The Institutionalization of Authority

A CERTAIN ‘utopianism’ which tends to minimize the significance of authority, coercive power, and physical force in human affairs has been a conspicuous feature of a large part of modern social and perhaps particularly economic thought. One of the most striking features of Weber’s sociological work is his continual and intensive concern with the problems of this field. Never does he treat an empirical problem without explicit inquiry into the bearing of power and authority factors on it. Indeed this constitutes one of the few major axes of this whole treatment of social phenomena.

The foundations of his treatment of authority are laid near the beginning of the logical unfolding of his conceptual scheme. After developing the broadest outline of his analysis, with the orientation of action to normative order, the basic types of relationship, etc., when he comes to the treatment of organized groups by far the most prominent place is given to a type to which an element of authority is fundamental, the Verband,¹ or ‘corporate group’ as that difficult term has here been translated. Whatever the content of its interests, the distinguishing feature of the Verband for Weber is an internal differentiation of roles with respect to authority. In a sense this derives from the very nature of the orientation of co-ordinated action to an ‘order,’ the terms of the order must be carried out and enforced, which in turn requires a responsible agency of administration and enforcement. Thus from the ordinary ‘members’ of a group Weber distinguishes those who carry responsibility and authority in this connexion. These in turn he subdivides into a ‘chief’ (Leiter) with the highest authority, and those who, though in certain respects under the authority of the chief, nevertheless at the same time exercise authority over the ordinary members, the ‘administrative staff’ (Verwaltungsstab). It is safe to hold that Weber considered this basic structure to be normal for groups of any size and complexity in all fields of routine human action. Departures from it were to him limiting cases which could only obtain under exceptional circumstances and, once realized, would tend to be particularly unstable. Perhaps the principal exception is the replacement in the status of ‘chief of an individual by a ‘collegial’ body of equals, such as a ‘board’ or committee. But even here there is a very strong tendency for one individual to become at least the primus inter pares if not the actual chief. Thus in the British system of cabinet government there has been a steady process of increase in the power of the Prime Minister so that in recent times he has far overshadowed the other members of the cabinet and had full powers of appointment and dismissal over them.

This basic uniformity in the structure of groups, however, is not incompatible with a fundamental range of variation in the character of the structure of authority within it. In developing the concept of Verband Weber lays particular stress on the importance of the relation of action within it to an order. It is hence not surprising that in differentiating types of organization of authority he takes as his point of departure variation in the nature of the claim to legitimacy which is necessarily involved in an institutionalized status for the chief and his staff, and which depends in turn on the character of the order to which they as well as the members are in some sense subject.

From this starting point Weber opens his analysis with the introduction of a classification of three basic types² which, with their analytical development and empirical use, must count as one of Weber’s few most important contributions to social science. As in other connexions, he is here particularly concerned with illuminating the character of authority in the modern Western World, and hence makes the type most conspicuous in our society his point of reference, defining and characterizing the others in terms of their specific contrasts with it.

This first type is what Weber calls ‘rational-legal authority.’³ The order in question then consists in a body of generalized rules, in the type case logically consistent and claiming to cover all possible ‘cases’ of conduct within the jurisdiction of the Verband as well as to define the limits of that jurisdiction. These rules are universalistic in that they apply impartially to all persons meeting the logically formulated criteria of their definitions, and impersonal in that the status and qualities of individuals are treated as a function of the application of the generalized rules to them and so far as they do not fall within them must be treated as irrelevant. The fundamental source of authority in this type is the authority of the impersonal order itself. It extends to individuals only in so far as they occupy a specifically legitimized status under the rules, an ‘office,’ and even then their powers are limited to a ‘sphere of competence’ as defined in the order. Outside this sphere they are treated as ‘private individuals’ with no more authority than anybody else. There is thus in principle a separation of the sphere of office and that of private affairs, corresponding closely to that between the enterprise and the budgetary unit in the economic sphere. Where authority includes, as it very generally does, command over the use of property, there is clear segregation between the property of the Verband, over which the incumbent of office has certain powers by virtue of his office, and his personal possessions which are controlled according to entirely different criteria. Often this extends to segregation of the premises of work from those of private life.

Where rational-legal authority involves an organized administrative staff, according to Weber it takes the form of a ‘bureaucratic’ structure. Here each member of the staff occupies an office with a specific delimitation of powers and a sharp segregation of the sphere of office from his private affairs. Remuneration is, in the type case, in the form of a fixed salary, preferably in money form. The different offices are organized in terms of a stringent hierarchy of higher and lower levels of authority in such a way that each lower level is subject to control and supervision by the one immediately above it. This control and supervision above all includes the power of appointment, promotion, demotion, and dismissal over the incumbents of lower offices. Fitness for an office is typically determined by technical competence, which in turn may be tested by such rational procedures as examination and very generally involves a long period of formalized training as a condition of eligibility. Bureaucracy in this sense, Weber says, is by far the most efficient instrument of large-scale administration which has ever been developed and the modern social order in many different spheres has become overwhelmingly dependent upon it.⁴

The second main type which Weber takes up is what he calls ‘traditional authority.’⁵ The term is not altogether happy since its traditionalism is only one of a rather complex combination of criteria which Weber attributes to the type. In this respect, however, in contrast with the order involved in rational-legal authority which is legitimized by having been ‘enacted’ or ‘imposed’ by a legitimate agency and procedure, the system of order is treated as having always existed and been binding. Even actual innovations are justified by the fiction that they were once in force but had fallen into disuse and only now are brought back to their rightful position of authority. One of the important consequences, and symptoms, of the existence of traditional authority is that there can be no such thing as new ‘legislation.’

