This content downloaded from on Tue, 11 Mar :51:19 PM All use subject to JSTOR Terms and Conditions

Transcription

1 Weber and the Classification of Forms of Legitimacy Author(s): Craig Matheson Source: The British Journal of Sociology, Vol. 38, No. 2 (Jun., 1987), pp Published by: Wiley on behalf of The London School of Economics and Political Science Stable URL: Accessed: 11/03/ :51 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Wiley and The London School of Economics and Political Science are collaborating with JSTOR to digitize, preserve and extend access to The British Journal of Sociology.

2 * * * * t Craig Matheson Weber and the classification of forms of ABSTRACT 'Weber and the classification of forms of legitimacy' This article examines the various ways in which relationships of command and obedience are legitimized and proposes an eightfold classification of the principles of legitimacy based upon the rationale for attributing legitimacy to command and obedience. Eight major sources of legitimacy are distinguished, namely, conventlon, contract, unlversa prlnclp es, sacrec ness, expertlse, popular approval, personal ties and personal qualities. This classification is then used as the basis for a critique of Weber's typology of forms of domination. It is argued that Weber's threefold classification of claims to legitimacy which formed the basis of his classification of types of domination actually comprises five separate principles of legitimation, namely, convention, sacredness, personal ties, personal qualities and rationality. Weber's threefold classification of forms of domination is not mistaken on this account however, since the three sets of principles which Weber's classification of forms of legitimacy comprise occur in conjunction with three specific types of command-obedience relationship. Contrary to Weber's argument however, the structure of dominatiorl in the cases of traditional and legal authority is not derived from the mode of claiming legitimacy, rather, the reverse is true. This article examines the notion of'legitimacy' as it is applied to relationships of command and obedience and seeks to identify the major ways in which these relationships are legitimized. Attention will be focused in particular upon Max Weber's well known typology of the forms of legitimate domination and an assessment of it will be offered. Weber's typology is only one of many which seek to classify authority relationships according to the particular rationale for attributing legitimacy to command and obedience. These rationales The Btitish Journal of Sociology Volume XXXVIII Number 2

3 200 Craig Matheson can exist as motivations for obedience on the part of power subjects or as legitimations of command advanced by power holders. Whether we are concerned with such rationales as motivations for obedience or as legitimations of command, we are in both cases dealing with 'explanations' or forms of understanding or belief which seek to 'legitimate' or make rightful the behaviour in question. Historically, these explanations are typically encountered in the form of legitimations advanced by power-holders to convince power-subjects that command and obedience are rightful. Power-holders have an interest in securing obedience motivated by a belief in legitimacy, for legitimate authority is less a 'costly' form of authority than either coercive or reward-based authority. In the case of coercive authority, only constant surveillance and supervision can ensure that subordinates completely comply with commands, for subordinates will comply only when they face the prospect of punishment for non-compliance. In the case of rewardbased authority, obedience has to be 'purchased' through the offer of rewards for compliance. Legitimate authority obviates the need for surveillance and rewards, since subordinates feel obliged to obey no matter whether there is a 'reward' for compliance or not. To secure obedience motivated by a belief in legitimacy power-holders must convince power-subjects that the command-obedience relation is 'rightful' or legitimate. Thus \Vrong argues that all legitimations are 'reasoned elaborations' intended to persuade someone that he ought t obey an authority. ' As he notes, legitimation exemplifies the relationship of persuasion passing into authority, since it involves the presentation of arguments as well as the issuance of commands. Nevertheless, a situation where the subject obeys commands because he is aware ofthepotentiality oftheir justification by rational argument is still an authority relationship, since persuasion here is not equivalent to the fact of obedience itself.2 As 'reasoned elaborations' legitimations constitute types of explanation or understanding of power relationships, for they are 'reasons why' one should obey someone. As such, they tend to form part of the prevailing form of explanation or understanding within a society. Thus Balandier notes that myths have the dual function of explaining the social order in historical terms and of justifying it on some moral basis.3 Similarly, Peter Berger notes that a legitimation consists of'knowledge' that serves to explain and justify the social order, and takes the form of proverbs, maxims, and traditional wisdom as well as abstract theories.4 If legitimations are explanations of why command and obedience are rightfui, then we can classify them according to the particular explanation or reason why command and obedience are said to be legitimate. The first of these is the legitimation of command and obedience on the basis of convention. Here, command and obedience are considered to be rightful or legitimate because they are conventional

