GAETANO MOSCA, in his work which has been translated as The Ruling Class, engaged in an extensive criticism of the democratic theory that stems in particular from Rousseau. In the light of the historical evidence which he considered important he concluded:
Among the constant facts and tendencies which are to be found in all political organisms one is so obvious that it is apparent to the most casual eye. In all societies — from societies that are meagerly developed and have barely attained the dawnings of civilization, down to the most advanced and powerful societies — two classes of people appear — a class that rules and a class that is ruled. The first always the less numerous, performs all political functions, monopolizes power and enjoys the advantages that power brings, whereas the second, the more numerous class is directed and controlled by the first….1
Mosca, of course, is but one of a long line of writers who have tried to show that as an explanation of the power processes democratic theory is absurd. At the same time, however, most of these anti-democratic writers grant that there is, in all human beings, a compelling sentiment that justice and equality should temper power; that the masses should enforce some restraint on their leaders; and that government always involves some kind of consent of the governed. Those who bring an empirical method to the study of power are frequently condemned on the ethical grounds that they seek to justify some kind of tyranny, but what they do more than anything else is to show up the realities of power. Only after some examination of a power system has been made, can it be determined whether or not that system conforms to some ethical ideal. The ideals which are widespread in the Western democracies are, first, that there is popular participation in decision-making; and, second, that the class system is open, that is, that recruitment to positions of power is the result of personal achievement in a competitive society. Undoubtedly most studies in stratification and mobility are motivated by these ideals. It is the task of philosophers to determine whether in industrial systems these ideals are tenable. It is the task of the social scientist to devise the best analytical scheme under which the facts of power can be arranged. The aim of the present paper is to suggest a scheme of elite groups which can stand as a model for the study of power stratification in Canada.
The term “power” is used here in much the same sense that Max Weber used the concept of authority—the probability that within a social system an individual (or a group) is in a position to carry out his will. This concept of power also requires that other actors in the system will obey the directives of the one who seeks to carry out his will. Power and legitimacy, both preconditions of social organization, are correlative terms, like rights and duties. Weber calls these preconditions “imperative control” and “imperative coordination.”2 Which of the actors in any given group, large or small, enjoys power depends on a variety of factors, but whatever those factors are, the power exercised is legitimate as far as the remaining actors are concerned. Some writers have suggested that the distinction between power and authority rests on legitimacy.3 The distinction seems a thin one, however, since where power is not legitimate it is unstable. If holders of power regulate members by force which it is not legitimate for them to possess, it is an indication that power relations within the group are changing. Thus in all systems where there is an unresolved struggle, such as in a civil war, in a factional dispute in a trade union, or in a cage of apes which has been disturbed by the arrival of females,4 power, in the sense in which the term is used here, is absent, except within the competing sub-systems. Unless the system is disintegrating there is no such thing as “naked” power5 unrestrained by the legitimate approval of subjects.
The principle of legitimacy which we take to be correlative with power is a psychological state, a complex of attitudes which people share about being subject to someone else’s power. It suggests that the governed do in fact give their consent. To a great extent consent and the sense of legitimacy spring from distortions of social realities. What Pareto calls “derivations,” Mosca “the political formula,” Bentham the ‘legal fiction,” Sorel “the myth” are much the same as Plato’s noble falsehood, except that many of those which have appeared in history are ignoble in that they would not satisfy ethical criteria. These systems of belief give the sense of rightness to the order of power within which masses of people live, but they need not reflect reality. They can be simply rationalizations. The psycho-analytic writers have done much to show the extent of rationalization in social affairs: the real world becomes distorted in the mind’s eye to conform to subjective needs. One of these basic subjective needs is to be treated with justice and as an equal—the sentiment that Pareto called the integrity of the individual. Hence the reality of power, however it may affect individuals, becomes rationalized as good, and the most extreme dictatorship can be subjectively felt as the epitome of justice while one’s neighbours are sent to concentration camps.
While it has been suggested that “naked power” does not exist, subtle methods are available to modern holders of power by which they can control some of the psychological factors which constitute legitimacy. The idea of “newspeak” and “double think” through which the mental processes of subjects can be influenced by implanting stereotypes favourable to a particular power-group, is now a popular subject for fiction. There is little doubt that modern media of communication do offer immense opportunities for those in power to create or strengthen the legitimacy on which their power rests. Human beings have always resisted or rationalized the use of force against their bodies, but they have not been greatly concerned in the modern period about the integrity of their minds or about protection against the force of ideas. The use of the public-address system in present-day China is perhaps the best example of highly developed coercion through communications. But even in this case it is doubtful whether a fully naked power has emerged. As yet the fictional accounts remain as prophecies or warnings. Except when power relations which have been relatively stable are changing, all de facto power, though it may not be de jure in the refined legal sense, is legitimate. Power and legitimacy being correlative, holders of power can violate civil, political, and social rights even when they are entrenched in constitutions, as long as it is widely held to be legitimate for them to do so. Thus as political attitudes change and new developments of power must be rationalized, bills of rights are reinterpreted by both the courts and the masses-a condition which makes it difficult for power to be confined within legal nets.