There is, however, not only the difference that in the one case rules are enacted, in the other traditionally received. The order has a different kind of content. It contains two main elements. One is a body of concrete rules. There may be elements of generalizations in these, but there is a great difference of degree from the rational-legal case. Conformity with the abstract logical requirements of a rationalized system is not as such a criterion of legitimacy, and many concrete prescriptions are involved with no other justification than that they have always been held to be binding. But in addition to rules governing the conduct of the members, the order underlying a system of traditional authority always defines a system of statuses of persons who can legitimately exercise authority. Such a status is different from an ‘office.’ It does not involve specifically defined powers with the presumption that everything not legitimized in terms of the order is outside its scope. It is rather defined in terms of three things. There are, first, the concrete traditional prescriptions of the traditional order, which are held to be binding on the person in authority as well as the others. There is, secondly, the authority of other persons above the particular status in a hierarchy, or in different spheres—as when a king assumes judicial authority when he is personally present—and finally, so long as it does not conflict with either of these sets of limitation, there is a sphere of arbitrary free ‘grace’ open to the incumbent. In this sphere he is bound by no specific rules, but is free to make decisions according to considerations of utility or raison d’état, of substantive ethical ‘justice,’ or even of sheer personal whim.

So long as the incumbent of such a status does not act counter to the traditional order, or infringe upon the prerogatives of his hierarchical superiors, loyalty is due, not to the ‘order’ as such, but to him personally. He is not restricted to specified powers, but is in a position to claim the performance of unspecified obligations and services as his legitimate right. There is thus notably no clear-cut separation between the sphere of authority and the individual’s private capacity apart from his authority. His status is a ‘total’ status so that his various roles must be far more stringently integrated than is the case with incumbency of an office.

Two implications of this situation are, perhaps, particularly important. Property tends not to be stringently distinguished as between the ‘means of administration’ which, in a rational-legal structure are strictly limited to use in an official capacity, and personal property which can be used for personal needs and desires. Furthermore the property aspect in either or both respects is far less likely to be segregated from other aspects of the individual’s status. Above all the enjoyment of property rights over things or persons generally carries with it personal authority, usually with at least an element of political jurisdiction, notably over persons and land. So far then, as the ‘means of production’ are appropriated within a system of traditional authority there is a strong tendency for it to be accompanied by various forms and degrees of unfree personal status for the persons subject to authority. Secondly there is a strong tendency for the status of authority to be fused with other aspects of status which in a rational-legal system are normally treated as part of the individual’s ‘private’ sphere. Perhaps the most important case of this is kinship. The result would be to treat status by kinship and in a system of authority as identical and hence inseparable. This involves, of course, the hereditary principle, which is always likely to develop in systems of traditional authority.

Weber presents a subclassification⁶ of the different kinds of traditional authority in terms of variation in the development and role of the ‘administrative staff.’ His base line is given in the concepts of ‘primary’ gerontocracy and patriarchalism where either a group or an individual as the case may be, occupy a position of authority but without a significant degree of independent control of an administrative staff. If they are called on for decisions they are dependent, for their enforcement, on the cooperation of persons whose status and functions are traditionally fixed, and do not to a significant degree fall within the scope of the ‘arbitrary’ grace of the incumbent of authority. One might say that the holder of authority is more of a ‘sage’ or wise man than an ‘executive.’ This case tends to be the one which is most stringently bound to the fulfilment of concrete traditional prescriptions and gives little scope for the large-scale development of power systems.

The sphere of arbitrary free choice, however, contains the seeds of a possible development. Where this is used to develop a complex administrative staff under the personal control of the chief, and not involved in a system of traditionally stereotyped statuses with which he cannot interfere, Weber speaks of ‘patrimonialism.’ There are, of course, indefinite gradations of degree. Weber lays stress on two crucial points of transition. One touches the recruitment of the administrative staff, the shift from the recruitment from ‘patrimonial’ sources, that is persons otherwise in a state of dependency on the chief through kinship, serfdom, a client relationship, etc., to recruitment of persons otherwise altogether independent of him. This of course greatly widens the sphere of arbitrary power since the chief is not bound in controlling them by the prescriptions of their independent traditional status. The other range of variation involves the attitude toward persons subject to authority. In one case the person in authority acts ‘on behalf’ of the collectivity and the typical individual is a ‘member’ of the collectivity with definite traditional rights and expectations, which define limitations on arbitrary will. The other case is that where they are treated as ‘subjects’ and the chief tends to treat his position of authority as a personal prerogative, almost as his private property, and the subjects hence as instruments in carrying out whatever projects he may have in hand. The extreme type case of this personal absolutism on a traditional basis of legitimation Weber calls ‘Sultanism.’

The effect of the transition from patriarchalism in the direction of patrimonialism is to emancipate an important part of the structure of authority from the direct control of tradition. The position of authority as such is still traditionally legitimized, but not the detailed structure of carrying it out which is, on the contrary, a ‘right’ of the chief to do what he will within his sphere of personal prerogative. One effect of such emancipation from detailed traditionalism is of course immensely to widen the area of free individual action and planning. But another in many respects equally fundamental consequence is to unloose a potential struggle for power which under ‘primary’ traditionalism could have no possibility of developing. The aspect of this struggle to which Weber pays particular attention is that between the chief and the members of the administrative staff. It is not common for a very high level of personal absolutism to be maintained for a long period. The rule is for important limitations on the chief’s power to be imposed through pressures exerted by the administrative staff.

