UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Toward a conception of educational leadership Coombs, Charlotte Rosalind Compton 1988

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1988_A2 C66.pdf [ 8.16MB ]
JSON: 831-1.0055792.json
JSON-LD: 831-1.0055792-ld.json
RDF/XML (Pretty): 831-1.0055792-rdf.xml
RDF/JSON: 831-1.0055792-rdf.json
Turtle: 831-1.0055792-turtle.txt
N-Triples: 831-1.0055792-rdf-ntriples.txt
Original Record: 831-1.0055792-source.json
Full Text

Full Text

TOWARD A CONCEPTION OF EDUCATIONAL LEADERSHIP by CHARLOTTE R.C. COOMBS A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION in THE FACULTY OF GRADUATE STUDIES Department of Administrative, Adult, and Higher Education We accept this thesis as conforming to the required standard THE UNIVERSITY OF April ® Charlotte R.C. BRITISH COLUMBIA 1988 Coombs, 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of / U m ^ - r m W ^ / U u l ^ /and. H ^ U - r e d u c t i o n The University of British Columbia Vancouver, Canada Date Af?n\ mf\ DE-6 (2/88) ABSTRACT The question which animates this thesis is the extent to which study of extant leadership theory and research is relevant to the fostering of educational leadership. Because this question cannot be answered without an adequate account of leadership in education, the thesis develops * a conception of educational leadership and assesses the relevance of the leadership literature in light of that conception. The foundations upon which this conception of educational leadership is built are as follows: (1) the ordinary language meaning of "leadership," (2) the ideals fundamental to a liberal, democratic society, and (3) our ideal of the educated person. Conceptual analysis of "leadership" reveals that it is used both descriptively (to denote positions) and normatively (to rate something positively). The normative use spans a continuum of standards from weak to strong. Weak-normative leadership is ascribed according to the extent to which a leader has been able to influence others to pursue a goal. Strong-normative leadership is ascribed according to the extent to which the leader's ends and means are worthwhile. It is argued that, in general, people who want educational administrators to exercise leadership want strong-normative leadership, i.e., they want administrators to envision worthwhile goals and to use morally appropriate means to influence their colleagues to pursue these goals. Thus, the sense of leadership upon which the conception of educational leadership is based is the strong-normative sense. ii The ideals taken to be fundamental to a liberal democracy are: (1) equal respect for persons, (2) use of intelligence in problem solving and promoting change, (3) cooperation and pooled experience in setting values and solving problems, and (4) respect for individual rights. The ideal of the educated person is characterized in terms of the acquisition of worthwhile knowledge through means which respect rational autonomy. A conception of educational leadership compatible with these ideals is explicated and defended. Rival conceptions of educational leadership are examined. It is argued that the conception of educational leadership developed here is preferable to its rivals in that it incorporates their strengths and avoids many of their weaknesses. Leadership theory and research are examined and shown to be generally incompatible with the conception of educational leadership explicated and defended in this thesis, and with the ordinary language concept of leadership. The thesis concludes that it is probably not appropriate to make such theory and research the central component of courses which aim to foster or improve the exercise of educational leadership; rather, study of works which attempt to clarify and justify educational goals and means would seem to be more promising. iii TABLE OF CONTENTS Chapter I. PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM 1 A. The Importance of Leadership to Administrators 5 B. The Relationship Between "Leadership" and "Administration" 9 C. Overview of the Thesis 11 Chapter II. THE CONCEPT OF LEADERSHIP 14 A. Conceptual Analysis 14 1. The Usefulness of Analysis 14 2. The Potential of Ordinary Language 16 3. The Nature of Conceptual Analysis 20 4. Conceptual Analysis Versus Other Means of Determining Ordinary Language Meaning 23 B. Conceptual Analysis of "Leadership" 26 1. "Lead," "Leader," and "Leadership" 26 2. "L is Exercising Leadership" 29 a. Model cases 30 b. Hypotheses 32 3. Testing the Conditions for Sufficiency 51 4. The Conditions of Exercising Leadership 56 5. "Followership" 57 6. Resolving Conceptual Confusion 59 C. Summary of the Senses of Leadership 60 D. The Centrality to Education of Strong-Normative Leadership 61 Chapter III. EDUCATIONAL LEADERSHIP 64 A. Developing a Conception of Educational Leadership 66 1. The Ideal of a Liberal Democracy 67 2. The Ideal of the Educated Person 71 B. Competing Conceptions of Leadership 78 1. The Democratic Conception of Leadership 79 2. The Congruency with Organizational Values Conception of Leadership 83 3. The Moral Conception of Leadership 86 C. Outline of a Conception of Educational Leadership 87 1. The Goal 88 2. The Means 89 3. The Conditions of Exercising Educational Leadership 97 4. Compatibility with the Democratic Principles 98 D. Fruitfulness of the Conception for Educational Administrators 99 Chapter IV. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH 103 A. Lakatos' Theory of Scientific Research Programs 104 B. Early Approaches to Leadership 105 iv 1. The Trait Approach 105 2. The Style Approach 109 3. The Behaviour Approach ^ I l l 4. Leadership in Groups 115 5. Summary 116 C. The Organizational Effectiveness Research Program of Leadership 117 1. The Contingency Approach 118 a. Fiedler's contingency theory 119 b. Hersey and Blanchard 121 c. House's path-goal theory 123 d. Contingency theory: future prospects 125 2. The Limitations of OERPL 126 a. Tendency toward the trivial 127 b. Failure to justify ends or means 129 c. View of persons 134 d. Summary 135 D. Other Leadership Theory and Research 137 1. Social Exchange Theory 137 2. Vroom's and Yetton's Decision-making Model 140 3. House's Theory of Charismatic Leadership 141 4. Kerr and Jermier 142 5. Summary 143 E. Generalizability of Research 143 F. Conclusion 146 Chapter V. WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS 148 A. Summary of the Thesis 148 B. Implications of the Thesis 151 1. Implications for the Preparation of Educational Administrators 152 a. The problem with current programs 152 b. What can be done? 154 2. Implications for the Selection of Educational Administrators 157 a. Criteria of selection 157 b. Leadership and power 160 3. Implications for Research 161 a. Resolving confusion 161 b. Future leadership research 163 C. Conclusion 166 REFERENCES 168 Appendix A: An Example of Conceptual Analysis 183 v ACKNOWLEDGMENTS Without the generous assistance of many people, completion of this dissertation would not have been possible. I am grateful to the members of my thesis committee for their careful reading of the many drafts of this thesis, their suggestions for improvement, and their untiring encouragement. My thanks go to my advisor, Dr. Jean Hills, Dr. Graham Kelsey, Dr. Earl Winkler, and Dr. Lome Downey (Professor Emeritus). To my husband, Jerry, and my colleagues in educational administration, particular^ Trisha Wilcox and Carol Gibson, my sincere thanks for your advice and support. My appreciation goes also to Pat Crehan, who encouraged me to test my ideas by presenting them to students in her leadership classes, and to Jay Handel, who helped me format the thesis for the computer. Finally, I would like to acknowledge the financial assistance of the Social Sciences and Humanities Research Council of Canada and the Isaak Walton Killam Memorial Fund for Advanced Studies. v i CHAPTER I. PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM Leadership is one of the most observed and least understood phenomena on earth. James MacGregor Burns Few concepts have sparked such heated debate, spurred as much research, or inspired as many volumes as that of leadership. Leadership is widely regarded as an important topic to include in preparation programs for educational administrators: recent textbooks in educational administration typically devote a chapter to leadership (e.g., Newell, 1978; Hanson, 1979; Silver, 1983; Snyder and Anderson, 1986; and Hoy and Miskel, 1987) and departments of educational administration throughout North America offer courses on it. Almost invariably, students are required to become familiar with theory and research in leadership, particularly organizational leadership. Presumably, learning about leadership theory and research is viewed as a means to enhance the ability of aspiring administrators to exercise leadership. The predominance given to leadership theory and research in administrative preparation programs is at odds with another frequently made assertion, namely that research results of leadership studies are widely regarded as confusing and disappointing. Mintzberg notes that decrying the state of leadership research is, in fact, the "establishment view" (1982:250). The following quotations are representative: We have masses of findings that no one seems able to pull together. They simply float around in the literature, providing nothing from 1 PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM / 2 which one can push off to anywhere (Lester, 1978:xii). Of all the hazy and confounding areas in social psychology, leadership theory undoubtedly contends for the top nomination (Bennis, 1959:259). Conceptually and methodologically, leadership research has bogged down (McCall and Lombardo, 1978:151). Four decades of research on leadership have produced a bewildering mass of findings (Stogdill, 1974:vii). There is perhaps no area of study in organizational behavior which has more blind alleys and less critical knowledge than the area of leadership (Salancik, Calder, Rowland, Leblebici, and Conway, 1975:81). The study of leadership in the last seventy years has resulted in little accumulated knowledge that permits one to understand or predict the effects of leadership approaches, or that provides a better understanding of how to be an effective leader (Melcher, 1977:94). The wisdom of the widespread practice of exposing would-be educational administrators to the leadership literature is questionable not only because of the paucity of useful research results in this area, but also because such practice appears to ignore very persuasive arguments by scholars in education (e.g., Graff and Street, 1957; Campbell, 1955; and Callahan, 1962) that some of the values, such as efficiency and profit, which motivate and direct much of the organizational leadership research, are inappropriate in the educational context, and that studies done in business or military settings cannot be readily generalized to education. Unlike Schriesheim, Hunt, and Sekaran (1982), who cavalierly liken adopting leadership definitions and models to selecting kinds or flavours of ice cream, these scholars argue that adopting certain definitions or models could have serious consequences. First, the adoption of an industrial model, for example, could lead one to liken teachers to workers in factories, and the adoption of a business model could lead to emphasis on the financial and PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM / 3 mechanical aspects of schools at the expense of the intellectual aspect. The meanings we ascribe to words like "leadership" shape our perception of what is, of what issues are important, and of what questions are worth asking (Coombs, 1980). Secondly, it would seem that having administrators study the organizational leadership literature is not having the desired effect. Administrators appear to be facing a crisis in leadership, or, at least, a "crisis in legitimacy" (Habermas, 1975). According to Foster, "public schools, highly visible and historically revered, are at the center of this crisis" (1980:496). Ott deplores what he takes to be the typical practice on the part of school administrators of resolving practical problems "without possessing firm convictions about educational ideals" which, he argues, leaves them vulnerable to "persuasive fads and half-baked ideas" (1985:2). Boyer would concur: After visiting schools from coast to coast, we are left with the distinct impression that high schools lack a clear and vital mission. They are unable to find common purpose or establish educational priorities that are widely shared. They seem unable to put it all together. The institution is adrift (1983:63). In an article entitled "Where are our leaders?" Williams (1984) claims that in Canada our commitment to and belief in education have declined over the past few years, and that we desperately need leaders with vision.t Administrators are viewed as not having any special knowledge which would enable them to exercise leadership. Maclntyre charges that administrative tit should be conceded that the picture is not uniformly bleak. In a recent Gallup poll (Gallup, 1985), ratings of principals and administrators in public schools were more positive than in 1981. Even so, fewer than half the respondents gave their community principals and administrators a grade of "A" or "B." PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM / 4 expertise is a myth used to "sustain and extend the authority and power of managers" (1981:72). And Robinson contends: Increasingly citizen groups are challenging the professional expertise of education administrators. In many cases, the administrators are found to be wanting in expertise. All this adds up to a loss of faith in administrators and a diminution of administrative authority (1981:6-7). The attention devoted to the study of organizational leadership theory and research becomes understandable, however, if one considers what is generally thought to be the only alternative — not studying leadership at all. Indeed, most professors of educational administration apparently believe that leadership is sufficiently important that it should be studied, however deficient and confused the current state of our knowledge. Assuming that most administrators have studied leadership, it seems reasonable to conclude that having administrators read the literature concerning organizational leadership is not having the desired effect. There are, of course, many ways to study the literature. It could be that studying leadership theory and research from a reflective and critical perspective may be useful, but acquiring such a perspective and assessing the extent to which such study is likely to be useful to fostering the exercise of educational leadership are contingent upon our understanding in rather specific terms what the exercise of leadership in education might be said to entail. Contributing to the development of such understanding is the purpose of this dissertation. In the remainder of this opening chapter, I shall: (1) defend the assumption that leadership is important to administrators, (2) examine the relationship between "administration" PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM / 5 and "leadership," and (3) present an overview of the thesis. A. THE IMPORTANCE OF LEADERSHIP TO ADMINISTRATORS It is widely assumed that administrators have the responsibility to exercise leadership in their organizations — administrative positions are often called leadership positions — and that their ability in this role will have important consequences for the welfare of persons and for the organization. As Vroom expresses it: The effective functioning of social systems from the local PTA to the United States of America is assumed to be dependent on the quality of their leadership (1976:1527). However, the importance of having administrators exercise leadership has been challenged in various ways. Some have argued that it makes no significant difference who occupies a leadership position, either because there is no real scope for the exercise of leadership, or because the exercise of leadership is unimportant. There is also an argument which grants the importance of leadership, but questions the usefulness of studying leadership research and theory in attempting to foster its exercise. Let us examine these arguments more closely. The first challenge to the importance of leadership concerns the situational constraints which can reduce leader discretion. For instance, Zaleznik argues: An organization is a system with a logic of its own and all the weight of tradition and inertia. The deck is stacked in favor of tried and proven ways of doing things and against the taking of risks and PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM / 6 striking out in new directions (1977:68). Bridges (1975) charges that it is not always possible for administrators to exercise leadership, and that it is dysfunctional to set up the expectation on the part of students of educational administration that they will be able to make meaningful changes, because the disillusionment which may result erodes their self-esteem. Pfeffer (1978) echoes this position with his contention that the discretion and behaviour of leaders are constrained once they attain leadership positions. There are always certain constraints on those in leadership positions, particularly those who are accountable to a governing agency; however, Stewart (1982), who has done considerable research in the area of administrative discretion, found that administrators have latitude in a number of areas, including the relative priority they attach to various aspects of their jobs, the amount and nature of what they delegate, and whom they seek to influence. She claims that the most effective administrators view their jobs in strategic terms and recognize opportunities for choice, rather than perceiving themselves as constrained. Similarly, Burch and Danley (1980) speculate that reports by supervisors that external restraints keep them from spending more time on improving instruction may in fact reflect their own failure to accept responsibility for how they allot their time. Perhaps there is some truth in Callahan's contention that administrators may lack autonomy because they are not of a calibre that the public will respect them and be willing to grant them this autonomy (1962:ix). According to Mintzberg, a good leader still has power, even if a poor one does not: PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM / 7 It has become fashionable among researchers (not practitioners) to argue that leadership does not matter. A well-known colleague of ours who has been saying these things, visited us at McGill recently and I asked him why it is that those who make this claim always seem to fight so hard when deans are being changed. Our colleague smiled, slightly embarassed (sic), admitting that he had in fact been deeply involved in such a process recently. When asked why, he answered, "Just in case!" A cute answer, for a cute hypothesis. But pure bull! In his practical mind, he doesn't believe it any more than I do. We both "know" very well that leadership matters, that while some situations are unmanageable, in the vast majority of cases, leadership can make an enormous difference. Both he and I have seen what has happened to different business schools under different deans (1982:253). The second argument claims that, because there is no conclusive evidence that leadership is causally related to organizational effectiveness, its importance has been greatfy exaggerated. Schriesheim and Kerr (1977a), for example, contend that most investigations have yielded non-significant results in terms of leader effects on subordinate performance. One might reasonably attack this argument on its underlying assumption that leadership is important only to the extent that it promotes organizational effectiveness, particularly when "organizational effectiveness" is interpreted as "productivity"; the effect of leadership on the welfare of persons might be significant. However, even if the research linking leadership to organizational effectiveness is inconclusive, the studies that lend support to leadership's making a difference to organizational outcomes far outweigh those that do not. For example, Stogdill (1974) reports that when leaders of groups who successfully complete a task and those of groups who are unsuccessful change places, formerly unsuccessful groups tend to gain in performance and morale, while formerly successful groups tend to decline in performance and morale. It might PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM / 8 also be the case that the various leaders studied in the research are not relevantly different, i.e., they do not differ with regard to the knowledge, abilities, and dispositions which may make a difference to the ability to exercise leadership. If Pfeffer (1978), March (1974), and others are right in their claim that the selection process tends to filter out creative candidates, leaving administrative positions to be filled by conventional thinkers, perhaps a different selection process would provide us with leaders who make a substantial difference to organizational outcomes. Even if it could be established empirically that certain "leaders" tend not to make a difference, one could not conclude that the exercise of leadership is unimportant unless it were also established that the purported leaders were actually exercising leadership as opposed to merely occupying leadership positions. A third argument against the importance of administrators exercising leadership is that there may be some contexts, such as when an organization has a clear sense of mission and the external situation is stable, in which leadership is redundant or unnecessary. Selznick (1957), for example, argues that leadership is most important in the setting-up stages of an organization, when the mission is not clearly defined (although he does not argue that leadership is otherwise unimportant); and Kerr and Jermier (1978) claim that in certain cases, such as those in which followers have training and ability to determine and reach their own goals, leadership is not very important. Such arguments do not establish that leadership in general is unimportant. A person's exercising leadership at just a few crucial times could prove to be PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM / 9 extremely significant to an organization. There is a further argument which, while accepting the importance of leadership, questions the usefulness of studying it, either because it is assumed that leadership ability is something innate, not enhanceable by training, or because it is held that studying leadership theory and research is not a useful way to prepare administrators. Until we have examined what it means to exercise leadership in the educational context, we are not in a position to determine the extent to which leadership ability is likely to be something that can be developed through training or education. And even if we grant that studying leadership theory and research has so far not had the desired effect, it is possible that the problem lies with the nature of the theorj' and research or with the methods by which it is taught rather than with studying leadership theory and research per se. Thus, this argument, rather than undermining the purpose of the thesis, highlights the need for understanding leadership in the educational context. B. THE RELATIONSHIP BETWEEN "LEADERSHIP" AND "ADMIN-ISTRATION" Is leadership most usefully conceived of as a function an administrator might perform in addition to administration? Is it best viewed as a way or style of administering? Or is it best seen as one aspect of administration? Writers take very different stances with regard to leadership and its interrelationship with PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM / 10 administration. And, indeed, the term "administration" itself has: two curiously opposite elements in its meaning: the element of "serving" and the element of "managing" or "directing." A connotation of inferiority and a connotation of superiority (Dunsire, 1973:38). Whereas "leadership" is usually viewed as unsettling the status quo, and "management," as keeping things running smoothly, "administration" seesaws in meaning from one to the other of these two terms. Some writers, such as Selznick (1957), view leadership as distinct from administration or management; the former concerned with affecting the basic character of the enterprise, and the latter concerned with the more routine function of joining available means to known ends (1957:135). Others, such as Hodgkinson (1978), view leadership and administration at one end of a continuum, and management at the other; the former being the art of influencing persons to accomplish organizational goals, and the latter being the science of specifying and implementing means to accomplish the same ends. Mintzberg (1973) and others (e.g., Kast and Rosenweig, 1974) regard leadership as one function of managers or administrators. For Mintzberg, leadership has to do with the manager's relationship with subordinates (as contrasted with the liaison role whereby the manager develops contacts outside the organization, and the role of figurehead whereby the manager carries out social and ceremonial duties). Zaleznik (1977), by contrast, claims that those who are good managers are unlikely also to be good leaders because very different types of people are required for each function. Although there seems to be considerable disagreement over the relationship PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM / 11 between leadership and administration, most would agree that there is no necessary relationship between these two types of activities. It is logically possible for one to exercise leadership without engaging in administration. Conversely, it is logically possible to engage in administration without exercising leadership, although it is an open question how good such administration would be. Despite the independence of the notions of leadership and administration, there are good reasons for regarding the exercise of leadership as one of the functions that a fully functioning administrator should be expected to perform. Persons in administrative positions are well placed for making important contributions to the welfare of persons and organizations. Actually making such contributions, however, requires the exercise of leadership. Moreover, the expectation that good administrators will make such contributions is widespread. C. OVERVIEW OF THE THESIS The concept of leadership is used to categorize, think about, discuss, and make claims about a capacity of persons (or a functional relationship between persons) which is regarded as highly valuable because it is thought to be instrumental in achieving human purposes — both individual and social. Before we can explore the question of what the exercise of leadership in the educational context entails, and assess the contributions of leadership research to it, we need a fuller understanding of the ordinary, non-technical meaning of the term "leadership." That is to say, before we can fruitfully assess and reconstruct our ways of thinking about leadership in education, we need to have a fuller and more perspicuous account of our shared, public concept of leadership. This is PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM / 12 vital, for any adjustments in our ways of thinking about the promotion of leadership must capture those aspects of the ordinary concept which make it important in our thinking about human affairs and the realization of values. The meaning of leadership is captured in the way we use the word "leadership" in language. Thus, in the following chapter the basic standards or conditions which are operative in the ordinary language sense of "leadership" are examined using techniques of conceptual analysis commonly employed in philosophical inquiry. The chapter begins with an explanation of this mode of inquiry. In the third chapter, a conception of educational leadership is outlined, based upon the nature of the educational enterprise in a liberal democracy. I argue that the conceptions of leadership underlying leadership theory and research must be compatible with this conception if such theory and research are to have more than peripheral relevance to the fostering of educational leadership. The fourth chapter examines the assumptions about the nature of leadership and its study which underlie past and current leadership theory and research. It suggests that the failure of this research to make satisfactory progress may be due to conceptual deficiencies in its underlying assumptions. It also argues that much of the research and theory is incompatible with the conception of educational leadership as outlined in Chapter 3. In the final chapter, I set. forth the implications of the thesis primarily for the preparation of educational administrators but also for the selection of educational administrators and for future leadership research. PROLIFERATION WITHOUT PROGRESS: THE LEADERSHIP PROBLEM / 13 To summarize, this thesis is concerned: (1) to articulate what it means to exercise leadership, particularly educational leadership, and (2) to explore the relevance of leadership theory and research to fostering educational leadership. CHAPTER II. THE CONCEPT OF LEADERSHIP The task of this chapter is to determine, by the techniques of conceptual analysis, what we mean when we say that someone is exercising leadership. An explanation of conceptual analysis precedes the analysis of "leadership." The chapter concludes with the argument that one particular sense of leadership is central for our purpose of increasing the abilitj' of educational administrators to exercise leadership. A. CONCEPTUAL ANALYSIS In this first section, I shall consider why conceptual analysis of the expression "L is exercising leadership" is warranted; why ordinary language, in the absence of evidence to the contrary, should be presumed to have special authority; what conceptual analysis is; and why it is the most useful way of reconstructing how competent language users use language. 1. The Usefulness of Analysis Analysis of a concept may be said to be useful when two conditions are met. The first is that the concept figures importantly in our area of interest, in such a way that the interpretation we give it makes a difference to how we treat persons, what programs are implemented, what research is undertaken, and 14 THE CONCEPT OF LEADERSHIP / 15 how research is conducted. The second is that there is confusion or disagreement over the word symbolizing the concept. That is, there is disagreement over the set of rules which govern our use of the word. These two conditions obtain in the case of "leadership." It is clear that leadership is considered to be important in the Field of educational administration. There are journals devoted to the topic; handbooks on leadership; and courses, seminars, and workshops on leadership. Leadership is virtually universally regarded as a good thing and important to educational administrators. However, the precise nature of this good thing is a matter of considerable disagreement. Examination of the organizational leadership literature leaves one more confused than enlightened about the concept of "leadership." One encounters such apparently diverse definitions of "leader" and "leadership" as the following: A leader is a person who possesses the greatest number of desirable traits of personality and character (Bingham, 1927; cited b3' Stogdill, 1978:8). Leadership, properly conceived, . . . serves the individual human goals that our society values so highly (Gardner, 1965:12). Leadership implies influencing change in the conduct of people (Nash, 1929:24). Leadership is the activity of influencing people to cooperate toward some goal which they come to find desirable (Tead, 1935:20). Leadership is social (i.e., interpersonal) influence exercised by a person in some position of superior authority . . . over some subordinate (Pondy, 1978:87). The essence of organizational leadership is the influential increment over and above mechanical compliance with the routine directives of THE CONCEPT OF LEADERSHIP / 16 the organization (Katz and Kahn, 1978:528). Leadership is the initiation of a new structure or procedure for accomplishing or changing an organization's goals and objectives (Lipham, 1964:122). In fact, Stogdill, who devoted many years to consolidating leadership research, concludes: There are almost as many different definitions of leadership as there are persons who have attempted to define the concept (1974:7). Examination of the empirical phenomena which are studied under the rubric of "leadership" is equally confusing, for different studies focus on such different things as traits, skills, behaviours, styles, power relationships, and decision-making. Moreover, some assume that all office holders exercise leadership; some, that only effective office holders. exercise leadership, and some, that office holding is irrelevant to leadership. Thus, "leadership" seems to be a term for which conceptual analysis would be useful. 2. The Potential of Ordinary Language One approach to clarification of the concept of leadership is through analysis of the uses of the word "leadership." Words are, after all, symbolic representations of the concepts we share; if this were not so, communication through language would be impossible. The concept of leadership is equivalent to the meaning of the word "leadership": "the meaning of a word is its use in the THE CONCEPT OF LEADERSHIP / 17 language" (Wittgenstein, 1963:20), that is, usage of a word identifies the category of phenomena to which the word applies. There are several reasons why it is preferable to begin with an analysis of ordinary discourse rather than the language of the research literature. First, the notion of "leadership" began as an ordinary language concept; it symbolized real life phenomena which intrigued scholars and gave impetus to their research. "Leadership" was not a concept developed by social scientists. Obviously, ordinary people regard leadership as very important and desirable in their lives. Only by examining the ordinary language meaning of the term "leadership" can we obtain a clear understanding of what is generally thought to be important when we speak of "leadership." Second, as Sartori (1984) points out, the bulk of our knowledge of ourselves is expressed in a natural language or, as Scriven phrases it, ordinary language was "tailored" to describe the complex problems of life (1966:6). Ordinary language is extremely rich and houses immeasurably subtle distinctions. Our common stock of words embodies all the [verbal] distinctions men have found worth drawing, and the connections they have found worth marking, in the lifetime of many generations: they surely are likely to be more numerous, more sound [than technical language], since they have stood up to the long test of the survival of the fittest (Austin, 1961:130). Third, analysis of ordinary language allows us to see a concept in relation to other concepts: What gives significance to an3^  concept, what allows us to use that concept to say something important is its web of interrelationships with other concepts [symbolized by and] embedded in our language. If we stipulate a meaning for a term, we run the risk of distorting our THE CONCEPT OF LEADERSHIP / 18 understanding (Coombs, 1978). Wittgenstein argued that it is dangerous to remove a word from its original context because then it becomes "like an engine idling" (1963:51) rather than doing work. Concepts must be examined not only in the context of language, but also in the contexts and environments in which language is used, and the larger constellations of activity, or forms of life, from which the smaller contexts derive. The concepts symbolized by terms such as "leadership" may have no exact counterpart in a foreign culture and language, in part because the supporting web of concepts and contexts are missing. For example, the French translation of leadership is "le leadership," possibly indicating that leadership has no precise equivalent in the French culture. In the case of leadership, to begin our analysis with the scholarly literature would be particularly ill advised: failure to make significant progress in research often bespeaks conceptual confusion. The leadership literature is replete with different stipulated definitions of leadership, but the scholars themselves admit that the field is floundering, as we saw in the last chapter. The plethora of specialized definitions in the literature is viewed as failing to capture adequately people's intuitions concerning what really matters about leadership. Mintzberg, for example, confesses that every leadership theory he encounters falls "with a dull thud": None that I can think of has ever touched a central nerve of leadership — approached its essence (1982:250). Unfortunate!