Power is a precondition of social organization because every social system requires individuals to take responsibilities for co-ordinating and directing the activities of the group. It would seem that even in very small and temporary groups, not all members participate in decision-making and directing.6 Somehow specific individuals are selected as leaders. There are in every system positions which, in the terminology of contemporary sociology, can be called “power roles.” In large and permanent social aggregates the selection of persons to fill these roles follows a well-established pattern; in small and temporary groups a pattern soon develops. The number and diversity of power roles depend mainly on the complexity of the social fabric or the division of labour within the system.
In simple types of social structure, such as the less-developed primitive group, the power roles tend to be fused with other roles. Although it is a common error to generalize about the political and economic institutions of primitive peoples, these decision-making and co-ordinating roles are, in the main, linked with the kinship structure or lineage systems or with religious and magical roles, or with skill in the performance of tasks which the group values highly, such as warfare and raids.7 The fusion of power roles with magical and religious functions can be illustrated by the example of the garden magician as a co-ordinator of economic activity. A similar case is the shaman who provides information on the likely movements of game and, in effect, decides on the direction of the hunt.8 It has been suggested that primitive groups holding property in common must, if they are to survive, institutionalize some kind of decision-making role to co-ordinate exploitation and to avoid exhaustion of their resources,9 if not a war of all against all.
(Anthropologists have, it is true, reported instances of primitive “democracies” where there seems to be a very minimum of co-ordination by individuals. However, since these are small groups near the subsistence level, it is probable that power lies in the interplay of personalities rather than in institutionalized roles. In such cases it is much more difficult for the field worker to isolate the element of leadership from the interaction.)
In simple structures, where the roles involving economic, political, and other forms of power are fused, the economic and political institutions are scarcely distinguishable. In his discussion of the Bantu of Kavirondo, Wagner says: “The assumption that each function in a culture must have its corresponding institution—religious, economic, political, etc.—would cut short an understanding of the way in which cultures are integrated into a body politic, the institutions of which are not yet clearly differentiated according to different aspects but which serve many functions at the same time.”10 Malinowski considered the kula as a half-commercial, half-ceremonial exchange,11 and, as Nadel has pointed out,12 the potlatch could be either a political or an economic activity.
Differentiation of power roles emerges only as the scale of social development is ascended, or after an increase in population. With the increased division of labour, it becomes necessary to co-ordinate the activities and exchanges of producers so that in some large primitive groups such as the Ashanti and other African native states there can be found the prototype of the manager and the bureaucrat and the military chief-of-staff working in a distinct and balanced power-system.13
In complex societies, such as modern industrial nation-states, power becomes separated into distinct but interrelated systems which correspond to essential social functions. “Function” is, as Merton has shown,14 a very ambiguous term in sociology: here it is used in the sense of a prerequisite to a persistent social structure. In general, these functions are the economic, the political, the administrative, the defensive, and the ideological. The last means the creation and articulation of myths, formulas, and derivations which aid in orienting the individuals in the system to super-empirical goals such as the Kingdom of God, the classless society, or Canada’s century.
It would be in keeping with intellectual tradition to select the political function as in some way superior to the others, but this “umpire” theory of the state as an all-embracing coercive apparatus tends to neglect the power which inheres in the other functions. As a result, political science is preoccupied with governmental organs and their formal operations. Recent sociological theory, if Professor Parsons’ is taken as an example, would seem to retain the traditional view of political power as a “master system.” In The Social System he argues:
While the structure of economic power is, as we have noted, lineally quantitative, simply a matter of more and less, that of political power is hierarchical; that is of higher and lower levels. The greater power is power over the lesser, not merely more power than the lesser. Political power is relational, not merely in reference, that is to n potential exchange partners, but in direct significance. This is perhaps another way of stating the diffuseness of political power, in that it is a mobilization of the total relational context as a facility relative to the goal in question.15
This view ignores the fact that the political system behaves according to the pressures brought to bear by competing social forces. The sovereignty which it claims, is subject to compromise with the other power systems with which it is interrelated in the society.
