Leadership as a Political Process

HomeArticlesLeadership as a Political Process

There is a common implication, in scholarly as well as popular literature, that although the leadership process may sometimes be political in nature the best leadership is somehow above politics. Indeed, the argument that leadership may be essentially and inherently political in nature has not been advanced in the scholarly literature. It is not surprising that this attitude should exist, since very few writers and scholars studying leadership are political scientists. But, even they are not as clear on the role of politics in leadership as they sometimes are on the role of leadership in public politics. Of those political scientists who do study and write about leadership most are constrained by disciplinary predilection to focus on leadership as elite behavior within established institutional frameworks such as the American presidency. Few approach leadership, as I will do here, as a political process occurring within human societies at all levels and in almost all (if not all) forms of society. Many recog nize that leadership often involves political characteristics, and, certainly, eminent political scientists have focused on leadership as a critical element in the success or failure of governmental office holders, party officials, and the like. Scholars in other disciplines conclude much the same thing in their studies of nongovernmental leaders, but no one has unambiguously argued, as I propose to do here, that the conceptualization of leadership may be redirected and refined with recognition that politics is the central, common element in all leadership.

To be sure, as concepts, both politics and leadership provide fertile ground for dispute and confusion. To consider them together requires a merger of disciplinary perspectives that may challenge some conventional ways of thinking. Such a merger requires some revisiting of well-worn ideas and some reinvention. It maybe fair (some may contest this assertion!) to say that most students of leadership have an inadequate appreciation and understanding of politics and most students of politics have a similarly inadequate appreciation and understanding of leadership. Although the latter may be in general agreement that politics, generically, is a social process through which contested values are distributed, they rarely pursue the implications of this basic conceptualization into the informal or nonpublic realm. The crux of the matter is that conceptual clarity and precision is at the heart of any theory, and leadership is a highly abstract concept extremely difficult to make either clear or precise. The latter is particularly significant because no attempt at conceptual clarity for leadership seems possible without dealing with other such concepts, such as politics, in the process.

On Politics

It is difficult to imagine two concepts more abstract and illusive than leadership and politics, yet we must deal with each and join them in productive and parsimonious ways if a conceptual basis for theoretical development is to be established. A start may be made by restating a defining conceptualization of politics and then proceeding to construct a conceptualization of leadership on that foundation. David Easton’s now classic definition of politics may be taken as starting point. Easton proposed what became a landmark definition of politics in asserting that “politics is that social process through which values are authoritatively allocated for a political system” (Easton 1971, 143-44). That definition has been criticized for being too “system oriented” in that it fixes the processes in question in the working of established, sovereign, governmental systems. Much of that criticism can be met by broadening the concept to consider politics as a generic phenomenon that occurs within all social structures or systems, however informal. As Adrian Leftwich put it, “the fact of the matter is that, unless one adopts a very narrow view of it, politics is a pervasive feature of collective human life” (Leftwich 1990, 3).

Thus, I suggest that politics should be understood as encompassing those social processes through which contested values are allocated. In this conception all that is required for politics to occur is a conflict over the allocation of values within any social set of two or more actors. A simple difference of preferences among interacting individuals is sufficient to trigger the political process. It does not require the existence of any particular level, form, or structure of formal “system” to exist. The notion of conflict, implicit in Easton’s formulation, is made explicit here in the form of the word “contest” but is not in any way restricted to any particular cultural notion of conflict. All that is required is some difference in perception among participants about preferred outcomes. Such an approa ch allows us to comprehend politics as a social process which, in all cases, has certain fundamental commonalties. Any finite, though probably varying, number of such processes then may construct or come to be components of those systems more conventionally recognized as the family, group, organization, or state.