}', articulating just what does approach the essence of leadership has THE CONCEPT OF LEADERSHIP / 19 proved to be much more difficult than specifying what does not. This is not to say that the definitions in the literature are all seriously misguided. Some may have political purposes and may not be meant to represent ordinary language meaning. Others may capture some aspects of our ordinary understanding of "leadership," but their propounders are like the three blind men who each feel one part of the elephant, and make the assumption that the trunk or the leg or the tail of the elephant is what characterizes it. These points should not be taken to suggest that there is no more to be said after analysis of the ordinary language concept has been completed: Certainly ordinary language has no claim to be the last word . . . If a distinction works well for practical purposes in ordinary life (no mean feat, for even ordinary life is full of hard cases), then there is sure to be something in it, it will not mark nothing: yet this is likely enough to be not the best way of arranging things if our interests are more extensive or more intellectual than the ordinary . . . Certainly, then, ordinary language is not the last word: in principle it can everywhere be supplemented and improved upon and superseded. Only remember, it is the first word (Austin, 1961:133). Analysis of the concept of leadership through examination of its use in ordinary language may prove to be merely a starting point in determining how we might fruitfully conceive of leadership in the context of educational administration. However, it may be a useful first step. Before one can argue defensibly for departing from ordinary language meaning, one must be aware of what change is being introduced into one's conceptual ecology. Specialized meanings of "leadership" in the research literature may prove to be more fruitful than the ordinary language concept, but we should accept such meanings only if THE CONCEPT OF LEADERSHIP / 20 we are convinced that they capture what is important in the ordinary concept and have advantages of precision or usefulness not possessed by the ordinary concept, or are free of undesirable assumptions implicit in the use of the ordinary concept. Should it be the case that the specialized conceptions prove to be unfruitful, or deficient in important ways, it will be difficult to diagnose the nature of the deficiency or to build a more satisfactory conception unless we begin with a clear understanding of our ordinary concept of leadership. The method by which such an understanding is achieved is called conceptual analysis. A brief description of conceptual analysis follows. 3. The Nature of Conceptual Analysis Conceptual analysis is a set of techniques commonly adopted by philosophers to map out the ways a word is used by competent language users. It consists of articulating whatever conventional regularities are to be found in the application of a term in some range of similar contexts (McClellan and Komisar, 1962). To speak of conceptual analysis as a method is somewhat misleading, for it involves no step-by-step procedure or recipe, although it does involve standard techniques. In essence, conceptual analysis attempts to determine the conditions under which a term is correctly used by careful examination of cases to which that term clearly applies, (i.e., cases in which competent language users would generally agree with the application in question), cases to which the term does not apply, and cases in which it is problematical whether or not the term THE CONCEPT OF LEADERSHIP / 21 applies. What is typically aimed at is something like a full account of the conditions which are necessarily associated with the use of the term by competent language users in the contexts relevant to our concern, and the conditions which are sufficient for such users to warrant the use of the term in these contexts. A necessary condition is one such that all cases of the concept or class X have it; nothing can be a case of the concept or member of the class X without that feature. A concept may have any number of necessary conditions. One or more necessary conditions are jointly sufficient if there is nothing which has all the necessary conditions which is not an instance of concept X.t In sum, one aim of conceptual analysis is to set out the necessary and sufficient conditions for the application of a word (symbolizing the concept in question) such that competent language users will not be able to think of any counter-example, i.e., a case to which the term applies, even though it lacks one of the conditions identified as necessary, or a case to which the term does not apply despite its having the set of conditions identified as sufficient. It should be pointed out that this method of analysis may not be useful for all concepts, such as "cluster" concepts and "defeasible" concepts. Cluster concepts are those which involve some cluster of features, the exact composition of which may vary from case to case. Wittgenstein (1963) decribes this phenomenon in terms of "family likenesses," that is, some uses of a term share tAppendix A is a demonstration of how one might proceed through an analysis by making hypotheses and searching for counter-examples. I use the example of a square for illustrative purposes (though "square" is a very rigidly defined concept and thus would not be a useful candidate for analysis). THE CONCEPT OF LEADERSHIP / 22 one important feature, and others another, just as members of a family may be readily identifiable by one or more features. Some of them may share the same build, or facial features, or gait, or temperament. Since certain features may not be necessary to the use of a concept in a particular context but majr be important in another, cluster concepts cannot be characterized by necessary and jointly sufficient conditions. Instead, each context must be analyzed separately. It may be the case that "leadership" can be usefully characterized as a cluster concept. That is, different conditions may characterize, say, religious leadership from those which would characterize educational leadership. Since this possibility should not be ruled out in advance of analysis, we must carefully consider features which seem to be common to many, though not all, cases we would describe in terms of "leadership." Nor, however, should we rule out in advance the possibility that necessary conditions of leadership can be fruitfully characterized in fairly abstract terms even though the conditions may have very different concrete embodiment in different contexts. There is also the possibility that leadership is, to some extent, what Hart (1955) refers to as a "defeasible" concept. According to Hart, defeasible concepts, such as "voluntary," are best defined in terms of what they rule out, i.e., what would defeat the claim that X is an instance of the concept, because they can never be specified in terms of a set of positive, jointly sufficient conditions. Hart claims that "voluntary" does not describe a positive state of affairs, but rather, the absence of such things as physical or mental coercion. If the concept of leadership has an element of defeasibility to it, we might find it useful to frame our discussion occasionally in terms of what would defeat the claim that some THE CONCEPT OF LEADERSHIP / 23 person was exercising leadership. Conceptual analysis is reflection about how competent language users use language. Operationally, the analyst takes himself or herself to be a competent language user. The main skills required are sensitivity to language, being able to tease out central features of a concept, and being able to submit One's conclusions to rigorous tests by searching systematically for counter-examples. An adequate conceptual analysis must withstand a critical search for counter-examples by members of the academic community, just as an adequate scientific conclusion must withstand a search for counter-evidence. A good analysis of a contentious concept will not necessarily be immune to all counter-examples, but will advance debate about the way a word or expression is used. 4. Conceptual Analysis Versus Other Means of Determining Ordinary Language Meaning One might agree that it is important to determine the ordinary language meaning of a word without thereby conceding conceptual analysis to be the best method of accomplishing this. One might argue, for example, that a dictionary definition would save a lot of time and effort, or that a survey asking people what they mean when they use a particular word would render a more accurate picture than one person's reflections. Dictionary definitions are often a good starting point, but they seldom attempt to locate concepts in a nest of related concepts or to examine the fuzzy THE CONCEPT OF LEADERSHIP / 24 "borderline" uses of a word. Wittgenstein (1963) argues that definitions of certain words will never coincide entirely with actual usage, because the "boundaries" of the concepts have never been mapped. Although conceptual analysis is meant to map out actual usage, the analyst may have good reason to regard some uses as anomalous, metaphoric, or parasitic upon central usages, and thus not a crucial part of the data base of the analysis. For example, if one were analyzing the meaning of the term "respect" in order to understand why it is thought that teachers should respect their students, one might be justified in regarding its use in the expression "I respect his temper" as parasitic upon more central meanings, such as "I respect my parents." This is because the word "respect," when used to describe an attitude toward someone's temper, lacks features essential to most other uses of "respect" (such as the implication that the object of respect has value). Whereas definitions attempt to delimit or sharpen the meaning of a word so as to reduce its vagueness, conceptual analysis begins with a careful account of how a word is used, including the kind of vagueness that attaches to its use: There is a general belief that analysis functions to make vague terms uniformly precise, ambiguous terms univocal, and generally to render all language concrete and specific. "To clarify" has come to be synonymous with "to make definite." These activities misconstrue the expectations of elucidation (McClellan and Komisar, 1962:viii). With regard to surveys, we cannot have confidence in the results of a questionnaire which asks people how they use a word because we cannot be certain that those surveyed, even if they use a word correctly, will be able to give an accurate account of its use. As Wittgenstein explains: THE CONCEPT OF LEADERSHIP / 25 Suppose it's the 'use of a word' we want to enquire about. You may say, how could I not know this? There is a point to this question — if you mean (1) Surety I've got mastery of the techniques using ['leadership']. You have. There is though (2) another task — to describe the technique of the use, and this is hard (n.d.:16). Hare, too, discerns a discrepancy between using a word in a certain way, and being able to articulate how that word is used: If we want to find out what ordinary people mean, it is seldom safe just to ask them. They will come out with a variety of answers, few of which, perhaps, will withstand a philosophical scrutiny or elenchus, conducted in the light of the ordinary people's own linguistic behaviour . . . (1981:80). Of course, where one's own intuitions and perceptions of what competent language users would or would not say in any given case are unclear, it is useful to ask others what they would say about the case. Thus, the linguistic intuitions of others can serve as useful data in the analysis of a concept. It could be argued that conceptual analysis is merely a philosophical term for what we all do quite naturally in trying to determine what a word means. Although there is a sense in which this is true, it should not lead one to conclude that systematic conceptual analysis is either easy or unimportant. The analogy of the "scientific method" is instructive — ordinary persons apply it intuitively. What ordinary persons do in solving problems and analyzing terms is a rough approximation of what scientists or philosophers do, but it is not as methodologically or conceptually sophisticated. Wilson argues that conceptual analysis is a specialized subject in its own right, having its own techniques, and that questions involving abstract concepts cannot be tackled without these THE CONCEPT OF LEADERSHIP / 26 techniques in any but the "most feeble and confused manner" (1966:viii). Let us now proceed with the analysis of "leadership." B. CONCEPTUAL ANALYSIS OF "LEADERSHIP" In this section, I shall compare briefly the terms "lead," "leader," and "leadership," and then attempt to construct a set of necessary and jointly sufficient conditions for the usage of "leadership" which seems to be central to the context of educational administration, i.e., "L is exercising leadership." 1. "Lead," "Leader," and "Leadership" In comparison with "lead" and "leader," "leadership" is a comparatively recent addition to the English language, coming into use sometime in the nineteenth century. Forms of the verb "lead" were evident as early as the ninth century, and the noun "leader," appeared in the sixteenth century. I have said that dictionary definitions are often a good point from which to launch analyses. In the Oxford English Dictionary, "lead" has over thirty dictionary definitions, "leader" has a somewhat smaller number, and "leadership," the fewest of all. Some of the definitions of "lead" may give us clues about what it means to exercise leadership: • direction given by going in front, example • to cause to go along with oneself, to bring or take to a place • to conduct, guide, especially by going on in advance THE CONCEPT OF LEADERSHIP / 27 • to conduct to a conclusion, to induce to do something • to precede, have the first place in "Leadership" is dealt with in a matter of lines: • the dignity, office, or position of a leader • ability to lead There would seem to be a wider range of cases to which we could apply "lead" and "leader" than "leadership." For example, if L and FFt were playing follow-the-leader, we would say that L leads FF but we would not (normally) say that L is exercising leadership; we would not normally attribute leadership to a tourist guide who is serving as leader of an expedition through a castle; and we would not generally use the word "leadership" in describing the leader in a marathon. In some contexts, to say that someone is leading or is a leader is to imply that he or she is exercising leadership. For example, persons who argue over whether or not someone is leading a school, or is an educational leader, are arguing neither about whether the person is physically, or in some other sense, ahead of others, nor about whether the person holds a particular position — it is the exercise of leadership which is at issue. Though not all leaders exercise leadership, all leaders lead; "lead" has a wider range of applicability than "leadership." For L to lead FF, L merely need be ahead of them. If L exercises leadership with regard to some FF, L is, for tl shall use the letter "L" to denote a person whom we are considering as a candidate for exercising leadership. "L" is not necessarily a leader in the position sense. I shall use "FF" to denote "followers" in the loose sense of "those who may be thought to be led by L." THE CONCEPT OF LEADERSHIP / 28 a time anyway, FF's leader and FF are in some sense followers. L can lead FF without being FF's leader, and FF can follow L without being followers. For example, someone might lead me to the nearest gas station but it would be odd to call that person my leader. This example raises another difference between "lead" and "leadership." It makes sense to speak of leading just one person, whereas it would be odd to speak of exercising leadership with regard to just one person (though this point does not matter for my purposes). L can lead FF for just a short time or on just one occasion, whereas "leader" tends to, but does not always, indicate a more stable, ongoing relationship. In this characteristic, "leadership" is more like "lead" than "leader"; one can exercise leadership for one brief episode. Conversely-, "leadership" does not always have the sense it has in the phrase "the exercise of leadership." "Leadership" serves as an umbrella term under which all research and theory having to do with leaders or leadership functions find shelter. We speak of leadership conventions or the leadership of a country in a value neutral sense, meaning merely "the incumbents of particular offices or positions." I shall call this sense of leadership position leadership. It is known in the leadership literature as status leadership. According to Newell, to say that Mr. So-and-so has been placed in a position of leadership is "simple recognition of the fact that leadership is associated with and expected from persons who occupy certain positions" (1979:223). This suggests that certain people, by virtue of the positions they hold, are regarded as having both a right and an obligation to exercise leadership. Persons in leadership positions do not necessarily exhibit or exercise leadership (though their tenure may be limited by THE CONCEPT OF LEADERSHIP / 29 their failure to do so). However, it is the judgment that "L is exercising leadership" that seems to get at the heart of what is held to be important about leadership. And, in the study of leadership in educational administration, the primary objective seems to be not to teach persons about every aspect of administration, nor to teach persons how to obtain leadership positions, but rather to prepare them to exercise leadership. Thus, let us consider the rules governing the use of the expression "L is exercising leadership." 2. "L is Exercising Leadership" In analyzing the expression "L is exercising leadership," I am attempting to understand what a person means when making the claim that L is exercising leadership, i.e., to extract from common usage the necessary and jointly sufficient conditions for using the expression "L is exercising leadership." Such an analysis is not meant to supply a checklist of observable phenomena by which we can determine whether some L is exercising leadership. In fact, I shall later suggest that such lists, constructed in the absence of adequate understanding of the concept of leadership, may have contributed to the current malaise in the field of leadership research. I am analyzing what competent language users mean when they ascribe the exercise of leadership to some L, not how they determine empirically that some particular L is exercising leadership. Spelling out how we can accurately determine whether or not the conditions are being met in any particular case is a further task. I shall begin by considering a number of model cases, i.e., cases in relation THE CONCEPT OF LEADERSHIP / 30 to which we would expect competent language users to agree that the expression "L is exercising leadership" applies. These will serve as the basis for formulating certain hypotheses concerning the necessary conditions for using the term "leadership" in accordance with the rules observed by competent language users. Having formulated all the hypotheses that seem plausible, I shall test each by systematically searching for counter-examples. Once I have identified the conditions necessarily (or, at least typically) associated with the use of the expression "L is exercising leadership," I shall consider what set of conditions is sufficient for applying the term. a. Model cases For the model cases, let us begin with some well known examples of persons who are widely regarded as having exercised leadership and then make up some clear cases of the exercise of leadership in the educational context. Few (other than racists) would dispute that Martin Luther King Jr. exercised leadership in his struggle for equal rights for blacks in the U.S. or that Bishop Desmond Tutu is exercising leadership in the fight against apartheid in South Africa. Most would say that Pierre Elliot Trudeau exercised leadership in repatriating the Constitution and in his successful fight to keep Quebec from separating from Canada; that Gloria Steinem exercised leadership in the feminist movement; that Mikhail Gorbachev is exercising leadership with regard to restructuring the Soviet Union; that Rene Levesque exercised leadership in the struggle for Quebec's independence; that Lee Iacocca exercised leadership in THE CONCEPT OF LEADERSHIP / 31 bringing the Chrysler corporation back on its feet in the U.S.; that Jesus exercised leadership with respect to his disciples and other adherents to Christianity; and that Jimmie Pattison exercised leadership in making Expo '86 a successful enterprise. Those who have exercised leadership in the educational sphere come less readily to mind. (In fact, Kaplan (1985) suggests that the identity of leaders in education has remained among the "better kept secrets" in the U.S. Those identified by Kaplan as education's top four leaders — George Anrig, Ernest Boyer, Bill Honig, and Albert Shanker — are hardly household names.) Perhaps among the more famous North Americans are Horace Mann, who initiated the public school movement; Mann's Canadian counterpart, Egerton Ryerson, who introduced to Upper Canada many of Mann's policies, such as universal free elementary education, property taxes for school funding, and secular schooling; and John Dewey, whose ideas on education continue to influence practice today. Let us flesh out this set of model cases with two cases in the context of education: • A superintendent of a large school district carefully examines a number of alternatives and becomes convinced that, as far as possible, all students should have a grounding in computer science. Following the superintendent's initiative and direction, educators discuss the pros and cons of providing students with computer science training and work out an appropriate program. The teachers, who were previously lethargic and unreflective about practice, become excited and implement a successful program of computer science in the district. THE CONCEPT OF LEADERSHIP / 32 • Mrs. X, mother of three school aged children, is concerned that the schools have no program of values education. She garners the support of a large group of parents who lobby the school board to develop a defensible program. Mrs. X also persuades a number of teachers of the merit of such a program and gets the parents and teachers to work together to develop a set of objectives for the program. Eventually, the schools adopt a program of values education. b. Hypotheses What features seem to be shared by most of these cases? (If a feature is shared by all but one case we may still hypothesize it to be necessary to the use of "leadership" in most contexts. This should keep our analysis from going astray should leadership turn out to be a "cluster concept.") In all the cases above, the leader provides some direction for the followers in terms of pursuing a goal, and persuades the followers to pursue the goal. In every case, too, the followers pursue the goal not because they are coerced or threatened, but because they believe it to be worthwhile. In only some of the model cases cited is the goal attained, but in every case, some progress is made. Let us separate these observations into discrete hypotheses. When leadership is being exercised: 1. L provides direction for FF in pursuit of a goal. 2. FF pursue the goal. 3. L persuades FF to pursue the goal. 4. L's goal is regarded by FF as desirable. 5. FF's compliance with L is voluntary. THE CONCEPT OF LEADERSHIP / 33 6. Progress toward the goal is made. We might also explore hypotheses not suggested by our model cases, but alluded to in definitions in the leadership literature, such as the following: 7. L has power or authority over FF. 8. L is charismatic. 9. L changes FF. Let us consider each of these hypotheses in turn. On the basis of our discussion, we may accept, modify, or reject each hypothesis. Hypothesis #1: To be Exercising Leadership, L Must Provide Direction for FF in Pursuit of a Goal Let us imagine a case in which L, who is wandering aimlessly through the desert, is followed by FF who mistakenly believes that L knows the way out. Would we say that L is exercising leadership? Clearly not. One thing that disqualifies this as a case of leadership is that L does not have a direction. If L is leading FF, then L must be going somewhere and FF must be going there too. Leadership involves some ideal or goal, some desired change. Suppose now that L and FF together determine a strategy for getting out of the desert. Since L in this case was never a formal leader, we have no basis now for calling L the leader, as L's contribution to the plan of action is no different in kind or degree from that of FF. If L had devised the plan, however, and led FF to safety, we might well say that L exercised leadership. THE CONCEPT OF LEADERSHIP / 34 Whatever else "leadership" implies, it seems to be concerned with giving FF direction. Usually, this is in the form of proposing goals for FF, but this need not be the case. Let us imagine that in a particular school district, policy requires that principals report to parents on certain aspects of the school's operation. In such a case, principals could take the conventional approach of sending out a monthly newsletter, or they could initiate a more creative program, including parent/student nights, telephone calls to parents, and even slide shows. Even though they did not initiate the goal of keeping parents informed, they could still exercise leadership by initiating creative means to accomplish the goal. The distinction between ends and means blurs as one considers levels of objectives — the means to a high level goal may itself be a lower level goal. It should be clear, however, that the plan of action must not be viewed by the competent language user as something routine or inconsequential, but rather as having significance in the same way that a goal is viewed as having significance. What is significant is, of course, a relative matter. A new office procedure initiated by a secretary for the secretarial staff may be regarded as significant in context. Let us now imagine a case in which L does not propose an idea or even a plan of action, but sets an example which inspires FF. For instance, suppose that a quarterback implements plays called by the coach, but does so with such skill, intensity, and determination that the rest of the team emulate him. I think that we might describe such a case in terms of the exercise of leadership. Whether L has a vision, exemplifies some ideal, proposes a goal, or devises a plan of action, L is setting or giving priority to certain values. In fact, Selznick THE CONCEPT OF LEADERSHIP / 35 (1957) and others speak of the choosing of key values as being an important element in leadership. I have said that the usual case of giving direction involves L's initiating a goal. Some, particularly proponents of "democratic leadership," have argued that leadership does not involve L's initiation of goals. Indeed, they maintain that it requires the active participation of FF in determining goals and priorities. Others, such as James (1951), claim that leadership involves making and proclaiming values for the masses who have neither the ability nor the disposition to make fundamental moral and aesthetic judgments. The most accurate view of how competent language users use the expression "exercising leadership" probably lies between these two extremes. There is nothing to prevent a leader from giving FF input into decision-making or allowing FF to criticize L's goals or programs of action, but if L does not assume special responsibility for goal identification, then L is not usually regarded as exercising leadership. In 1859 Mill argued: The initiation of all wise or noble things must come from individuals, generally at first from one individual. The honour and glory of the average man is that he is capable of knowing that initiative; that he can respond internally to wise and noble things, and be led to them with his eyes open (1956:81). If FF were capable of coordinated, goal-directed, and responsible action without direction from L, then the call for leadership would likely not be heard. But FF do not always know the best course of action for themselves, in the sense of "having a vision," or FF may have far less knowledge or expertise than L, but appreciate the appropriateness of the direction set by L. THE CONCEPT OF LEADERSHIP / 36 In the usual case which we would describe in terms of the exercise of leadership, L would not provide direction for FF unintentionally. However, let us consider what I take to be a "borderline" case of leadership. In this case, L is a morally good person who sets an example for others. FF are inspired by L, and become morally good persons, but L is not attempting to set an example or to provide direction for FF. The way the first hypothesis is worded, this case is not clearly ruled out, but neither is it clearly included. The matter hinges on the extent to which L can be said to be providing direction unintentionally to FF in pursuit of a goal. My intuition is that we would not describe this case in terms of the exercise of leadership. Although we would not ascribe the exercise of leadership to L in cases in which L just happens to be heading somewhere, this does not mean that leadership need be proactive and deliberative. As Hare argues, "leadership can and does enter into situations in which one is forced into responding in an ad hoc spontaneous fashion" (1986:54). Many would argue that J.F. Kennedy exercised leadership in his response to the Cuban Missile Crisis. Would we describe a case in which L does not believe in a goal but is able to inspire others to act, in terms of the exercise of leadership? Suppose, for instance, that L is paid to get a group of people enthusiastic about something he or she regards as immoral; or just for the fun of it, persuades union members to strike. I do not think that we would be inclined in such cases to say that L is exercising leadership, though we might describe the cases in terms of L's having done a good job of instigating group action. Or suppose that L THE CONCEPT OF LEADERSHIP / 37 agrees to help a person learn to read, or to help a team develop their skating skills. In these instances, we might say that L was "tutoring" or "coaching," but not "exercising leadership" (except, perhaps, in extreme instances where L becomes devoted to the task, and initiates creative plans of action for FF). Thus, we need to amend our hypothesis to reflect the idea that L must be committed to the goal for himself or herself and not just for FF, by adding the word "shared": L must provide direction for FF in pursuit of a non-trivial, shared goal.t There appear to be important counter-examples to the thesis that the exercise of leadership necessarily involves L's personal commitment to the goal. For example, Robert E. Lee is widely regarded as having exercised leadership in defence of the Confederacy in the American Civil War, yet he did not believe in the cause of the Confederacy. But whether this is taken to be a valid counter-example depends upon what we take Lee's goal to be. Lee did not preach to his followers the cause of the Confederacy, but he did believe in defending the honour and integrity of his state — his goal was to serve the political leadership of his home territories. We might argue that, although Lee's leadership served a broad political purpose, he exercised military leadership with regard to the army, not political leadership. That the cause of the Confederacy was served under his leadership was incidental to his exercise of leadership. To the extent that his followers shared his goal, Lee exercised leadership. t"Goal" should hereafter be interpreted loosely to include "goal" and "goal-oriented activities." THE CONCEPT OF LEADERSHIP / 38 Our first hypothesis, with certain amendments, seems to be borne out. I shall call it the Direction Condition. Hypothesis #2: For L to be Exercising Leadership, FF Must Pursue the Goal Clearly, FF do not need to attain a goal for L to be said to be exercising leadership. We can and do ascribe the exercise of leadership to persons even when their goals have not yet been realized. However, if FF did not even attempt to pursue the direction provided by L, the ascription of leadership would not occur. For L to be said to be exercising leadership, L must ensure that FF attempt to attain the shared goal. Even if L initiated laudable goals of which FF approved, we would not say that L was exercising leadership unless FF made some move to pursue them. For example, in a newspaper editorial, Jack Munro is praised for his unsuccessful efforts in trying to get the forestry industry and senior governments to participate in a program of silviculture and reforestation. The writer labels Munro's actions not "leadership" but "an imaginative attempt at innovative leadership." In order for leadership to be said to be exercised, FF must pursue the goal in question. I shall call this the Adherence Condition. Hypothesis #3: For L to be Exercising Leadership, L Must Persuade FF to Pursue the Goal It is clear that in many cases of L's exercising leadership, L persuades FF to pursue a goal. Can we imagine a case of the exercise of leadership in which L does not persuade FF to follow? Suppose that L does not need to persuade THE CONCEPT OF LEADERSHIP / 39 FF to follow him or her because FF are so taken by L's charisma (as opposed to being persuaded on the basis of reasons) that they follow him or her. Or, take the case of the quarterback mentioned earlier. In each case, we might attribute the exercise of leadership to L. However, suppose that FF head for L's goal because they just happen to be heading in the same direction as L. To modify our earlier example, FF could just happen to follow the North Star out of the desert. There is no exercise of leadership in this case. Or, to take another example, perhaps F l wants to be near someone else in L's group, say F2, because F l likes F2, and will do whatever is L's bidding for the privilege of working with F2. If we knew Fl's motive for following L, we would not say that L was exercising leadership with regard to F l . Clearly, L's persuading FF is not a necessary condition of exercising leadership, but FF's pursuit of a goal cannot be totally independent of L either. We might employ a term which encompasses persuasion but has broader application by hypothesizing that L influences FF to pursue the goal, rather than that L persuades FF. In order for L to be said to be exercising leadership, FF must be following L, that is, striving for a goal because of L's influence, t Balderson's (1980) claim that leadership involves getting people to do what they would not do without the leader's influence, exemplifies this idea that L plays a part in getting FF to pursue what comes to be a shared goal. It is only when FF pursue the goal or plan of action in collaboration with L because of L's fit seems preferable to speak in terms of "influence" rather than "cause," as the latter term tends to suggest involuntariness on the part of FF whereas "influence" suggests that FF are rational agents of sorts. I shall later argue that the idea of FF's rational autonomy is embedded in the concept of leadership. THE CONCEPT OF LEADERSHIP / 40 influence that we speak of "leadership." Kerr's and Jermier's characterization of effective leadership as "the ability to supply subordinates with needed guidance and good feelings which are not being supplied by other sources" (1978:400), captures the idea of L's making a difference. I cannot imagine a case of exercising leadership in which L has no influence over FF's pursuit of the goal. Let us thus tentatively set out the Influence Condition: L influences FF to pursue the goal. Hypothesis #4: For L to be Exercising Leadership, L's Goal Must be Regarded by FF as Desirable Suppose that FF join Greenpeace solely because they are attracted by its leader's personal magnetism, air of authority, or even ability to advance their careers, and are influenced by the leader to pursue a goal even though they do not regard it to be worthwhile. We would be unlikely to describe such cases in terms of the exercise of leadership. If FF's sole reason for pursuing a goal is L's charisma (in the sense of personal magnetism), authority, or promise of extrinsic reward, then we do not generally regard it as a case of leadership. What seems to be missing from such cases is that FF are not pursuing the goal, even in part, because they regard the goal as something desirable or worthwhile for them to pursue. As Tead expresses it, "The leader points the way but equally the followers decide that the way is good" (1935:209). That is, FF must believe that the goal or plan of action is a good thing not just for L, but for themselves as well. Follett comes close to capturing this idea: The leader must make his co-workers see that it is not his purpose which is to be achieved, but a common purpose, born of the desires THE CONCEPT OF LEADERSHIP / 41 and activities of the group (1940:261-262). However, I would disagree that the purpose need necessarily be born of the desires and activities of the group. Leaders can sometimes get FF to pursue a goal which was previously foreign to them but which they come to view as appropriate. Truman's well-known definition of leadership as "the ability to get men to do what the3r don't want to do, and like it" perhaps captures this aspect better. However, in leadership situations, FF do not do what they don't want to do; rather they must in some sense want to or think that they ought to pursue the goal. This is not to say that FF need feel enthusiastic about pursuing it. Suppose that L persuades reluctant FF to pursue a particular goal, the pursuit of which will be very difficult and will cause FF a certain amount of hardship (e.g., staging a political protest that may put FF in jail temporarily but may also result in policy changes that will benefit some minority group). Even though FF may have certain misgivings, we might call this an example of leadership. FF need not pursue a goal or plan of action with great enthusiasm, but FF must perceive the achievement of the goal as a change for the better. Nor need FF pursue a goal solely because they believe it to be worthwhile. They may have additional motives for following L. It could be argued that although the leader who is able to "take a tough decision," such as firing two or three staff members, may be hated and feared by the staff members, he may be credited with strong leadership by the Board. In this case, it seems that the Board members are crediting L with the exercise of leadership even though FF are not committed to L's goals. It seems very unlikely, however, that were staff members in general still uncommitted to L's THE CONCEPT OF LEADERSHIP / 42 goal after the firings, the Board would regard L as having exercised leadership. Of course, a Board may, in order to disguise the use of naked coercion, engage in a bit of newspeak, but such uses of "leadership" need not concern us here. Since I have been unable to discover more creditable counter-examples to this hypothesis, I conclude that it is a necessary feature of the exercise of leadership that FF view the goal to be worthwhile. It should be pointed out that not all FF who follow L need feel commitment before L can be said to be exercising leadership with regard to the group; otherwise very little leadership would be