This is usually a matter of a struggle for the control of ‘advantages’ the most important of which are power as such, the ‘means of administration’ as Weber calls them, and remuneration. In a traditional framework this tends to take the form of a determination of the total status of the individual concerned, not merely of his office. If the imposition of a limitation on the chief is successful there is a strong tendency for it to become traditionalized as an established, even appropriated, right of the incumbent of a status in the structure of authority. Under most conditions appropriation will tend to extend beyond the lifetime of the particular incumbent, to become hereditary, and thus to fuse the structure of authority with the class structure. Under other conditions, as in China, the right established may be that of the members of a class with a wellintegrated cultural tradition, but without individual appropriation beyond the period of office.

Under traditional conditions the tendency is for all three elements of advantage to be fused in the perquisites of a single status, and all together to be either under the control of the chief, or in various ways and degrees kept beyond his control. There are, however, two main types which, with reference to the property interests involved, Weber calls ‘benefices’ and ‘fiefs’ respectively.⁷ A benefice is a bundle of rights granted to the individual on a ‘personal’ basis, that is without hereditary succession, and as a free grant of the chief. A ‘fief’ on the other hand is granted formally by virtue of a contract, which implies specific reciprocal obligations on the part of the chief, and a standard of the honour of the holder which the chief must respect. A feudal contract is, as Weber points out, peculiar in that it involves a total status and diffuse mutual obligations of loyalty. It is thus not like a modern business contract.⁸

The contrast between rational-legal and traditional authority is associated for Weber with that between formal and substantive rationality. A system of rational-legal authority is highly favourable to the development of a formalized legal system, while that of traditional authority favours two types of substantive ethical norms, the traditionally bound embodied in concrete traditional precepts, and the free, exercised within the sphere of arbitrary personal will of the chief. One of the obstacles to the development and maintenance of rational-legal authority is the extent to which this legal formalism offends the sentiments of ‘substantive justice’ in a population.

Another important connexion is that involving the status of money. Command over money funds, both in the means of administration and in remuneration, is the economic mode of provision most appropriate to a system of rational-legal authority precisely because it is the most mobile and because it is most easily dissociated from the individual’s status in the other, the ‘private’ respects. A system of provision by benefices or fiefs in kind strongly favours the pattern of traditional authority.

Both the first two types of authority are, for Weber, modes of organization appropriate to a settled permanent social system. Though subject, like all human arrangements, to change, they are of specifically ‘routine’ character. The third, the ‘charismatic’ type, differs in precisely this respect. It is, by definition, a kind of claim to authority which is specifically in conflict with the bases of legitimacy of an established, fully institutionalized order. The charismatic leader is always in some sense a revolutionary, setting himself in conscious opposition to some established aspects of the society in which he works. Furthermore this Ausseralltäglichkeit, this emancipation from routine, is a note which runs all through Weber’s treatment of charismatic authority⁹ and its organization, notably with respect to modes of provision with the means of administration and to the status of members of the administrative staff.

The ‘deviance’ of the charismatic leader is not, however, by any means either mere eccentricity or the indulgence of purely personal wishes. He introduces, rather, a pattern of conduct conformity with which is treated as a definite duty. By virtue of it the leader claims moral authority and hence legitimacy for giving orders to his followers, or whoever falls within the scope of the pattern.

As Weber treats charisma in the context of authority, its bearer is always an individual ‘leader.’ His charismatic quality has to be ‘proved’ by being recognized as genuine by his followers. This is not, however, as Weber is careful to point out, the ordinary case of leadership by ‘consent’ of the led, in the usual democratic meaning. The authority of the leader does not express the ‘will’ of the followers, but rather their duty or obligation. Furthermore, in the event of conflict there can in principle be only one correct solution. Majorities, if employed at all, are given authority only because they are thought to have the correct solution, not because a greater number have as such a greater right to prevail. And the leader does not compromise with his followers in a utilitarian sense. Recognition by them is interpreted as an expression of the moral legitimacy of his claim to authority.

Above all this claim is one to impose obligations in conflict with ordinary routine roles and status. Hence the member of the administrative staff cannot occupy either an office in the rational-legal sense or a traditionalized status. He is usually a personal disciple, actuated by enthusiasm for the ‘cause’ and by personal loyalty to the leader or both. The leader usually attributes charismatic qualities to him, and assigns him particular ad hoc missions to perform. On the one hand he can have no established ‘rights’ in his status, particularly against the leader. On the other there can be no inherent limitations on the scope of his authority or functions such as are essential to the pattern of office. The limitations can only be defined by the leader, and are inherently unstable, being settled in terms of momentary exigencies.

Corresponding to the two great fields of charismatic activity for Weber, religious proselytizing and the use of force, there are two primary sources of support of charismatic movements, both for provision with the means of administration and for remuneration of the followers, namely formally free gifts and ‘booty.’ Both of these forms are specifically outside the range of routinized economic provision and can only become permanent and stable sources of income through a profound change in their character.¹⁰

In Weber’s treatment perhaps two points stand out about charisma besides the fact that it is a source of legitimate authority, namely that it is a revolutionary force, tending to upset the stability of institutionalized orders, and that in the nature of the case it cannot itself become the basis of a stabilized order without undergoing profound structural changes.¹¹ As a result of these changes it tends to become transformed into either the rational-legal or the traditional type.