The nature of the politics of these familiar entities can be seen in terms of certain basic common characteristics as well as in terms of their differences. The differences in the allocative processes of the state and those of an elemental pair (“dyad”) are fundamentally differences of magnitude and not of kind. Politics structures political systems and political systems may be transient or persistent and may acquire characteristics having varying degrees of formality and institutionalization. From this perspective any two or more persons may compose a political system while they are engaged in interactions through which values that are in contest among them are distributed. To the extent that those interactions produce second-order values, such as the rules of interaction which themselves are contested and distributed, and do so on a repeated or habitual basis, the system or systems become more or less institutionalized. Embedded in such processes must be a significant set of values that deal with the emergence, claim and recognition, and performance of roles within the interacting set. Principal among those roles must be leadership.

Linking Politics and Leadership

With few notable exceptions, such as James MacGregor Burns’s Leadership (1978), attempts to pursue an explicit understanding of the political nature of leadership have been conspicuously absent from the literature. Effectively, Bums saw leadership as a sociopolitical process enacted by individuals in interaction with others, but even he seemed focused on leadership in public politics rather than on the politics in leadership. (Or at least, Bums seems to have been convinced that the most meaningful forms of leadership are those that operate at the highest levels of public politics.)

Bums’s work has significantly influenced serious leadership studies in the last 20 years, particularly as exemplified in the efforts of Bemard Bass and others to adapt and operationalize Bums’s notion of “transformational leadership.” Bums argued that there is a dichotomy of leadership process types ranging from the transactional, which emphasizes exchanges of values having the net tendency to maintain the status quo, to the transformational, which emphasizes the advancement of both leaders and followers in the Maslowian hierarchy of values. The value bias inherent in Bums’s formulation is revealed in his refusal to accept the possibility that individuals who interact with others within a different, undesirable, structure of values (which he called “power wielding”) should be considered leaders at all. Ultimately, for Bums, leadership is unidirectional in, at a minimum, maintaining the existing order of values or, at a maximum, transforming that order into a higher, more moral form which produces better indi viduals and better communities.

Although initially accepting Burns’s dichotomous view, Bass eventually responded to Bums by taking up the notion of transformational leadership as the seminal concept while modifying it to recognize that transformational leadership can be multidirectional and need not be tied to a particular set of norms and expectations and, as well, is apt to be simultaneously transactional in nature, so that the leader who seeks to transform may also (necessarily) be a cutter of more mundane deals in the process (see Bass, 1990 and 1996). Interestingly, for my purposes, Bass and many others who have taken up the theme have largely ignored the contextual framework of Burns’s argument–that leadership is always a political process embedded in a social and cultural environment–while continuing to search for knowledge about how individuals enacting the role of leaders can transform that environment. Some useful insight might be produced by trying to be more straightforward in identifying and explaining leadership as a fundam entally political phenomenon and, thus, bring its most fundamental nature and function into the light of day. There is some danger, however, that some political scientists may think the argument banal while some students of leadership may think it outrageous to consider the political processes embedded within leadership processes to be essential to the definition of the phenomenon itself.

The groundwork for a conception based on the relation of leadership and politics does seem to have emerged. There is general agreement on some basic characteristics of leadership. These include: (1) it exists; (2) it is asocial process that occurs wherever society occurs; and (3) it has some necessary bearing on the performances of groups and organizations within those societies. In addition, it plays a significant role in the communications processes though which those groups, organizations, and larger societies emerge and change. A recent text identifies three common themes in leadership theory that focus on (1) the exercise of influence, (2) the group context, and (3) the collaborative or reciprocal nature of the relationships between leaders and followers (Hack-man and Johnson 1996, 12-13). All of these may be taken to suggest that leadership is always a particular form of social process embedded in all social sets which is characterized by, above all else, the reciprocal interplay of influence among mem bers of those sets in which one or more of those members exercises disproportionate influence on the choices and acts of the group (set) and its members. In simple, leadership is a political process; why not try to define, study, and understand it as such?