exercised. It is not clear just what percentage of committed FF would be necessary, but when the ascription of leadership is made, the competent language user probably does not assume that everyone who follows L is being led in the requisite sense. It should also be noted that those FF who follow L purely for motives other than the worthwhileness of the goal (e.g., peer pressure, rewards), might reasonably deny that L had exercised leadership with regard to them. One further case should be discussed. Suppose that, even though FF disapprove of a goal, they follow L because of coercion. Later, they come to appreciate and approve of the goal. Could we say that L exercised leadership? I think that it very much depends upon what options L had open. If, for reasons of secrecy (say, in a military manoeuvre) or time constraints, L was unable to spend time convincing FF of the merits of a goal, it is possible that we would ascribe the exercise of leadership to L, even though FF did not follow L willingly. However, just as FF would come to appreciate the worth of the goal THE CONCEPT OF LEADERSHIP / .43 only after the fact, the ascription of the exercise of leadership would be ex post facto. A case somewhat parallel to this situation arose recently over Canada's free trade deal with the U.S. Although coercion was not involved, Brian Mulroney implied that leadership could be exercised even when FF do not accept the worth of the goal at the time. In vowing to pursue the deal, Mulroney declared, "As prime minister, my obligation is to provide national leadership whether it is popular or not," adding that he would accept the verdict of the people at the polls (as reported in The Vancouver Sun, November 27, 1987:A7). We need a condition which states that FF pursue the goal or plan of action, at least in part, because they regard the goal as desirable or worthwhile and the plan of action as legitimate. I shall call this the Commitment Condition. Hypothesis #5: For L to be Exercising Leadership, FF's Compliance with L Must be Voluntary If FF were forced or coerced into doing what L wanted, would talk of exercising leadership arise? Jacobs argues: A key requirement for leadership is that the influence target probably must always have the option of deciding for or against compliance with the leader's wishes, without incurring coercive penalties (1974:203). Schenk carries this idea to the extreme with his definition of leadership as "the management of men by persuasion and inspiration rather than by the direct or THE CONCEPT OF LEADERSHIP / 44 implied threat of coercion" (1928). Burns (1978) contends that in the context of leadership, not only must FF have the choice not to follow L, but also a choice of L to follow. But how far is this idea of voluntariness borne out in the language? Although competition among "leaders" may be desirable, it does not appear to be necessary to the exercise of leadership. Persons can and do exercise leadership in cases in which there are no rival leaders. Sometimes there is a leadership "void" such that virtually anyone who volunteers to be leader is willingly followed. "Persuasion" and "inspiration" may figure importantly in certain kinds of leadership, such as educational leadership, but may have less significance in other kinds of leadership. If a principal coerced teachers into implementing an educational program, we might hesitate to ascribe the exercise of leadership to him or her whereas it is possible that some might attribute the exercise of leadership to an army general who makes occasional use of coercion and coercive sanctions for non-compliance. Jacobs' (1974) condition of leadership, that FF must have the option of not following L without incurring coercive penalties, may be somewhat too strong to reflect ordinary language usage. Ordinarily, we would allow that a certain element of coercion can be part of the exercise of leadership; however, the greater the element of coercion, the less likely we are to regard the person as exercising leadership. It might be argued that Hypothesis #5 is otiose in that it is subsumed by the Commitment Condition. If FF pursue a goal in part because the}' are convinced that the goal is worthwhile, then is not their pursuit ipso facto THE CONCEPT OF LEADERSHIP / 45 voluntary? This issue is more complex than it might seem. Suppose that FF are committed to a goal, but the autonomy of their commitment has been compromised in that they have been brainwashed, indoctrinated, or conditioned into certain beliefs or actions by L, or because L has deceived them into believing a goal desirable. In such cases, L would not be said to be exercising leadership. Brainwashing, indoctrination, and, to some extent, conditioning all restrict FF's freedom because they impair FF's capacity to decide for themselves whether a goal is worthwhile. Deceit similarly obstructs FF's free, rational choice. FF's decision to follow L need not be rational in the sense of being well thought out, but it must be rational in the sense of being FF's own choice based on reasons that are genuine reasons for FF. What is required is a rephrasing of Hypothesis #5 which will rule out significant reliance on either physical or mental coercion on the part of L. The idea of autonomy rather than voluntariness might better serve this purpose, because it allows for a certain degree of coercion while ruling out means which impair FF's capacity to decide rationally. It seems to be a necessary condition of the exercise of leadership that FF have autonomy in determining whether or not to pursue the direction provided by L. I shall call this the Autonomy Condition. Notice that the Autonomy Condition does not rule out the possibility of position leaders exercising leadership within an organization even though FF are subject to coercive penalties for non-compliance, t However, if L were to gain tin fact, having position authority may actually facilitate the exercise of leadership. Having authority empowers the leader to coordinate tasks in the accomplishment of some end, and to give assurances to some followers that if THE CONCEPT OF LEADERSHIP / 46 compliance solely through the use of coercive penalties, we would be unlikely to regard L as exercising leadership. Hypothesis #6: For L to be Exercising Leadership, Progress Must be Made Toward Attainment of the Goal We noted earlier that L may be said to have exercised leadership even when the goal is not attained. For example, a political leader can exercise leadership during an election campaign even if his or her party does not win. When the goal is not achieved, however, we will generally ascribe to L the - exercise of leadership only when the plan of action he or she initiated was at least a sensible way of striving to achieve the goal. Someone who succeeds in galvanizing heroic effort in the service of a stupid or nonsensical plan of action toward a worthwhile goal would not be thought to be exercising leadership. Even if the goal were attained, we might deny that leadership had been exercised if it were perceived that the goal was achieved, for example, by grossly immoral means or by mere chance. Although it might be reasonable to ascribe the exercise of leadership to someone in a particular instance even if no progress was made toward a goal, we are unlikely to. say that someone was providing leadership if, over time, no progress was made toward the attainment of goals. The ascription of the exercise of leadership is not merely a description of behaviour, but also a judgment of achievement, and one aspect of achievement is goal attainment. If L mobilized a sizeable following to pursue his or her visions but over time no one made progress toward the goals, we might say L came t(cont'd) they do their part, others will do theirs, i.e., that the efforts of the followers will not be wasted due to the unreliability of others. THE CONCEPT OF LEADERSHIP / 47 up with ideas, but we would not likely describe L in terms of the exercise of leadership. In sum, although progress toward attainment of goals is important to the exercise of leadership, the relationship is not one of entailment. That is, it is not a necessary condition of the exercise of leadership. Hypothesis 47: To Exercise Leadership, L Must Have Power or Authority Over FF The concepts of power and authority and their interrelationship are complex, and beyond the purview of this thesis to explore in depth. However, they warrant some attention because they are commonly thought to figure importantly in the exercise of leadership. Janda, for example, characterizes leadership as: a particular t3>pe of power relationship characterized by a group member's perception that another group member has the right to prescribe behavior patterns for the former regarding his activity as a member of a particular group (1960:358). A number of writers (e.g., Newell, 1978; Kast and Rosenweig, 1974) define "power" as the ability to influence, and "authority" as the right to use power. If we were to adopt this definition of power, then L would have power over FF by virtue of the Influence Condition. However, "power" is often used to denote a relation that is clearly not necessarj' to the exercise of leadership, and may even be antithetical to it. First, "power" tends to suggest a unilateral relationship between L and FF whereas leadership may sometimes require a reciprocal relationship in which L is open to being influenced by FF (Jacobs, 1974). THE CONCEPT OF LEADERSHIP / 48 Second, "power" suggests a difference in standing between two parties — one decides and one complies — whereas "influence" allows that a dialogue between equals is possible. Finally, power is generally associated with having the means to coerce compliance (Jacobs, 1974; Kast and Rosenweig, 1974), whereas leadership is not. In short, where "power" is equated with "influence," L must have power over FF to exercise leadership, but this characteristic is already covered by the Influence Condition; where "power" is used in a more specific sense, it is not a necessary condition of the exercise of leadership that L have power over FF. Let us consider the necessity of L's authority to the exercise of leadership. In discussing Hypothesis #5, I argued that L's having authority was compatible with L's exercising leadership. Here we must determine whether it is a necessary condition. Weber (1964) defines authority as the willing compliance of people based on the belief that it is legitimate for the designated leader to impose his or her will on subordinates. Benn (1967) makes a distinction between de jure and de facto authority which, I think, will be useful for our purposes. De jure authority presumes a set of rules according to which certain persons are authorized to make decisions with which others must comply or suffer negative sanctions. (This may be more familiar to some as "position" authority.) De facto authority is when, in the absence of legitimizing rules, a person recognizes another as entitled to command him or her. It seems clear that L need not have authority over FF in the de jure sense. A student in a class could exercise leadership with regard to the Other students without having any formal authority. But must L have de facto authority in the sense that FF view L as having the THE CONCEPT OF LEADERSHIP / 49 right to command them? FF often follow L only because they find L's views persuasive, and not because they also regard L as having the right to command them. They may feel that L has the right to attempt to influence them, but not to dictate a course of action. As John Stuart Mill expressed it, "All [the leader] can claim is, freedom to point out the way" (1956:81). If this argument is correct, then the definition cited at the beginning of this discussion characterizes authority but not leadership. L's having power or authority over FF is not a necessary condition of the exercise of leadership. Hypothesis #8: To Exercise Leadership, L Must be Charismatic The word "charisma" is often linked with "leadership." Burns (1978) characterizes charisma as belief in leaders because of their personage alone, aside from their tested capacities, experience, or stand on issues; faith in the leader's capacity to overcome obstacles; and readiness to grant leaders power. Weber characterizes charisma as: a certain quality of an individual personality by virtue of which he is set apart from ordinary men and treated as endowed with supernatural, superhuman, or at least specifically exceptional powers or qualities (1964:358). Parsons (1951) defines charisma in terms of the ability to activate value commitments as opposed to the ability to persuade on the basis of common interest. All of these definitions share the idea of a person who is able to inspire dedication on the part of followers because of personal characteristics rather than on the basis of rational persuasion. THE CONCEPT OF LEADERSHIP / 50 Can we imagine a case of some non-charismatic L who exercised leadership? If so, then L's having charisma is not a necessary condition for the exercise of leadership. Although exemplars of the exercise of leadership tend to involve charismatic individuals, there have been those with little personal magnetism who have exercised leadership. For example, Harry Truman was able to command great support in the United States in his bid to halt the spread of Communism, and John Dewey profoundly influenced educators with his arguments about democracy and education. Neither is reputed to have had charisma. Charisma is not a necessary condition of the exercise of leadership, but it is certainly worthy of attention in the study of leadership. Hypothesis #9: For L to be Exercising Leadership, FF Must Change Some writers believe that L's exercising leadership brings about some change in FF's attitude or behaviour. If we take "change in attitude or behaviour" to mean merely that FF pursue a goal or activity, then this is covered by the Adherence Condition. But "change in attitude or behaviour" is suggestive of a significant, even permanent, change in the character of FF; some would even argue that in the sense of leadership which concerns us here, L and FF are mutually transformed through leadership. Burns, for example, says that (transforming) leadership ultimately becomes moral because it "raises the level of human conduct and ethical aspiration of both leader and led, and thus has a transforming effect on both" (1978:20). Empirically, it seems true that leaders are often influenced by their relationship with their followers, and that followers THE CONCEPT OF LEADERSHIP / 51 often undergo a transformation, sometimes for the better. I do not think that this is necessarily the case, however. It makes sense to say that L exercised leadership where L and FF mutually attain a goal, even if neither undergoes significant change. We often ascribe the exercise of leadership to L if something gets accomplished even if we have no knowledge of any resultant change in behaviour or attitude in FF or even if we believe that FF's behaviour and attitudes have not changed. Hypothesis #9 is thus disconfirmed. 3. Testing the Conditions for Sufficiency We have now considered all of our initial hypotheses. The list of necessary conditions for the exercise of leadership which has evolved from these hypotheses is as follows: 1. The Direction Condition: L provides direction for FF in pursuit of a non-trivial, shared goal. 2. The Adherence Condition: FF pursue the goal. 3. The Influence Condition: L influences FF to pursue the goal. 4. The Commitment Condition: FF pursue the goal, at least in part, because they regard it as desirable or worthwhile. 5. The Autonomy Condition: FF have autonomy in determining whether or not to follow L. It is probably useful to view leadership not as an all-or-nothing concept with regard to these conditions, but as being on a sliding scale. That is, the degree to which the conditions are met determine the degree to which leadership THE CONCEPT OF LEADERSHIP / 52 is exercised. Moreover, the more fully these conditions are met, the more likely we are to speak of the exercise of leadership. Given that we have identified a range of characteristics which seem to be present in all cases of the exercise of leadership, it would not seem to be fruitful to pursue the question of whether leadership is to some extent a cluster concept. For the purpose of testing the conditions, let us consider the conditions of leadership to be necessary as stated. Are these necessary conditions also jointly sufficient? That is, can we conceive of a case to which we would not ascribe the exercise of leadership in which L gives FF direction in terms of non-trivial shared goals, and influences FF without impinging on FF's rational capacity, to pursue the goals for the reason that FF view them to be worthwhile? In testing these conditions for sufficiency, one may discover a puzzling phenomenon: although all would agree that some persons, such as Martin Luther King Jr., exercised leadership in cases where the above conditions were fulfilled, there is disagreement as to whether other persons, such as Adolf Hitler, exercised leadership in cases where the above conditions were fulfilled. Burns, for example, argues that Hitler was not a leader, but its "polar opposite" — a tyrant (1978:2-3). Tucker (1981) claims to refute Burns on this point, with his argument that Hitler had many followers, the implication being that if a person has followers, he or she must thereby be exercising leadership. However, I suspect that Burns would agree that Hitler had many followers, but maintain that he did not exercise leadership. I do not think that resolution of this issue lies in saying that Hitler exercised leadership with regard to some and was a THE CONCEPT OF LEADERSHIP / 53 tyrant with regard to many. Those who deny that Hitler exercised leadership would not grant that he exercised leadership even with regard to those who prospered under his rule. Nor does resolution lie in saying that persons disagree over the facts. Clearly, Burns and Tucker are not quibbling over the facts. Even where persons agree that the five conditions above have been met, they may still disagree as to whether leadership has been exercised. A plausible resolution of this puzzle is that "L is exercising leadership" is used not merely to describe a state of affairs, but also to evaluate it. Thus, we could explain disagreements over the ascription of the exercise of leadership in terms of different standards being used to evaluate what a leader is doing in a situation. Many writers have pointed out the evaluative force of the term "leadership." Maxcy, for example, holds that the term leadership has a "halo effect, conferring worthwhileness and value to the designate and high regard to the program or plan" (1984:330). He designates "leadership" a "value-added" term. That is, to ascribe leadership to L in some situation is not merely to describe something but also to express approval, just as to call a killing "murder" is to evaluate it negative^. Hare notes that "not every change will be deemed progressive and valuable, and [in such cases] there may be no inclination to speak of leadership" (1986:52). The Hitler case can be resolved by saying that those who ascribe the exercise of leadership to Hitler view the ability to influence others to pursue shared goals as admirable in itself. Whatever else leadership implies, it seems to imply this. Those who deny that Hitler exercised leadership believe that such an THE CONCEPT OF LEADERSHIP / 54 ability is not enough; that ascribing the exercise of leadership to some L implies more general approval of L's behaviour, including the appropriateness of L's direction or goals, and L's means. Cronin's definition of leaders as "people who perceive what is needed and what is right and know how to mobilize people and resources to accomplish mutual goals" (1984:26) (emphasis added) embodies this view. Because any activity which fulfills the more restrictive standards will necessarily also fulfill the more inclusive standards, this hypothesis about the different standards also explains why (virtually) everyone agrees that Martin Luther King Jr. exercised leadership. It will be useful to call leadership ascribed on the basis of the more inclusive standards "weak-normative leadership," and leadership ascribed on the basis of the more restrictive standards "strong-normative leadership." These should not be viewed as two "kinds" of leadership but, rather, as two representative points near the poles of a continuum of standards for determining whether or not something counts as the exercise of leadership. Persons normally use only one set of standards in ascribing leadership in a given context. For example, few persons would say, "Hitler exercised leadership in one way, but not in another." Approval in the strong sense is based primarity upon the appropriateness of L's goal or direction, and the means by which L is effective in mobilizing FF and getting something accomplished. Let us consider these standards. The nature of the goal is generally considered very important in the ascription of leadership in the strong-normative sense. As Tead argues: THE CONCEPT OF LEADERSHIP / 55 No idea of leadership which slights the crucial importance of sound objectives gets beyond the primary stages. For in the last analysis the leader is only as strong as his objectives are sound. A leader is known by the objectives he espouses (1935:53). The speaker may assess the goal in terms of such standards as its moral worth, its ambitiousness, its novelty, or its appropriateness in relation to the values of the collectivity within which leadership is exercised. For example, a principal who mobilized teachers to keep a tidy classroom but failed to initiate any educational goals would be criticized for his or her failure to establish appropriate priorities, whereas a custodian who did the same thing might be praised as exercising leadership. In terms of how L mobilizes FF, the speaker may appraise the morality of the means or the difficulty of mobilizing FF. A leader who organizes people who are already willing may not be assessed as highly as one who mobilizes reluctant and disorganized FF. The importance to a speaker of the various aspects of leadership may vary across situations, but the ascription of the exercise of strong-normative leadership to some L will usually involve overall assessment of L's actions. To ascribe the exercise of leadership to some L in either the weak-normative or the strong-normative sense is to evaluate L positively. In the weak-normative sense, it involves crediting L with the ability to mobilize others to pursue shared goals. In the strong-normative sense, it involves evaluating L positively with regard to ends and means. It is not merely to say, "I personally approve of what L is doing," but to say, "What L is doing merits approval." Let us add the Merit Condition to the other conditions for the exercise of leadership: L's actions merit approval. This will allow for both weak-normative THE CONCEPT OF LEADERSHIP / 56 and strong-normative leadership. It may well be the case that many competent language users share certain standards for judging whether or not leadership is being exercised, but I am not, at this point, attempting to argue for any particular standards. That is the task of the next chapter. 4. The Conditions of Exercising Leadership As far as I can determine, there is no case of exercising leadership which includes all the conditions we have discussed but which a knowledgeable language user would not want to call "exercising leadership," and no case which manifests additional conditions not covered by the above. Let us review the conditions: 1. The Direction Condition: L provides direction for FF in pursuit of a non-trivial, shared goal. 2. The Adherence Condition: FF pursue the goal. 3. The Influence Condition: L influences FF to pursue the goal. 4. The Commitment Condition: FF pursue the goal, at least in part, because they regard it as desirable or worthwhile. 5. The Autonomy Condition: FF have autonomy in determining whether or not to follow L. 6. The Merit Condition: L's actions in some sense merit approval. It should be pointed out that this analysis does not assume that all the conditions will necessarily be met by the same L. An organization could THE CONCEPT OF LEADERSHIP / 57 conceivably have a leadership team in which one person had great ideas and another inspired FF to carry them out. Although we might not designate either as a "leader," we might well say that leadership was being exercised in that organization. 5. "Followership" Throughout this analysis I have used cases which competent language users would describe in terms of exercising leadership and cases sharing some of the same characteristics in which they would not, to help clarify the phrase "exercising leadership." At this point it will be useful to consider briefly the idea of "followership" to see whether it corroborates the analysis or casts doubt on it: Leadership and followership are linked concepts, neither of which can be comprehended without understanding the other (Heller and Til, 1982:405). It would be a mistake to assume too close a relationship between "exercising leadership" and "following," however; "following" often has connotations of "obeying," and many FF would be reluctant to describe themselves in terms of following L. For example, in a university department, leadership is exercised but department members "work with" the dean rather than "follow" him or her. Some of the dictionary definitions of "follow" seem directly related to "leadership": 1. to go, proceed, come after 2. to accept as authority, obey THE CONCEPT OF LEADERSHIP / 58 3. to copy after, imitate 4. to result or occur as a consequence, effect or inference (Webster's New Collegiate Dictionary). The first definition of "follow" corresponds to the Direction Condition to the extent that L initiates a direction for FF. L shows the way, and FF proceeds afterwards. The second definition of followership parallels the Adherence Condition in that to accept as authority or to obey is to do what someone wants one to do; in the case of leadership, to pursue certain goals or activities. The sense of "obey" which is relevant here is that in which we obey a rule that we accept, rather than the sense of obeying a person or a command, in that FF do not follow L because they are "obeying" him or her. The third definition, "to copy after, imitate" suggests that the leader has initiated something, and that both the leader and the follower are doing the same thing, or share a goal. The fourth definition, "to occur as a consequence," finds its counterpart in the Influence Condition, in which following L is a consequence of L's influence. The dictionary lists several definitions under "follower," such as "one in the service of another," "one that follows the opinions or teachings of another," and "one that imitates another," and notes that the shared element of meaning in each is "one who attaches himself to another." This suggests voluntariness, captured in the Autonomy Condition. "Followership" is defined as "capacity or willingness to obey," again reinforcing the voluntary aspect of leadership. Overall, a superficial examination of "followership" seems to support our analysis of "leadership." THE CONCEPT OF LEADERSHIP / 59 6. Resolving Conceptual Confusion McLellan and Komisar (1962) maintain that analysis is useful only to the extent that it can resolve problems; the conceptual analysis of "leadership" in this chapter offers plausible resolutions to a number of problems. First, distinguishing among two uses of "leadership" (the position use and the normative use) enables us to see how leadership seems both to encompass everything that position leaders and non-position leaders do, and to mean something fairly specific. It also enables us to understand how it can be treated as both value-neutral and value-laden. In particular, the distinction between weak-normative and strong-normative leadership explains the dispute over whether such persons as Hitler exercised leadership. By recognizing that "leadership" is a value term for which different people may have different standards, we can understand more clearly the nature of certain disagreements. Clarifying the conditions of exercising leadership enables us to pinpoint the shortcomings of various definitions of leadership. We can see that there is a strong tendency to focus on one aspect of leadership to the exclusion of others. For example, Lipham's definition of leadership as "the initiation of a new structure or procedure for accomplishing an organization's goals and objectives or for changing an organization's goals and objectives" (1964:122) does not make explicit the idea of FF's following L because they view the new structure or procedure to be worthwhile. Nash's definition of leadership as "influencing change in the conduct of people" (1929:24) omits the idea of the change being for the THE CONCEPT OF LEADERSHIP / 60 better. The analysis also pinpoints for us deficiencies in arguments such as the following: . . . [Principals have to be seen as leaders. The word implies adopting a posture of power over others and influencing them to achieve stated goals. According to some philosophers the very act of exerting influence is by definition immoral. At times, the intense ambiguity of this position reduces the conscientious administrator/ principal to a condition bordering on catalepsy (Watts, 1985:33). Our analysis reveals that leadership does not necessarily involve adopting a posture of power over others. Moreover, it is not exerting influence that is immoral, but using means of influence which fail to respect the personhood of FF. Definitions such as "Leadership is the ability of one person to influence another to act in a way desired by the first," a definition which Tosi (1982) claims has wide acceptance in the field, fail to take into account the importance of FF's autonomy. Such definitions also omit the idea of "goal" or significant change. Convincing someone to have Cheerios for breakfast rather than Cornflakes is not, for most of us, an example of the exercise of leadership. In other words, to the extent that it is a correct account of how we use language, the conceptual analysis gives us a standard by which to assess other definitions. It enables us to see what is being given up in certain definitions of leadership, and to make informed decisions about the relevance to us of theory or research based on such definitions. C. SUMMARY OF THE SENSES OF LEADERSHIP To recapitulate, there appear to be two different ordinary language uses of the term "leadership." The first is position leadership, a nominal usage indicating THE CONCEPT OF LEADERSHIP / 61 offices or positions of power or authority. We might say that the leadership of a country is in trouble, or that a leadership convention is being held. The second usage is a normative one, in which the speaker is evaluating positively some aspect or aspects of a leader's giving direction to followers in pursuit of a shared goal, influencing them in a manner which respects their rational autonomy, to view the goal as worthwhile. To ascribe the exercise of leadership to some L primarily on the basis of L's ability to mobilize FF or to accomplish what L sets out to do is to use the weak standards of "leadership." To ascribe the exercise of leadership to some L primarily on the basis of the worthwhileness of the goals, and the appropriateness of L's means of influence is to use the strong standards of "leadership." Let us consider now which kinds of standards should guide our efforts to equip educational administrators to exercise leadership. D. THE CENTRALITY TO EDUCATION OF STRONG-NORMATIVE LEADERSHIP It seems reasonable to suppose that position leadership, weak-normative leadership, and strong-normative leadership are relevant and important to educational administrators. As potential holders of leadership positions, aspiring educational administrators have some interest in position leadership; weak-normative leadership is relevant to educational administrators in their role as managers in that they are required to get their subordinates to perform various duties; and strong-normative leadership is relevant in terms of their role in shaping educational programs. The call for leadership is, however, not for THE CONCEPT OF LEADERSHIP / 62 someone merely to fill leadership positions, nor for someone who can convince FF to follow him or her; it is, rather, a call for persons of vision who can make legitimate value choices, initiate valuable, well-considered plans of action for educating persons and lead others toward some change for the better — it is a call for strong-normative leadership. To quote Burns once more: The call for leadership is one of the keynotes of our time . . . for moral, uplifting, transcending leadership, a leadership of large ideas, broad direction, strong commitment (1978:451). Rosenbach and Taylor speculate that our concern with leadership reflects a need to be assured that our leaders will "take us where we want to be, individually and collectively" (1976:297). Indeed, we may prefer an educational administrator with educational vision and only adequate ability to mobilize, to an excellent mobilizer with limited vision. It is possible that strong-normative leadership is more important in the context of education than in other contexts, such as business, because the goals of education are not straightforward; rather, they are indeterminate ideals which need to be given particular extension. Having a clear idea of what goals should be pursued in education is important. As Ott argues: It is not inconsequential what values are held as ideals . . . They give us our stamp of who we are, what we stand for, what we work towards (1985:2). Neither position leadership nor weak-normative leadership takes into account the nature of the goals. Thus they do not speak directly to our concern with increasing the ability of educators to exercise leadership. Because we want to THE CONCEPT OF LEADERSHIP / 63 prepare administrators to lead others to worthwhile pursuits, the most useful and relevant sense of leadership on which to focus is strong-normative leadership. This will direct our attention to the leader's ends and means, and appropriate standards for assessing them. But the concept of leadership cannot tell us which ends and means are worthy of approval. Thus we need to argue for what would be defensible ends and means in the context of educational leadership — we need a defensible conception of educational leadership. Building such a conception is the task of the next chapter. CHAPTER III. EDUCATIONAL LEADERSHIP Chapter 2 explicated the use of the expression "L is exercising leadership" and argued that the widespread concern with promoting leadership among educational administrators is most reasonably construed as a concern for promoting strong-normative leadership. However, we do not yet have an account of the exercise of educational leadership which is adequate for assessing the usefulness of leadership theory and research to educational administrators. Leadership may be exercised in many different contexts — in business, religion, politics, sports, and the military, as well as in education, t We cannot merely assume that leadership theory or research developed in a business or military context will be relevant to the educational context. In fact, there are reasons for supposing such research may not be generalizable across contexts. It seems very unlikely that the same kinds of goals or plans of action will merit approval in every context or that the same sorts of means for inducing persons to pursue the goals will be acceptable in every context: Leadership, like teaching, takes different forms in different contexts, and there is no reason to think that being effective in one area will make one effective in some other. Because contexts require very different sorts of goals, for example, having a sense of what is worth pursuing in one field will not necessaril}'' mean that one will have a sense of what is worthwhile elsewhere (Hare, 1986:4). tA content analysis of the popular press revealed that these are the most common modifiers of "leadership" (Lord, 1982), and presumably what ordinary language users take to be the primary leadership contexts. 