The initial source of its revolutionary character lies in setting up the authority of an individual against the established order, the office or traditionalized status of those originally in authority. But if the ‘movement’ secures sufficient recognition to have the prospect of permanent organization the successors of the original leader cannot in the nature of the case base their claim to legitimacy on the same grounds. Hence the problem of succession,¹² both of who shall succeed and of the pattern of determination of the legitimacy of his status, is crucial for all charismatic movements. The functional problem is that of maintaining the authority of the original point of reference—as in a divine mission—and yet meeting the changed conditions. With respect to the pattern it may take the form of hereditary succession (Gentilcharisma), of succession in an office (Amtscharisma), or a succession by an unbound process of individual selection. There is an important functional interest in eliminating arbitrary elements so far as possible. In another connexion the leader himself may have the primary role in designating his successor and the pattern of succession, or in varying ways and degrees the decisions may be participated in by the members of the administrative staff or the total membership of the group of followers.

Though Weber does not analyse it in detail here,¹³ of course the character of the system of ideas in terms of which the charismatic claims of the movement is formulated has an important influence on the way in which this routinization works out, especially on the pattern in which authority is held by succeeding individuals. Thus the fundamentally important development of the charisma of office in the Roman Catholic Church would hardly have been possible if grace, embodied in the sacramental authority of the priesthood through apostolic succession, had not been conceived as something impersonally separable from particular individuals or lines of descent. This objective character of sacramental authority was an indispensable condition of the development of priesthood as an office rather than as an hereditary status. Similarly in the political sphere the Roman imperium was an objective power which could be transferred from one individual magistrate to another, it was not conceived as the prerogative of a status as such.

The same fundamental alternatives which arise in connexion with the problem of succession for the leader are involved all through the structure of authority. For Weber the fundamental question is whether the administrative staff will tend to take on more the character of a band of patrimonial retainers and hence lead in a traditional direction, or that of a group of officials in a rational-legal direction.

Another basic field of routinization is that involved in the relation of the movement to economic affairs, especially to provision with the means of administration and remuneration of the administrative staff.¹⁴ Provision by free gifts is inherently unstable. To meet routine needs it must be regularized in the form of a system of obligations on the part of the donors which from being merely ethically obligatory become to a greater or less extent factually compulsory. Thus they tend to take on either the character of taxes, as in the established Christian churches, or of some kind of ‘liturgies,’ of compulsory payments or services. The need of insuring the regularity of such sources of income often exerts a pressure in the direction of limiting the personal freedom of the obligated groups.

Finally there is a fundamental need for regularizing the status of the members of the administrative staff—it cannot long continue just a group of persons who are from time to time assigned ad hoc missions by the chief. This takes the familiar directions of developing offices or of developing traditionalized statuses with benefices or fiefs as their mode of support. It should also be remembered that precisely because of the instability inherent in the position of a successful charismatic movement, the way is open for a struggle for power within the movement, and for the appropriation of important rights by the members of the administrative staff.

In the process of routinization the charismatic element does not necessarily disappear. It becomes, rather, dissociated from the person of the individual leader and embodied in an objective institutional structure, so that the new holders of authority exercise it at second remove as it were, by virtue of an institutionally legitimized status or office. This points to certain difficulties in Weber’s analysis which will have to be briefly discussed later.

What is, for Weber, characteristic of the modern institutional order, is the relative predominance of the pattern of rational-legal authority. This is above all true of the modern state. Only in Rome has there been an at all comparable development, and this was less fully elaborated in a number of respects. Furthermore, this predominance of rational-legal authority is not a fortuitous curiosity of modern Western civilization— it is one of its fundamental characteristics, closely interdependent with a great many others. The free market economy could not function without it on an at all comparable scale. Such fundamental things as personal freedom, and the most important liberties as of speech, scientific investigation, the press, are basically dependent upon it.

Moreover, as is true of all the main institutional characteristics of the modern order, rational-legal authority is relatively unstable. If there is one safe generalization of sociology it would seem to be that no structuralization of human relationships is ‘foolproof’ and immune to change. Such institutions exist only in the fantasies of Utopian dreamers such as the heralds of the Tausend jähriges Reich. But to a considerable extent each particular institution has its own peculiar sources of instability and hence its specific tendencies to change into other specific forms. Of the two forms of ‘routine’ organization of authority Weber certainly considered the traditional to be the more stable, and that there was a basic tendency for the rational-legal form to break down into the traditional, either directly or through the intermediate influence of charismatic movements.

One approach to the analysis of this problem, for which there is much material in Weber’s discussion and observations, starts with the fact that a system of rational-legal authority can only operate through imposing and enforcing with relative efficiency, seriously frustrating limits on many important human interests, interests which either operate, independently of particular institutions, in any society, or are generated by the strains inherent in the particular structure itself. One source of such strain is the segregation of roles, and of the corresponding authority to use influence over others and over non-human resources, which is inherent in the functionally limited sphere of office. There are always tendencies to stretch the sanctioned limits of official authority to take in ranges of otherwise ‘personal’ interests. In other words this form of institutionalization involves a kind of ‘abstraction’ of a part of the human individual from the concrete whole which is in a certain sense ‘unreal’ and hence can only be maintained by continual discipline. A particular case of this is the tendency of persons in authority to claim obedience, and for this claim to be recognized, on a personal basis, as John Jones, rather than on the basis of office as such. It is then not the impersonal order which is being obeyed, but the personal prestige of the incumbent. The segregation required by rational-legal authority is subtle and difficult to maintain.