4 Weber and the classification offorms of legitimacy or prescribed by norms. In this sense, the right of command and the obligation of obedience can be said to be grounded in norms, for these are what legitimize expectations of command and obedience. Thus Benn and Peters note that the right of command and corresponding duty of obedience can be derived from rules, whether these are legal or customary, and that this occurs in so far as the rules prescribe certain forms of behaviour.5 The second basis of legitimacy which we can identify is contract. Here, command and obedience are considered to be legitimate because they are specified by a contract in which the power-holder and the power-subject assume mutual rights and obligations. The right of command and the obligation of obedience will be among these rights and obligations, and thus can be said to be grounded in a contract rather than in social norms. That is to say, expectations of command and obedience are legitimate because they are specified in a contract, or have been agreed to. Benn and Peters identify this type of rationale for obedience with the theory of consent as the basis of political obligation, a theory whose origins can be traced back to the social contract theorists.6 The third identifiable basis of legitimacy is that of conformity with universal principles. Here, command and obedience are considered to be legitimate if they are in conformity with certain universal principles which are discoverable by all human beings. Historically, this basis of legitimacy is to be encountered in those theories of'natural law' which assert the existence of a moral order which is superior to all man-made laws. The idea is to be found in political philosophy from the earliest times in the form of the distinction between 'nature' and 'convention'. Thus Benn and Peters note that we can distinguish 'moral norms' and their accompanying rights and obligations from both custom and law, together with their accompanying rights and obligations, on the basis that moral norms have been critically examined in accordance with criteria other than the fact of their being conventional.7 In addition to the three principles of legitimacy which we have just outlined there exist other principles of legitimation in which the right of command and the obligation of obedience can be grounded. Historically the most important of these is undoubtedly the principle of the sacredness of authority or of the norms which prescribe command and obedience. Here, command and obedience are considered to be legitimate either because the power-holder is himself considered to be sacred, or because the norms which prescribe command and obedience are considered to be sacred. The former belief is most commonly encountered in some version of the theory of the 'divine right of kings', in which a right of command and a duty of obedience is claimed on the grounds that the monarch rules by divine sanction or appointment, or is himself a divinity. In this form, as monarchy, the sacredness of authority represents an instance of what Weber termed 'familial charisma', or that extraordinary or supra- 201

5 202 Craig Matheson human quality that is attributed to a family in the belief that this is transmitted through blood ties. This quality can also be attributed to a specific person and as such form the basis for a claim to a right of command and an obligation of obedience in its pure form as charismatic leadership. As well as being an attribute of a family or person, charisma or sacredness can also be an attribute of an office or set of norms. For Weber, charisma in this sense appeared in the form of 'institutional charisma' in which it forms an attribute of an office rather than of a person. In the process of the 'routinization' of personal charisma into its familial or institutional forms a special affinity emerges between charisma and tradition, since tradition is commonly regarded as being sacred. It is this sanctity of tradition that forms a second principle of legitimation for Weber, and it is found wherever rulers claim a right of command on the basis of the sanctity of immemorial tradition. In this instance we have a case where legitimacy is claimed on the basis of the sanctity of norms, as distinct from the sacredness of the person of the authority-holder. The reason why the sacred forms such a widespread instrumentality of legitimation is, as Peter Berger notes, because the sacralization of social institutions makes them seem beyond human caprice.8 As many students of comparative religion have noted, the essence of the sacred is its 'otherness' as a power confronting man to which he must submit. This principle of legitimation is what Sjoberg terms the 'appeal to absolutes' or a claim to legitimation by forces independent of human action, such as we see embodied in the theory of the divine right of kings.9 Benn and Peters similarly note the existence of a principle of legitimation, such as we see embodied in the theory of the divine right of kings, which refers to a force independent of human choice and custom. Such a principle of legitimation has its secular parallels in those theories which assert that a leader has been marked out by fate or history for a special mission.l Another reason why the sacred is so often invoked in order to legitimate authority is because authority itself bears a certain resemblance to the sacred which makes for a natural predisposition towards legitimation by the sacred. As Balandier observes, the notions of the sacred and the political interlock since both are forces for order superior to the individual which arouse feelings of respect, fear, attraction and repulsion.ll The fifth identifiable principle of legitimation is that of expertise. Here, command and obedience are considered to be legitimate because the power-holder possesses some type of expertise which makes him the person who is best suited to exercise power. This is an argument which, considered as a justification in terms of purely 'technical' expertise, was first advanced by Saint-Simon and is to be encountered in the modern era in the form of Taylorian theories of management and various technocratic theories. Considered as a justification in terms of 'knowledge' generally, such a principle of