Many students of leadership will admit that “the leadership process as a whole is highly political, requiring the exercise of political skills by organizational leaders as well as extended periods of negotiations and adjustments among participants” (Robertson and Tang 1995, 9). But few then attempt even a marginally sophisticated explanation of what they mean by “political.” Perhaps, as the previous quote suggests on closer examination, there is a confusion born of compound meanings, as in the confounding of leadership process with hierarchical position or in the lack of useful distinction between a political “process” and “political skills.” These authors are not alone in this dilemma and, in fact, are quite representative in their usage of these terms. This has long been a difficulty encountered in leadership studies. The political scientists’ understanding of politics as a complex social process through which conflict emerges, is managed, and resolved in a continuing process not only suggests a new way of identifying leadership but flies in the face of the so often implicit notion that leadership must be beyond politics. Indeed, nearly forty years ago, Sydney Verba identified what he called the “‘no-conflict’ assumption: i.e., that there is a single group goal or a single method of attaining a group goal that is in the best interests of all concerned –both leaders and followers” (Verba 1961, 222).

Gregory Yukl, in a widely used text, asserts that “a prolonged, highly political decision process is likely when decisions involve important and complex problems for which there are no ready-made, good solutions, there are many affected parties with conflicting interests, and there is a diffusion of power among the parties” (Yukl 1998, 20). This statement unquestionably implies, or assumes, that there might be such a thing as a nonconflicted and thus nonpolitical decision process in any group setting (which is inherently preferable for being nonpolitical!). It may be too easy to suggest that error lies in the fact that a decision is no less political when made authoritatively by virtue of concentrated power, regardless of the amount or level of conflict, than it is when made as a result of prolonged conflict; nor even when it is made by unspoken consensus, it is only differently political, not less political! Yukl does place his comment in the context of the notion that decision processes in groups and organ izations tend to be “disorderly and political” but I think the point is made well enough: decisions and leadership may be more or less conflicted but they, by proper definition, cannot be less “political” or less than “political”! These are the sorts of issues that need more attention.

Clearly, leadership must be understood to involve more than the exercise of formal authority. It must also be understood as a critical element in the process by which authority is both created and sustained (Weaver 1991, 161). The “office holder” may not be the locus of leadership in a social structure. A concern with leadership qua power and authority in formal organizations effectively diverts attention from the structures and processes of informal power and authority and, therefore, leadership within those organizations or in other social structures (Weaver 1991, 162). Leadership must involve more than performing an office; it must define and be defined “by virtue of intricate reciprocities of behavior and perceptions” (Weaver 1991, 162). 1 will repeat here a position that I have made before that “leadership is a generically political role that has something important to do with initiative in the definition, articulation, and/or authoritative allocation of values in any social construct” (Weaver 1991,162) . An old, apocryphal riddle asks “What is the difference between a politician and a statesman?” and is answered with the observation that “a statesman is a politician with whom one agrees.” So too, with most students of leadership, “a nonleader is anyone who acts the same as a leader except that we disagree with her or him in some significant way.”

By extension, many would say leadership is that which produces a desired outcome or effect rather than an outcome that we do not desire. That leadership is substantially in the eye of the beholder is an unsurprising realization. Yet, the perception of the act of leadership is seemingly inextricably to be confused with the act itself. The conceptual and theoretical difficulty with these tendencies is that they are rooted in some notion that there is a separate arena in social life reserved for the political and that, therefore, there are arenas of social life in which leadership operates but politics does not. This is not a tenable position, yet it is one to which t he bulk of leadership studies cling. That leadership and politics are abstract labels for a complex sociopolitical processes of significance in all human social activity may be taken for granted by political scientists, but they are not so taken by most students of leadership.

It may be helpful to return to basics in understanding what should be meant by the use of the term “political.” As Peter Corning put it in his powerful analysis, The Synergism Hypothesis: A Theory of Progressive Evolution: “…political systems include the subset of all imaginable cybernetic systems that are social organizations of some sort. Thus politics is not at heart a separate and specialized sphere of social life; it is an aspect or dimension of all organized social life” (Corning 1984, 6). Or, again from Corning, “…politics is not an epiphenomenon, not a distillate of economic activities, or of the class struggle, or of the machinations of ambitious leaders. Politics is a natural and necessary process of social life, a process that occurs when two or more individuals come together to work out a shared problem or to coordinate their efforts toward some shared goal, such as raising children or making war” (Corning 1984, 7).