In our society we distinguish sharply between ‘personal’ and ‘official’ capacities. Within a considerable area the personal sphere is to us one which is thought of as altogether uninstitutionalized. But this is by no means even predominantly the case. It also includes some elements of institutionalized status which, like those of kinship and marriage, are formally sanctioned in the law, but do not constitute incumbency of an office, or such other functionally specific roles as those of employee. But still more it includes a very wide area of ‘informally’ institutionalized status such as that of social class and ethnic origin in the United States. There is reason to believe that the scope of ‘purely personal’ behaviour in the uninstitutionalized sense is narrowly limited by the functional requirements of social systems, and that what we in Western society have of it, which is exceptionally broad, is to a considerable extent made possible only by the ‘protection’ of the individual by rational-legal norms. The tendency to break down the segregation between the official and the personal spheres will not, therefore, probably result mainly in the permanent increase of individualistic freedom, but rather in the increase of institutionalization of individual functions and status in the ‘total’ status form, in other words, of features Weber treats as typical of traditional authority.

A second fundamental consideration is that discipline and authority, probably always in any large-scale permanent system, generate various forms of resistance and resentment. It is naturally a condition of the continued existence of an institutionalized system of authority that it should, with relative efficiency, ‘take care’ of this resistance and prevent it from undermining the system of authority. But this control is probably always relatively precarious, and under favourable conditions the loopholes in the system present opportunities for cumulative change. A system of rational-legal authority is not, as the history of such things goes, necessarily particularly severe in its repressive measures—indeed it generally involves important mitigations of the severity of other forms. To some extent this is sometimes a source of strength in spite of the fact that it is deprived of otherwise effective instrumentalities of discipline. But at the same time it probably also in other directions generates tensions peculiar to itself. The very fact that its sphere of authority is functionally limited deprives it of the support of certain motives which contribute to the solidarity of other systems, notably motives of personally loyal attachment to particular individuals and Gemeinschaft groups. Above all, for a variety of reasons, it tends to generate widespread feelings of insecurity.

One immediate effect of this may well be to emancipate individuals or groups of them from the control of rational-legal structures.¹⁵ There are certain possibilities that this emancipation may result in a more or less stable institutionalization of anti-authoritarian patterns which protect the freedom of the individual. But this is only one possibility and in most cases not the most likely. There are, as has just been pointed out, functional reasons why limitations on the sphere of personal freedom tend to become institutionalized, in terms of authority as well as other forms. The probability is then that a break-down of rational-legal authority through successful defiance of it on the part of its ‘objects’ will result in its gradual replacement by other forms, notably the traditional. This may take the form of a gradual change in the character of authority relationships within the Verband, or in the subjection of the individual to authority in a rival organization. Thus the industrial worker may escape the authority of his employer only to fall under that of union leaders.

This functional tendency undoubtedly is reinforced by important elements of motivation, which Weber did not directly analyse. One way to mitigate the anxiety associated with a state of psychological insecurity is to abdicate individual responsibility in favour of dependence on a source of authority. This is especially likely to be of the traditional type since on these deeper psychological levels the prototype of a state of security is that of dependence on the parent. A rational-legal officeholder is not likely to be an effective parent substitute. In general modern clinical psychology has taught us that attitudes toward authority may be deeply ambivalent. The same individual who in one context is notably rebellious against some forms of authority will often readily submit to even more stringent control if it occurs in a somewhat different context.

The direct transition to traditional patterns is not, however, the only path by which a system of rational-legal authority can change fundamentally—it can also give way to charismatic movements. Weber had relatively little to say directly about the conditions which favour the development of charismatic movements—he was more concerned with their character and consequences. But a good deal of evidence has accumulated on this subject which fits admirably into his analysis. Any situation where an established institutional order has to a considerable extent become disorganized, where established routines, expectations, and symbols are broken up or are under attack is a favourable situation for such a movement. This creates widespread psychological insecurity which in turn is susceptible of reintegration in terms of attachment to a charismatic movement. In addition to relatively generalized and diffuse ‘anomie’¹⁶ and insecurity, there are generally specifically structured sources of strain and frustration which may have much to do with the definition of the specific content of effective charismatic appeal.

There is reason to believe that a social order of which a system of rational-legal authority is an important part is considerably more subject to this kind of disorganization than is a highly traditionalized society. The rationalized structures are never exhaustive of the whole social structure, and there is often important strain between them and the parts which are predominantly traditionalized, a strain which seems to provide the principal basis in fact for the ‘cultural lag’ theory which has been so popular in recent years. ‘Reason’ is, as Weber several times¹⁷remarks, an inherently dynamic force subject to continual change, and hence has a strong tendency not to permit the development of settled routines and symbolic associations which would minimize psychological strain. At any rate there is ample empirical evidence of the susceptibility of our society to the type of movement which Weber describes, all the way from the prevalence of innumerable fads through the proliferation of many kinds of religious cults¹⁸ to the Communist and National Socialist movements themselves which are grand scale movements involving charismatic authority in the political field.

The immediate result of a charismatic movement will be, so far as its influence extends, to undermine the stability of the prevailing institutional order, in this case of rational-legal authority. This is partly because its sources of legitimacy are challenged, partly because the movement itself requires forms of organization which are in conflict with the established. But the charismatic basis of organization is inherently unstable and temporary. The question of its long-run effects will depend on whether the process of routinization takes the traditional or the rational-legal direction. This in turn depends on a complex variety of factors, but on general grounds it may be said that there is a presumption in favour of the path of traditionalization, both because it is in general more stable and because the specific strains generated in a rational-legal system would, by contrast, strengthen the forces biassed in that direction.