64 EDUCATIONAL LEADERSHIP / 65 We need criteria for differentiating research which is relevant to the exercise of leadership in education from that which is relevant to the exercise of leadership in other contexts. To establish these criteria we must develop, as Miner urges, "a strong link between leadership on the one hand and the particular organizational context in which leadership behavior occurs on the other" (1982:295). That is, we need to develop a conceptiont of leadership for the educational context which outlines at least the basic criteria in terms of which educational administrators should choose goals and plans of action, and which could serve as one basis for developing programs for educational administrators. This chapter has three parts: (1) an explication of the values that should inform our conception of educational leadership; (2) a consideration of the strengths and weaknesses of three conceptions of leadership which have been influential among educational administrators; and (3) an outline of a conception of educational leadership which, there are reasons to believe, is the most defensible tRawls (1971) introduces the notion of a conception in A Theory of Justice. One builds a conception by interpreting, or giving determinate meaning to, a concept. For example, although we might agree that "justice" involves a proper balance between competing claims, we may not agree on particular instances of justice because we interpret "proper balance" differently. It is not that we mean different things by the expression "proper balance," but that we disagree as to what is to count as proper balance in a particular instance. Thus conceptions of justice may vary even when the general concept is agreed upon. Similarly, we may agree that the concept of leadership includes the leader's influencing others in some appropriate manner toward some appropriate end. Yet we may differ in our conceptions of leadership because we may make different judgments about what ends or means are appropriate. This is not to say that all conceptions are equally defensible or that adopting a conception is a matter of mere preference. Indeed, there may be good reasons to adopt one conception over another. Not all concepts are sufficiently vague or indeterminate that persons have importantly different conceptions of them. The concept of water, for example, is relatively straightforward. However, most value concepts, such as "equity," "excellence," and "knowledge" are sufficiently indeterminate that persons might disagree on how the conditions should be interpreted in general, or in particular instances. EDUCATIONAL LEADERSHIP / 66 for addressing the problem of promoting the exercise of leadership among educational administrators. A. DEVELOPING A CONCEPTION OF EDUCATIONAL LEADERSHIP There are basically two sources of values we need to examine to develop a conception of educational leadership for our society. Since we would expect a system of public education to be organized in such a way as to support the fundamental principles upon which its society rests, examination of values embodied in a liberal democracy would be one relevant consideration. We might also usefully examine the values implicit in our society's shared ideal of the educated person. Crittenden elucidates the relationship between education and democracy: Since education is in part a process of initiation into the political life of a society, it is not surprising that there would be connections, both conceptual and empirical, between the theory and practice of politics and education. Any comprehensive theory of what education should be like must, to some extent at least, include an ideal of society. The precise extent of the overlap of educational and democratic (or other social) theory depends inevitably on how we understand the concepts of education and democracy. Political theory becomes much more significant when, for example, 'education' is interpreted predominantly as a process of socialization, or 'democracy' is thought to refer not just to a form of government, but to a comprehensive style of life (1973:129). Let us turn now to what some have argued about the ideal of a liberal democracy. EDUCATIONAL LEADERSHIP / 67 1. The Ideal of a Liberal Democracy Like "leadership," "democracy" often functions as a value term. Because many countries with very different political arrangements have laid claim to a democracy, it is not clear that the term entails any particular descriptive state of affairs. However, there seems to be considerable agreement as to important principles which underlie liberal democracy in the twentieth century. Various attempts have been made to characterize the democratic way of life in this century. For Dewey (1916), a society was democratic to the extent that it: (1) provided for participation in its good of all its members on equal terms and (2) secured flexible readjustment of its institutions through interaction of the different forms of associated life. In other words, there had to be a variety of interests which were shared, and social change had to come about through dealing with problems encountered through varied interaction among social groups. In a speech given in 1937, Dewey (1974) argued that the foundations of democracy were faith in the capacities of human nature, in human intelligence, and in the power of pooled and cooperative experience. In Dewey's interpretation of democracy, each person had the right to be heard, and to have his or her judgment assessed on its own merit rather than on the basis of prior status. In building a value framework for educational administration, Graff and Street summarize what they take to be the ideals of democracy (though no existing democratic society approaches their ideal): 1. Paramount value is placed on the dignitj' and inherent worth of each EDUCATIONAL LEADERSHIP / 68 individual. 2. All who are influenced by a decision should have an appropriate part in its determination and in its implementation. 3. Every individual is obligated to become reliably informed concerning social problems and to act with others in their solution. 4. Actions, both individual and group, should be based in the method of intelligence rather than upon intuition, revelation, authoritative decree, or impulse. 5. Both social and individual development of the best kind is realized through calculated evolutionary means rather than through expedienc}' or revolutionary violence. 6. Freedom of action is not laissez-faire license, but rather is earned as the result of increasing individual and group responsibility for the results of action (1957:148). More recently, Crittenden has identified a number of features he takes to be critical to a democracy: 1. Decisions affecting the public interest, or the common good, or the settlement of conflicts between special interest groups within the society, are determined by the will of a majority of citizens directly or of a majority of their representatives. 2. The election of representatives is done by a majority vote of the citizens. 3. Beyond a common age of maturity, every human being in the society is equal as a citizen. Every citizen is free (at least in the sense that there is no legal obstacle) to seek election and to have an equal vote in the EDUCATIONAL LEADERSHIP / 69 election of representatives. 4. The government protects extensive freedom of speech and association. Interest groups are not impeded from organizing for political purposes. 5. The only acceptable method of gaining support for a policy is non-violent persuasion. 6. The conditions of a general election are such that - the ruling group may be dislodged. 7. The government does not have a monopoly on the channels of information. 8. At least in the traditions of the society, if not in an explicit constitution, there are constraints on the sphere of government action and minorities are protected (1973:131). Finally, Scheffler eloquently expresses his view of the democratic ideal: The democratic ideal is that of an open and dynamic society: open, in that there is no antecedent social blueprint which is itself taken as a dogma immune to critical evaluation in the public forum; dynamic, in that its fundamental institutions are not designed to arrest change but to order and channel it by exposing it to public scrutiny and resting it ultimately upon the choices of its members. The democratic ideal is antithetical to the notion of a fixed class of rulers, with privileges resting upon social myths which it is forbidden to question. It envisions rather a society that sustains itself not by the indoctrination of myth, but b}' the reasoned choices of its citizens, who continue to favor it in the light of a critical scrutiny both of it and its alternatives. Choice of the democratic ideal rests upon the hope that this ideal will be sustained and strengthened by critical and responsible inquiry into the truth about social matters. The democratic faith consists not in a dogma, but in a reasonable trust that unfettered inquiry and free choice will themselves be chosen, and chosen again, by free and informed men (1973:137). Underlying all these accounts are basic principles about persons and EDUCATIONAL LEADERSHIP / 70 how they ought to be treated. First is the principle of equality of persons. This does not mean that persons are equally talented or of equal worth, but that they have equal right to be treated with respect. This includes equal rights as citizens (e.g., one person, one vote); equal opportunity; and equal consideration of their interests. Second is the principle of the use of intelligence. In solving problems, the use of intelligence is to be preferred to reliance upon authoritative decree, revelation, or intuition. This ideal is reinforced in the ideal of a liberal state that persons are to be autonomous in conducting their lives (so long as they do not infringe on the rights of others). According to Strike (1982), this entails the ability to function rationally. In promoting change, rational persuasion is to be preferred to violence or force. Third is the principle of cooperation and pooled experience in setting values and in problem solving. Not all conceptions of democracy allow for limits on governments, once elected. Western liberal democracies, however, embody the principle that individuals and minorities should be protected against possible injustices by governments or majorities: The freedom we enjoy in the Western democracies is a precious commodity, as well as a fragile one, and it represents Western man's idea, evolving since classical times, of the just society (Berger, 1982:xiii). Thus we might add a fourth fundamental principle, the idea that individuals have certain rights, such as those entrenched in the Constitution, which cannot be sacrificed for the good of the majority. EDUCATIONAL LEADERSHIP / 71 Any defensible conception of educational leadership should be true to the fundamental principles upon which our society is based. The principles widely regarded as important to a liberal democracy may be summarized as follows: 1. Equal respect for persons 2. Use of intelligence in problem solving and promoting change 3. Cooperation and pooled experience in setting values and solving problems 4. Respect for individual rights. 2. The Ideal of the Educated Person The ascription of educational leadership to some L is always open to debate because what count as legitimate goals and plans of action are not settled issues. Still, there is substantial agreement among philosophers (e.g., Peters, 1966; Frankena, 1973; and Crittenden, 1973) as to the nature of the educational ideals generally held in our society. Moreover, there is considerable agreement concerning the procedural principles that must govern education in a liberal democracy. Together these provide the basis for developing standards relevant to judging whether or not educational leadership is being exercised. Peters' (1966) analysis of the "educated man" was a landmark in philosophy of education, and has greatly influenced the field for nearly two decades. Even his critics acknowledge the significance of his analysis: I think it fair to say that the traits Peters claims one must possess EDUCATIONAL LEADERSHIP / 72 to be a truly educated person and the kind of education he assumes one must have in order to acquire those traits would, with minor variations, be cited by any number of people today if they were to describe their own conception of the ideal (Martin, 1981:97-109). Peters purported to be explicating the ordinary language concept of education or, at least, the sophisticated language user's differentiated concept of education rather than proposing an "ideal" of education. In any case, his analysis seems to be a useful starting point for determining how we may defensibly conceive of educating persons. Peters points out that "education" can be used to describe both a task and an achievement. Being educated is the achievement relative to a family of tasks or processes of education. The sense of "education" of concern here is not the value-neutral sense in which we might speak of the educational system of country X; nor is it the sense which equates education with schooling. Rather, it is the evaluative sense in which to say that P is an educated person is typically to say something good about P. Peters looks at what we mean when we call someone an educated person and at the standards to which processes must adhere before we acknowledge them as "educational." According to Peters' analysis, education is the intentional transmission or acquisition of worthwhile knowledge or understanding by means which allow for voluntariness and wittingness on the part of the learner. Let us unpack this definition. The intentional transmission of worthwhile knowledge is central to Peters' EDUCATIONAL LEADERSHIP / 73 account of education. Characterizing education as intentional establishes it as a goal oriented activity rather than as a sequence of behavioural steps (Scheffler, 1966). Moreover, the "knowledge" must not be a mere collection of facts, but must be tied to a conceptual scheme in the sense that it has explanatory or interpretive power. Thus, someone who can answer every Trivial Pursuit question is not necessarily educated. Scheffler's characterization of "knowledge" seems to be related to Peters' idea of education and may help to elucidate it. According to Scheffler, acquiring knowledge is neither a matter of storing bits of information nor of being prompted to search in oneself for realities which exist in one's mind but which have been hitherto unexplored. Rather it is acquiring beliefs through employing principles of rational judgment; that is, public standards of evidence, such as the rules of logic and consistency. These principles or standards of evidence are not immutable but reflect what we acknowledge as the best principles we have at present. Scheffler elaborates further that in order for a person to have knowledge, that person's "autonomy must be evidenced in the ability to construct and evaluate fresh and alternative arguments, the power to innovate, rather than just the capacity to reproduce stale arguments earlier stored" (1966:110). Neither Scheffler nor Peters argues that it is illegitimate to teach persons isolated pieces of information in the process of educating them, or that all learning must be the result of rational reflection. This is legitimate, but it is insufficient, for having such bits of information counts as having knowledge only in the weak sense of that term. Having knowledge in this sense is compatible with holding beliefs for which one has no evidence or grounds. If this sort of knowledge is all one has, according to Scheffler and Peters, he or she is not an educated person. EDUCATIONAL LEADERSHIP / 74 For Peters, an educated person must have both "depth" and "breadth" of knowledge. Moreover, this knowledge cannot be inert. It must affect the person and transform his or her outlook. The person must care about the standards of evidence implicit in this knowledge and have some sort of commitment that "comes from being on the inside of a form of thought and awareness" (1966:31). Also important in Peters' account of "the educated person" is that what is transmitted is thought by the user of the term "educated" to be worthwhile. Peters (1972:8) concedes that the worthwhileness criterion may be part of the knowledge criterion rather than a necessary condition in its own right; that is, education is thought to be worthwhile because knowledge is thought to be worthwhile. In a society which did not value knowledge, it is possible that there could be cases of education in which nothing of value was transmitted. However, central cases include both knowledge and worthwhileness. Lastly, for Peters, the process of education must be witting and voluntary on the part of the learner, and be morally unobjectionable. That is, the learner must be aware of the teacher's intention and voluntarily learn what is taught. This criterion would rule out at least some procedures of transmission, such as conditioningt and brainwashing. (Indoctrination is ruled out on the grounds that an indoctrinated person does not have knowledge and understanding.) Underlying this third criterion of education is the idea of "respect for persons." To respect persons is to treat them seriously as agents or as determiners of their own tin cases where a person desires to be conditioned to behave in a particular way, conditioning that person would not signify lack of respect, but nor would it count as educating the person. EDUCATIONAL LEADERSHIP / 75 destiny, as well as to take seriously their feelings and views of the world. Thus, to try to brainwash, indoctrinate, or in some cases, condition persons is not to respect them as persons: When it is said that a man who brainwashes others, or who settles their lives for them without consulting them shows lack of 'respect for persons', the implication is that he does not treat others seriously as agents or as determiners of their own destiny, and that he disregards their feelings and view of the world. He either refuses to let them be in a situation where their intentions, decisions, appraisals and choices can operate effectivelj', or he purposely interferes with or nullifies their capacity for self-direction. He ensures that for them the question, 'What ought I to do?' either scarcely arises or serves as a cork on the tide of events whose drift derives from elsewhere. He denies them the dignity which is the due of a self-determining agent, who is capable of valuation and choice, and who has a point of view about his own future and interests (1966:210). In Peters' view, education is contrasted with socialization, with training, and with schooling. Education differs from socialization in that only worthwhile knowledge is conceived to be educative; many things perceived not to be worthwhile are learned as part of socialization. It differs from training in that it necessarily involves breadth of knowledge; training may involve only very narrowly focussed learning. It differs from schooling primarily in that much schooling is not seen as involving the transmission of worthwhile knowledge, and becoming educated need not occur in the context of schooling. It differs from indoctrination (and other forms of transmission such as conditioning and brainwashing) in that where education respects the personhood of the learner, the purpose of indoctrination is to get someone to believe something without regard for their coming to believe it' as the result of considering reasons. It should be pointed out that this account, though it contains interpretive EDUCATIONAL LEADERSHIP / 76 elements, is essentially an explication of the concept of education; it does not propose a particular conception of education. According to Frankena (1973), every activity of education can be represented by the following formula: In it X is fostering or seeking to foster in Y some disposition D by method M. For the normative concept of education, X = those doing the educating; Y = those being educated; D = dispositions it is desirable Y should have; and M = the methods that are satisfactory. Frankena argues that answering the question of what dispositions are desirable and what methods of fostering them are satisfactory is not a matter of asking what is regarded as desirable or satisfactory by our society, but is to be answered by looking at more universal premises about what is good and right and factual premises about the psychology of learning, methods of fostering dispositions, etc. He views the desirable dispositions as consisting of a mastery of forms of thought and action with their respective standards, together with responsibility and autonomy. This view of education is consistent with that of Peters, but Frankena notes that missing from Peters' account is the aim of creating individuals who can and will make advances within these traditions and even begin new traditions. Crittenden claims that the notion of education is inextricably bound up with EDUCATIONAL LEADERSHIP / 77 the notion of schooling. He views education as: systematic initiation into the main public modes of thought, undertaken primarily because of the skills and conceptual perspectives for understanding and appreciation that these modes of thought afford (1973:xi). For Crittenden, schooling is viewed as a process of initiation with the following dimensions: 1. The acquisition of basic linguistic, logical, and mathematical skills; 2. An understanding of the best available bodies of theory and methods of inquiry for explaining human and natural phenomena; 3. An understanding of the main varieties of belief and theory for interpreting human life and guiding action; 4. An appreciation of the broad range of ways in which human beings express themselves imaginatively; and 5. The attainment of some proficiency in the art of using concepts, theories, methods of inquiry, of evaluating, making practical judgments, expressing oneself imaginatively, and living as a moral agent. Like Frankena's view of education, Crittenden's view of schooling could, without much distortion, be summed up as the transmission of worthwhile knowledge in a morally acceptable way. Developing a composite of these analyses, we might take the following conditions to hold in usual cases which would be characterized by the expression "X is an educated person": 1. Person Y transmits knowledge to X. EDUCATIONAL LEADERSHIP / 78 2. The knowledge is worthwhile, including such things as listed by Crittenden (1973) above. 3. The knowledge is intentionally transmitted by Y and intentionally acquired by X. 4. X has depth and breadth of understanding. 5. X cares about the knowledge or is transformed by it in some way. 6. The means of transmission is morally acceptable, i.e., it respects and fosters the rational autonomy of X. Someone who is exercising educational leadership is not necessarily educating FF; thus, the standards for judging whether someone is engaged in educating do not translate directly into standards for judging whether educational leadership is being exercised. Nevertheless, the values implicit on our democratic ideals and those implicit in our ideal of the educated person do suggest standards for assessing the goals and the means of educational administrators, and thus for determining the degree to which they are exercising educational leadership. For ease of reference in the following discussion, I shall call a person who is committed to these ideals of democracy and education a "democratic educator." B. COMPETING CONCEPTIONS OF LEADERSHIP Understanding the ideals of the educational enterprise in a liberal democracy is essential in building a defensible conception of educational leadership. Rather than try to fashion a conception out of whole cloth, however, we might first consider three conceptions of leadership to determine whether they might be EDUCATIONAL LEADERSHIP / 79 usefully adapted for our purposes. These conceptions are implied in various works on leadership. For ease of reference, we shall call these conceptions the Democratic conception of leadership, the Congruency with Organizational Values conception of leadership, and the Moral conception of leadership. 1. The Democratic Conception of Leadership The Democratic conception of leadership was popular in the mid-century, at which time educational journals teemed with articles on how to translate the democratic ideals to the educational context. Those in the democratic leadership movement abhored the mechanistic view of persons which underlay such movements as that of Scientific Management. They argued that each individual had worth and that persons should be allowed to set goals for themselves. At a meeting of the National Conference of Professors of Educational Administration (NCPEA) held in 1948, members attempted to reach agreement as to basic ideals of a democratic society upon which all value frameworks for educational administrators should be based. They stated those ideals as follows: The first declares the dignity and worth of the individual. Man is placed first; things are subjugated to the welfare of man . . . The second ideal of democracy places reliance upon the method of intelligence. This ideal conceives that man's problems can be solved through his own intellectual efforts . . . It is diametrically opposed to the concept that man's problems may be solved by appeals to authority. The third ideal of democracy places reliance on the cooperative use of intelligence in the solution of problems common to the group . . . Cooperative action is action which will bring the individual intelligence of each member of the group to bear most fully and appropriately in EDUCATIONAL LEADERSHIP / 80 the solution of a common problem . . . (1948:5-6). According to the NCPEA, Democratic leadership always exercises its function toward the achievement of two ends: First, society itself is improved. Things get done. Toward this end efficiency is the criterion. Second, those who get things done are themselves improved . . . Toward this end growth is the criterion (1948:16-17). The NCPEA developed a Democratic conception of educational leadership which included the inward convictions of democratic leadership and its outward signs. The inward convictions included the following: 1. The welfare of the group is assured by the welfare of each individual. 2. Decisions reached through the cooperative use of intelligence are, in total, more valid than decisons made by individuals. 3. Every individual is entitled to a fair hearing. 4. Every person can make a unique and important contribution. 5. Growth comes from within the group rather than from without. 6. Democracy is a way of living. 7. Democratic methods are efficient methods. 8. Individuals are dependable. 9. Persons merit love. The outward signs of democratic leadership, which grow out of these inward convictions, were characterized as follows: 1. Its processes increase the powers of individuals to adjust, to solve problems, to gain satisfactory expression, to maintain emotional poise, and to grow in EDUCATIONAL LEADERSHIP / 81 attitudes and mature in behavior. 2. Its effectiveness is measured by what happens to people. 3. It grows out of the action of a group working on a problem and does not belong to any one individual as a privilege. 4. It comes from within the group and not from some outside source. 5. It develops and uses for the common good the potentialities of each member of the group. 6. It shares the formulation of policies and decisions with every person concerned with or to be affected by the decison, insofar as possible. 7. It assists the group in arriving at a consensus. Graff and Street also propounded a Democratic conception of educational leadership: . . . [I]n a democracy a person in an administrative situation dare not permit himself to act in such manner that the people blindly follow him as a person. It is his duty and obligation to manage the group in such wise that the goals set by the group become objectives for all; and that the individuals in the group give their full allegiance to these goals, until they change or modify them, rather than to the administrator or any other member of the group. Thus it may be said that in a democratic situation people follow or are directed by the ideas (goals) which they themselves have created, rather than following the leader (1957:151). In addressing the problem of lack of educational leadership, the Democratic conception, as characterized here, shows little promise of being fruitful. First, in requiring that FF set goals for themselves, it places administrators in the position of being held accountable for goals not of their making. It is unrealistic to expect them to embrace a mere coordinating role, given their hierarchical EDUCATIONAL LEADERSHIP / 82 position. Second, this conception focusses on the process at the expense of the goal or plan of action. Although the NCPEA gives token acknowledgment to the goal of the betterment of society, in their conception the worthwhileness of goals is secondary to the leader's means. If L makes decisions in a democratic manner, then those decisions are seen to be good. The idea of the leader intentionally initiating a goal or influencing FF offended the moral sense of those in the Democratic leadership movement, resulting in such attenuation of the concept of leadership that Caswell, at Eisenhower's inauguration dinner, felt compelled to denounce the "cult of discussion leadership" in which persons with expertise were expected to defer to those whose minds were "untrammeled by knowledge": The splendid tide of civic interest in education deserves sterner stuff than the leader who never makes a decision that will displease anyone, appoints committees for every detail, and will not say whether the sun is shining without counting the yeas and nays first (1955:14). A third problem with this conception is that FF are expected to become better in some way, which places excessive demands on the exercise of leadership. Finally, to require leaders to believe that persons merit love is to require them to believe something for which the evidence is at best inconclusive. It is regrettable that the democratic ideals were translated so restrictively into conceptions of Democratic leadership. If the democratic ideals had been interpreted differently, the conception might have proved more fruitful, and thus not been abandoned. There is nothing in the democratic ideals which requires that FF set goals for themselves. As long as FF have good reason to believe EDUCATIONAL LEADERSHIP / 83 that their interests have been taken into account, and come to accept the goals initiated by L as their goals, democracy is being served. James (1951) believes that the widespread belief that all persons should participate in every decision relevant to them has been dysfunctional. He maintains that the only conditions required for leadership in a democratic community are that it is open to men and women of any class; that it is based as far as possible on persuasion; and that there is a wide diffusion of responsibility. Also, the democratic ideals do not require that the effectiveness of leadership be measured in terms of what happens to FF. As long as FF are rationally committed to a goal, they need not be transformed in any way. Finally, the democratic ideals demand the belief that persons should be treated with respect but not that they are entitled to love. Such a belief rests not upon the nebulous inherent goodness or badness of human beings, but upon their having rational wills, with purposes and plans of their own. In building our conception of educational leadership, it will be useful to keep in mind that the democratic ideals have typically been interpreted more restrictively than necessary, often resulting in unrealistic expectations of administrators. 2. The Congruency with Organizational Values Conception of Leadership The Congruency with Organizational Values (COV) conception of leadership is implicit in much current talk about leadership. The main idea behind this view is that leadership consists in promoting whatever an organization was set EDUCATIONAL LEADERSHIP / 84 up to achieve. One's exercise of leadership is assessed by the degree to which one maintains the integrity of the organizational value system and thus those who impose values which are not part of the collectivity are not viewed as exercising leadership. A simple example will illustrate why this conception of leadership is unlikely to be fruitful for our purpose of improving the exercise of educational leadership. Let us imagine a school in which all the staff are committed to getting the students to memorize a great deal of information; in fact, the school is set up for this very purpose, and the principal does an excellent job of devising methods which work well. Although the principal is mobilizing persons to achieve the organization's values, our hypothetical "democratic educator" would not ascribe the exercise of leadership to her. Now suppose that a new principal convinces the teachers that memorizing material is not very useful or productive, and as a result the teachers become committed to getting the students to understand the various fields of knowledge. Our "democratic educator" would judge this principal to be exercising educational leadership. The COV conception of leadership would lead us to regard the first principal as exercising leadership and the second principal as not exercising leadership, despite the fact that the second principal is pursuing goals of far greater educational worth. The general point is that the COV conception of leadership does not give importance to the initiation of new, more defensible values for the organization. Yet the ability to provide more valuable direction is precisely what is seen to be lacking in our educational leaders. EDUCATIONAL LEADERSHIP / 85 Those holding the COV conception of leadership would probably maintain that it is not the educational administrator's job to initiate goals but, rather, to adhere to the goals set by the public, the client.t Even if we were to accept the legitimate role of the public in determining educational goals, it is clear that the public is not univocal; there is confusion and dispute over goals and their relative importance. Is it the school's job to educate students morally? To teach them job training skills? Leadership requires that the competing goals be molded into a synthetic, unified view of an educated person — that goals of education be seen in broad rather than narrow perspective. Often, in response to particular situations, some segment of the public, perhaps even the government, will demand educational arrangements that are antithetical to our society's basic ethical commitments, such as justice, respect for persons, or equality. Educational administrators who want to exercise educational leadership cannot just adopt whatever goals are currently fashionable among dominant or vocal groups in our society. As members of the National Education Association expressed the problem: [I]t is not that we do not think hard, but rather that we do not think critically about those values that determine our choices. In education we are satisfied to be swept along with each new movement, hoping that we may be ingenious enough to earn for ourselves some little attention from the crowd (1933:8). It should be noted that the COV conception of leadership may or may not impose restrictions on a leader's means, depending upon whether the organization's value structure includes such restrictions. For example, it is likely that most organizations disallow physical coercion on the part of L. The problem tNotice that the word "client" predisposes one to the view that the public have the right to decide what services thejr wish to have provided. EDUCATIONAL LEADERSHIP / 86 is that this conception does not guarantee any such restrictions; L's goals and means must be congruent with the organization's values, whatever they are. 3. The Moral Conception of Leadership In characterizing what he calls "transformational leadership,"t Burns (1978) was attempting to explicate what I have called strong-normative leadership, rather than to build a conception of leadership. Although I do not think that his characterization adequately captures how we use "leadership," it can be usefully considered as a conception of educational leadership. For Burns, transformational leadership is moral in that leaders and led have a relationship not only of power, but also of mutual needs, aspirations, and values. The followers have adequate knowledge of alternatives and the capacity to choose among them. Leaders take responsibility for their commitments, that is: leaders must accommodate followers' wants and needs without sacrificing basic principle (otherwise they would not be leaders); they must mediate conflict without becoming mere referees without purpose of their own; they must be "with" their followers but also above them (1978:78). For Burns, the ultimate test of leadership is its capacity to transcend the claims of the multiplicity of everyday wants, needs, and expectations to respond to the higher levels of moral development. Interestingly, he sees an intimate connection between education and leadership: tBurns argues that transforming leadership ultimately becomes moral; thus, this explication of his characterization of transformational leadership includes his characterization of moral leadership. EDUCATIONAL LEADERSHIP / 87 Ultimately education and leadership shade into each other to become almost inseparable but only when both are defined as the reciprocal raising of levels of motivation rather than indoctrination or coercion (1978:448). In terms of potential fruitfulness, Burns' conception of transformational leadership probably demands too much in requiring that FF must be "transformed" in some way. Although a conception should allow for the transformation of FF, it should not require it. As noted earlier, L's contribution in exercising leadership ma}' be not in changing FF or raising FF's levels of motivation, but in helping FF to translate goals to which they are already committed into programs of action. In education, many of the goals are agreed upon at a general level, but there is need for persons who can synthesize the various abstract educational goals into a concrete program of action. Burns' conception would lead us to regard such desirable accomplishments as insufficient to count as the exercise of leadership. C. OUTLINE OF A CONCEPTION OF EDUCATIONAL LEADERSHIP In this section, we shall consider the kinds of goals that some L who is exercising educational leadership might initiate, and the means the leader might appropriately use to influence FF. In the analysis of "leadership," and to a lesser extent, "education," I tried to set forth the conditions which governed the use of the terms by competent language users. Here, my task is not to analyze how language is used, but to set forth the standards which would be invoked by a democratic educator (i.e., a person committed to democratic ideals and to the epistemological and moral values implicit in our ideal of education) in ascribing EDUCATIONAL LEADERSHIP / 88 the exercise of educational leadership to some L. This discussion will take into account the ideals of liberal democracy, the ideals of education, and the three conceptions of leadership advanced so far. 1. The Goal If Peters' analysis of how the differentiated language user uses the term "education" is essentially correct, then a goal or plan of action would not count as "educational" unless it were in some way connected to the transmission of worthwhile knowledge in a morally acceptable way. We should keep in mind that not all goals of an educational administrator have to be educational ones, but it would be odd to say that L was exercising educational leadership! if he or she were pursuing some non-educational goal. Suppose that a principal gets teachers to teach astrology, alchemy, and problematic doctrines such as Creationism, and to represent them as true. Because our democratic educator subscribes to standards which do not allow such subjects to count as knowledge, he or she would not view such actions as the exercise of educational leadership. Similarly, a superintendent could decide that students require a wide range of facts at their disposal, and initiate a program tAlthough my focus in this thesis is on educational leadership, it is likely that educational administrators will also be called upon to exercise political leadership, entrepreneurial leadership, managerial leadership, moral leadership, and business leadership. It is at least plausible that what is to count as leadership in each case (and even what is to count as educational management) will be shaped to some extent by the nature of the educational context. For example, political leadership within the educational context may be somewhat different from political leadership in a governmental context. EDUCATIONAL LEADERSHIP / 89 in which students are taught all the answers to Trivial Pursuit. Again, our democratic educator would want his or her conception of educational leadership to rule out such cases — what is being taught is tied neither to knowledge in any strong sense nor to explanatory principles. A democratic educator would also want a conception which rules out the case of a principal who inspires teachers to instigate a program of electric shock treatment to ensure that the students acquire worthwhile knowledge. In sum, our democratic educator will ascribe leadership to an educational administrator only when that administrator is influencing persons to undertake plans of action which are relevant to the enterprise of transmitting worthwhile knowledge through morally acceptable means. 