If leadership is to be considered a social process, in the basic and generic sense, then it must follow that it is rooted in the political aspect of all social life. That political aspect might be approached by its degree of formality, but the degree of formality cannot determine its existence. There is a certain safety in restricting our study of politics to public government and there may be illusory assurance in studying leadership from a deliberately nonpolitical perspective, but there is too little truth. It would seem therapeutic, at least, to give further consideration to politics in the full generic sense as a universal property of all leadership. And, if politics is ubiquitous and necessary to all organized social life, then we might be able to suggest that leadership also is a necessary process in organized social life. Ralph Stogdill argued that “leadership is an aspect of organization” (Stogdill 1950, 1-4). It should be taken as given that all organized groups (pardon the redundancy!) are political systems. Such logic, however, does not permit us to draw any conclusions ab out what kinds of leadership better serve such groups, only that it helps enact their organization.

House and Aditya have made a substantial case for both the achievements and the difficulties of leadership studies and noted that one of the major problems is that the “current study of leadership … continues to focus excessively on superior-subordinate relationships to the exclusion of several functions that leaders perform and to the exclusion of organizational and environmental variables that are crucial to effective leadership performance” (House and Aditya 1997, 465). They also assert that there are generic leadership functions which “represent broad classes of specific leader behaviors” (House and Aditya 1997, 449). The idea is that a generic function can be enacted by diverse behaviors in different settings, e.g., group maintenance may be seen as such a generic function which is performed by leaders in most, if not all, settings. The notion of generic functions is useful if not applied too selectively. After all, the same role may be served in different cultures by different behavior. The variety of behavior that serves the political nature of leadership will not deny the fact that a fundamental function (it may best be called a metafunction) of all leadership, regardless of setting, is to be political. Once this step is taken, it may become more possible to identify the behaviors which serve that metafunction.

Much of the leadership literature, however, concentrates on trying to discover what behaviors constitute effective organizational leadership while assuming the metafunctional context and considering only instrumental functions such as group maintenance. The tendency to focus on the latter suggests some avoidance of the more fundamental question of what is leadership in all cases; that is, if there are generic functions performed by leaders in all possible social settings, what are those functions, not just what are their functions in “effective organizations.” Interestingly, House and Aditya “believe that there are indeed several leadership functions generic to the exercise of leadership” and “research needs to be directed toward establishing the generic functions, the conditions requiring their performance, and the specific leader behaviors required to enact these functions” (House and Aditya 1997, 451). Effective leadership may be no more nor less than the perceived result of the practice of effective poli tics.

Usage of leadership concepts which derives from the leadership desired, or the results desired, rather than from a common generic function betray a misleading normative bias. To say that leadership is manifested in a political environment may avail us of little if we then become enmired in locating the criteria for “good” leadership and “good” politics! The approach to this sort of dilemma may best begin with an acceptance of the objective coexistence of leadership and politics and work from there toward a conception that is less value-bound.

There have been some extensive discussions of the conception of leadership as a political phenomena such as Burns’s Leadership, and Robert Tucker’s Politics as Leadership. Both propose that leadership be placed at the center of politics. No one, however, has turned the issue fully on its head by considering politics as the essence of leadership! In some ways it may be unremarkable to claim that leadership is always a political process, but unless one becomes comfortable with the understanding that all social systems are also political systems, regardless of other characteristics, it may offend certain sensibilities, Words like politics and power give rise to unpleasant cognitive associations which deter many from accepting such an approach. Perhaps this is a reason why a conceptual discussion along these lines has not heretofore been given the attention it deserves.