While these systems are interrelated, they may not be geographically co-extensive. Economic power will spill over political boundaries, and conversely political considerations can impose anti-economic and irrational activities on economic systems, for example by preventing trade with potential enemies or by requiring the allocation of resources to areas which, in the opinion of military chiefs, are of strategic importance. Similarly religious and ideological systems are not confined to individual states despite the emergence of nationalism as a secular ideology. It is of course convenient in the contemporary world to regard nation-states as the widest effective social groupings, but possibly economic systems, since they extend beyond political boundaries, are the widest effective groupings. If economic systems could be put on a map they would not in their entirety coincide with political systems. (The doctrines of historical materialism and imperialism hold of course that the effective groupings are indeed the economic ones. This materialist view holds that the economic function has a primacy over the others. The view here taken is that not only must all five functions be undertaken, but also that in any given case economic, political, bureaucratic, defensive, or ideological considerations might dominate.) Many ideologies are anti-economic. This is particularly true of prestige economies at the primitive level, and, at the complex level, the entire leisure-class culture was considered by Veblen, for example, to be anti-economic. However, for the purpose of analysing power it is most convenient to use political boundaries because they are more clearly established and more stable than international economic activity as it ebbs and flows. But it is a mistake to regard the political system as the sole determinant of all power relations. There is, as Merriam16 puts it, a family of power, any member of which can exert a greater influence than the others at a particular time and place.
Power, then, is a generic term. Its species in the complex structure are economic, political, bureaucratic, military, and ideological. If these five functional systems of the complex society are regarded as very large role-systems, there must be in them decision-making and co-ordinating roles. It should be the task of the sociology of power to isolate these power roles and to examine the way in which persons are selected to fill them. The circulation of persons in and out of power roles is the process of power itself.
Power roles, as Parsons has pointed out,17 have two aspects: one is the control of facilities, the other the rewards which accrue to those in power positions. The term “facilities,” which seems to be a useful one, means those objects, physical and non-physical, which an individual must control if he is to perform the task allotted to him in the social system. The rewards are necessary because the task itself does not yield pleasure—a conception which is, of course, the old notion of the disutility of labour. Frequently, however, power brings its own satisfactions apart from any economic advantage. Differential remuneration may be an important factor in motivating individuals to undertake different jobs or roles in the middle range of occupational hierarchies, but for those who are in a position to direct giant corporations, or modern armies, or political systems, increments in monetary rewards can yield little utility and a great deal of income tax. The satisfaction of organizing and controlling large systems, and thereby the destinies of masses of human beings, has long been recognized. The distinction, therefore, between rewards and the control of facilities begins to fade at the level of the power roles which are being considered here. (Of course, the avant-garde in sociological theory recognize that the distinction does not always exist, or as they say, “. . an object which is useful as a facility comes to be cathected directly so that its possession is also interpreted by the actor and by others as a reward.”)18
Those persons in power roles can be designated as an elite. Thus in each of these interrelated systems of power there is a corresponding elite. The fused power-roles of primitive societies which result in a relatively simple hierarchical arrangement give way in the industrial system to separate elite groups. Historically the first power-specialists were administrative staffs (usually a priesthood), and a military caste. In the modern period the exercise of power requires more specialization. Economic, political, bureaucratic, military, and ideological power-roles tend to be specialized, and each of the various elites has the task of directing a functional power-system.
We have so far been considering power in industrial systems. In the twentieth century these have developed into two types, the democratic and the dictatorial. Some would argue that as industrial systems become more complex, the difference between the two types becomes less noticeable, that is, that democratic systems are marked by oligarchy in organization, widespread apathy, and an inner compulsive conformity. Professor Aron19 has, however, made an important distinction between what he calls the “Soviet” type and the “Western” type, a distinction which rests on the structure of the elite groups. These groups, it has been suggested, are the result of functional differentiation. In the “Western” type the elite groups are separate and compete for power, with the result that control tends towards an equilibrium of compromise. In the “Soviet” type, on the other hand, the elite groups, while still retaining some functional identity, are unified by their membership in the Communist party. They thus serve a higher loyalty. In two other industrial dictatorships of the twentieth century, Italy and Germany in the interwar years, the elite groups were similarly unified. In Germany, the one elite group with a long tradition of independent superiority, the general staff, was finally unified with the other elites by taking the oath of personal loyalty to Hitler, and by placing party generals, such as Keitel and Jodl, in key roles. The more gloomy writers predict that because of recurring crises and the new interpretations of loyalty and national interests, the Western types will lose their distinctive features.
