceptualized, developed, and sold to organizations throughout the world. While this vision may be somewhat idealized, we intend to extend considerable effort in examining the theoretical model that we have outlined in this article.