While there has been a near universal avoidance of the notion that the key to the leadership phenomenon is its fundamental political nature, the problem has been compounded by the failure of most analysts successfully to distinguish “leadership as process” from “leadership as property.” As the presumed property of those possessed of status, rank, or position, leadership is reduced to a subordinate aspect of preexisting social structures. Hence, overwhelmingly, research on leadership is conducted on the presumption that leadership is a property associated with formal position within formal organization. In another sense, if leadership is considered the result of the personal properties of an individual actuated in a situation, then the leadership process is dependent on those properties and is apt to produce a skewed pattern of investigation which seeks to identify the most desirable traits and characteristics according to criteria of desired outcomes. Approached as a generic process independent of such presu mptions, leadership can be understood to be a political phenomenon which, like any other political cum social process, has no necessarily desirable outcome but has significant influence on the kind and nature of the outcomes that do eventuate. That leadership is a relational phenomenon is a common observation, but seeing that phenomenon as having identifiable processes and effects that occur in all cases should allow a greater degree of separation between subjective desire and objective observation. After all, we should be able to identify what processes are characteristic of leadership in all cases and treat that as an issue distinct from what may be the preferred leadership process in a given instance.

It may be fair to say that the modern industrial paradigm has made it natural for most to “assume the necessity of hierarchical structures and the certainty of limited power assumptions” (Murrell 1997, 35). It may also be fair to say the modern political paradigm has made it natural and convenient to think that all political processes take place within hierarchical structures of power. Both points are underscored by the popular myth that “complex social entities can be lead through the acts and will of a single individual” (Murrell 1997, 36). Similarly reinforcing of many views is the “ancient myth about leadership that it is the source where significant rewards and punishments are parceled out in order to get people to do what is good for them” (Murrell 1997, 37). Thinking of both leadership and politics as relational phenomena does not, however, eliminate the individual as a critical component of the leadership process but, rather, places the individual in a more realistic perspective as a participant in a social setting which has certain identifiable characteristics marking a political process. Perhaps most notably, these characteristics include the identification of a conflict of values, the definition of that conflict, and its articulation.

As Murrell suggests, studying the “social context for leadership allows for research and theory building to go well beyond just the leader-follower exchange relationship” and allows us to look into “the whole process by which social systems change and to see the socially constructed roles and relationships developed that might be labeled leadership” by putting the emphasis of study “squarely on the human processes of how people decide, act and present themselves to each other” (Murrell 1997, 39). Seeing a group or organization as a sociopolitical system also is to see them as communities but this does not fit easily into traditional, conventional frames of reference. It is not easy to get past conventional assumptions about hierarchy, ownership, and relationships which tend to take authority and position for granted.

Insufficiently examined conventional assumptions have brought many to accept the conundrum that leadership defies conceptual precision and easy understanding. Some then attempt to address the subject anyway, others take the meaning of the concept more or less for granted, and still others reject it as unsolvable, then proceed to address whatever collateral topic forms their primary interest (the list is too long but some examples are “Presidential leadership,” “administrative leadership,” “managerial leadership,” or “religious leadership”–any and all of which may be taken to mean the study of elites within particular settings). Such studies can be useful and may add incrementally to the growing fund of knowledge about leadership, but there may be reason to doubt their ultimate, larger utility. Most have as their aim, if not their subject, the elucidation of methods of leadership which may be effective given preexisting organizational goals and thus have more to do with identifying successful tactics within necessarily idiosyncratic settings. I would suggest that students of leadership need to reconsider a straightforward question: how are we to identify leadersh=ip when it appears without taking for granted, as noted by House and Aditya, that it must be revealed in formal “superior-subordinate relationships” (House and Aditya 1997, 465)?