2. The Means Should our conception of educational leadership include any restrictions upon the educational leader's means of influence other than those implicit in the concept of leadership itself? Even weak-normative leadership embodies certain moral restraints, in that L must not coerce FF unduly, and FF must follow L autonomously. But the concept of leadership does not rule out FF's believing a goal to be worthwhile because of being caught up in L's enthusiasm or seduced by L's charismat; that is, FF's belief does not have to be rationally based. Do we want to allow reliance on personal attraction alone to be a legitimate means tPerhaps, in the future, greater use of rational persuasion will be required of political leadership than in the past. It is interesting to note that even in political leadership, personal charm has its limitations. Persons seemed willing to follow Ronald Reagan on the basis of his charm until the stock market crash in October, 1987. At that time, as Morrow describes it, "the nation callfed] for leadership and there [was] no one home" (1987:22). EDUCATIONAL LEADERSHIP / 90 of influence in our conception of educational leadership, or do we want to restrict means of influence to rational persuasion? Before answering this question, I should elaborate on the notions of rationality and rational persuasion as used here. By "rationality," I do not mean some restricted set of rules for making logical deductions, or some technical means-end relationship. A rational decision may include political, economic, and moral considerations, to name just a few. Following Scheffler (1973), I view being rational as the ability to participate in critical and open evaluation of rules and principles in any area of life, treating arguments of others on an equal basis as one's own and being open to change on the basis of reason. "Rational persuasion" entails certain criteria such that the action or belief being propounded has good reasons in its favour, that it is morally unobjectionable, and that the persuader rely predominantly on adequacy of evidence and soundness of argument (Crittenden, 1973:116). This is not to say that L must rationally persuade FF of the merits of every detail of a plan of action; only significant changes in direction require rational persuasion. It could be objected that although the leader's goals must conform to educational ideals, there is no reason to assume that the leader's means should conform to educational ideals. Educational administrators, so the objection would go, educate students only indirectly through leading teachers; it is not their responsibility to educate teachers. This objection seems plausible, but even if we accept that it is not part of L's job to educate FF (which I shall question) it should be pointed out that one need not be "educating" FF by emplosang rational persuasion as a means of EDUCATIONAL LEADERSHIP / 91 influence. Moreover, there are numerous reasons for questioning the adequacy of personal charm as a means of influence in the context of education. Let us begin by considering what our democratic educator would say about a case in which FF become committed to a course of action without understanding it. The principal, rather than rationally persuade the staff of the merits of a new language development program, stimulates them with her enthusiasm. The teachers do not understand how the program works, but follow the directions to the best of their ability. Their efforts result in their pupils attaining higher scores on standardized tests, t Our democratic educator may not want to allow such cases to count as educational leadership because to respect FF's rational autonomy fully would entail allowing them to form their beliefs on the basis of reason and evidence. Moreover, our democratic educator would probably not regard the achievement of higher test scores as justifying the means of influence. The exercise of leadership is not merely a matter of achieving ends, but the means must also adhere to the democratic ideal of respect for persons. Pondy's insight regarding effective leadership captures part of our democratic educator's thinking: My perception of research in the field is that effectiveness is typically conceptualized as "performance of the subordinate group" — usually some kind of output measure — or perhaps as compliance or tSuch a result seems unlikely because students are so variable that we would not expect a particular technique to work for them all; unless teachers had a clear understanding of what they were doing, they would be unable to tailor their teaching to accommodate individual differences. EDUCATIONAL LEADERSHIP / 92 adherence to the leader's directives. In any case, the effectiveness concept and measure is invariably a behavioral one. The "good" leader is one who can get his subordinates to do something. What happens if we force ourselves away from this marriage to behavioral concepts? What kind of insights can we get if we say that the effectiveness of a leader lies in his ability to make activity meaningful for those in his role set — not to change behavior but to give others a sense of understanding what they are doing, and especially to articulate it so they can communicate about the meaning of their behavior (1978:94). It is not only our democratic educator who would want included in his or her conception of educational leadership the idea that FF follow L because they see the merit of the goal rather than because of L's personal charm. In building its conception of democratic educational leadership, the National Conference of Professors of Educational Administration (1948) argued that because a "leading" person may rely on personal attraction alone (or force, delusion, or position!), leadership described in these terms was not an adequate concept for education. The National Education Association (NEA), too, deplores the idea of L's personal charm: One sometimes finds an organization in which there is complete harmony of interest and desire, due to the charming quality of the leader. He is worshiped by the group. His slightest wish is law. No sacrifice for him is too great. Service is a pleasure. The minds of the workers never rise above nor stray beyond the thought and plan and purpose of the master. Here one finds willing, wholehearted harmony, consecration to a common purpose, but in it there is no respect for individuality, no cultivation of uniqueness, no fostering of self-respect, no critical attitude, and little if any growth. Some call it loyalty; others might call it intellectual and spiritual slavery (1933:94). Haller and Strike explain why they find offensive the conception of tThe NCPEA failed to distinguish between "leading" and "exercising leadership." According to our analysis, only the first of these means could be included in the exercise of leadership. EDUCATIONAL LEADERSHIP / 93 leadership which emphasizes leader personality and style rather than the adequacy of his or her view: It is incompatible with the values of autonomy, reason and democracy, which we see as among the central commitments of our society and of our educational system. Of course educational administrators must be leaders, but let them lead by reason and persuasion, not by force of personality (1986:326). And Balderson implicitly accepts the idea of rational persuasion for educational leadership in his set of ethical guidelines for administrative leadership: 1. Use methods of influence which support the self-respect and dignity of those who are being influenced. 2. Do not use methods of influence which are destructive to the ability of individuals to reflect rationally on their interests. 3. Do not use methods of influence which [adversely] affect in fundamental ways the personal dignity of individuals. 4. Do not use methods of influence which rely on deception or on ignorance of relevant facts. 5. Avoid using physically intrusive methods of influence, if possible. 6. Ensure that changes are reversible, if possible. 7. Promote the active cognitive and affective involvement of recipients, rather than the passive acceptance of changes (1980:4). Balderson justifies the appropriateness of the guidelines on the grounds that they are consistent with the notion that the purpose of education is to foster autonomy. Whereas it is not the administrator's primary mandate to educate EDUCATIONAL LEADERSHIP / 94 teachers, it can be argued that treating teachers in such a way as to foster their autonomy will increase the probability that teachers will, in turn, treat their students in a similar manner. The argument is often made that it is reasonable to assume a domino effect of sorts in attaining certain educational goals, including those concerned with fostering a climate of democracy. That is, "if these values [of scientific attitude, spirit of democracy, zeal for social betterment, and creativeness in attacking the problems of teaching] are manifest in pupil-teacher-supervisor-administrator relationships the probability of securing inspired teaching is greatly increased" (NEA, 1933:18). Giving FF reasons for pursuing a goal and ensuring that FF pursue the goal on rational grounds rather than on blind faith is to respect their rational autonomy. With regard to rational persuasion, an educational administrator who is not interested in providing teachers with good reasons for pursuing a goal could not reasonably expect them to respect the rational autonomy of their students. By exhibiting lack of respect for autonomy, they may well be undermining commitment to this ideal. It seems likely that teachers acquire commitment to such an ideal not merely through lectures or textbook reading, but through working with others who have that commitment. Teachers who are given little intellectual or moral respect are unlikely to pass on such respect to their students. Dewey argues that teachers with low esteem and little motivation can adversely affect the learning process; they may "pass on, perhaps unconsciously, what they feel to be arbitrary treatment received by them to their pupils" (1937:461). Although Dewey was advocating democratic methods of dealing with teachers, his arguments would hold for respecting FF's rational autonomy through use of rational persuasion. Scheffler, too, supports this idea: EDUCATIONAL LEADERSHIP / 95 Rational character and critical judgment grow only through increased participation in adult experience and criticism, through treatment which respects the dignity of the learner as well as teacher (1966:111). (Emphasis added.) I think that one should not accept unquestioningly the idea that it is not part of an educational administrator's job in exercising leadership to educate FF, particularly if one agrees with the NEA that "an educational organization has for its purpose the growth and development of all the members of the organization, including administrators, supervisors, teachers, and pupils" (1933:19). Some would hold that the best leadership involves the education of FF; for them, the value restrictions imposed on educational means would be highly appropriate for a conception of educational leadership. According to Bennis: Above all, the task for today's leader is to create not only a climate of ethical probity but a climate in which it is possible for the jpeople around him to grow and continue learning, in which contributions are prized and independence and autonomy encouraged. Forgive an educator if he puts in first place the leader's obligation to encourage the ability to learn. For more than 20 years, people in the Institute for Social Research, at the University of Michigan, have been trying to discover just what it is that gives a person satisfaction in the job. And they have finally concluded it is, above all, the opportunity and capacity to learn (1976:139-40). Tead, too, argues that the good leader is a good teacher: His role is like the teacher's in helping others through experiences which bring a changed mind and motive. Emphasis upon this view of his task would be helpful if for no other reason than that it keeps to the front the complete difference between leading and bossing. A good teacher is never a boss. He is a guide helping to start and hold the students' interest toward mastery in a particular field. And this is no less true of the leader (1935:140-1). EDUCATIONAL LEADERSHIP / 96 Although Selznick overstates the case for our purposes in speaking of "transforming" FF, he makes the point well that initiating goals and getting FF to identify with them is an educational process: The inbuilding of purpose is a challenge to creativity because it involves transforming men and groups from neutral, technical units into participants who have a peculiar stamp, sensitivity, and commitment. This is ultimately an educational process. It has been said that the effective leader must know the meaning and master the techniques of the educator . . . . The leader as educator requires an ability to interpret the role and character of the enterprise, to perceive and develop models for thought and behavior, and to find modes of communication that will inculcate general rather than merely partial perspectives (1957:149-150). And, to the extent shaping educational goals, public as to how to play that the public have a part of L's leadership role that role responsibly. legitimate role to play in would be in educating the Given these arguments, it seems reasonable to include in our conception of educational leadership the idea that L must rely chiefly on rational persuasion as a means of influence. This is not to deny the importance of other means of influence, such as exemplifying an ideal, as concomitants to rational persuasion. We noted earlier Jacobs' (1974) argument that in the exercise of leadership, L must be open to influence attempts by FF, i.e., L must in some sense be prepared to change a goal or plan of action if some F convinces him or her that such change is warranted. Should this restriction be included in our conception of educational leadership? EDUCATIONAL LEADERSHIP / 97 It would be very odd indeed if we expected L to use rational persuasion with regard to FF, but not to be open to change if good evidence could be adduced that such change was desirable. Part of having respect for persons is giving serious consideration to their views (which is not to imply acceptance of those views). Thus, it would seem that L must be open to being rationally persuaded by FF or others to modify or abandon a goal or plan of action. 3. The Conditions of Exercising Educational Leadership The ascription of the exercise of educational leadership to some L entails judgments about the desirability of goals and plans of action and about the suitability of methods of influencing people. The goals of one who is exercising educational leadership will have to contribute to the education of persons, i.e., the transmission of knowledge in a morally acceptable way and to be justified in terms of a democratic society's basic commitments, including both goals such as equal opportunity and means such as rational persuasion. The leader should rely mainly on rational persuasion as a means of influence rather than on charisma or subtle manipulation. In fleshing out this conception, one might argue for how priorities should be set in education, what constitutes worthwhile knowledge, what counts as rational persuasion, and the like. But any morally and epistemologically defensible conception must operate within these constraints. Let us now consider the conditions of exercising educational leadership: EDUCATIONAL LEADERSHIP / 98 1. The Direction Condition: L provides direction for FF in pursuit of a non-trivial, shared goal that leads or contributes to the transmission to students of worthwhile knowledge by morally acceptable means. 2. The Adherence Condition: FF pursue the goal. 3. The Influence Condition: L influences FF, in a manner which respects the rational autonomy of FF (chiefly by means of rational persuasion), to pursue the goal. L is open to rational persuasion to change. 4. The Commitment Condition: FF pursue the goal because they regard it as educationally desirable or worthwhile. 5. The Autonomy Condition: FF have autonomy in determining whether or not to follow L. (The Merit Condition is not included as a separate condition in this conception because it is fulfilled when all of the above conditions are met.) 4. Compatibility with the Democratic Principles To what extent is this conception compatible with the principles identified earlier as underlying the ideal of a liberal democracy? First, equal respect for persons is embodied in the idea that L must give serious consideration to the views of all followers, regardless of their status. Respect for persons is also taken into account in the ideas that L must rely chiefly on rational persuasion, and that FF must pursue a goal in part because they regard it to be worthwhile. EDUCATIONAL LEADERSHIP / 99 Second, the use of intelligence rather than authority in solving problems and the use of intelligence rather than violence in initiating change are compatible with this view of educational leadership. FF follow L for good reasons rather than because of L's charisma or authority. Also, where L relies on coercion or rewards and punishment to initiate change, leadership is not exercised according to this conception. Third, this conception ensures some measure of cooperation in setting goals in that FF must be rationally persuaded that the goals are appropriate before FF adopt the goals. Moreover, FF always have the opportunity to attempt to rationally persuade L to change. This conception also allows for the possibility of pooled experience. A good leader will coordinate the talents of the followers to take advantage of their experience and talents in concerted action. Finally, several of the conditions of exercising educational leadership ensure that FF's rights are not violated, including the Influence Condition, which states that L must respect FF's rational autonomy; the Commitment Condition, which states that FF must view the goal as educationally desirable; and the Autonomy Condition, which gives FF the choice not to follow L. D. FRUITFULNESS OF THE CONCEPTION FOR EDUCATIONAL ADMINISTRATORS Is the conception of educational leadership which I have outlined likely to be fruitful for educational administrators? EDUCATIONAL LEADERSHIP / 100 Some scholars criticize those who would advocate normative conceptions of leadership which restrict occasions when administrators may be said to lead. Tyack, for example, argues that to portray an administrator as either a "routinized bureaucrat" or an "institutional Moses who parts the Red Sea" is to undervalue the important everyday work of running the schools, and to elevate leadership to something unattainable by the ordinary person (1982:265). Although the conception of educational leadership outlined here does restrict occasions on which administrators may be said to lead educationally, it is not overly restrictive in that it does not require great vision or supernatural talents on the part of educational administrators, and it in no way precludes the acknowledgment of the exercise of a variety of other kinds of leadership within the context of the schools. All that is required of educational administrators is that they have a clear understanding of the nature of education, as characterized here, and act with reason and with respect for the persons they lead. Granted, even these abilities are often developed only through long and difficult study, but this does not mean that we should broaden the meaning of the expression "L is exercising educational leadership" to encompass the everyday work of running the schools. This is not to claim that only the leadership aspect of administration deserves attention or recognition, but rather that because managing requires more technical expertise, it is usefully kept distinct from leadership. A useful conception of educational leadership will allow for the exercise of leadership at all levels of the hierarchy. There is nothing in the conception of educational leadership outlined in this chapter which is specific to any level of administration. The exercise of educational leadership may involve mega-projects EDUCATIONAL LEADERSHIP / 101 and high order goals, but it may also involve initiating plans of action. In a paper presented to the Philosophy of Education Society, Maxcy (1984) argued that educational leadership is correlated with policy operation, because it is only at that level that there is capacity to introduce uniqueness; other tasks do not have that capacity. In response, Sokiloff perspicaciously observed: The leadership that can be exercised outside of the policy making authority delegated to the board of school directors is considerable. Significant leadership can be exercised through collecting and interpreting data, in the policy recommendations made to the lay board, in the way professional prerogatives are exercised, and in the way policy is interpreted and implemented (1984:341). The conception of educational leadership outlined herein is preferable to the Democratic conception of leadership on several counts. First, it takes into account the worthwhileness of the goals, rather than merely the means. Second, it is less restrictive than Democratic conceptions which require growth on the part of FF as an aim. This conception allows for FF's growth but does not require it. It also does not require that administrators believe that persons merit love, but merely that they are entitled to be treated with respect. Finally, in this conception, it is not necessary that FF set their own goals: they need only be accorded the respect due a rational agent. Given the present situation in which administrators do have authority over teachers and are held accountable for the goals, it allows administrators a role more in keeping with their hierarchical status, in that they provide direction, rather than merely coordinate. This conception is preferable to the Congruency with Organizational Values conception in that it recognizes the importance of critically assessing and revising EDUCATIONAL LEADERSHIP / 102 the values and visions according to which educational institutions currently operate, in light of our more fundamental democratic and educational ideals. It also allows for L to initiate goals rather than to take them as given. Restricting the means to (primarily) rational persuasion ensures that FF are viewed as persons; the COV conception has no such safeguards. This conception also has some advantages over the Moral conception of educational leadership, if only that it is more realistic in terms of what we can reasonably expect of educational leaders. It does not require that people are "lifted into their better selves" (Burns, 1978:462) or even that they change at all. All that is required is that they understand why a goal or plan of action is appropriate. There would seem to be good reasons for educational administrators to take seriously the conception of educational leadership I have outlined in general terms in this chapter. Clearly, it avoids many of the disadvantages of its competitors. In the next chapter we shall review leadership theory and research to determine its relevance to fostering educational leadership as outlined in this chapter. CHAPTER IV. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH Mintzberg (1982) argues that unless research serves the practitioner, it does not serve leadership. If leadership theory and research fail to shed some light on the penumbra of how to foster the exercise of leadership, there may be little point in including it in preparation programs for educational administrators. The task of this chapter is to determine the relevance of existing leadership theory and research with respect to educational leadership as explicated in the previous chapter. Lakatos' (1970) influential and highly regarded theory of scientific research programs t provides a useful lens through which to view leadership theory and research. Although research approaches used by those studying leadership may not in any strict sense qualify as research programs, viewing them from this perspective allows us to focus on what is taken to be essential to leadership in each program, and what is open to modification. I shall first review early approaches to leadership — trait, style, behaviour, and leadership in groups; I shall then explicate what I take to be the predominant research program of leadership, which I call the Organizational Effectiveness Research Program of Leadership (OERPL), and assess it for its potential usefulness to the fostering of the exercise of educational leadership, as tFor those familiar with Kuhn's work, "research programs" in Lakatos are similar to "paradigms" in Kuhn except that, unlike paradigms, numerous research programs can be accepted by members of the scientific communit3' at the same time. 103 ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 104 conceived here. Next, I shall consider and assess current leadership research which is not classifiable as OERPL. Finally, I shall consider the question of generalizability of research. Let us begin with a brief account of Lakatos' theory of scientific research programs. A. LAKATOS' THEORY OF SCIENTIFIC RESEARCH PROGRAMS The features of Lakatos' theory which are relevant to my purpose are as follows: 1. Research programs are based upon a fundamental "hard core" of propositions which cannot be revised without destroying the program. The hard core tells us what will not be revised even in the face of contrary evidence. In addition, research programs include a "protective belt" of propositions which can be altered in light of contrary evidence. The protective belt tells us what factors and variables to investigate in our research. 2. The proper unit of appraisal is neither a proposition nor a single theory, but a connected series of theories called a research program. These theories share the hard core but may hold different views about the importance of particular variables. For example, theories might have in common a view of leadership as traits, but place importance on different kinds of traits. 3. A theoretically progressive research program is one which makes novel predictions about the course of nature; it is empirically progressive if these ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 105 predictions are confirmed. By contrast, there are theoretically and empirically degenerating research programs. 4. It is not irrational to support a particular research program despite damaging evidence against it. Research programs are not suddenly abandoned because crucial experiments show them to be false. Tests in science are three-way contests among rival theories and nature, rather than two-way contests between a theory and nature. When a clash occurs between a particular theory and nature, that theory is not given up unless there is another to replace it. Let us now review early leadership research. B. EARLY APPROACHES TO LEADERSHIP In this section, I shall review early research on leadership traits, styles, and behaviours, as well as leadership in groups. 1. The Trait Approach Early trait studies grew out of the belief that leaders are persons endowed with superior qualities which differentiate them from followers. Consequently, studies were conducted to try to determine the nature of those qualities. In terms of Lakatos' theory, the hard core was that leadership was something about L (in this case, traits) that allowed L to be a leader. The protective belt consisted of several assumptions, the most important of which was that anyone ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 106 who holds a "leadership" position is a leader. The appropriate methodology was to find traits which correlated with holding such a position. However, the subsequent recognition that leadership is not equivalent to office holding led to a shift in the protective belt, and trait studies began to look for correlations between traits and various measures of leadership performance, such as observer ratings, peer ratings, and self ratings (Mann, 1959). In a review of trait research covering the years 1904 to 1947, Stogdill concludes that: A person does not become a leader by virtue of the possession of some combination of traits, but the pattern of personal characteristics of the leader must bear some relevant relationship to the characteristics, activities, and goals of the followers (Reprinted in 1974:63-64). Reviewing trait research from 1948 to 1970, he reports that a number of leader characteristics have been claimed variously to differentiate leaders from followers, effective leaders from ineffective leaders, and higher echelon from lower echelon leaders: A strong drive for responsibility and task completion, vigor and persistence in pursuit of goals, venturesomeness and originality in problem-solving, drive to exercise initiative in social situations, self-confidence and willingness to accept consequences of decision and action, readiness to absorb interpersonal stress, willingness to tolerate frustration and delay, ability to influence other persons' behavior and capacity to structure social interaction systems to the purpose at hand (1974:82). These seem to be desirable traits for leaders to possess, but rather than showing consistency, the findings of the studies are contradictory. The studies ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 107 that show that leaders in general (as opposed to followers) possess these traits are inconsistent with the studies that show that ineffective and lower echelon leaders do not possess them. The general consensus is that trait studies failed to identify a core of leadership traits. As Fielder sums it up: It is probably fair to say that almost every conceivable personality trait has been related at one time or another to leadership behavior, status, or performance (1971:4). Moreover, researchers found the idea of traits too nebulous to pin down and measure. Lastly, because leadership "traits" connoted innate qualities which could not be taught, some researchers felt that this approach smacked of elitism. Trait research was unfashionable for some years, but recently it has resurfaced in various forms. Handy (1976), for example, has developed what he calls the "differentiated trait approach." He suggests that a core of traits, such as high tolerance for ambiguity, clear self-concept, and high energy, are important for any leader. To this core, additional traits will be useful according to the particular requirements of the situation. Rather than studying leaders to see what traits they have in common, he examines the situation to determine what traits would be useful. Such an idea conceivably could be useful in educational administration, particularly if the notion of "trait" were broadened to include knowledge, abilities, dispositions, and value commitments. We could determine what traits would seem ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 108 to be important in the exercise of educational leadership, devise means for determining how to assess the presence of these traits, and then see the extent to which educational administrators widely regarded as exercising educational leadership possess these traits. Although few would now take the position that leadership consists of some set of traits, it may be the case that certain traits are preconditions for the exercise of leadership in certain contexts. One advantage of focussing on traits is that this reflects our ordinary language usage to some extent, in that we speak of persons having leadership potential and leadership ability. Another branch of trait research focusses on the values which motivate managers. Miner (1978), for example, has developed a test which he claims has predictive power in terms of who will attain management positions and be promoted to higher levels of management. The test, the Miner Sentence Completion Scale, tests positive attitudes toward authority figures, desire to compete with peers, desire to exercise power, desire to be actively assertive, desire to stand out from the group, and willingness to carry out rather routine administrative functions. Another researcher in this area, McClelland (1975), concludes that the need for power is the dominant motive of most successful organizational managers, particularly when the managers are sufficiently mature emotionally to channel that power toward the benefit of others. Consider the kinds of behaviours which he and Winter (1973) cite as typical of expressing the need for power: 1. Reading books or watching films with an emphasis on violence, explicit sexuality, or competitive sports. 2. Collecting prestige possessions as symbols of influence and status. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 109 3. Engaging in competitive sports, especially those with a "one-on-one" situation where a player tries to outwit or dominate a particular opponent. 4. Taking alcohol or drugs, or participating in mystical-religious rituals to heighten experience of personal strength and influence over events. 5. Helping others or giving advice in a way that demonstrates personal superiority and the weakness or dependence of others. 6. Joining organizations and assuming a leadership role in them. (Cited by Yukl, 1981:78.) Since these are not trainable traits, this kind of research would be relevant to our concerns only if we were to suppose that it identifies traits which should guide our selection of potential leaders. Given our rejection of power as a necessary condition of the exercise of leadership, we have no reason to suppose that it does identify such traits. 2. The Style Approach One of the more influential studies on leadership was conducted by Lewin, Lippitt, and White in 1938 (White and Lippitt, 1960) at the Iowa Child Welfare Station at the University of Iowa. The researchers tried to determine the extent to which leadership style affected task performance of eleven-year-old children. It was found that under autocratic leadership, the children were productive, but tended to have low morale; that under democratic leadership, the children exhibited high work quality and high morale; and that under laissez-faire leadership, the children exhibited low productivity, low work quality, and low morale. This study spawned a great deal of research on various leadership styles. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 110 The question of the extent to which leadership style is trait-related (and thus not readily changed), and the extent to which it is behaviour-related (and thus readily modifiable), has not been answered. It is difficult to judge whether this research differed significantly from trait research. In one sense, it did not. Leadership was viewed as something about L (style), and the appropriate methodology was to correlate that with some measure of leadership effectiveness. In another sense, it was very different. Leadership effectiveness was measured, at least in part, by subordinate performance.t This may have marked the beginning of concern with "organizational" leadership, in which leadership effectiveness is equated with attainment of organizational goals. Research on leadership styles has not been notably successful in identifying styles which correlate strongly with indicators of desired subordinate performance. Such conclusions as have been suggested by this research program are not particularly useful for the purpose of promoting the exercise of educational leadership, unless we can satisfy ourselves that using the recommended style is consistent with respecting the autonomy of followers, and acting in a morally responsible manner, and can be effective in bringing about worthwhile educational goals. We must not suppose that everything identified in the literature as a "leadership stj'le" passes these tests. We shall have to look closely and critically at the range of activities taken to be exemplars of each style in individual studies. tHowever, it should be noted that the researchers did not make any judgments about the most "effective " leadership style. It was those reading about the studj' who made the judgments. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 111 3. The Behaviour Approach In the late 1940's, leadership researchers shifted their attention to leader behaviour. Some aspired to study the behaviours of those who had attained managerial positions, whereas others attempted to study the behaviours of those who were considered to be effective managers. It would perhaps be fitting to begin with a review of the Ohio State leadership studies because they influenced research on leadership for several decades. Approximately 1800 items describing leader behaviour were sorted into nine different categories. Of these, 150 were used to develop the first form of the Leader Behavior Description Questionnaire. Two of the categories accounted for much more variation than the others — consideration and initiating structure: Initiating structure refers to the leader's behavior in delineating the relationship between himself and members of the work group and in endeavoring to establish well-defined patterns of organization, channels of communication, and methods of procedure. Consideration refers to behavior indicative of friendship, mutual trust, respect, and warmth in the relationship between the leader and members of his staff (Halpin, 1966:86). Led by such scholars as Halpin (1957) and Hemphill (1957), large scale studies were conducted in the attempt to discover leader behaviours which correlate with organizational effectiveness. The shift in interest from what leaders do to what "effective" leaders do, was subtle but radical. Concern with leadership was replaced with concern for the attainment of organizational goals. As we shall see, this is characteristic of OERPL research. Early studies indicated that the most effective leaders scored high on both consideration and initiating structure ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 112 indicators. Subsequent research showed consideration to be correlated positively with subordinate satisfaction but negatively with production (satisfaction and production being typically viewed as indicators of organizational effectiveness); and initiating structure to be correlated positively with production but negatively with satisfaction. More recently, the findings of the early studies have been corroborated. The dimensions of consideration and initiating structure influenced later leadership research in contingency theory, but were too general to be useful descriptors of leader behaviour. This was remedied by Stogdill, Goode, and Day (1962) with the Leadership Behavior Description Questionnaire XII which tests for twelve types of "behaviours" — consideration, initiating structure, demand reconciliation, tolerance of uncertainty, persuasiveness, tolerance of freedom, role retention, predictive accuracy, production emphasis, integration, and influence with superiors and persons outside the immediate organization subunit as well as with subordinates. Even if the items which comprised each factor were originally characterized in behavioural terms, few of these twelve factors are articulated in such terms. Some "behaviours," such as demand reconciliation, role retention, and influence with others are expressed as achievements rather than behaviours; others, such as tolerance of uncertainty and of freedom, are expressed as traits. There is little point in dealing with similar studies in great detail. Research conducted at the University of Michigan resulted in somewhat different categories (supportive behaviour, group method of supervision, setting high performance goals, and serving as an intermediary between one's work group and higher ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 113 management); Bowers and Seashore (1966) developed the categories of support, facilitating interaction between members, emphasizing goals, and facilitating work; Yukl (1971) developed a taxonomy of nineteen behaviour categories, and many others have found different ways of categorizing leader behaviour. Sometimes categories of behaviour were based upon observation, sometimes reflection, sometimes simulation studies, and sometimes questionnaires, including open ended questionnaires wherein subjects were asked to describe "critical" incidents in which leadership was effective or ineffective. Some scholars were interested in what office holders do, whereas others tried to ascertain which behaviours were associated with effective leaders, t The hard core is that leadership is something about L (in this case, behaviour), and the appropriate methodology is to correlate that something with holding a position or with being effective in that position. In the end, those working in the behaviour approach became frustrated because there appeared to be no behaviours which were common to all leaders or effective leaders and no behaviours which only leaders exhibited. Also, categorizing behaviours in terms neither so general as to include almost anything (like "'consideration" and "initiating structure") nor overly specific, proved to be a daunting task. The reasons for the failure of the behaviour approach are fairly obvious, given our earlier analysis of the concept of leadership. First, leadership is a tThose who undertook these studies did not necessarily view them as leadership studies. Mintzberg (1973), for example, was concerned with managerial behaviour in general. The important point, however, is that their studies have been viewed as leadership studies, and as contributing to our knowledge of leadership. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 114 term denoting intentional activity, and like many such terms it may denote an indefinite range of behaviours depending on the context in which the activity takes place. Leadership would seem to be a concept more like practicing than running. One cannot observe and categorize all the behaviours of persons who are practicing (e.g., playing the violin, playing basketball, singing) and be able, on the basis of these observations, to give an account of practicing in terms of behaviours. Whether or not playing the violin, for example, counts as practicing depends upon the context. If Isaac Sterne is playing the violin before an audience in Carnegie Hall, he is probably not practicing. Thus our conceptual analysis provides good reason for believing that the search for a particular set of behaviours constitutive of leadership was doomed to fail. Secondly, it was doomed to failure because, as our analysis shows, "leadership" is an evaluative term. Recognizing this, Campbell argues: It is a truism to say that we must have some idea of what leadership is before we can decide how much we know about it. What is, perhaps, not so obvious is that this is for the most part a value judgment or a matter of personal preference. It is not an empirical question (1977:222). (Emphasis added.) Although I am not in agreement with the apparent equating of "value judgment" with "personal preference," I agree with the main point of the argument, which is that leadership is not a descriptive concept. When a variety of researchers, making their own implicit value judgments, give different descriptive content to the term, the resulting body of research is likely to be confusing. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 115 4. Leadership in Groups There has been a long tradition of studying the phenomenon of group leaders. It is primarily concerned with understanding the dynamics of leading a group rather than with organizational effectiveness. Some of this research deals with "emergent" leadership. It attempts to understand how a person becomes leader of a group rather than to understand the exercise of leadership. It tells us, for example, that the person who sits in the most prominent chair, best represents the group norms, and talks frequently but not incessantly, tends to gain control over a group in the short run. Exercising leadership is seen as equivalent to having position power in a group whether or not such power is used to influence others to pursue worthwhile goals. Such studies are of interest only to the extent that they are concerned with how one attains position power which is not vested in an office. Other group research, such as the work of Bales (1950) at Harvard, studied the social behaviour of leaders in groups. Bales concluded that there are two leadership roles, one in initiating ideas and keeping a group on task, and the other in giving group members support and cohesion. He found that the two roles were often performed by different individuals, a finding compatible with my claim that leadership functions may be performed by more than one person, i.e., that L may represent more than one person in any given instance of the exercise of leadership. Although this research was conducted with small groups in laboratory settings, and thus may not necessarily generalize to larger, real-life contexts, it suggests a hypothesis for researchers in educational leadership to test: ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 116 L is more likely to persuade FF to follow some course of action in cases where L creates a supportive environment in which FF feel valued as persons than in cases where L does not create such an environment (other things being equal). 5. Summary In some ways, early trait, style, behaviour and group leadership studies shared basic assumptions about leadership; whether they viewed the important thing to be traits, behaviours, or styles, they sought to discover something which would characterize effective leadership in all situations. This proved to be a fruitless endeavour, and the way was paved for contingency theory, in which researchers seek to discover the situational variables which make a difference to the traits, behaviours, or styles which will be effective in a given context. It is not clear whether Fiedler's shift in focus from leadership to leadership effectiveness was responsible for the radical shift in leadership research programs, or whether industrial psychologists concerned with managerial effectiveness changed the nature of trait and behaviour research. In any case, much current research, particularly the contingency approach, represents a different research program — what I call the Organizational Effectivenness Research Program of Leadership (OERPL). Let us now consider the basic concerns of this reseach program. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 117 C. THE ORGANIZATIONAL EFFECTIVENESS RESEARCH PROGRAM OF LEADERSHIP The overall concern of OERPL is to improve organizational effectiveness.! "Organizational effectiveness," as used here, is the degree to which an organization achieves its acknowledged goals rather than a judgment that an organization itself is effective in some generalized sense. For example, although we may disapprove of an organization in which dealers buy used goods cheaply and resell them at high profits, if the organization achieves its purpose, it is "effective." In OERPL, leadership effectiveness is tied to organizational effectiveness. As phrased by Allen, "Any evaluation of organizational effectiveness is ultimately a measure of managerial effectiveness" (1977:24). "Leadership effectiveness" is not differentiated from "leader effectiveness" or "managerial effectiveness." Leadership effectiveness is measured chiefly in terms of productivity, although sometimes it is measured in terms of subordinate satisfaction, and some studies make use of others' perceptions of the leader's effectiveness. Using Lakatos' terminology, OERPL's "hard core" has two central features. The first is the underlying conception of leadership as something the leader is or does which promotes organizational effectiveness. The typical guiding definition is "Leadership is the exercise of influence to achieve a goal." Fiedler and Garcia, T OERPL may be just one arm of a larger scientific research program concerned with organizational effectiveness. Other arms might include the effects of the external environment upon organizational effectiveness, and the effects of organizational structure upon organizational effectiveness. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 118 for example, define leadership effectiveness as "the degree of success with which a group performs the primary assigned task" (1987:4).t Such definitions impose no restrictions on the goals or the means to achieve them. Who initiates the goals, their legitimacy, and whether or not the means to achieve them are morally acceptable are not relevant questions from the point of view of this hard core. The second important feature of the hard core is the methodology deemed appropriate for studying leadership — correlating leader traits, behaviours, styles, or modes of influence with measures of organizational effectiveness. OERPL's "protective belt," open to modification, consists of (1) hypotheses concerning what it is about the leader which promotes organizational effectiveness, and (2) hypotheses concerning the range of situations in which given leader traits, behaviours, or styles are thought to promote effectiveness. These are seen as the appropriate factors to investigate. The majority of current leadership research could be classified as the contingency approach to leadership. I shall try to show that some of the more influential contingency theories belong to OERPL, and I shall argue that OERPL research is not useful in promoting the exercise of educational leadership. 1. The Contingency Approach The Contingency Approach represented a major breakthrough for leadership researchers because it laid to rest the endless quest for traits or behaviours of a tThe authors note that a person may be an outstanding leader even though the organization fails to perform its task, but their model does not appear to take this insight into account. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 119 leader which would contribute to effectiveness in all situations. Instead, it sought to ascertain the situations in which particular leader traits or styles would prove to be associated with organizational effectiveness. I shall review a number of the more influential contingency theories because they tend to receive more attention in educational administration than trait or behaviour theories, and on the surface, at least, the various contingency theories appear to be quite different. a. Fiedler's contingency theory By far the most influential contingency theory has been that proposed by Fiedler in the 1960's. Fiedler (1971) developed the Least Preferred Coworker (LPC) Instrument which he claims measures a person's leadership style. Style, it is important to note, is not behaviour. According to Fiedler, it is variously what motivates one to act, and a relatively consistent way of interacting with subordinates. The instrument directs respondents to identify mentally that person with whom they have had most difficulty in completing a task, and then to rate that person in terms of a number (usually eighteen) of "bipolar adjectives" on a scale from one (very low) to eight (very high). Respondents' scores are then classified as "high" or "low" (a number of different cut-off points for these categories are to be found across various studies) and those with a high score are called relationship-motivated (high LPC) while those with a low score are called task-motivated (low LPC). ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 120 Fiedler's theory postulates that the effectiveness of a group leader will be contingent upon the optimum match of style and situation. High LPC (or relationship-motivated) leaders will fare best in terms of group effectiveness where there is a moderate amount of situational control (favourableness). Low LPC (or task-motivated) leaders are best suited to high control (very favourable) situations or low control (very unfavourable) situations. Whether or not a situation provides for control (favourableness) is determined by three criteria: leader-member relations, task structure, and leader position power. In a high control (very favourable) situation, leader-member relations are good, the task is highly structured, and leader position power is strong: in a low control (very unfavourable) situation, leader-member relations are poor, the task is loosely structured, and position power is weak. Other combinations constitute situations of moderate control (favourableness). Because Fiedler maintains that leadership style is relatively inflexible (despite the fact that LPC scores have been shown to alter dramatically in test-retest situations (Schriesheim and Kerr, 1977b)), he advocates that organizations should fit the job to the person rather than the person to the job. Organizations should restructure the task, adjust the authority vested in a position, or change the composition of the subordinate group to alter leader-member relations. In terms of OERPL, the hard core of Fiedler's theory is that leadership is something about L (style) which promotes group performance in a given situation. The methodology relies on correlational studies, but now the focus is on correlating styles of leadership with organizational effectiveness under certain ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 121 conditions rather than in all circumstances. Fiedler's notion of leadership is not a normative one. He states that we "cannot really talk about a good or a poor leader" (1971:13) because the data show that the effectiveness of the leader is determined in large part by the situation. His theory cannot account for the fact that we do indeed speak of good and poor leaders. Fiedler's view of leadership continues to influence leadership debate. According to Hodgkinson: . . . it can be said that anyone can lead. And so they do. The point of dissension is not about leadership itself but about leadership effectiveness. In the ordinary language it is positive effectiveness which has come to be equated with leadership (1978:196-97). Failure to distinguish between "lead" and "exercise leadership" might induce one to conclude mistakenly that the debate is not really about leadership. "Anyone can lead" does not mean that "anyone can exercise leadership" — an important fact often overlooked in leadership debate. b. Hersey and Blanchard Hersey's and Blanchard's (1976) contingency theory of leadership, once called "life cycle theory" and now called "situational leadership theory," is based on the amount of direction (task behaviour) and the amount of socio-emotional support (relationship behaviour) which a leader needs to provide in order to maximize organizational effectiveness! given the level of maturity of a subordinate with tAlthough Hersey and Blanchard argue that effectiveness involves not only objective performance, but also human costs and psychological conditions, this ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 122 regard to the task. Maturity is defined in terms of the capacity to set high but attainable goals, willingness to take responsibility for one's behaviour with regard to a specific task, and education or experience. Maturity involves both job maturity or requisite skills, and psychological maturity, or high self-respect. A subordinate can be (1) neither willing nor able (to perform a task), (2) willing but not able, (3) able but not willing, or (4) both willing and able. Hersey and Blanchard postulate that the most appropriate leadership style for "neither willing nor able" is telling or high task, high relationship; for "willing but not able," selling or high task, low relationship; for "able but not willing," participating, or low task, high relationship; and for "both able and willing," delegating or low task, low relationship. These are actually four points on a continuum. The greater the maturity, the less task oriented behaviour is required on the part of the leader in order to be effective. Very mature subordinates also require less relationship oriented behaviour. The model also provides for L's increasing FF's maturity by gradually decreasing directive behaviour toward FF. Moreover, it introduces the notion of contingency contracting, or negotiating an agreement with regard to subordinate duties and L's style. If FF regress, L can adjust his or her style accordingly. Hersey and Blanchard (n.d.) have extended their model to include the kind of power base that a leader should use in order to induce compliance. If the followers are of above-average maturity, the emphasis is on influence (expert t (cont'd) insight seems to get lost in practice, where all that counts is task completion. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 123 power), whereas if the followers are of below-average maturity, the emphasis is on compliance (coercive power). Basically, Hersey and Blanchard take the leader's goals as givens, and determine that the leader's means for influencing the follower to work toward these goals should depend on the maturity of the subordinate. However, to link maturity to willingness to perform a task is questionable; sometimes unwillingness to perform a task is a sign of maturity. In terms of OERPL, the theory adopts the hard core assumptions that leadership is something about L which enhances organizational effectiveness, and that the proper method of studj' is to correlate data on leaders with measures of organizational effectiveness. All that changes in the protective belt are the situational variables of interest (maturity) and the use of leader behaviour rather than style. c. House's path-goal theory A clear and concise explication of House's Path Goal Theory (1974) would be difficult, partly because of its complexity and partly because of ongoing revisions to it. However, because it is viewed as a major contribution to the study of organizational leadership, it is important to examine at least its basic elements. House attempts to explain how leader behaviour affects subordinate motivation and satisfaction. His theory is based on Vroom's expectancy theory, which explains motivation to work in terms of expectations and valences. Briefly, ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 124 in expectation theory, a worker calculates the probability that a given effort will result in successful completion of a task (effort-performance expectancy); considers the desirability of the work itself (intrinsic valence); and determines the value and probability of extrinsic outcomes (performance-reward expectancy), such as higher pay, promotion, dismissal, and rejection. A worker is satisfied when the various outcomes are worth the effort expended. House postulates that leader behaviours affect subordinate expectations and valences which, in turn, affect subordinate effort and satisfaction. The sensitive leader will take into account subordinate characteristics (needs, abilities, and personality traits) and characteristics of the task (structure and mechanization) and environment (the formalization of the organization) and will vary his or her behaviour accordingly. House has identified four categories of leadership behaviour: • Supportive — creating a friendly climate, concern for subordinates' well-being. • Directive — letting subordinates know what they are expected to do. • Participative — consulting with subordinates and taking their opinions into account. • Achievement oriented — setting challenging goals, seeking performance improvements, emphasizing excellence. According to the theory, directive leadership will increase subordinate effort and satisfaction when there is role ambiguity, but not when the task is highly ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 125 structured. Directive leadership includes the manipulation of rewards according to FF's performance. Supportive leadership is always desirable: in dull structured tasks, it increases the intrinsic valence of the task, and in unstructured, difficult tasks, it increases the subordinates' self-esteem. Achievement oriented leadership will give subordinates greater confidence in their ability to attain challenging goals, i.e., it increases effort-performance expectancy. It would be otiose in the case of routine tasks. Participative leadership increases subordinate effort in unstructured tasks. Through participation, the subordinates gain knowledge of the situation and thus increase their effort-performance expectancy. House's theory is not readily identifiable as belonging to OERPL because it seems to take into account FF's reasons for following L, and it does not use subordinate performance as an indicator of leadership effectiveness. However, the theory does not escape the OERPL net entirely. Leadership is viewed as something about L (in this case, behaviour) and the methodology is to correlate that something with measures of leadership effectiveness. One of these measures for House is subordinate motivation to perform tasks; another is worker satisfaction (which is in turn based on being rewarded for efforts to perform tasks). These measures are not really very different from subordinate performance. d. Contingency theory: future prospects Contingency researchers have attempted to characterize aspects of the leadership context which make a difference in the relationship between leader ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 126 traits, behaviours, or styles and organizational effectiveness. However, there is no agreement regarding which variables are the critical ones. Although the Contingency approach once opened new vistas with its assumption that each kind of situation requires a different style or pattern of traits or behaviours in order to enhance organizational effectiveness, optimism that a comprehensive theory would be developed has turned to disillusionment in the face of limited progress. Researchers have been unable to devise a theory which is complex enough to be meaningful but simple enough to be understood. Thejr have failed to agree upon the crucial elements of the situation which make a significant difference in the relationship between the traits, behaviours, or styles and organizational effectiveness. Individual researchers may claim to have isolated the crucial variables; however, work in this area is so method dependent that they must agree that no universal set of crucial variables has been found. Thus, even on their own terms, their research program appears to be degenerating. 2. The Limitations of OERPL Because OERPL has dominated leadership theory and research for the past several decades, it is important to consider its relevance to educational leadership in light of the conception developed in Chapter 3. OERPL research has tended to be conducted as though leadership were a purely descriptive term. Such research has taken place in the larger context of what is sometimes called the theory movement in educational administration, a movement which has its roots in ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 127 logical positivism. Culbertson traces the beginnings of the movement to a seminar held in 1957 at the University of Chicago. Two of the core ideas of that seminar he identifies as follows: (1) Effective theory development in administration requires the use of social and behavioral science concepts and modes of inquiry; and (2) Theory and research concentrate on the description, explanation, and prediction of administrative and organizational behavior; generalizations prescribing courses of action or specifying what administrators or organizations should do are beyond the purview of researchers and the capacity of science (1981:27). In OERPL, observable traits, behaviours, or styles are correlated with successful completion of tasks (or some other measure of effectiveness) in some context. Such a procedure seems innocuous enough, until we probe its implications. In this section we shall consider why educational administrators should be skeptical of OERPL. theory and research. In particular, we shall look at (1) its focus upon what ultimately reduce to trivial aspects of leadership, (2) its failure to justify the worthwhileness of the ends or the morality of the means, and (3) its implicit view of persons. a. Tendency toward the trivial Researchers working within the OERPL tradition have a particular view of social scientific research. Briefly, they see the point of such research as being not merely to understand the phenomenon of leadership, but to be able to predict and control human behaviour. Thus, they use correlational studies to point the way toward the discovery of general causal principles. One hallmark of such studies is that they tend to deal only with variables that can be fairly readily ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 128 • and easily measured or observed, for otherwise the research would be too time consuming and costly. In leadership research, as indeed in administrative research generally, this has led to a neglect of factors which are not easily measured or observed. In 1933 the National Education Association felt uneasy with what they saw as a growing trend toward such research in education: In their anxiety to be scientific, the research departments have narrowed their studies to those problems in which the data are objective, the treatment impersonal, and the conclusions demonstrable. In education, however, the most significant data are not objective; the treatment cannot be altogether impersonal, for the problems are personal, i.e., the hopes and ambitions and desires of persons are involved; and the soundness of the convictions on which educators act cannot always be demonstrable . . . Often this narrow conception of research has drawn the organization away from a common-sense consideration of the problem into a pseudo-scientific program of measurement, and thinking has been replaced by a routine performance from which trivial and invalid conclusions have been drawn (1933:136-37). Nearly fifty years later, Mintzberg argues that the conclusions are still trivial, and that researchers throw up a "smokescreen" of jargon to hide the emptiness of their findings. He makes his point by paraphrasing a complex research-based model of Multiplexed Supervision as "Sometimes leaders must treat their employees the same and sometimes differently," and an Opponent Process model as "How [leaders do things] is at least as important as how often they do them" (1982:240). Pondy, too, is concerned that viewing leadership "scientifically" leads to a focus on the superficial: Sounds, actions, and surface "expressions are observable; they constitute behavior that can be "scientifically" measured in reproducible ways. But deep structures, and especially meanings, are elusive concepts that have no physical, behavioral counterpart. They cannot be observed. But if leadership is to be studied scientifically, attention must ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 129 therefore be limited to the observable, surface, stylistic components (1978:89-90). Finally, in a recent review of leadership research, Immegart acknowledges certain misgivings about quantitative scientific research in this area: A nagging concern throughout the course of this review activity was whether scientific inquiry itself was getting in the way of what is known about leadership. Although research has contributed to understanding . . . it is also possible that normal processes of inquiry have unduly delimited and restricted what has been investigated. . . . Perhaps what has been studied has been explored in part because of convenience or because it has been more readily operationalized and measured. If so, other things, in spite of their import, may have been neglected or avoided because they are not as amenable to scientific inquiry and are, in fact, more subjective and elusive (1988:269). Viewing leadership as a descriptive term has more serious consequences than diverting our attention away from the more significant aspects of leadership. As Bernstein points out, "When we examine those empirical theories that have been advanced, we discover again and again that they are not value-neutral, but reflect deep ideological biases and secret controversial value positions" (1976:228). Let us consider how this is true of OERPL. b. Failure to justify ends or means In OERPL, and indeed, in much other administrative theory and research, the basic focus is on processes. Because determining what is valuable is thought to be beyond the reach of science, the problem which motivates OERPL research is reduced to one of finding the most effective or efficient means of achieving organizational goals; assessing the legitimacy of either the goal or the means is ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 130 beyond their purview. Many writers have noted this phenomenon. Maclntyre, for example, argues: Managers themselves and most writers about management conceive of themselves as morally neutral characters whose skills enable them to devise the most efficient means of achieving whatever is proposed. Whether a given manager is effective or not is on the dominant view a different question from that of the morality of the ends which his effectiveness serves or fails to serve (1981:71). Miklos makes a similar point: [W]e tend to treat administrative theory as being in large part morally neutral and to equate what is effective in administrative action with being good (1977-78:2). And Enns characterizes the administrative outlook as one in which: the controlling criteria [a]re effectiveness and efficiency in achieving the stated goals, rather than the Tightness or wrongness of either the goals or the means employed to achieve them (1981:3). Science can tell us only which of some set of alternative means is the most effective and efficient for achieving a given end but it cannot tell us which ends are good, or which means are morally acceptable. Accordingly, we should expect OERPL research to focus on how leaders can most effectively and efficiently achieve the results they want. However, by limiting leadership to the effective and efficient attainment of goals, this research encourages persons to believe that concern for the worthwhileness of goals is beyond the purview of leadership, i.e., not something that need concern leaders. Not only is the "goal" initiated by someone other than L in this research, but it usually consists of some task to be completed. We have noted that leadership often involves the ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 131 initiation of worthwhile goals, especially in education. This kind of research does nothing to illuminate this aspect of leadership. OERPL research focusses on the effectiveness and efficiency of means to the exclusion of their morality. L's means may include both L's form of influence and L's plan of action in pursuit of some end. OERPL researchers tend to study the "effectiveness" of any form of influence. Yukl, for example, reviews eleven forms of influence ranging from physical coercion to rational persuasion, and concludes that, "Under the right conditions, a leader may use any of the 11 forms of influence" (1971:21) (though, to be fair, he acknowledges that in most organizations, certain forms of influence may be prohibited). The implicit message emanating from this type of research is that one should select whatever means of influence is most effective in each situation. Trait research is particularly indeterminate with regard to means. In correlating traits with certain outcomes, no indication is given of L's means of influence. Yet we have seen that certain forms of influence are outside the concept of leadership. Similarly, what L gets FF to do in pursuit of a given end has no restrictions in OERPL research. Effective or efficient achievement of results is apparently the only criterion of desirability. It could be argued that leadership research itself is morally neutral — that it is only the purposes for which we use such research that can be properly regarded as moral or immoral. This is true, to the extent that leadership researchers are saying, "Given that you want compliance in situation X, this is the best means for achieving it." But to the extent that they purport to be ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 132 studying leadership, a positive value concept, we can legitimately criticize their conception as lacking important moral dimensions. OERPL's tendency to measure leadership effectiveness solely or chiefly in terms of productivity highlights its inappropriateness for the educational context in terms of both ends and means. First, in a profit oriented enterprise, such as business, it may be acceptable to assume that productivity is the most important result, but in education, the goals are often vague and indeterminate.t In fact, a central problem for the educational leader is to develop a comprehensive view of education including the specific nature of the goals which should be pursued. And there is no good reason to assume that what will be effective with regard to business productivity will also be effective with regard to responsible educational goals. Second, it is not even clear how one could measure the product of education, even if productivity were a well accepted goal. Third, in focussing on productivity, unintended consequences of L's actions are overlooked. For example, L might make short term gains in productivity at the expense of workers' safety or of maintaining equipment. In education, gains in standardized achievement scores might be noted while a resultant increase in student anxiety might be overlooked. And those consequences which are intended, but not concerned with productivity, may be given short shrift. For example, a basketball coach may not tFor example, consider just two of the goals of the Vancouver School Board. To create an environment in which students will achieve: • the ability to comprehend and to communicate effectively for social, scholastic, business, professional, and vocational purposes • a positive self-concept. No one would dispute the worthwhileness of these goals, but translating such goals into specific knowledge, abilities, and dispositions, and devising a unified plan of action to achieve them is not a straightforward matter, but requires educational leadership. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 133 get recognition for helping team members to increase their sense of fair play and their self-respect, especially when his team loses a few more games than last season. Surely in education such consequences must be viewed as important. Fourth, in assessing leadership through productivity, L's means are left out of account. In one study, for example, Fiedler (1971) measured the leadership effectiveness of coaches by the number of games their teams won in a season. This tells us nothing about whether the leader encouraged cheating, or whether he coerced or bribed the team members. In educational settings, the morality of the means is very important. Moreover, if one purpose of education is to enable children and youth to make responsible value choices, the success of such endeavours may depend in part upon setting an example. Finally, it is clear that not all outcomes can be attributed to the leader. In sum, educational leadership should not be judged primarily in terms of outcomes. This is not to say that L should never be judged in terms of attaining measurable goals, but that this is only one of many standards, and that emphasis on this standard may lead to the neglect of other important standards. Though morally admirable, the less frequently used indicator of leadership effectiveness, worker satisfaction, may also be unsatisfactory in terms of assessing the exercise of leadership. Followers may be happy and complacent, even when the plans of action initiated by L are highly inappropriate. Some of the religious cults would be a good case in point. Conversely, L may be exercising leadership even when FF are not satisfied. In cases where L does "uplift" FF, FF may be ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 134 spurred to action by a kind of dissonance or dissatisfaction with the state of things. Or, FF might be following L rather reluctantly, even though they feel they are doing the right thing. In such cases, FF's morale or satisfaction might be quite low. Thus, L can be exercising leadership even when FF are dissatisfied, or failing to exercise leadership even when FF are satisfied. The view of leadership underlying OERPL is not unlike the Congruence with Organizational Values conception outlined in the previous chapter. However, there are certain differences worth noting. Whereas OERPL has no built-in safeguards as to the ends or means of the exercise of leadership, COV requires that L act within the parameters of the organization's values, which are typically a derivation of societal values. Those working in OERPL may actually be building such restrictions into their research, but they are failing to make them explicit. This leaves their work open to the interpretation that there are no behaviours on the part of L which would be unacceptable, and no limits on the kinds of goals L might legitimately pursue. c. View of persons OERPL theory and research fail to take into account two important features of exercising leadership as outlined in Chapter 2: FF's reasons for following L (The Commitment Condition), and FF's autonomy in following L (The Autonomy Condition). In ignoring FF's reasons for following L, such theory and research are vulnerable to the charge that they harbour an implicit view of persons as things to be manipulated in order to achieve desired behaviour, what Burns calls ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 135 "the most pernicious and inhuman concept of all" (1978:446). I think it would be fair to say that those engaged in OERPL theory and research do not view themselves as Machiavellian or think of persons as beings to be manipulated. But the very nature of their research — finding out what L can do to make FF successfully pursue tasks or goals — contextually implies this view of persons. If OERPL researchers had considered FF's reasons for following L to be important, it is unlikely that they would have systematically ignored this dimension of leadership. d. Summary OERPL has failed to take into account almost all of the conditions associated with exercising leadership even in the weak-normative sense. First, in many studies, the goal is a given task rather than a non-trivial, shared goal (a violation of the Direction Condition). Second, no account is taken in these studies of why FF pursue the "goal" (a violation of the Commitment Condition). Third, there are often no built-in restrictions on what kind of influence L may use. In cases where leadership is measured solely in terms of productivity, L's use of deceit, coercion, etc., would not be ruled out (a violation of the Autonomy Condition). Finally, although OERPL researchers probably think that whatever they are studying is a good thing, it is not part of their mandate to study more than the descriptive aspects of leadership; they cannot explore the standards used in ascribing leadership in the strong-normative sense — appropriate ends and means. As Foster sums it up, "[T]he major charge that might be leveled against these various theories of leadership is that they do not ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 136 address leadership at all" (1986:176). Thus, it seems that most OERPL research is of only peripheral relevance to leadership as ordinarily understood; a fortiori it has only peripheral relevance to anything that would be recognized as educational leadership by the democratic educator. < Given the analysis of leadership presented in the second chapter, there is little reason to regard OERPL researchers as actually studying leadership, but rather as investigating task completion, or productivity, or worker satisfaction. These studies might have some bearing on administration, and even leadership, but this would have to be determined by careful analysis of the individual studies. In sum, avenues of inquiry which might prove to be fruitful are blocked by the OERPL hard core of assumptions about the nature of leadership and its study. OERPL can ask only what leader traits, behaviours, styles, or means of influence correlate best with given outcomes. It cannot ask whether these outcomes are worthwhile, whether the means of influence are legitimate, or why FF are pursuing the goals, precisely the kinds of questions with which we, as educators, should be concerned. Thus, we should probably feel gratified that OERPL has failed in its quest to find the secret of getting people to pursue whatever goal one wants — and like it. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 137 D. OTHER LEADERSHIP THEORY AND RESEARCH In this section, I shall consider social exchange theory, Vroom's and Yetton's (1973) decision-making model, House's (1977) theory of charismatic leadership and Kerr's and Jermier's (1978) idea of substitutes for leadership, with regard to their relevance to educational leadership. 1. Social Exchange Theory Social exchange theory is concerned with the reciprocal nature of influence between leaders and followers, and seeks to identify and explain the kinds of exchanges made by leaders and followers over time which contribute to the maintenance of their relationship. Hollander (1979), Burns (1978), Dansereau, Graen and Haga (1975), and Jacobs (1974) have developed versions of social exchange theory. Hollander (1979) views the phenomenon of leadership as an exchange relationship between leaders and followers. In return for a contribution involving the leader's unique control over scarce resources, access to vital information, and/or skill and expertise in dealing with critical task problems, he or she is accorded higher status, increased influence, and freedom to deviate from non-essential norms and traditions. Hollander contends that members who demonstrate competence and loyalty to the group accumulate credits that contribute to their relative status in the ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 138 group and allow them to deviate from group norms, with impunity. If, on the one hand, a leader's innovative proposal proves to be successful, he or she will be accorded greater status and influence; if, on the other hand, the leader's proposal is implemented and fails, then the terms of the exchange relationship will be reassessed according to the extent of the failure and the original status of the leader. A very high status leader will lose more status than a lower status leader, because more was expected of him or her. According to Burns' theory of transactional leadership (which he differentiates from the more valuational "transformational" leadership), leadership is an exchange relationship in which leaders approach followers with an eye to exchanging one thing for another. Although a transactional leader must understand and appeal to FF's motivations, his or her goals may be totally independent of FF's. In transactional leadership, the object is "a bargain to aid the individual interests of persons or groups going their separate ways" (1978:425). One example of this type of leadership which is cited by Burns is the Dutch colonists who gave beads to the native Indians in exchange for real estate. Dansereau, Graen, and Haga (1975) have developed a social exchange theory which they call the Vertical Dyad Linkage Theory. This theory holds that L usually develops a special exchange relationship with a small number of subordinates who assist and advise L. In exchange for their loyalty, commitment, and sharing of administrative tasks, these "in-group" FF are rewarded with such outcomes as assignment to interesting tasks, pay increases, and support and ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 139 approval by L. These theories may be of some use to persons concerned with exercising leadership in that they identify possible means of influencing persons. However, they do not address our primary concern which is influencing persons to achieve worthwhile goals through morally acceptable means. Jacobs' (1974) theory incorporates many important features of leadership which the other social exchange theories do not. He conceptualizes leadership in terms of an interaction between persons such that the influence target becomes convinced that his or her outcomes (benefits/cost) ratio will be improved if he or she behaves as the influence initiator desires. It produces behaviour that is self-motivating, and which therefore requires no supervision for compliance. Because Jacobs views the essence of leadership as the development of a new state of knowledge, belief, or attitude in the influence target, he believes that communication skills are important to a leader. For Jacobs, FF do not follow L merely for social approval; they follow L also because they have the conviction that compliance will produce intrinsic benefits; that is, the act desired by the leader is really best for the group. If FF do not believe this, then they have the option not to follow L without fear of reprisal. Underlying Jacobs' theory appears to be what I have termed weak-normative leadership. In general, social exchange theory does not belong to OERPL because it tends to be only indirectly concerned with organizational effectiveness. Promoting organizational effectiveness is only of concern to such theory to the extent that ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 140 it allows L to stay in power. According to the theory, leadership is occurring so long as both L and FF are satisfied. But although the colonists and the native Indians may have been happy with the deal they struck in Burns' example, we would normally call such an exchange bartering, not leadership. Leadership is not necessarily occurring if FF are satisfied with L. In terms of fostering the exercise of educational leadership, social exchange theory is unlikely to be useful. First, such theory may allow L and FF to have different goals. Second, it appears not to place limits on L's means of influence. Finally, FF's approval is viewed as sufficient for ascribing leadership to L. Our analysis showed that the requisite sense of leadership requires the fulfilling of standards concerning L's ends and means. 2. Vroom's and Yetton's Decision-making Model Vroom and Yetton (1973) focus upon delegation of authority in decision making in their contingency model. The decision procedures include solving a problem oneself, getting information from subordinates before making a decision, sharing the problems with only those subordinates to whom the decision is relevant before deciding, sharing the problems with one's subordinates as a group before deciding, and acting as a chairperson, attempting to reach consensus. Vroom and Yetton attempt to understand how a leader's decision procedure affects both decision quality and subordinate acceptance of that decision. Their concern with organizational effectiveness seems to be secondary to their concern ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 141 with the quality of decision making, so I have not included their theory under the OERPL tradition. They have developed a decision tree which maps out the "best" procedure to use given various circumstances such as time available to make the decision, adequacy of available information, and importance of subordinate acceptance of the decision. Information on the psychological aspects of decision making can be extremely useful in terms of helping leaders to improve the quality of their decisions. The ability to make defensible decisions is clearly important to the exercise of leadership. However, one could make good decisions without exercising leadership and conversely, one could exercise (weak-normative) leadership without making good decisions. It is not really a theory about "leadership" as characterized here, but about "decision making." 3. House's Theory of Charismatic Leadership Although House's (1977) theory of charismatic leadership is a trait theory of sorts, it does not belong to the OERPL tradition, as it does not attempt to correlate the presence of various charismatic traits of leaders with measures of organizational effectiveness. It consists of a number of propositions which identify charismatic leader traits and behaviours. Among them are the claims that charismatic leaders are likely to have high self-confidence, a strong conviction in their own beliefs and ideals, and a strong need to influence people; that they are likely to articulate goals for subordinates and to define follower roles in ideological terms that will appeal to them; that they are likely to use role ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 142 modeling; and that they are likely to behave in ways that arouse motives relevant to the accomplishment of a group's mission. House's theory appears to be as much an account of what charisma means as it is an empirically testable theory. It does, however, raise the interesting and important issue of the extent to which L's example plays a part in FF's acceptance of the goals. Because charismatic influence may be used concomitantly with rational persuasion in educational leadership, it would be useful to determine whether administrators who exemplify the characteristics of an educated person, as explicated here, would be more likely than those who do not exemplify these characteristics, to influence FF to pursue educational ends. 4. Kerr and Jermier Although Kerr's and Jermier's (1978) model might be construed as a variant of OERPL, it seems to lack the hard core which views leadership as something about L which affects organizational effectiveness; rather, it views leadership in terms of functions. Their model identifies aspects of the situation that nullify the influence of a leader's traits or behaviours. Kerr and Jermier distinguish between supportive and instrumental leadership, which correspond roughly to consideration and initiating structure. In certain situations, leadership by an administrator is unnecessary because the leadership role is being taken over by "substitutes," i.e., characteristics of the subordinates, the task, or the organization which make relationship or task leadership behaviour on the part of the administrator unnecessary or redundant. For example, FF may have the ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 143 training and ability to determine and reach their own goals; the task may be very unambiguous or satisfying and not require L to devise a means to complete it; and the organization may have a cohesive work group. Other situations may counteract the effects of L's behaviour or prevent L from acting. For example, if FF are indifferent toward rewards or if the organization has extremely rigid rules and procedures, L may have little control. These aspects of the situation Kerr and Jermier call "neutralizers." This theory is interesting because, by outlining the conditions where leadership is not necessary, Kerr and Jermier make assumptions about the functions of leadership, i.e., determining goals, and supervising and mobilizing FF to reach the goals. 5. Summary Although pockets of non-OERPL research might have some application in promoting our understanding of educational leadership, it would appear that none would be more than peripherally relevant. Like OERPL research, such research tends not to take into account the normative nature of leadership. E. GENERALIZABILITY OF RESEARCH Much of the theory and research to which educational administrators are typically exposed is concerned with leadership in a non-educational context. Thus, it might be useful to consider the extent to which non-educational leadership theory or research may be relevant to the educational context. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 144 This question has been pondered by many educational scholars, as part of a larger debate about the generalizability of theories. Griffiths notes that numerous discussions of this issue "have not affected the widespread practice of borrowing theories, concepts, and research regardless of the source" (1975:13). He claims that there must be a reasonable degree of isomorphism between organizations before research can be legitimately generalized from one to another. Griffiths endorses Katzell's (1962) criteria for determining isormorphism, which he summarizes as follows: 1. Size 2. Degree of interaction and dependence 3. Personalities of organizational members 4. Degree of congruence between organizational goals and goals and needs of members 5. Who has ability and motivation to take action to further organization's objectives (Griffiths, 1975:13). These criteria do not seem to me to represent what must be relevantly similar between organizations before we could legitimately generalize research findings from one to the other. It seems likely that even if all these criteria were met, we would have no good grounds for generalizing the findings. Given the conception of educational leadership outlined here, it seems reasonable to suppose that the goals and means of influence between two organizations should be compatible before we could expect leadership research findings to generalize from one organization to the other. Of course, no other ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 145 institution shares the specific goals of education, but some institutions in the public sector, such as health centres, may share the general goal of helping persons to lead productive lives. And certain studies of other kinds of leadership, such as political leadership, may be based on a conception of leadership in which L respects the rational autonomy of persons. Because leadership theory and research from the business context are most often assumed to be useful for educators, I shall consider the likelihood that such theory and research would be relevant to the exercise of leadership in the educational context. The appropriateness of extrapolating business theory and research for the educational context has been questioned by a number of scholars. Callahan (1962) has documented the fervour with which business values were embraced in education from 1900 to 1930. He argues that this was a serious mistake because it undermined the integrity of the educational enterprise at its core, turning the superintendents into business managers who viewed teachers as workers rather than as professionals. Campbell (1955), too, has argued that the framework of the manufacturing plant, with its operative value of efficiency and its view of persons as cogs in the plan of mechanized production! is inappropriate in the educational context. The National Education Association (1933) argued that there are important tSome authors have begun to question the suitabilitjT of these values even for business or industry, writing on such topics as employee rights, reverse discrimination, and whistle blowing (Beauchamp and Bowie, 1983). ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 146 differences between industry and education, although educational organizations have tended to be patterned on industry. One important difference they underscore is that the educational purpose is not profit, but the integrated personality of human beings. Thus, the "product" is not standardized, but infinitely variable, and the techniques cannot be determined in advance. From these observations, it would seem that theory and research conducted in the context of business and industry cannot be assumed to be relevant to educational leadership.! Although the issue awaits empirical confirmation, given the uniqueness of the educational enterprise, it seems unlikely that leadership theory generated from studies of business organizations can provide much guidance in fostering the exercise of educational leadership, even to the limited extent of telling us which morally acceptable means of influencing persons is likely to be effective in securing responsible pursuit of an educational goal. F. CONCLUSION Our survey of leadership theory and research reveals that leadership in the service of genuinely educational goals has rarely been studied. At best, most of the leadership theory and research are only peripherally related to the exercise of leadership as conceived here and, at worst, they appear to imply by their omissions a morally suspect view of persons. Because so little extant leadership tit must be emphasized that I am referring only to leadership research and not to research on management. The field of educational administration shares many problems with business and industry which may well benefit from research in these areas. ANALYSIS OF LEADERSHIP THEORY AND RESEARCH / 147 theory and research appears to be demonstrably relevant to the fostering of educational leadership, the practice of making the study of such theory and research a major component of preparation programs for educational administrators, at least for the purpose of fostering the exercise of leadership, should be seriously reconsidered. In the final chapter, I shall summarize the thesis and point out its implications for the selection and preparation of educational administrators as well as for research. CHAPTER V. WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS One of the most universal cravings of our time is a hunger for compelling and creative leadership. James MacGregor Burns The problem that has motivated this thesis is the extent to which the study of literature concerning "organizational leadership" is likely to be efficacious in enhancing the ability of educational administrators to exercise educational leadership. The main conclusion it reaches is that such study is of questionable relevance, and may even be deleterious in that it encourages a narrow and distorted view of the requirements of educational leadership. This chapter will summarize the major features of the argument supporting this conclusion, and elaborate the implications of the argument for (1) the preparation and selection of educational administrators, (2) understanding the difficulties bedeviling organizational leadership research, and (3) conducting research that could be fruitful for educational administrators seeking to exercise educational leadership. A. SUMMARY OF THE THESIS This thesis has two central purposes. The first is to articulate what it means to exercise leadership, and how we might usefully conceive of educational leadership, and the second is to explore the relevance of extant leadership theory and research to fostering educational leadership. I began this inquiry with the 148 WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 149 assumption that the most useful conception of leadership would capture what competent language users mean when they use the word "leadership" (i.e., the publicly shared meaning of leadership) because interest in leadership arose in the context of real life concerns rather than in the context of social science theorizing. Through conceptual analysis, I distinguished two uses of "leadership" in ordinary language: a value-neutral usage having to do with positions of authority or power, which I called position leadership; and a normative sense which has the function of expressing a positive appraisal of a leader's directing others in the pursuit of shared goals. According to my analysis, there are six necessary and jointly sufficient conditions of exercising (normative) leadership: 1. The Direction Condition: L provides direction for FF in pursuit of a non-trivial, shared goal. 2. The Adherence Condition: FF pursue the goal. 3. The Influence Condition: L influences FF to pursue the goal. 4. The Commitment Condition: FF pursue the goal, at least in part, because they regard it as desirable or worthwhile. 5. The Autonomy Condition: FF have autonomy in determining whether or not to follow L. 6. The Merit Condition: L's actions merit approval. I argued that the standards by which persons assess whether or not leadership has been exercised fall into roughly two categories: those having to do with the extent to which the leader is able to mobilize others to pursue a shared goal, and those having to do with the extent to which the goals and means are appropriate. I called these weak-normative and strong-normative WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 150 leadership, respectively. I argued further that strong-normative leadership is what people typically have in mind when they call for leadership on the part of educational administrators. Analyzing the meaning of the term "leadership" enabled us to see what is common to the exercise of leadership in various situations, but could not tell us specifically the kinds of actions which might merit approval in the educational context and which might not. Thus, in the third chapter, I outlined a conception of educational leadership based upon the basic tenets of a liberal democracy and upon the standards implicit in "educating" persons in such a context. This conception included moral and epistemological standards which L's actions must meet if he or she is to count as exercising educational leadership. I argued that this conception is preferable to the plausible alternatives. In the fourth chapter, I explored organizational leadership research and theory to which aspiring educational administrators are typically exposed, in order to determine the extent to which the conceptions of leadership implicit in them would be compatible with the conception of educational leadership developed earlier. Viewing the various approaches from the perspective of Lakatos' (1970) theory of scientific research programs, I argued that the vast majority of leadership theory and research does not take seriously the idea that leadership requires influencing persons to pursue worthwhile goals or plans of action by defensible means. In fact, most leadership theory and research focusses on identifying effective and efficient (but not necessarily moral) means for getting people to pursue any sort of goal or plan of action. I suggested that this WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 151 conception was inadequate for helping educational administrators to exercise educational leadership, because it is not clear that these findings can be generalized across contexts. B. IMPLICATIONS OF THE THESIS The general conclusions of this thesis are as follows: 1. "Leadership" is used in ordinary language in both a position sense and a normative sense. The standards used for ascribing the exercise of leadership in the normative sense range from weak to strong. 2. The most relevant sense of leadership to the exercise of educational leadership is strong-normative leadership. 3. The conception of educational leadership developed in this thesis is preferable to the available alternatives. 4. Given the conception of educational leadership developed and defended in this thesis, study of current leadership theory and research is likely to be of very little use in the fostering of the exercise of educational leadership. McClellan and Komisar (1962) maintain that those who analyze the conceptual underpinnings of educational practice have an occupational mandate to culminate their work in recommendations. I shall attempt to fulfill this mandate by setting out what I take to be the implications of this thesis for the preparation and selection of educational administrators and for research. WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 152 1. Implications for the Preparation of Educational Administrators It has been argued by many that our educational system reflects a paucity of leadership. I suggest that, were our preparation programs to make some fundamental changes, many of those educational administrators currently not exercising leadership would be more likely to do so. In this section, I shall point out (1) what I take to be the main reason that current programs are failing, and (2) what would seem to be more useful ways of preparing educational administrators to exercise educational leadership. a. The problem with current programs I have maintained that viewing leadership as a descriptive concept has led to disillusionment and lack of progress in leadership research. Such a view seems also to have led many to believe that teaching administrators to exercise educational leadership is a matter of teaching them skills and techniques. The problem is not that these skills are useless in the exercise of leadership but, rather, that they are emphasized at the expense of examining educational values and aims. If I am correct that the strong sense of leadership, which focusses attention on defensible ends and means, is central to educational leadership, then it may be vacuous to teach leadership skills without ensuring that prospective educational administrators have some grounding in educational issues. Although others do not argue this point on the basis of a conception of educational leadership, they sense that current training programs are inadequate.! tA survey of forty-eight departments of educational administration conducted in WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 153 Sergiovanni, for example, expresses reservations about programs which downplay substance and emphasize skills: I wouldn't want to have to choose between the two emphases, but one consequence nevertheless of focusing on the managerial or behavioral aspects of leadership is that the substance of leadership decisions can be slighted. Leadership skills are important, but they cannot bring genuine leadership if the leader does not have a sense of purpose and direction (1979:393). Eisner shares this concern: One might hope that schools of education that prepare school administrators would provide the kind of professional education that would enable them to think critically about the virtues toward which education aims. One might hope that such people would be encouraged to think deeply about the aims of education and to provide leadership and educational services to the community on whose support the schools depend. Unfortunately, as schools become industrialized, the training programs for administrators focus more and more on the development of skills of labour negotiation and on courses offered in business schools, departments of economics, and the like. Such courses might have utility for some aspects of educational administration, but they are essentially technical studies. Embedded within technique are implicit visions of what is important, and these visions are seldom appraised by criteria emanating from a conception of education itself (1979:14). And Burns fears that such training may promote the view that it is acceptable to manipulate persons in various ways in order to achieve desired ends: Machiavelli has had countless imitators. The vogue of "how to" manuals still thrives today . . . . While few of them emulate the master in offering Machiavellian advice on how to coerce, control, or deceive persons, many do seek to train persons to manage and manipulate other persons rather than to lead them . . . . Worse, the manuals treat persons as things, as tools to be used or objects to be stormed like a castle (1978:446). t(cont'd) 1978-79 by Farquhar (1981) gives some empirical backing to their intuitions. Of eighteen returns, only four universities include courses or programs focussing on ethics in their educational administration preparation programs. WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 154 These misgivings about our current programs are understandable, given my conception of educational leadership. b. What can be done? If professors of educational administration were to take seriously the conception of educational leadership presented in this thesis, they would probably revise their programs, encouraging students to take more courses dealing with educational concepts and value issues. Such a change would probably not be met with enthusiasm by students, because these matters are complex and it is time-consuming to study them. However, my analysis leads to the same conclusion that Hodgkinson reaches by a different route, that "No matter how great the complexity may be, leadership can never be understood unless the problem of value is incorporated into its study" (1983:190-91). Although Hodgkinson's (1978) value model is conceptually problematic,! I believe he is correct in emphasizing that administration is essentially a philosophical endeavour concerned with the making of defensible value judgments. tin the value model which Hodgkinson proposes, there are three general types of values. Type III values are based on personal predilections and preferences; Type II values are grounded in consensus and concurrence with the will of the majority, or alternatively, on a reasoned analysis of consequences; and Type I values "whether they derive from a postulated moral insight, an asserted religious revelation, or an aesthetic sense of individual drama" (1978:113) cannot be justified by logical argument. According to Hodgkinson, Type I values are superior to the others. What his thesis boils down to is that values held strongly, on whatever grounds, are superior to those held less fervently. I would take issue with this analysis, arguing that the strength with which one holds values is not what makes them better or more justified; it is, rather, the defensibility of the reasons for which they are held. WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 155 The value of philosophy may not always be readily apparent, but it can affect one's broad educational perspective and, ultimately, the way in which one exercises leadership: It would be silly to contend that in the day-to-day business of administration . . . problems are solved in the light of conscious theoretical principles. But it would be equally foolish to deny that the discussion of such principles has a real effect upon the criteria by which courses of action are judged, and it is probable that at moments when really great decisions are being considered the lessons of Aristotle or Burke or T.H. Green may quite consciously be of value (James, 1951:76). Taking courses in philosophy of education and discussing value issues with knowledgeable professors would be promising avenues for those aspiring to exercise educational leadership, but we should not rule out the possibility that important lessons can be learned through working with those who exemplify educational leadership, or those who set an example in terms of important leadership abilities, such as value reasoning. As Dunsheath argues: The value of the example set by a university teacher is insufficiently appreciated. There is a far greater opportunity of influencing a student in the characteristics of good leadership by the intangible things in personal conduct than by any amount of specific instruction (1948, as cited in James, 1951:24). Indeed, Livingstone maintains that educational leadership is best epitomized in a universitj' setting: Personal magnetism counts for little or nothing, and those who possess it usually seek more exciting and crowded scenes for the exercise of their gifts than the humdrum roads of education. There is no element of crowd psychology to help. Further, the scholar rarely desires to be a leader; if he did, he might lose something of his integrity and proper virtue. He is naturally and proverbially aloof from the world, a man pursuing his bent or his vision for its own sake. Were he WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 156 otherwise, there is no compulsion to follow him. Those who follow, do so of their own free will, drawn on by some inner urge (1950:23-24). To the extent that the ability to exercise leadership is acquired through role-modeling rather than through textbook study, internships with carefully chosen educational leaders might be useful. We might also explore the usefulness of other alternatives to textbooks, such as biographies of those who exercised leadership, or literary works in which persons exercised leadership, because they might furnish exemplars of the exercise of leadership more than the "current leadership literature. This possibility is echoed by Immegart (1988), who concludes that biographical material on political figures such as Winston Churchill and Lyndon Johnson often speak to the topic of leadership more directly than books about leadership. One of the primary tasks of this thesis was to explore the relevance of leadership theory and research to fostering the exercise of educational leadership. My conclusion at the end of the previous chapter was that leadership theory and research is of questionable relevance for educational administration. I suggested that we needed to reconsider seriously the practice of exposing educational administrators to this theory and research, especially when this is done uncritically. Exposing educational administrators to a wide range of leadership theories and models is unlikely to have the effect of improving their understanding of educational leadership, as conceptualized here, or of increasing their ability to exercise it. In fact, such exposure is more likely to have a detrimental effect, by leaving prospective administrators with the impression that acquiring leadership ability is a matter of picking up a few human relations WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 157 skills. The problem, of course, lies with the nature of the leadership theory and research rather than with the study of it per se. Were we to have a core of theories and studies directly relevant to educational leadership appropriately conceived, it is possible that studying such theory and research could be fruitful in fostering educational leadership. It is important to note, however, that the relevant theories will be normative theories, and the relevant empirical research will be determined by the values built into the theories. Finally, aspiring administrators should reflect upon the strengths and weaknesses of the competing conceptions of educational leadership, and understand the implications of adopting a particular conception. This would provide them with tools for critically assessing leadership theory and research. 