It is characteristic of power in industrial societies, particularly in the West, that it is exercised, not so much by individuals as by collectivities. Legally speaking, the fiction of incorporation considers a group of individuals as a responsible person with rights: sociologically speaking, these collectivities are the incumbents of power roles. Cabinets, committees, chiefs-of-staff, and boards of directors all operate by collective responsibility. This principle of collegiality is in part demanded by specialization in the direction of operations, and in the wide range of experience and knowledge that must be brought to the making of decisions. The board of directors of a large corporation, for example, will include some of the senior executives as well as legal and financial experts. The range of decisions which they must make as a collectivity requires a variety of skills and experience. The empirical problem, therefore, of determining the individuals who occupy power roles can be solved by identifying them with collectivities which are legally acknowledged as being responsible for and having rights in facilities.
The elite in each power system is given some homogeneity by cross-membership in collectivities. Although interlocking directorships in the economic elite are perhaps the best example, other examples are the cross-membership of high-ranking bureaucrats in committees, and the cross-membership of church hierarchies in many advisory bodies. Even intellectuals interlock on editorial boards, learned societies, and councils. Elite groups interact also in their clubs, and associations. Sociologists and anthropologists20 have, for a long time, pointed out the importance of associations in linking various groups together by a web of cross-membership. In the sense that elite associations, particularly clubs, tend to be exclusive, they become a means of consolidating power. As Mr. McWilliams21 has observed: “Social power is organized by exclusion.” A large number of honorific posts, such as governor-ships of hospitals, universities, and the like constitute the cognate roles for members of elite groups.
Associations may also bring together the elites of the various power systems. However, the elite of the Western type do not often fill power roles in more than one system at the same time, and it is often considered undesirable for them to do so lest power should be concentrated. Mr. Wilson had to relinquish his interest in General Motors before being accepted as a member of President Eisenhower’s cabinet, and in other systems cabinet ministers are expected to forgo their directorships in corporations. (It would seem that trust companies must be excepted here.) A high degree of simultaneous membership in the elites of the various power systems, a condition which can be determined by empirical investigation, would imply a homogeneity of the elite of the entire society-or an exclusive ruling class. There is a long tradition, too, that the bureaucracy should be out of politics. Because the bureaucracies require experts, they recruit highly intelligent persons who may take up a dangerously paternal attitude to their fellow creatures. Perhaps the bureaucrat with his schemes and plans, rather than the politician, is the new Machiavelli. The principle that a bureaucratic career and a political career should not be mixed is a defence against the power of expertise encroaching on another power-system which is its most effective check.
Elites tend to merge and lose their functional identity owing to the practice of individuals moving from one elite group to another during their careers. A degree of exclusiveness is reached when directorships, and ministerial and bureaucratic posts are made available for the same person or group of persons. The avenue to elite roles in one functional area should normally be through a career in that area only. Sir Anthony Eden’s remark about Sir Winston’s “passion for the political life” serves to illustrate the principle that political institutions should be the sphere of activity for a recruit to the political elite. The vigour of parliamentary institutions and the extent to which political power can counter other power-systems depend on these institutions’ being the ladders to and training-ground for political power-roles. When these power-roles are filled primarily from other power-systems there comes a diminution of political power, and a drabness to the political processes.
Mosca suggested22 that the elite—or what he called the ruling class—should represent a balance of the various social forces in the society. He criticized the notion that a balance of power is achieved in any society by separating the organs of government, since the same victorious political party can capture all the organs. It may be argued that a true separation of power results when elite groups represent functionally separate systems of the society as a whole. That is, if elite groups remain separate, they are in a position to “countervail” each other, to borrow a term from Professor Galbraith.23 When they begin to lose their separate identities the checks on power begin to disappear. Such is the danger, for example, in a governmental bureaucracy, which may encroach on the economic system or become the arbiter of culture and of the intellectual life of a nation. What Walt Whitman spoke of as the “never-ending audacity of elected persons”24 can be found, too, in the bureaucrat, the militarist, the corporation director, and the ideologist.