This essay takes an approach to the concept of leadership which is both heuristic and theoretic. The attempt is to identify the fundamental, generic qualities of leadership as a political process embedded within social processes. I take the elementary question of what is leadership to be a necessary predicate to any effective answer to derivative questions such as how to lead or what forms of leadership are most valuable and thus choose to concentrate on the former. Great care must be taken in this enterprise, to be sure, since attempts to comprehend leadership often run afoul of subjective biases, confused thinking, and inaccurate assumptions. Kenneth Janda discussed the nature of the problems which occur when conventional words are taken into the specialized “vocabulary of those attempting to construct a systematic body of knowledge about social behavior” (Janda 1960, 345-47). He identified “at least two” of those problems as (1) “the delusion of sufficiency” and (2) “confusion by similarity.” The first re fers to a “premature satisfaction with the analytical utility of the concept being proposed.” The variety of meanings normally associated with a word may not account for built-in contradictions and inconsistencies, for example. In other cases the word in question might be used too hastily in a taken-for-granted manner which may not support rigorous analysis. The “delusion of sufficiency produces concepts which are not analytically tight and are therefore inadequate for exacting study” (Janda 1960, 346). The second, confusion by similarity, “relates to the entanglement of a carefully formulated concept with one or more other analytically distinct concepts that share the same label” (Janda 1960, 346).

I here suggest four primary theoretical assertions which hold that leadership (1) is always, in every case, a political phenomenon; (2) is a phenomenon which is necessary but not necessarily sufficient to group syntality (the various performances exhibited by the group in an effort to achieve a goal); (3) may best be approached as an emergent and contingent process phenomenon within all social systems; and (4) within any given social system, in particular as its complexity increases beyond the most primitive levels, leadership roles and functions will be distributed among and circulate in kind, degree, and character among the actors within the system. Again, all social systems are explicitly understood also to be political systems, whether formal or informal, implicit or explicit, in nature. Nearly all organizations will be understood to have multiple leadership actors and performances within them which may or may not coincide with their formal, institutional structure. Leadership, thus, may be treated as a political role-process, or pattern, that occurs at many points, and often concurrently, within any social system.

One fundamental proposition here is that leaders emerge (that is, leadership processes are engaged) in a decision incident resulting from conflict within the social set.

Each such incident may be treated as a discrete event in which one or more individuals have dominated in decisions about the distribution of contested values. Such domination occurs when someone succeeds in imposing, or having the greatest influence in fixing a definition of the situation or interpretation of the perceived environment among the members of the system. As Wilson and Rhodes comment in the Journal of Conflict Resolution in 1997, “a leader is someone who provides a signal around which others rally” (Wilson and Rhodes 1997, 768). It is the repetition of such events that, in varying degrees, results in the informal and/or formal constitution of leadership within the system. The persistence of leadership as both role and process, which sometimes develops into formally constituted roles or “offices,” may be seen as a function of patterns of social reciprocity in which a rising frequency of deference to these individuals in subsequent incidents would play an increasingly persuasive part. Nevertheless, the essential act of leadership by an individual is best approached as a discrete incident, not as a continuous role.

Ultimately, leadership needs to be understood as more than a transient political process since leadership as an ongoing social process performs the vital functions of maintaining or redefining a social system’s linkages among its past, present, and future. Only rarely should social systems be understood to rely for their continuation on a single individual playing the leader role. Specific acts of leadership may be necessary but rarely will they be sufficient. Social systems will survive and prosper, decline and fail as the web of leadership is woven from the emergent acts of their members.

As a process within a process, leadership first must be approached in its preinstitutionalized manifestations. The formalization of authority into “public,” institutional, or organizational roles is, in strictest terms, a post-emergent phenomenon; the identification of roles that transcend a given situation as conventionally conceived leadership roles follows from the nature of leadership in its most basic form as an emergent and discontinuous behavioral phenomenon occurring in one or more structures of social interaction. The basic and universal political nature of leadership as a discrete and discontinuous phenomenon lies in the emergence of one or more actors within a set who give form and direction to the contest of values which defines the set in the first place. This notion arises from thinking similar to Michael Foucault’s when he suggested that participation is the initiating moment in the formation of new networks of power (in Bachrach and Botwinick 1992, 42). The ongoing formation and reconfigurati on of networks of power based on participation in the social set may be considered the continuous process embedded in which is the discontinuous emergence of leaders. This supports an important point made by Mughan and Patterson in that “leadership cannot inhere in individuals because it is a relational phenomena” (Mughan and Patterson 1992, 8). The confounding of leadership, authority, and hierarchical position is not uncommon, of course, but the idea that authority is an inherent characteristic of leadership which derives from the relationship among members of the social construct, and is not an attribute of the identified leader, points us to the need for a conception of process which separates the two. As Welsh observed “there is no necessary relationship between the occupants of high positions in which formal power is concentrated (the political elite) and the identity of political leaders who mobilize human resources for task achievement” (Welsh 1979, 19). It may be therapeutic simply to remove the word “political” from Welsh’s sentence more clearly to see how the thought applies to any human organization.