In the first instance there is the question of the ideal patterns involved in the sources of legitimacy of the charismatic movement. In this case, as between the two most important charismatic political movements of our time, there seems to be an important difference. Communism is, on the whole, in this respect a child of the rationalistic enlightenment of modern times—its basic values are associated with science, human equality, technology, and contain important elements of ethical universalism, and give indeed some basis for specificity of functional roles. National Socialism, on the other hand, represents to a far greater degree a ‘fundamentalist reaction,’ a reassertion in revolutionary form of precisely those traditional values which have been most injured and threatened by the development of the rational-legal institutional order itself in the first place. Hence its character would seem to point to a far greater likelihood of its leading in a traditionalistic direction.

But the character of the relevant ideal patterns does not stand alone. Above all there is the question of economic provision and of the status of the administrative staffs. In National Socialism, for instance, it is clear that provision was very largely in terms of ‘booty,’ for the first few years derived mainly from the expropriation of the established classes and organizations within Germany.¹⁹Since then the spoils of conquest played a tremendous part. It would seem altogether possible that, if the movement escapes collapse through military defeat, this may well develop into a system of benefices in the hands of the magnates of the party. Similarly, there is the question of what is to become of the vast organization of party functionaries.²⁰There were several indications of a bitter struggle for power within the party—most dramatically in the blood purge of June 1934—even during the lifetime of the original charismatic leader.

The prevention of appropriation of important rights on the part of the administrative staff, especially after the problem of succession arises, would presuppose a unity and effectiveness of the highest central authorities which, considering the heterogeneity of its composition and the intensity of the personal rivalries known to exist, would seem most unlikely. Above all it is crucially important that some elements of this administrative staff have control over armed force more or less in their own right. The Gestapo and the SS might well turn out to play a role analogous to that historically of the Praetorian Guard or of the Janissaries. For the administrative staff to eventuate in a pure rational-legal bureaucracy would presuppose a continuity of central control which is most unlikely to be realized.

On the background of his general analysis of authority, Weber, though in fragmentary fashion, outlined a significant analysis of certain aspects of modern democracy.²¹In the first place he calls attention to the fact that two of the most important types of check on centralized authority, the separation of powers and the presence of collegial bodies in place of monocratic positions of authority, are primarily associated with aristocratic rather than democratic regimes. They have originated largely in the process of appropriation by groups in the administrative staff, of powers and rights in the course of a struggle with the chief. Thus above all, these famous features of the British constitution were the result of the ability of the nobility and gentry to put checks on the power of the Tudor and Stuart monarchies. In another field, the peculiar supremacy of law in England was, though partly the work of Parliament, also in no small measure a result of the influence of the organized legal profession—a specifically privileged group—in restraining the arbitrary action of the crown.

The development of rational-legal authority, with bureaucratic administration, is both dependent on the breakdown of traditionalized particularistic privileged groups and in turn itself has a levelling influence, in that it treats social class by birth and other privileged statuses as to a large degree irrelevant to status in the system of authority. Along with this levelling influence, both the struggle for power in the Western national states, and the pressure of certain of the ideological patterns have tended to weaken the control of a centralized authority over an administrative machine. Above all, as Weber illustrates from the United States in particular, elective officials in the nature of the case cannot be subject to the same order of discipline that appointive officials are. And the democratic system has, through the spoils system, on occasion made great inroads on bureaucratic organization even in the sphere of appointive administration.

But the combination of a relatively levelled undifferentiated mass upon which to base political support, rather than groups with fixed and differentiated status, and the democratic principle of appeal to the electorate for legitimation, has opened the way to the emergence of leaders with a wide personal appeal. This in turn is associated with the party system.²²A party leader can, to a greater or less degree, approach the type of a charismatic leader, and use of the methods of appeal and the forms of organization of his power appropriate to that type. Though, for the most part as in England and the United States, modern party leadership has only in a minor degree taken a revolutionary charismatic form, elsewhere it has done so pre-eminently. To Weber the demagogic party leader, once he became genuinely charismatic with the implied denial of the legitimacy of the position of any rivals, either within or outside his party, could readily become a dictator with a consequent shift in the character of the authority system. This need not involve a drastic break with the democratic reference of legitimacy. As Weber pointed out,²³the plebiscite could be used as a major symbol of the democratic legitimation of such a leader, and would tend to be effective relatively regardless of its actual genuineness as an expression of the popular will. Weber saw this already happening in the case of the two Napoleons in France, and evidently expected it to happen on a larger scale again. Though Weber died when Hitler was no more than a recently demobilized and dissatisfied German war veteran, he went far to predict the political pattern in which the Nazi movement eventually developed out of the traditional patterns of Western democracy. Indeed, it can scarcely be doubted that this link with democratic sentiments, through continuity with the ‘party’ system and even including the specific instrument of the plebiscite, was essential to the elevation of the Fascist dictatorships to their position of power. Up to the end Hitler liked to refer to the allegedly arbitrary interference of the British with the ‘self-determination’ of the populations of Europe. To incorporate the will of the German people is a claim which the Nazi movement made with some plausibility—when it came to those of the occupied territories it requires a stretching of the meaning of categories which only a highly emotional devotion to the Nazi faith could make possible.