In the Western systems the principle of countervailing power can be seen operating among and even within the power systems. Absolute power in the hands of the economic elite, for example, is limited, not only by the other elites in the society, but because it is made up of trade-union leaders as well as corporation directors. A balance of power is maintained if both groups retain their separate identity. As Chancellor Kerr has observed25 in a discussion of the positive value of industrial conflict, “the union which is in constant and complete agreement with management has ceased to be a union.” Similarly in the political elite which would include the leaders of rival political parties, the real limitation on power stems from a strong opposition who can take over the function of government. In the parliamentary system the parties tolerate each other and work by mutually acceptable rules within which they can “safely afford to bicker” in much the same way that the economic elite of big business and big unions generally accept the rules of collective bargaining and mediation. A strong opposition is of even greater importance where the political and bureaucratic elite begin to take over the economic system, because the balance provided by the economic elite tends to be removed, thus making for a greater concentration of power which can only be overcome by strengthening the rival faction in the political elite. The weakness of Oppositions in the Canadian Parliament is frequently taken as a sign of political ill-health, but it need not be considered pathological where control of facilities by Government is limited. The large number of organizations which annually take their briefs to Parliament Hill indicates that other elite groups, brought together in such associations as the Chamber of Commerce, trade-union congresses, church councils, and the like, impose a check on political power. If Government operations were more extensive it would be necessary to have a much stronger Opposition within the political system.
The argument is, then, that power tends towards an equilibrium of competing elites. The checks and balances which are everywhere considered desirable do not come from control by the masses or from membership in organizations or from legal restrictions, but from the tradition of independence built up by elite groups and the jealousy with which they guard the spheres of activity which they consider to be theirs. In this respect rights which are the rational-legal limitations on power roles, constitute, in the main, the rights of organizations to exist and to carry out their activities. If these rights are violated by one elite, the power of another is challenged. The right of association and the right to strike, for example, are essential weapons without which trade-union leaders would be powerless; the rights of property are the rights through which directors of large corporations control vast economic resources; the rights of assembly and worship are the rights of ecclesiastical hierarchies to expound their own particular interpretations of the world. How often does the individual enter litigation on his own to enforce his rights, and how often is it that the courts must decide between the claims of organizations and corporations? The individual, more often than not, has to wait until some organization, interested in a test case, enters the legal process on his behalf. His rights thereby become the subject of a tactical move in a power conflict. The ability to enforce rights is itself largely a question of power. In modern industrial systems the power of the individual, the common man, the ordinary employee, the shareholder, consists simply in withdrawing his loyalty, reneging on his fees, switching his support to a contending faction, or bargaining his votes in a proxy battle. As H. G. Wells said of the political function: “In Great Britain we do not have elections any more; we have rejections. What really happens at a general election is that the party organizations—obscure and secretive conclaves with entirely mysterious funds—appoint about twelve hundred men to be our rulers, and all that we, we so called self-governing people are permitted to do is, in a muddled angry way, to strike off the names of about half these selected gentlemen.”26
The task now is to show how this theoretical scheme of elite groups may be applied to the study of power in Canada, as an example of an industrial system of the Western type. If power roles can be satisfactorily identified in the social structure and the social characteristics of their incumbents studied, a clearer picture of social mobility and power stratification would emerge. (By this approach, moreover, the problems of determining classes by subjective criteria, or non-functional objective criteria such as statistical groups, could be avoided.) As Miss Barbara Wootton has observed in her review of Social Mobility in Britain,27 “… it would be fascinating to see how far the small circle of those who man the highest positions in industry and in the public and semi-public services constitute a self-recruiting community. Certainly, any classification which treats ‘company director’ as a single category must miss vital aspects of our social structure through its inability to distinguish real industrial tycoons from small taxi-cab proprietors.”
The economic elite is a reflection of the oligopoly and monopolistic competition which exists in all industrial systems, a condition which has received attention in both theoretical and applied economics. In the United States, following the original investigation of Berle and Means into the concentration of economic power, there have been the studies by the Temporary National Economic Committee and later inquiries by American government agencies.28 Somewhat similar studies have been undertaken in other countries. Although these studies have some deficiencies, they do indicate a very high degree of concentration in the more developed industrial systems. A further concentration can be seen by tracing interlocking directors. A sociologist interested in recruitment to power roles can begin his studies, then, by identifying individuals with the collectivities that control the most important corporations in the economy. Taussig’s and Joslyn’s early study of American business leaders29 was from a sample of the directory of directors in which it was difficult to ensure that those selected controlled large companies. Since the Canadian directory of directors includes directors of very small firms, a random sample from it would hardly be a sample of the economic elite.
There is no thorough study of the concentration of economic power in Canada. Since a study of the economic elite would be pointless without one, I have tried to show30 that a group of 183 “dominant” corporations were responsible, in the period 1948-50, for 40 to 50 per cent of the gross value of production in manufacturing, 63 per cent of the total value of metal production; 90 per cent of railway transportation; 88 per cent of the gross earnings of telegraph and cable services; 82 per cent of the total revenue of Canadian air carriers; 83 per cent of telephone revenues; and 60 to 70 per cent of the hydro-electricity produced by privately owned companies, as well as a large but undetermined proportion of other industries such as industrial minerals, fuels, water transportation, and retail distribution. In 170 of these 183 corporations (in 13 of them the identity of the directors could not be established) there are 1,613 directorships, 296 of which are held by non-Canadians, the majority of whom are Americans. The 1,317 remaining directorships are held by 922 Canadian residents. I have taken these 922 persons, many of whom also serve on the boards of the principal banks and insurance companies, as that part of the economic elite in Canada represented by directors.