One of the most significant difficulties to be encountered with thinking about leadership in the manner suggested here is that it does little to relieve us of a major underlying conceptual problem: where are the meaningful and observable boundaries of the political system in question and of the leadership processes emergent within that system? Clearly, we can be tempted to use the idea of leadership to identify acts which are so distant from the response that the definitions and functions we have described become unusable in any practical way. A great artist may set the stage for a new stylistic school and thus seem to fir our criteria but not provide us with conclusive relation between leaders and followers. A political system can be observed, as defined above, in indirect, distant, and illusive relations but become so vague as to defy rigorous study. The problem can be approached with some parsimony if it is framed as a question of social boundaries. It may be plausible to propose that just as the fundamen tals of all political processes first occur within small groups so too must leadership manifest itself most clearly in those groups. Here I will borrow an old idea well expressed by Karl Deutsch (1980, 134-38) in the notion that the boundaries of a political system can be identified in terms of frequencies of transaction. A social group, and its attendant internal political process) can be identified in terms of the relative frequency of transactions as among one set of actors and another set identified by the same means. Thus, membership in a group is enacted by the behavior of a set of actors in terms of each other. In this way, the effective boundary of a group might be established by comparing the distributions of frequencies of transaction among individuals or, potentially, among sets of groups since small groups are subsystems of larger social systems. Nevertheless, it may be that the appropriate place to begin leadership study is at the level of small groups wherein the frequency of transactions is sufficiently high to provide a discernible boundary and the pattern of those interactions reveal a leadership process.

Leadership emerges from relationships but it is insufficient to say that leaders emerge from group interaction alone. There must be a pattern as well as a frequency to that interaction, but neither tells us anything about the content of the interaction. The pattern should be triggered by the primary political process in which the functions of conflict processing and resolution are performed. The intensity of interaction–a function of time and frequency taken together–may reveal even more. Small groups which have a stable leadership structure may be able to resolve conflict more efficiently than those with unstable structures which must spend time searching for leadership in the conflict resolution process (see Verba 1961, 159).

Any formal organization can be expected to be a structure built of many small groups, some formally defined, some not; some transient, some persistent. But one might be best advised not to start with the most formal and largest structure of an organization to understand its leadership processes, rather with the small formal and informal groups that exist within that structure. If leadership is understood to begin (emerge) in any setting as someone engages in the combined acts of identifying, defining, and articulating a situation in a way which elicits positively correlative behavior from others in the setting then it may be concluded that the essential political criterion of leadership has been fulfilled. By accepting and acting upon a definition of a situation proffered by someone else some subset of the group has accepted the distribution of values contained therein. That identifies the essential pattern of leadership in its most basic form.

It is not within the intention of this essay to go further than making an argument for seeing leadership as a political phenomena. However, some general lines of attack seem to recommend themselves. First, a revisiting of what was once called the “small-group” approach would help deal with two of the major problems posed by the study of leadership: (1) the boundary problem — are there effective limits on the size (“population”) of the system in which effective leadership can occur; and (2) the levels of analysis problem — at what level of both size and complexity can effective leadership occur? In other words, if leadership is a small-group phenomenon, then how can we apply the idea to large groups? One response might be to look for the structure (pattern) of structures (patterns) within complex systems. The complex reciprocities which exist among individuals in small groups may be the pattern on which the structure of larger groups is formed. Multiple group memberships combined with the inevitable cross-c utting reciprocities and strata of participation maybe found to have their primary expression in the political processes of small groups. One may hypothesize that the interactive leadership processes among and between small groups create, in an on-going way, the leadership processes and political systems of larger, formal organizations. Perhaps the most poignant difficulty in our thinking about leadership may be found in the notion that its purpose is to produce order rather than to help us confront the unknown from which order produces itself.