2. Implications for the Selection of Educational Administrators The arguments of this thesis have certain implications for the selection of educational administrators. The first implication concerns the kinds of knowledge, abilities, and dispositions which we should look for in selecting educational administrators. The second has to do with gender equality in selecting for administrative positions. a. Criteria of selection If educational leadership involves L's directing FF toward educational ends by means which respect their rational autonomy, then what kinds of knowledge, WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 158 value commitments, abilities, and dispositions would seem to be germane to selecting educational administrators who can exercise educational leadership? The knowledge most needed by prospective educational leaders is knowledge of the nature of those goals which are fundamental to realizing the ideals of education in our democratic society. These include such educational goals as autonomy, critical thinking, creativity, and moral character. Although these goals are generally taken to be important, just what they entail is often unclear or poorly understood. Prospective leaders must have the kind of understanding that enables them to develop and communicate defensible conceptions of these goals. They must, for example, have answers to questions such as the following: Is creativity best conceived of as a general ability or as context-specific? Is it best conceived of as a distinct process of thought or as a distinct kind of achievement? Is developing moral character best conceived of as inculcating the moral values of one's society, as helping persons to clarify their values, or as fostering the development of justifiable moral principles? In addition to understanding educational goals, prospective leaders will need operational understanding of the fundamental moral commitments of our society. Educational administrators will work in a milieu mined with moral hazards. Unless they have reflected on such moral issues as equality of educational opportunity, adjudication of interests in a pluralistic society, indoctrination, education of the "gifted" and the handicapped, confidentiality of information, corporal punishment, and student rights, they are unlikely to be able to provide FF with appropriate direction. Moreover, simply understanding such issues is not WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 159 enough; they need to understand the value of realizing our fundamental commitments in our educational practices. Knowledge of psychological theories of learning and development, such as those of Piaget (1952) and Kohlberg (1984), are also germane to the exercise of educational leadership. As long as administrators have due regard for the limitations of such theories, the theories could be useful in the development of realistic educational programs. Finally, knowledge of current research on teaching and learning, as well as on effective schools, might be useful in formulating ways of achieving goals. Abilities important to the exercise of educational leadership include the ability to mount and articulate rationally persuasive arguments for a particular goal or course of action, the ability to coordinate the talents of followers in order to realize a goal most effectively, and the ability to employ careful practical reasoning, particularly value reasoning, in resolving problems and in making decisions. Such abilities are not entirely generalizable across contexts. In order to choose defensible goals in a particular context, one needs to understand the context, and in order to influence persons rationally, one needs to have some understanding of their background knowledge and commitments. In terms of dispositions, the exercise of educational leadership would require the dispositions to respect persons, to conform to the democratic. ideals, and to use the aforementioned knowledge and abilities. WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 160 There is no easy way to assess these qualities in a prospective administrator. If we accept the conception of leadership outlined in this thesis, then sensitive, intelligent interviewers who themselves have insight into fundamental educational issues will be required. Asking candidates what they would do in various hypothetical situations, i.e., simulated decision-making, would seem to be one useful way of eliciting the extent to which important educational issues have been reflected upon critically. Since exercising leadership is only one aspect of the educational administrator's role, leadership ability should not be the sole criterion of selection. However, if the exercise of educational leadership is important to a particular position, such ability should be weighted heavily, for improving management techniques would seem to be easier than improving one's ability to exercise leadership. Moreover, it may be useful to consider carefully the extent to which the exercise of leadership is desirable in a particular position. Administrators who exercise leadership may have very different views from others in the chain of command. Sometimes a little "shaking up" can be healthy, but one should be aware of the possible dysfunctions of such a move. b. Leadership and power The analysis of "leadership" showed that it is conceptually independent of a power or authority relationship between L and FF, yet many researchers focus on power (what Yukl (1981) calls the "Power-influence approach") to the exclusion of other aspects of leadership. Since power tends to be unevenly WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 161 distributed between males and females, this emphasis on power may lead to the stereotype of leaders as male, with the consequence that women may be passed over in favour of men for leadership positions. Although Burns did not derive his conclusion from a conceptual analysis of the concept of leadership, he recognized the source of the problem: Male bias is reflected in the false conception of leadership as mere command or control. As leadership comes properly to be seen as a process of leaders engaging and mobilizing human needs and aspirations of followers, women will be more readily recognized as leaders (1978:50). Thus, focussing our attention on educational leadership as conceptualized in this thesis could lead to more women being chosen for administrative positions. 3. Implications for Research This thesis has implications both for resolving much of the confusion surrounding leadership research, and for future directions which leadership research might take. a. Resolving confusion A number of writers have puzzled over the fact that "leadership" is rarely studied in leadership research. Prentice (1982), for example, notes that our intuitive sense of the "essence" of leadership is not captured in research. According to Tosi: The irony that permeates the current established paradigm is that it WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 162 is charisma in which the field is interested yet it is ordinary managers, quite often in very low level jobs, who are studied (1982:6). I think that it is strong-normative leadership rather than charisma that interests the field, but Tosi hovers around an important point: the "leadership" we talk about and write about is not the "leadership" we study. Isolating the ways in which we use the term "leadership" helps to explain the puzzle. Basically, there are three considerations which could explain the failure of researchers to come to grips with the essence of leadership. First, they have conflated the position sense of leadership with both the weak-normative and strong-normative leadership. Since the position sense of leadership includes both the managerial and the leadership aspects of the administrative role, research based on this conflated notion of leadership will not differentiate between doing a good job as an administrator in general, and exercising leadership. The problem is not that this research is misguided; rather, it is that confusion results when persons assume that this research is about leadership in the normative sense. Second, because the promulgation of what is worthwhile is not normally in the purview of scientific investigation, leadership research has tended to focus only on the descriptive aspects of leadership. Since the strong-normative sense captures the essence of leadership, i.e., the sort of leadership everyone wants, it is to be expected that this research will be perceived as wide of the mark. Finally, the conception of leadership with which researchers have worked views the ability to exercise leadership as the ability to deploy discrete traits or skills which are efficacious in securing the pursuit of any sort of goal or action plan. This ignores the possibilitj' that efficacy in securing pursuit of a goal may depend on WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 163 knowing and being able to articulate why that particular goal is significant or worthwhile. If this is a real possibility, then what one knows about a particular context of action and the values to be realized by possible courses of action in that context will be more important to leadership than will skills which generalize over contexts. b. Future leadership research This thesis explains McCalPs and Lombardo's reluctant conclusion: Improvement of our understanding of leadership apparently does not lie in pursuing existing trends or in attempting to integrate existing research (1978:172). Or, as Foster expresses it, "For those looking at leadership through scientific lenses, the study has dead-ended" (1986:171). If I am correct that leadership is centrally a value concept, the prospects for the kinds of "scientific" research now being conducted are limited. For example, since determining the goal is an important part of leadership, it should no longer be viewed as sensible to study leadership in terms of getting groups to complete given tasks successfully (though it is possible that such studies might prove to have some bearing on leadership). Nor should it be viewed as sensible to measure the exercise of leadership solely in terms of productivity or FF's satisfaction. Because the exercise of leadership should reasonably be assessed by such things as the overall appropriateness of the goal or plan of action, intended and unintended consequences (where apparent) of pursuing it, the efficiency and morality of the means of WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 164 influencing persons to pursue the goal or plan of action, and the improvement of education over time, it is unlikely that there is much scope for quantitative research in this area. If, as I have argued, it is strong-normative leadership that is our central concern, and that leadership serves more as an ideal than as a descriptive, empirical concept, then empirical researchers will have to confine themselves to studying aspects of leadership delimited by our normative theories. For example, there would appear to be some scope for investigation of the sorts of rational influence most likely to invoke autonomous commitment on the part of FF. Since researchers are already doing work in the area of the leader's means of influence, this might be a good place to start. And, because rational persuasion involves the abilitj' to reason well, the study of pitfalls in practical reasoning, including psychological blocks to making good decisions (such as the work of Janis and Mann, 1977), might also be useful. If social scientists are able to become sensitive to moral concerns and to revoke their claims to value neutrality, it is possible that they might be able to do some useful work in the area of leadership. For example, if they were to base their inquiry on a defensible conception of leadership in a particular context, and study persons who exercise leadership in that context, then some of their findings regarding effectiveness and efficiency could inform practice. WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 165 Much conceptual work remains to be done in distinguishing among different contexts of leadership and building conceptions for each context. Although I disagree with Immegart's conclusion that the "continued advancement of the understanding of leadership is . . . contingent on sound empirical activity" (1988:275), I agree with his assessment that conceptualizations of leadership tend not to approach the full picture and that improved conceptualizations of leadership seem critical for further advancing inquiry and the understanding of leadership. The conception of educational leadership outlined in this thesis could be further refined, on the basis of continual critical discussion concerning the nature of worthwhile knowledge, the forms and limits of rational persuasion, and defensible educational goals. The new conception could serve as the basis of a new research program of leadership which would, I believe, hold greater promise of being progressive than OERPL. The conception of educational leadership conceptualized here would likely best be illuminated through qualitative, case study research. For example, Kaplan's (1985) descriptions of the "top" four educational leaders in the U.S. emphasize certain abilities which may be important to the exercise of leadership in some contexts. These include moral suasion, information gathering, network building, and legislative manipulation. Research could also be conducted on specific aspects of the exercise of leadership, such as the nature of the goals and means of influence of persons widely regarded as educational leaders. WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 166 Finally, I concur with Immegart that "some effort ought to be invested in adjudicating the quality and adequacy of investigations" (1988:272), rather than merely summarizing and collecting leadership research. Individual studies could be assessed, in part, according to their conceptions of leadership. It may be the case that some studies have taken into account at least FF's reasons for following L, and these might have relevance for educational leadership in some contexts. C. CONCLUSION In the first chapter, I speculated that professors of educational administration encourage the study of organizational leadership literature because they regard leadership as too important to ignore, despite the fact that they have no clear criteria for determining which theories and models are defensible in the context of educational administration and which are not. This thesis offers some fairly clear criteria for determining the relevance of organizational leadership theory and research studies. At the very least, it shows the need for presenting leadership theory and research in a more critical light. There is currently much talk about the importance of values in educational administration (e.g., Balderson, 1980; Miklos, 1977-78; Farquhar, 1981; and Enns, 1981). Dimock's (1958) plea to let the pendulum swing in the direction of the humanities rather than the sciences, has now become a groundswell. Fostering the ability to exercise leadership is one context in which the humanities assume considerable importance. Our conception allows us to be fairly specific in terms of WHERE DOES ALL THIS LEAD?: IMPLICATIONS OF THE THESIS / 167 how we might employ such subjects as ethics, social theory, and philosophy of education to understand and improve the exercise of educational leadership. This thesis is a call to recognize that preparing administrators for educational leadership is a defensible enterprise only to the extent that we are guided by a conception of educational leadership which embraces the values of education in a liberal democracy. Until we appreciate this point, we are unlikely to equip our prospective leaders with the insight they will need in order to exercise educational leadership. REFERENCES Allen, Richard K. 1977 Organizational Management through Communication. San Francisco: Harper and Row. Austin, J.L. 1961 A plea for excuses, pp. 123-52 in J.L. Austin Philosophical Papers. Oxford: Clarendon Press. Balderson, James H. 1980 Guidelines for ethical leadership. Challenge in Educational Administration: The CSA Bulletin, 19(3):4-5. Bales, R.F. 1950 A set of categories for the analysis of small group interaction. American Sociological Review, 15:257-63. Beauchamp, Tom L., and Norman E. Bowie (eds.) 1983 Ethical Theory and Business, (2nd ed.). Englewood Cliffs, N.J.: Prentice-Hall. Benn, Stanley I. 1967 Authority, pp.215-18 in Paul Edwards (ed.), Encyclopedia of Philosophy, vol.1. New York: MacMillan. Bennis, Warren 1976 The Unconscious Conspiracy: Why Leaders Can't Lead. New York: Amacom. 1959 Leadership theory and administrative behavior: The problem of authority. Administrative Science Quarterly, 4:259-260. Berger, Thomas R. 1982 Fragile Freedoms: Human Rights and Dissent in Canada, (rev. ed.). Toronto: Irwin. 168 / 169 Bernstein, R.J. 1976 The Restructuring of Social and Political Theory. New York: Harcourt Brace Jovanovich. Bowers, D.G. and S.E. Seashore 1966 Predicting organizational effectiveness with a four-factor theory of leadership. Administrative Science Quarterly, 11:238-63. Boyer, E.L. 1983 High School: A Report on Secondary Education in America. New York: Harper and Row. Bridges, Edwin M. 1975 The nature of leadership. Paper presented to the conference, Educational Administration, Twenty Years Later: 1959-74. Ohio State University. Burch, Barbara G. and W. Elzie Danley Sr. 1980 The instructional leadership role of central office supervisors. Educational Leadership, 37:636-37. Burns, James M. 1978 Leadership. New York: Harper and Row. Callahan, Raymond E. 1962 Education and the Cult of Efficiency: A Study of the Social Forces that Have Shaped the Administration of the Public Schools. Chicago: University of Chicago Press. Campbell, Clyde M. 1955 Human relations techniques useful in school administration. School Board Journal, 77:31-32. Campbell, John P. 1977 The cutting edge of leadership: An overview, pp.221-34 in J.G. Hunt and L.L. Larson (eds.), Leadership: The Cutting Edge. Carbondale: Southern Illinois University Press. / 170 Caswell, Hollis L. 1955 Inaugural address at Eisenhower's inauguration dinner, pp. 12-14 in K.D. Benne and W.O. Stanley (eds.), Essays for John Dewey's Ninetieth Birthday. Report of a Conference on Education and Philosophy at the University of Illinois. Coombs, Jerrold R. 1980 Discipline maintained, pp. 197-208 in D. Cochrane and M. Schiralli (eds.), Philosophy of Education: Canadian Perspectives. Don Mills, Ontario: Collier MacMillan. 1978 Concepts. Talk to Educational Graduate Students' Colloquium, University of British Columbia. Crittenden, Brian 1973 Education and Social Ideals. Don Mills, Ontario: Longman. Cronin, Thomas 1984 Thinking and learning about leadership. Presidential Studies Quarterly, 12:22,33-34. Culbertson, Jack A. 1981 Antecedents of the theory movement. Educational Administration Quarterly, 17:25-47. Dansereau, F.; G. Graen; and W.J. Haga 1975 A vertical dyad linkage approach to leadership within formal organizations — a longitudinal investigation of the role making process. Organizational Behavior and Human Performance, 13:46-78. Dewey, John 1974 Freedom, equality and intelligence, pp.460-66 in Charles M. Sherover (ed.), The Development of the Democratic Idea: Readings from Pericles to the Present, (rev. ed.). New York: New American Library. 1937 Democracy and educational administration. School and Society, 45(1162):457-62. / 171 1916 Democracy and Education: An Introduction to the Philosophy of Education. New York: Macmillan. Dimock, M. 1958 A Philosophy of Administration. New York: Harper. Dunsire, Andrew 1973 Administration: The Word and the Science. London: Martin Robertson. Eisner, Elliot W. 1979 The Educational Imagination: On the Design and Evaluation of School Programs. New York: Macmillan. Enns, Frederick 1961 Some ethical-moral concerns in administration. The Canadian Administrator, 20:1-8. Farquhar, Robin 1981 Preparing educational administrators for ethical practice. The Alberta Journal of Educational Research, 27(2): 192-204. Fiedler, Fred E. 1971 Leadership. General Learning Press: 1-24. and Joseph E. Garcia 1987 New Approaches to Effective Leadership: Cognitive Resources and Organizational Performance. Toronto: Wiley. Follett, Mary Parker 1940 Leader and expert, pp.247-69 in Henry C. Metcalf and L. Urwick (eds.), Dynamic Administration: The Collected Papers of Mary Parker Follett. New York: Harper and Row. Foster, William 1986 Paradigms and Promises: New Approaches to Educational Administration. New York: Prometheus Books. 1980 Administration and the crisis in legitimacy: A review of Habermasian thought. Harvard Education Review, 50(4):496-505. / 172 Frankena, William K. 1973 The concept of education today, pp. 19-32 in James F. Doyle (ed.), Educational Judgments: Papers in the Philosophy of Education. Boston: Routledge and Kegan Paul. Gallup, Alec 1985 The 17th annual Gallup poll of the public's attitudes toward the public schools. Phi Delta Kappan, 67:35-47. Gardner, J.W. 1965 The antileadership vaccine. Annual Report of the Carnegie Corporation of New York:3-12. Graff, Orin B. and Calvin M. Street 1957 Developing a value framework for educational administration, pp. 120-52 in Roald F. Campbell and Russell T. Gregg (eds.), Administrative Behavior in Education. New York: Harper. Griffiths, Daniel E. 1975 Some thoughts about theory in educational administration — 1975. UCEA Review, 17:12-18. Habermas, J. 1975 Legitimation Crisis, (T. McCarthy, trans.). Boston: Beacon Press. Haller, Emil J. and Kenneth A. Strike 1986 An Introduction to Educational Administration: Social, Legal, and Ethical Perspectives. New York: Longman. Halpin, Andrew W. 1966 Theory and Research in Educational Administration. New York: Macmillan. and Winer, B.J. 1957 A factorial study of the leader behavior descriptions, pp.39-51 in Ralph M. Stogdill and A.E. Coons (eds.), Leader Behavior: Its Description and Measurement. Columbus: Bureau of Business Research, Ohio State University. Handy, Charles B. 1976 Understanding Organizations. Markham, Ontario: Penguin. / 173 Hanson, E. Mark 1979 Educational Administration and Organizational Behavior. Toronto: Allyn and Bacon. Hare, Richard M. 1981 Moral Thinking: Its Levels, Method, and Point. Oxford: Clarendon Press. Hare, William 1986 Looking for leadership: Some false leads. Journal of Educational Administration and Foundations, l(2):52-55. Hart, H.L.A. 1955 Ascription of responsibilities and rights, pp. 145-67 in A. Flew (ed.), Logic and Language. Oxford: Blackwell. Heller, Trudy and John Van Til 1982 Leadership and followership: Some summary propositions. The Journal of Applied Behavioral Science, 18:405-14. Hemphill,. J.K. and A.E. Coons 1957 Development of the leader behavior description questionnaire, pp.6-38 in R.M. Stogdill and A.E. Coons (eds.), Leader Behavior: Its Description and Measurement. Columbus: Bureau of Business Research, Ohio State University. Hersey, Paul and Kenneth H. Blanchard 1976 Situational leadership. Center for Leadership Studies. Undated Situational leadership, perception, and the impact of power. Center for Leadership Studies. Hodgkinson, Christopher 1983 The Philosophy of Leadership. Oxford: Blackwell. 1978 Towards a Philosophy of Administration. London: Blackwell. / 174 Hollander, E.P. 1979 Leadership and social exchange processes, pp. 103-15 in K. Gergen, M.S. Greenberg, and R.H. Willis (eds.), Social Exchange: Advances in Theory and Research. New York: Winston-Wiley. House, R.J. 1977 A 1976 theory of charismatic leadership, pp. 189-207 in James G. Hunt and Lars L. Larson (eds.), Leadership: The Cutting Edge. Carbondale, Illinois: Southern Illinois University Press. and Gary Dessler 1974 The path-goal theory of leadership: Some post hoc and a priori tests, pp.29-55 in James G. Hunt and Lars L. Larson (eds.), Contingency Approaches to Leadership. Carbondale, Illinois: Southern Illinois University Press. Hoy, Wayne K. and Cecil G. Miskel 1987 Educational Administration: Theory, Research, and Practice, (3rd ed.). New York: Random House. Immegart, Glenn L. 1988 Leadership and leader behavior, pp.259-77 in Norman J. Boyan (ed.), Handbook of Research on Educational Administration: A Project of the American Educational Research Association. New York: Longman. Jacobs, T.O. 1974 Leadership and social exchange, pp.193-210 in K.W. Tilley (ed.), Leadership and Management Appraisal. New York: Crane, Russak. James, Eric 1951 Education and Leadership. London: Harrap. Janda, K.F. 1960 Towards the explication of the concept of leadership in terms of the concept of power. Human Relations, 13:345-63. Janis, Irving L. and Leon Mann 1977 Decision Making: A Psychological Analysis of Conflict, Choice, and Commitment. New York: Free Press. / 175 Kaplan, George R. 1985 Shining lights in high places: Education's top four leaders and their heirs. Phi Delta Kappan, 67:7-16. Kast, Fremont E. and James E. Rosenweig 1974 Organizations and Management: A Systems Approach, (2nd ed.). New York: McGraw-Hill. Katz, Daniel, and Robert L. Kahn 1978 The Social Psychology of Organizations, (2nd ed.). New York: Wiley. Katzell, Raymond 1962 Contrasting systems of work organization. American Psychologist, 17:102-08. Kerr, Steven, and John M. Jermier 1978 Substitutes for leadership: Their meaning and measurement. Organizational Behavior and Human Performance, 22:375-403. Kohlberg, Lawrence 1984 The Psychology of Moral Development: The Nature and Validity of Moral Stages. San Francisco: Harper and Row. Lakatos, Imre 1970 Falsification and the methodology of scientific research programs, pp.91-193 in I. Lakatos and A. Musgrove (eds.), Criticism and the Growth of Knowledge. Cambridge: Cambridge University Press. Lester, James 1978 Preface, in Morgan W. McCall Jr. and Michael M. Lombardo (eds.), Leadership: Where Else Can We Go? Durham, North Carolina: Duke University Press. Lipham, James 1964 Leadership and administration, pp. 119-41 in D.E. Griffiths (ed.), Behavioral Science and Educational Administration. Chicago: University of Chicago Press. / 176 Livingstone, R.W. 1950 Leadership in Education. Walker Trust Lectures on Leadership No. 10. London: Oxford University Press. Lord, Robert G.; Roseanne J. Foti; and James S. Phillips 1982 A theory of leadership categorization, pp. 104-21 in James G. Hunt, Uma Sekaran, and Chester A. Schriesheim (eds.), Leadership: Beyond Establishment Views. Carbondale, Illinois: Southern Illinois University Press. Maclntyre, Alisdair 1981 After Virtue. Notre Dame: University of Indiana Press. Mann, R.D. 1959 A review of the relationships between personality and performance in small groups. Psychological Bulletin, 56:241-70. March, James G. 1974 Analytical skills and the university training of educational administrators. The Journal of Educational Administration, 12:17-43. Martin, Jane Roland 1981 The ideal of the educated person. Educational Theory, 31(2):97-109. Maxcy, Spencer J. 1984 Leadership, administration, and the educational policy operation, pp.329-38 in E.E. Robertson (ed.), Philosophy of Education 1984: Proceedings of the Fortieth Annual Meeting of the Philosophy of Education Society. Normal, Illinois: Philosophy of Education Society. McCall, Morgan W. Jr. and Michael M. Lombardo 1978 Where else can we go?, pp. 151-73 in Morgan W. McCall Jr. and Michael M. Lombardo (eds.), Leadership: Where Else Can We Go? Durham, N.C.: Duke University Press. McClellan, James E. and B. Paul Komisar 1962 Preface, in CD. Hardie, Truth and Fallacy in Educational Theory. New York: Teachers College Press, Columbia University. / 177 McClelland, D. 1975 Power: The Inner Experience. New York: Irvington. Melcher, Arlyn 1977 J. Leadership models and research approaches, pp.94-108 in James G. Hunt and Lars L. Larson (eds.), Leadership: The Cutting Edge. Carbondale: Southern Illinois University Press. Miklos, Erwin 1977-78 Ethical aspects of administrative action: Implications for research and preparation. Administrator's Notebook, 26(5): 1-4. Mill, John Stuart 1956 On Liberty, (originally published in 1859), (ed. C.V. Shields). New York: Bobbs Merrill. Miner, J.B. 1982 The uncertain future of the leadership concept: Revisions and clarifications. The Journal of Applied Behavioral Science, 18:293-307. 1978 Twenty years of research on role motivation theory of management. Personnel Psychology, 31:739-60. Mintzberg, Henry 1982 If you're not serving Bill and Barbara, then you're not serving leadership, pp.239-59 in James G. Hunt, Uma Sekaran, and Chester A. Schriesheim (eds.), Leadership: Beyond Establishment Views. Carbondale: Southern Illinois University Press. 1973 The Nature of Managerial Work. San Francisco: Harper and Row. Morrow, Lance 1987 Who's in charge? The nation calls for leadership and there is no one home. Time Magazine, 130(19):22-25. Nash, J.B. 1929 Leadership. Phi Delta Kappan, 12:24-25. / 178 National Conference of Professors of Educational Administration 1948 Educational Leaders: Their Function and Preparation. A Report of the Second Work Conference of the NCPEA held at Madison, Wisconsin, Aug.29-Sept.4, 1948. National Education Association 1933 Effective Instructional Leadership: A Study of the Problem of Integration. NEA 6th Yearbook (chair Rudolph L. Lindquist). New York: Teachers College Press, Columbia University. Newell, Clarence A. 1978 Human Behavior in Educational Administration. Englewood Cliffs, N.J.: Prentice-Hall. Ott, Walter 1985 Ideals in educational decision making. The Canadian Administrator, 24(6): 1-6. Parsons, Talcott 1951 The Social System. New York: Free Press. Peters, Richard S. 1972 Education and the educated man, pp.3-18 in R.F. Dearden, P.H. Hirst, and R.S. Peters (eds.), Education and the Development of Reason. London: Routledge and Kegan Paul. 1966 Ethics and Education. London: Allen and Unwin. Pfeffer, Jeffrey 1978 The ambiguity of leadership, pp. 13-33 in Morgan W. McCall Jr. and Michael M. Lombardo (eds.), Leadership: Where Else Can We Go? Durham, N.C.: Duke University Press. Piaget, Jean 1952 The Origins of Intelligence in Children. New York: International University Press. Pondy, Louis R. 1978 Leadership is a language game, pp.87-99 in Morgan W. McCall Jr. and Michael M. Lombardo (eds.), Leadership: Where Else Can We Go? Durham, N.C.: Duke University Press. / 179 Prentice, W.C.H. 1982 Understanding leadership. Reprinted from Harvard Business Review, September-October 1961, Number 61511. Rawls, John 1971 A Theory of Justice. Cambridge, Mass.: Harvard University Press. Robinson, Norman 1981 Surviving and flourishing in the turbulent world of educational politics. The Canadian School Executive, l(6):5-8. Rosenbach, William E. and Robert L. Taylor 1976 Contemporary Issues in Leadership. Boulder, Colorado: Westview Press. Salancik, Gerald R.; Bibby J. Calder; Kendrith M. Rowland; Huseyin Leblebici; and Mary Conway 1975 Leadership as an outcome of social structure and process: A multidimensional analysis, pp.81-101 in James G. Hunt and Lars L. Larson (eds.), Leadership Frontiers. Kent, Ohio: Kent State University Press. Sartori, Giovanni 1984 Guidelines for concept analysis, pp. 15-85 in Giovanni Sartori (ed.), Social Science Concepts: A Systematic Analysis. Beverly Hills: Sage Publications. Scheffler, Israel 1973 Reason and Teaching. London: Routledge and Kegan Paul. 1966 Philosophical models of teaching, pp.99-114 in Israel Scheffler (ed.), Philosophy and Education, (2nd ed.). Boston: Allyn and Bacon. Schenk, C. 1928 Leadership. Infantry Journal, 33:111-22. / 180 Schriesheim, Chester A.; James G. Hunt; and Uma Sekaran 1982 Conclusion: The leadership-management controversy revisited, pp. 290-98 in James G. Hunt, Uma Sekaran, and Chester A. Schriesheim (eds.), Leadership: Beyond Establishment Views. Carbondale: Southern Illinois University Press. and Steven Kerr 1977a Theories and measures of leadership: A critical appraisal of current and future directions, pp.9-45 in James G. Hunt and Lars L. Larson (eds.), Leadership: The Cutting Edge. Carbondale: Southern Illinois University Press. 1977b RIP LPC: A response to Fiedler, pp.51-56 in James G. Hunt and Lars L. Larson (eds.), Leadership: The Cutting Edge. Carbondale: Southern Illinois University Press. Scriven, Michael 1966 Primary Philosophy. New York: McGraw-Hill. Selznick, Philip 1957 Leadership in Administration: A Sociological Perspective. White Plains, New York: Row, Peterson, and Co. Sergiovanni, T.J. 1979 Is leadership the next great training robbery? Educational Leadership, 3(6):388-94. Silver, Paula 1983 Educational Administration: Theoretical Perspectives on Practice and Research. New York: Harper and Row. Snyder, Karolyn J. and Robert H. Anderson 1986 Managing Productive Schools: Toward an Ecology. New York: Academic Press. Sokiloff, Harris J. 1984 Response to Professor Maxcy, pp.339-41 in E.E. Robertson (ed.), Philosophy of Education 1984: Proceedings of the Fortieth Annual Meeting of the Philosophy of Education Society. Normal, Illinois: Philosophy of Education Society. / 181 Stewart, Rosemary 1982 The relevance of some studies of managerial work and behavior to leadership research, pp. 11-30 in James G. Hunt, Uma Sekaran, and Chester A. Schriesheim (eds.), Leadership: Beyond Establishment Views. Carbondale: Southern Illinois University Press. Stogdill, Ralph 1974 Handbook of Leadership: A Survey of Theory and Research. New York: Free Press. O.S. Goode, and D.R. Day 1962 New leader behavior description subscales. Journal of Psychology, 54:259-69. Strike, Kenneth A. 1982 Educational Policy and the Just Society. Urbana: University of Illinois Press. Tead, Ordway 1935 The Art of Leadership. Toronto: McGraw-Hill. Tosi, Henry J. Jr. 1982 Toward a paradigm shift in the study of leadership, pp.222-33 in James G. Hunt, Uma Sekaran, and Chester A. Schriesheim (eds.), Leadership: Beyond Establishment Views. Carbondale: Southern Illinois University Press. Tucker, R.C. 1981 Politics as Leadership. Columbia, Mo.: University of Missouri Press. Tyack, David and Elisabeth Hansot 1982 Managers of Virtue: Public School Leadership in America 1820-1980. New York: Basic Books. Vroom, Victor H. 1976 Leadership, pp.1527-51 in M.D. Dunnette (ed.), Handbook of Industrial and Organizational Psychology. Chicago: Rand McNally College Publishing. / 182 and P.W. Yetton 1973 Leadership and Decision-Making. Pittsburgh: University of Pittsburgh Press. Watts, Bert 1985 Principals' Principles. The Yellow Papers: Administration and Morality, 5:30-36. Weber, Max 1964 The Theory of Social and Economic Organizations, (trans. A.M. Henderson and Talcott Parsons). London: Free Press of Glencoe. White, Ralph and Ronald Lippitt 1960 Autocracy and Democracy: An Experimental Inquiry. New York: Harper and Row. Williams, R.B. 1984 Where are our leaders? The B.C. Teacher, 63(3): 141-42. Wilson, John 1963 Thinking with Concepts. Cambridge: Cambridge University Press. Winter, D.G. 1973 The Power Motive. New York: Free Press. Wittgenstein, Ludwig 1963 Philosophical Investigations, (2nd ed.), (trans. G.E.M. Anscombe). Oxford: Basil Blackwell. Undated Philosophical Psychology. Unpublished notes. Yukl, Gary 1981 Leadership in Organizations. Englewood Cliffs, N.J.: Prentice-Hall. Zaleznik, Abraham 1977 Managers and leaders: Are they different? Harvard Business Review, 55:67-78. APPENDIX A: AN EXAMPLE OF CONCEPTUAL ANALYSIS Suppose we are trying to discover the necessary and sufficient features for the identification of phenomena that are squares. Let us suppose that the examples of squares we think of all happen to be red. We might begin by hypothesizing that all squares are red, and that all squares have exactly four closed sides. In testing these features for necessity, we try to imagine cases of things which competent language users would call "squares" but which are either not red or do not have four closed sides. Since we can readily think of non-red shapes that we would ordinarily refer to as "squares," "redness" is not a necessary feature (condition) of squares. But we cannot think of something we would call a "square" which does not have four closed sides. Thus, we may tentatively assume that having four closed sides is a necessary condition of a square. To test closed four-sidedness for sufficiency, we try to think of a shape having four closed sides which we would not call a "square." If we can think of such a case, then having four closed sides is not the only necessary feature of a square. It is easy to imagine a quadrilateral figure with closed sides that is not a square, such as a trapezoid (a plane figure with four sides, two of which are parallel). We might hypothesize that a square must also have opposite sides parallel. But a rhomboid (a parallelogram with oblique angles and unequal adjacent sides) has four closed sides with opposite sides parallel, and it is not a square. We might hypothesize that another necessary condition of a square is that the sides must be equal in length. There would seem to be no squares with unequal sides. But a rhombus (an equilateral parallelogram with oblique angles) has four equal closed sides and opposite sides parallel, and it is not a square. Thus our list of necessary conditions is still incomplete. The difference 183 / 184 between a rhombus and a square is that only a square has ninety-degree angles, another necessary feature of a square. If we cannot imagine a figure with four equal closed sides, in which the opposite sides are parallel, and four ninety-degree angles which is not a square (and I cannot) then we may assume that our list of necessary conditions is also sufficient, but this assumption is always tentative, and holds only so long as no case of a square is discovered which is not fully circumscribed by the set of necessary and sufficient conditions. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items