An important assumption in taking the directors of large corporations as members of the economic elite is that they do in fact control the large enterprises which they legally head. There is the view, of course, that directorships are nominal and that real control rests with managers. (In the recent spectacular proxy battles in the United States, however, it was directorships that were the goals of the contending factions.) In Canada, however, in almost all the dominant corporations, the boards of directors include some of the senior executives or managers.
Another assumption which applies to all elite collectivities is that all members are equally powerful. One member might dominate a board of directors because of his personality or because of the interests he represents. If the more powerful members of all of the 170 boards of directors could be determined, an even smaller elite might emerge. However, given secrecy and the collective responsibility which operates in all these elite collectivities, it must be assumed that all members have some power. If the interaction of board meetings could be recorded with the Bales apparatus we might get a clearer idea of decision-making in such groups, but there is little doubt that most of them would regard such activity as an intrusion and an indignity. No doubt that rather contradictory principle of primus inter pares applies, but it is not always possible to know who is first.
The other institution of economic power which has grown up with the giant corporation is the trade union. Although the instruments of production are not legally controlled by trade unions, their basic reason for existing, as one English trade-union leader asserted, “is to exercise that power which arises from the refusal to work.”31 Unlike corporations, however, in which there is a general recognition of control from the top down, trade unions have traditionally upheld the doctrine of democratic control by membership. It may be that the doctrine of control by shareholders is the counterpart of this democratic theory, but while no one seriously accepts the view that shareholders do control, trade unions claim to be democratic organizations. Yet Robert Michels long ago argued with considerable force that trade unions, like political parties, became victims of the “iron law of oligarchies.”32 He attributed the failure of democratic control to the centralization and specialization that are required in any large organization bent on challenging power, as well as to the tendency on the part of leaders of radical movements to become separated from the masses which they lead and to undergo a psychological transformation, a kind of megalomania. A recent study by Dr. Goldstein33 of the Transport and General Workers Union in Great Britain, which has been criticized particularly by trade-union leaders, tries to show that oligarchic control, apathetic membership, and a complicated constitution eliminate any chance of democratic organization. It may be doubted, of course, whether the T.G.W.U., or the branch in which Goldstein became a participant observer, is typical of trade unions.
There is a great deal of literature which tries to show that trade unions are run by moguls, just as some writers allege that corporations are operated by self-seeking tycoons. Whatever their personalities, trade-union leaders are not very often thrown out of office by the rank and file. Professor Mills34 has shown how long the hierarchies persist. Perhaps more than any other power organization, a trade union becomes dependent on the skill and specialization of its leaders. As unions become larger and centralization increases, their leaders, particularly when they operate on the national level, are remote from the local activity through which unionism takes on meaning for the rank and file. In his recent study of strikes in Great Britain, Mr. Knowles35 points out how the membership can exert pressure on leaders through unofficial stoppages. It would seem, however, that challenges to union hierarchies, as in other power organizations, come mainly from a contending faction led in turn by dominant personalities who must try to win over the membership by offering something more than the old leaders.
As has been shown already, there is considerable justification for assuming that trade-union leaders, in the sense that they are co-ordinators and decision-makers, constitute an elite within their own power systems. In selecting the trade-union elite in Canada it would be logical to choose those persons who direct unions with shops and locals in the 183 dominant corporations, as well as the directing groups of the two congresses and the Catholic Confederation. Trade unions vary a great deal in their structure and one would have to take the directing groups of the exclusively Canadian unions, and of the largest Canadian units of the internationals, and, where they are large, those of regional units.
It is not intended to lengthen this paper with anything more than a brief indication of how the other elites may be selected from the institutions in which they function. For the political elite both federal and provincial systems must be taken into account. At the federal level, it is suggested, power roles are held by all members of the cabinet and their parliamentary assistants, the shadow cabinet of the Conservative Opposition, the leaders of the other opposition parties, and other prominent members of the Commons, prominence to be determined by party activities, chairmanship of committees, and the like. Professor Ward’s study36 of representation in the Canadian House of Commons has gone some way toward showing the career patterns of M.P.’s, but in the chapter on “The Personnel of Parliament” he makes no distinction between leaders and ordinary members. The political elite would also include members of the Senate, not so much because of the Senate’s latent political power, but because of the link which it represents with other elite groups. Perhaps senatorships should be considered more as honorific than as power roles. It would seem reasonable to include also the national ex-executives of the major political parties. The judiciary must be taken into consideration also. Not only do its members make effective decisions for the social system, but they also arbitrate and investigate a wide range of problems. It is suggested, therefore, that all members of the federal courts and the provincial supreme courts should be included in the political elite. Political power roles at the provincial level could well be restricted to members of provincial cabinets.