Clearly, any effort to understand leadership faces daunting problems. Both the virtue and the vice in the way of thinking which has been suggested in this essay lies in its generally objective attitude; it does not immediately help us to discover the secrets of how to lead but it may hold some promise for helping us to see why and how all leadership processes take form and, thus, allow us to ask better questions. Hopefully, it will help bring leadership study more prominently into the range of vision of social and political scientists.



BACHRACH, P., and A. BOTWINIK. 1992. Power and Empowerment: A Radical Theory of Participatory Democracy. Philadelphia: Temple University Press.

BASS, B. 1996. A New Paradigm of Leadership: An Inquiry into Transformational Leadership. Alexandria VA: U.S. Army Research Institute for the Behavioral and Social Sciences.

BURNS, J. M. 1978. Leadership. New York: Harper and Row.

CORNING, P. 1984. The Synergism Hypothesis: A Theory of Progressive Evolution. New York: McGraw-Hill.

DEUTSCH, K. 1980. Politics and Government: How People Decide their Fate. Boston: Houghton-Mifflin.

EASTON, D. 1971. The Political System: An Inquiry into the State of Political Science. 2d ed. New York: Knopf.

FOUCAULT, M. 1980. Power/Knowledge. Ed. Colin Gordon. New York: Pantheon Books.

HACKMAN, M., AND C. JOHNSON. 1996. Leadership: A Communication Perspective. 2d ed. Prospect Heights, IL: Waveland Press.

HOUSE, R., and R. ADITYA. 1997. The Social Scientific Study of Leadership: Quo Vadis? Journal of Management 23:409u73.

JANDA, K. 1960. Towards the Explication of the Concept of Leadership in Terms of the Concept of Power. Human Relations 13:345u63.

LEFTWICH, A., ed. 1990. New Developments in Political Science. Brookfield, VT: Gower Publishing Co.

MUGHAN, A, and S. PATTERSON. 1992. Political Leadership in Democratic Societies. Chicago: Nelson Hall.

MURRELL, K. 1997. Emergent Theories of Leadership for the Next Century: Towards Relational Concepts. Organization Development Journal 15 (3):35-42.

ROBERTSON, P., AND S. TANG. 1995. The Role of Commitment in Collective Action: Comparing the Organizational Behavior and Rational Choice Perspectives. Public Administration Review. 55:67-80.

STOGDILL, R. 1950. Leadership, Membership, and Organization. Psychological Bulletin 47:1-4

TUCKER, R. 1981. Politics as Leadership. Columbia, MO: University of Missouri Press.

VERBA, S. 1961. Small Groups and Political Behavior: A Study of Leadership. Princeton, NJ: Princeton University Press.

WEAVER, D. 1991. Liberalism and Leadership: Lockean Roots. Leadership Quarterly 2(3): 157-74.

WELSH, W. 1979. Leaders and Elites. New York: Holt, Rinehart and Winston.

WILSON, R., AND C. RHODES. 1997. Leadership and Credibility in N-Person Coordination Games. Journal of Conflict Resolution 41(6): 767-91.

YUKL, G. 1998. Leadership in Organizations. 4th ed. Upper Saddle River, NJ: Prentice Hall.

COPYRIGHT 2000 Michigan Academy of Science Arts & Letters


This article was originally published with the same title in Michigan Academician, 2000 (Gale, Cengage Learning).

David R. Weaver

Written by

David Weaver is Professor Emeritus of Political Science at Saginaw Valley State University. He received his Ph.D. from the University of Cincinnati in Political Science. His major areas of interest include foreign relations and political development, political thought and theory, and, primarily, leadership and democratic theory and practice.