6 Weber and the classification offorms of legitimacy 203 legitimation is present whenever leaders claim a right of command on the basis of their knowledge of the historical process, such as in the case of contemporary communist governments. This principle of legitimation has been identified by Benn and Peters as that of the appeal to the 'historic mission' that occurs wherever leaders seek to justify their authority in terms of their insight into 'historic laws' which govern the future development of society.l2 Historically, legitimation on the basis of expertise is to be typically encountered in the case of paternal authority and the authority of the teacher over the pupil. 'Expertise' in these cases consists of the ability of the powerholder to determine what is in the best interests of the power-subject or of the superior knowledge of the power-holder vis-a-vis that of the power-subject. In both cases, but particularly in the case of paternal authority, the interests of the power-subject are served by compliance with the commands of the power-holder. Wrong notes that we can identify as 'competent authority' those types of power relations in which the subject obeys out of a belief in the authority's superior competence or expertise to decide which actions will best serve the subject's interests and goals.l3 As he notes, competent authority has often been seen as being the most benign and desirable form of authority since it resembles persuasion. Hence we find that paternal authority is the typical form of gemeinschaft-like authority and is characterized by an I-thou relationship in which the interests and needs of the subordinate are an object of the superordinate's concern. The sixth identifiable basis of legitimacy is the approval of the exercise of power that power-subjects extend to power-holders. This can assume the form of a claim to popular support in the form of a claim to a popular 'mandate', or, as in the case of an admixture with conventional or contractual basis of legitimacy, of a claim to democratic election in accordance with constitutional procedures. In either form, this type of claim to political legitimacy is the most common one to be found in the modern world, and has supplanted the theory of the divine right of kings as the most common form of political legitimation. As Carl J. Friedrich notes, democracy or popular majorities have increasingly become the only basis of legitimate government in a process that has involved the fading of the legitimacy of monarchical government.'4 The modern ground of legitimacy is what Friedrich terms 'constitutionalism', or the idea that a regime is legitimate only when it is based on the will ofthe people.l5 This popular base of legitimacy is what Nisbet sees as being one of the distinctive features of the political order introduced by the French Revolution, which replaced heredity, divine right, and tradition with popular approval as the sole source of legitimacy.'6 With respect to legitimation on the basis of popular support, sse can distinguish between regimes which base their legitimacy upon the claim to popular support as such, irrespective of whether there exist

7 204 Craig Matheson constitutional mechanisms for ensuring effective popular control of government, and those which base their claim to legitimacy upon the fact of popular election in accordance with constitutional procedures which ensure the permanent possibility of popular control of government. In the latter case, the ground of legitimacy is not simply popular support or majority rule but the fact that the power-holders have been elected in accordance with constitutional procedures. This type of basis of legitimacy therefore comprises a mixture of popular approval and of legal or contractual bases of legitimacy. The distinction drawn here roughly corresponds to that which Friedrich draws between populist democracies, in which the will of the majority is sovereign, and constitutionalist democracies, in which the will of the majority is limited by a constitution.l7 The seventh identifiable basis of legitimacy is the 'personal relation' or tie that exists between the power-holder and the power-subject. Wrong identifies this relationship where the power-subject obeys because of the 'personal significance' of the power-holder to him as 'personal authority'.l8 What is termed personal relation or significance here may be equated with the shortness of distance between perspectives that exists between power-holder and power-subject, so that a power-holder will claim a right of command and the powersubject acknowledge a duty of obedience on the basis that the other person in the relationship is 'close' to him. This type of basis of legitimacy is typically encountered in paternal authority relationship, and in those types of authority relationships which are modelled on paternal authority, such as, for example, the relation of master and servant. In these types of relationships we find, as Weber noted in his analysis of'traditional domination', that Versonaloyalty constitutes a ground of the obligation of obedience,' for it is in precisely these types of authority relationships that the distance between powerholder and power-subject is short. Obedience to a command here is interpreted by both power-holders and power-subjects as a manifestation of solidarity or 'loyalty', although such a relationship can also be the source of reciprocal obligations on the part of the powerholder towards the power-subject. Hence we find that the feudal relation of master and serf was, in contrast to slavery, characterized by a reciprocity of rights and obligations. A 'depersonalization' of these types of authority relationships can occur either because the group of superiors and subordinates grows so large that the degree of personal intimacy which is necessary in order for this ground of obligation to be sustained is no longer present, or because the rules which govern the relationship forbid the exercise of authority in accordance with 'particularistic' criteria. In the latter case, the ground of obligation ceases to be the personal relation of power-holder to power-subject and becomes instead the fact that command and obedience are prescribed by norms. Hence we find that for those for