The bureaucratic elite is more difficult to determine because the structure of bureaucracies is, at least theoretically, perfect in terms of spheres of competence and authority. It is thus somewhat arbitrary to draw a line across a hierarchical structure and say that those above the line are the elite and those below are not. Despite this obvious problem it is realistic to assume that those at the top of the civil service have far more power than those at the bottom. This power, in part, stems from the role of the expert in advising his political chief or chiefs, since the higher civil servant frequently occupies his position longer than does the cabinet minister. Power also rests in the function of the civil servant to administer broad lines of policies once they have been laid down. As the bureaucratic elite, therefore, those persons should be selected who actually participate in policy-making and are responsible for its implementing. Undoubtedly, to be absolutely sure of the power roles within the civil service, one would have to examine within each department the informal social structure which makes it possible for formal systems to operate. In the bureaucratic elite would be deputy ministers, their associates, and assistants—or their equivalents. As well, this elite group should include the directors of crown corporations many of which are economic organizations, though the men who direct crown corporations are probably closer in their careers to Weber’s “official” than to the entrepreneur. They are “public servants” and thus their role expectations are quite different from those of the director of privately owned corporations.
Like bureaucracies, military organizations are formal and hierarchical. There is a range of power from the lance-corporal to the chief-of-staff. Once again an arbitrary line must be drawn. It would seem logical to consider all those holding general rank, that is, brigadiers and up in the army and their equivalent ranks in the other services, as making up the military elite. It can well be argued that in Canada, as in other Western systems, neither the bureaucracy nor the military have power because of their traditional subservience to the political system. However, it is worth observing, first, that their function gives them power inevitably and, second, that in some cases they constitute the dominant elite group. It is not intended to suggest that all elite groups are equally powerful at all times. Obviously as the conditions of survival change, certain functions become more important than others. In the permanently garrisoned state the military elite would have more power than they had in Canada in the interwar years, for example, when military organization was scanty.
The last category of elite roles has been termed the ideological. In this group are to be included those who have power over the minds of men, those, that is, who have the function of orienting members of a society to a body of beliefs by means of which all members acquire solidarity. In a sacred (or “ideational”) society these roles are assumed by a priestly caste. In an industrial society the super-empirical goals tend more to be contained in a secular ideology, and the functions of articulating them are assumed by professional propagandists, journalists, and educators. Although the Western type of society has become secular, there is still a great deal of religious appeal, so that the ideological elite in Canada would include the leaders of the various religious denominations. Other important sectors of the social structure are the communication and educational systems. In Canada, those who direct the policies of the major newspapers, the publishing houses, and the C.B.C. can be considered members of this elite group. In the educational systems the dominant roles are assumed by those in the institutions of higher learning. Thus the senates of the Canadian universities, or at least the larger ones, should be included. Because of their specialized and extensive knowledge, these individuals often influence the policies and decisions of other elite groups, by belonging to special commissions, boards, and councils. Power through ideas is a very nebulous thing and one cannot be at all sure at this stage that the ideological elite as it has been here outlined actually reflects the real situation. It may be necessary, therefore, to reconsider whether the roles of this group are really power roles.
The scheme of elite groups which has been presented here is intended as a set of analytical categories by which to study power stratification. In this way it may be possible to show how, in Canada, power roles, facilities, and rewards are allocated to different members of the society. This problem of allocation must be solved in all social systems, and the two polar types of solution have been by ascription and achievement. All societies of course have some ascriptive criteria, if only age and sex, but undoubtedly other criteria, such as kinship and ethnic relationships, religious affiliation, territorial location, membership in associations, educational experience, and the like, are also important.
The North American ideology strongly favours role allocation on the basis of achievement. The principle that the career should be open to the talented can be upheld on ethical grounds as well as on grounds of efficiency since the recruiting process should place the most able persons in elite roles. When these elite roles have been identified, and their incumbents studied, it should be possible to show to what extent allocation is by ascription or achievement and just what ascriptive criteria are important.
1IG. Mosca, The Ruling Class, tr. H. D. Kahn (New York, 1939), 50.