As in the case of his analysis of economic institutions, there are a number of points in the analysis of authority where difficult critical problems are raised. Perhaps the most general is the question of how far the several different variables involved in the differential criteria of the three types of authority in fact necessarily vary together to the extent implied in making this classification exhaustive of the major empirical possibilities. Thus the functional specificity of roles in rational-legal authority is contrasted with its diffuseness in the other two types. This criterion of distinction seems to be satisfactory but, as has been pointed out above, it is questionable whether functional specificity of roles is always associated with bureaucratic organization. Similarly it is questionable whether, in routine structures, diffuseness is necessarily associated with traditionalism or charisma. A non-traditional diffuseness seems to be a most important element of our patterns of friendship, marriage, and even other aspects of kinship. It is true that the pattern of romanic love has many charismatic features, but marriage is by no means exhausted by this pattern. In quite another connexion it seems questionable whether a charismatic appeal is necessarily associated with such particularistic patterns of loyalty to an individual leader as Weber tends to indicate. There seems to be an important distinction between the pattern of legitimacy and the individual bearer of authority under it, which Weber does not sufficiently work out and emphasize.

The concept of charisma in particular seems to involve important difficulties. In most of his explicit treatment of it Weber associates it most specifically with the claim to authority of an individual personal leader and treats it on this structural level, as a matter of the differentiation of roles of persons, not an element of the structure of systems of action.²⁴In another connexion, however, where he attempts to systematize his categories for the generalized analysis of religious phenomena,²⁵he treats charisma as a quality, not necessarily only of persons, but of non-empirical aspects of the situation of the action, of, in a special technical sense, a ‘supernatural’ order, recognition of which underlies the moral legitimacy of normative rules generally. The concept, that is, becomes exactly equivalent to Durkheim’s ‘sacred.’ There are indications of this ambivalence²⁶ in Weber’s treatment of charismatic authority in that in the routinization of charisma, the charisma of the original leader does not disappear, but becomes ‘objectified’ as a quality of the order developing from a charismatic origin, as the charisma of office or of a ruling house. What seems to have happened is that, in formulating the concept for the treatment of authority Weber mixed two different levels of analysis, one the concrete structure of a certain class of movements of social change, with special reference to the role of a type of person, the other, analytically the more basic level, of the analysis of systems of action, of the basic elements of orientation of the action of the individual to his situation and to the normative patterns governing his action. What on the first level is a characteristic only of certain specific types of leaders of specific kinds of movement, becomes on the second an element of all systems of action. From the second point of view all authority has a charismatic basis in some form. The special type Weber calls charismatic is then characterized by specific kinds of content, modes of embodiment and relation to the basis of legitimacy of the established institutional order in conflict with which it stands.

This suggests a further difficulty, parallel to that encountered in Weber’s treatment of economic institutions. All of the most important theoretical elements of Weber’s analysis of authority are of generalized significance for the whole field of social relationships. Yet he tends to treat the sphere of the organization of authority as analytically autonomous in a way which obscures this continuity of pattern throughout the social system as a whole. What Weber seems to have done is illegitimately to hypostatize a certain mode of structuring of social systems as an independent entity. To be sure he is continually calling attention to its complex interrelations with other structures, such as the economic, but still in such a way as to presuppose a kind (not necessarily a degree) of independence which is unreal. This is a particular example of the consequences of Weber’s failure to employ systematically the concept of a generalized social system on all the main levels. A generalized account of the principal variables patterning social relationships is logically prior to the treatment of such relatively specialized structures as those of authority and of economic allocation. The analytical continuity which runs through them in Weber’s treatment is not the outcome of such a careful systematic analysis, but rather of the ad hoc necessities of his empirical work.

These strictures should, however, in this case as in others, not be permitted to bias the perspective in which Weber’s work is seen. If, along with many others equally possible, they were systematically taken account of, they might well lead to considerable refinement of Weber’s analysis, to the elimination of a number of particular difficulties, and to some alteration of the perspective in which his empirical results were seen. This would, however, constitute going on from where Weber left off, not ‘refuting’ him and substituting another theory. Probably Weber’s analysis of authority even as it stands constitutes the most highly developed and broadly applicable conceptual scheme in any comparable field which is available, not only in the specifically sociological literature, but in that of social science as a whole.

1 Chap, i, sec. 12, pp. 145 ff.

2 The classification is given in sec. 2, pp. 328-9. This scheme has been very little discussed in the literature in English. The fullest discussion is in Goldhamer and Shils, ‘Types of Power and Status,’ American Journal of Sociology, September 1939.

3 Sees. 3-5, pp. 329 ff.

4 Weber’s formulation of the characteristics of bureaucratic organization, which has become a classic, raises some serious analytical difficulties in the treatment of social structure. It is the present writer’s opinion that he has thrown together two essentially different types which, though often shading into each other, are analytically separate. Taking account of this distinction would considerably alter the perspective of Weber’s analysis on a number of empirical problems.

The problem may be approached by noting the importance Weber attaches to technical competence (p. 335) as a basis of bureaucratic efficiency, and his statements that bureaucratic administration is ‘essentially control by means of knowledge’ (p. 337). Now the terms ‘knowledge’ and ‘technical competence’ immediately suggest the ‘professional’ expert such, for instance, as the modern physician. But it is furthermore suggestive that Weber nowhere calls attention to the fact that medical practice like various other professional functions in the Western World has not been predominantly organized in bureaucratic form, but rather, at least to a large extent, in that of ‘private practice.’ In private practice the physician does, to be sure, exercise a kind of authority, he issues, as we often say, ‘orders’ to his patients, and there is a rather high probability in most cases that these orders will be followed. This authority rests, fundamentally, on the belief on the part of the patient that the physician has and will employ for his benefit a technical competence adequate to help him in his illness. That is he has knowledge and skill which the patient does not have, and cannot criticize in detail so that the patient must take his doctor’s advice or orders ‘on authority.’