2R. Craig McIlvor and John H. Panabaker, “Canadian Post-War Monetary Policy, Henderson (Edinburgh, 1947), 139, 297 ff.
3E.g. S. F. Nadel, The Foundations of Social Anthropology (London, 1951), 169. Weber also made this distinction. See the discussion of the editor in Weber, The Theory of Social and Economic Organization, 189, on the difficulty of translating Herrschaft and Macht.
4See the discussion of S. Zuckerman’s The Social Life of Monkeys and Apes (London, 1932) in E. F. M. Durbin and G. Catlin, eds., War and Democracy (London, 1938).
5B. Russell, Power: A New Social Analysis (London, 1948), chap. vi.
6See R. F. Bales et al., “Channels of Communication in Small Groups,” American Sociological Review, XVI, Aug., 1951; and T. M. Mills, “Power Relations in Three-Person Groups,” American Sociological Review, XVIII, Aug., 1953. Also W. F. Whyte, “Small Groups and Large Organizations” in J. H. Rohrer and M. Sherif, eds., Social Psychology at the Cross Roads; and the discussions of leadership in G. Homans, The Human Group (New York, 1950).
7Cf. M. Fortes and E. E. Evans-Pritchard, eds., African Political Systems (Oxford, 1950), especially Preface and Introduction.
8As a diviner, the Tlingit Shaman “was often consulted as to weather, the proper time to start on a hunt, whether a certain venture would meet with success or failure and about other things.” (L. F. A. Jones, Study of the Thlingets of Alaska (New York, 1914), 159, quoted in J. J. Honigmann, The Kaska Indians: An Ethnographic Reconstruction (New Haven, Conn., 1954), 106.)
9H. Scott Gordon, “The Economic Theory of a Common-Property Resource: The Fishery,” Journal of Political Economy, LXII, April, 1954, 134.
10G. Wagner, “The Political Organization of the Bantu of Kavirondo” in Fortes and Evans-Pritchard, eds., African Political Systems, 201.
11B. Malinowski, Argonauts of the Western Pacific (New York, 1950), 510.
12Nadel, The Foundations of Social Anthropology, 131.
13Fortes and Evans-Pritchard, eds., African Political Systems. See also M. Herskovits. Man and His Works (New York, 1951), chap. 20.
14R. Merton, Social Theory and Social Structure (Glencoe, Ill., 1949), 21 ff.
15T. Parsons, The Social System (London, 1952), 126.
16C. E. Merriam, Political Power (Glencoe, Ill., 1950) chap. I.
17Parsons, The Social System, 119 ff.
18T. Parsons and E. Shils, eds., Toward a General Theory of Action (Cambridge, 1951), 202.
19R. Aron, “Social Structure and the Ruling Class,” British Journal of Sociology, March, 1950.
20E.g. Max Gluckman, “Political Institutions” in The Institutions of Primitive Society (Oxford, 1954), 77.
21C. McWilliams, A Mask of Privilege (Boston, 1948), 116.
22Mosca, The Ruling Class, chap. v.
23J. K. Galbraith, American Capitalism: The Concept of Countervailing Power (Boston, 1952), particularly chap. ix.
24“Where the populace rise at once against the never-ending audacity of elected persons . .” from Song of the Broad-Axe.
25Clark Kerr, “Industrial Conflict and Its Mediation,” American Journal of Sociology, LX, Nov., 1954, 231.
26Quoted in M. Ginsberg, The Psychology of Society (London, 1951), 163.
27Barbara Wootton, “Social Prestige and Social Class,” British Journal of Sociology, V, Dec., 1954, 373.
28A. A. Berle and G. C. Means, The Modern Corporation and Private Property (New York, 1932); monographs of the Temporary National Economic Committee (Washington); National Resources Committee, The Structure of the American Economy (Washington, 1939). See also A. R. Burns, The Decline of Competition (New York, 1936); A. B. Levy, Private Corporations and Their Control (London, 1950).
29F. W. Taussig and C. S. Joslyn, American Business Leaders (New York, 1932), particularly Table, p. 35.
30In a forthcoming article in this JOURNAL.
31Mr. George Hicks quoted in K. G. J. C. Knowles, Strikes: A Study of Industrial Conflict (Oxford, 1954), 20.
32R. Michels, Political Parties, tr. E. and C. Paul (Glencoe, Ill., 1949).
33J. Goldstein, The Government of British Trade Unions (London, 1952).
34C. W. Mills, The New Men of Power (New York, 1948), 64.
35Knowles, Strikes, 80 ff.
36N. Ward, The Canadian House of Commons: Representation (Toronto, 1950), chap. VII.