It is true also that this position of authority is not a matter wholly of the individual impression made by the particular physician, but is institutionalized. Its possessor is socially categorized through such instrumentalities as formal training, the M.D. degree, a license to practice, etc. But the distinctive thing is that this institutionalization does not carry with it coercive powers and the physician does not occupy an office in Weber’s sense. His getting his orders obeyed depends entirely on securing the voluntary consent of his patient to submit to them.

Weber, however, started from the organization of authority within a corporate group. The fundamental model he had in mind was that of legal ‘powers,’ particularly powers of coercion in case of recalcitrance. The position of the exerciser of authority of this sort is legitimized by his incumbency of a legally defined office. It is not logically essential to it that its exerciser should have either superior knowledge or superior skill as compared to those subject to his orders. Thus the treasurer of a corporation is empowered to sign checks disbursing large funds. There is no implication in this ‘power’ that he is a more competent signer of checks than the bank clerks or tellers who cash or deposit them for the recipient. Legal ‘competence’ is a question of ‘powers’ in this sense, technical competence is of a different order.

Of course persons occupying positions of legal authority, the higher the more so, generally are in some important sense and degree ‘superior’ to those under them in respect to ability and achievement. They can successfully do things which the others, if the places were changed, could not in fact accomplish. But this is not logically essential to the definition of the type of structure, and holds of only a part of the actual functional content of office. Furthermore it does not follow that typically the superior ability and potentiality of achievement of the ‘executive’ is of the same order as the technical competence which is decisive in such fields as medicine or engineering. The differentiation of the two kinds of functional superiority, even only on an ideal type basis is, however, a difficult and subtle problem.

There can obviously be no such thing as the ‘private practice’ of the functions of bureaucratic office. But professional services are often, indeed increasingly, carried out in complex organizations rather than by independent individuals. There is, however, considerable evidence that when this is the case there are strong tendencies for them to develop a different sort of structure from that characteristic of the administrative hierarchy which Weber has, in most respects, classically described in his discussions of bureaucracy. Instead of a rigid hierarchy of status and authority there tends to be what is roughly, in formal status, a ‘company of equals,’ an equalization of status which ignores the inevitable gradation of distinction and achievement to be found in any considerable group of technically competent persons. Perhaps the best example of this tendency, which Weber curiously enough seems to have overlooked in its bearing on this problem, is to be found in the universities of the modern Western World. Much the same will, on close examination, be found to be true of the professional staffs of such organizations as hospitals or law firms.

It is probable that Weber’s neglect to analyse professional authority is associated with a tendency to overemphasize the coercive aspect of authority and hierarchy in human relations in general, important as it is in particular cases. This was a healthy reaction against the common utopianism in these respects of so much of social thought. But working out the implications of this point would, as has been noted, considerably alter the perspective of many of Weber’s empirical generalizations.

The above selects only one among several sources of difficulty in Weber’s formulation. In fairness to him it should, however, be remembered that the exposition of his views in the text is highly schematic, neglecting many of the complications he himself called attention to.

5 Sees. 6-9, pp. 341 ff.

6 Sec. 7 (a), pp. 346 ff.

7 Sec. 8, pp. 351 ff.

8 It is a Statuskontrakt not a Zweckkontrakt. See Rechtssoziologie, Wirtschaft und Gesellschaft, p. 413.

9 Secs. 10-12, pp. 358 ff.

10 P. 362.

11 Pp. 363 ff.

12 Pp. 364 ff.

13 This is a principal theme of the studies on the Sociology of Religion.

14 Cf. especially sec. 12, pp. 367 ff.

15 The restriction of production in modern industry is an excellent example of this kind of resistance to authority. Though doubtless considerably increased by the deliberate policies of labour organizations, there is evidence that it goes considerably deeper than that and appears spontaneously where change imposed by rational measures from above upsets the settled routines and informal social relationships of a working group. See Roethlisberger and Dickson, Management and the Worker.

16 A term used by Durkheim, especially in Le Suicide.

17 Cf. for instance p. 363.

18 For instance, Christian Science, Buchmanism, Father Divine, etc.

19 Not only the property of the ‘rich’ but, for instance, the funds of trade unions have been, formally or informally, expropriated.

20 On the sociological structure of the Nazi party see Hans Gerth, ‘The Nazi Party: Its Leadership and Composition,’ American Journal of Sociology, January 1940.

21 Cf. especially secs. 14 ff. Cf. also the essay Politik als Beruf in Gesammelte politische Schriften, and the editor’s articles ‘Max Weber and the Contemporary Political Crisis,’ Review of Politics, January and April 1942.

22 Cf. sec. 18 ff., pp. 407 ff.

23 Pp. 387-8.

24 In the relevant sections of the present volume and in Wirtschaft und Gesellschaft, part iii, chaps, ix and x.

25 Religionssoziologie, Wirtschaft und Gesellschaft, part ii, chap. iv.

26 The problem is more fully discussed in the editor’s Structure of Social Action, chap. xvii.