8 Weber and the classiyication offorms of legztimacy 205 whom authority is 'institutionalized power', it is often defined as being, in contrast to 'leadership', a type of power relationship in which command and obedience are not affected by personal loyalties. Thus Robert Bierstedt argues that in the ideal case the exercise of authority is 'wholly objective, impartial, impersonal and disinterested', so that a person exercises the same kind of authority over friends and intimates as he does over enemies and strangers.20 The eighth identifiable basis of legitimacy is some 'personal quality of the power-holder, by virtue of which he can claim a right of command, and the power-subject acknowledge a duty of obedience. Wrong identifies this type of motivation for obedience as being, together with the personal significance of the power-holder to the power-subject, the basis of'personal authority'.2' The two motivations for obedience are, however, distinguishable, for a power-holder need not be 'personally significant' in the sense of standing in a personal or intimate relation to the power-subject in order for the latter to recognize some personal quality of the former as constituting a ground for an obligation of obedience. Benn and Peters note that we can distinguish authority relationships where the right of command is bestowed by rules from those in which the right of command is derived from 'personal factors'.22 For both Wrong and Benn and Peters the outstanding example of authority based on personal qualities is Weber's category of charismatic domination. In this case, the right of command and the duty of obedience is said to be derived from belief in the extraordinary quality of a specific person.23 It is not only 'extraordinary' qualities that can entitle a person to command, for there exist a wide variety of personal qualities which constitute potential bases of legitimacy. The most notable of these is the quality of'leadership', for this bears an immediate relation to the task of command. Thus Benn and Peters note that the presence of the quality of leadership can be a basis of legitimacy.24 We may conclude then, that it is possible to distinguish eight separate bases of legitimacy, namely, convention, contract, universal principles, sacredness, expertise, popular approval, personal ties and personal qualities. Such a list is not exhaustive, for it is possible to conceive of a large number of reasons why a person should consider a particular relationship of command and obedience to be legitimate. The eight reasons that have been outlined here however are the most common ones. Other classifications of bases of legitimacy correspond at least in part to the classification that I have offered here. Sjoberg for example, distinguishes four bases of legitimacy, namely, the appeal to absolutes, the appeal to tradition, the appeal to experts and the appeal to the governed.25 Friedrich distinguishes five bases of legitimacy, namely, the religious, the juristic or philosophical, the traditional, the procedural or democratic, and that of success.26 Benn and Peters distinguish rules, personal qualities, divine sanction and

9 206 Crazg Matheson consent as bases of political obligation.27 The most famous classification of bases of legitimacy is that offered by Weber, who distinguished three principles of legitimation, namely, the traditional, the rational or 'legal', and the charismatic.28 According to Weber, each of these principles of legitimation corresponds to a specific 'apparatus' of domination, or structure of the relationship between rulers, ruled and administrative staff. Weber's classification does not fit easily into the classification that I have outlined here however, and there is some basis for supposing that what Weber conceived to be three single principles of legitimation are in fact composites of a number of separate bases of legitimacy. In the following pages I intend to undertake an analysis of Weber's classification of principles of legitimation to demonstrate that this is indeed the case. Weber classifies types of authority relationships according to the claim to legitimacy advanced by power-holders. According to Weber, we can distinguish action which is motivated by self-interest and that which is guided by a belief in a 'legitimate order'. In the latter case, individuals perform actions not merely because such actions are expedient, but because they feel a sense of duty to perform such actions.29 Authority can, according to Weber be based upon selfinterested compliance, but this is usually not sufeciently reliable. Instead, rulers will seek to cultivate a belief in legitimacy on the part of the ruled. According to the type of legitimacy which is claimed, the type of obedience, the kind of administrative staff and the mode of exercising authority will all differ fundamentally.30 Weber argues that in such authority relationships subordinates may obey out of selfinterest or because there is no alternative but that these considerations are not relevant to classification. What is important is that the claim to legitimacy is treated as valid.3l This seeming contradiction between the insistence upon the difference between obedience out of selfinterest and obedience motivated by a belief in legitimacy, and the assertion that obedience in the case of legitimate authority may be motivated by self-interest, is an instance of what has been termed the 'ruler-centric' orientation of Weber.32 As Merquior notes, Weber defined legitimacy from the viewpoint of the rulers, not of the ruled. Such ruler-centricism does not vitiate the analytical utility of Weber's classification of principles of legitimacy however, for what is a rationale for command constitutes a potential rationale for obedience as well. Weber himself noted that certain motivations for obedience were found in conjunction with certain types of domination, and these motivations for obedience simply mirror the rationales for command that are offered by power-holders. The first principle of legitimation identified by Weber is that which rests on traditional grounds. Here, command and obedience are legitimized on the basis of the sanctity of immemorial traditions which govern the authority relationship.33 Commands are legitimate in that

10 Weber and the classification offorms of legitimacy they are are in accord with custom and power-subjects obey out of a feeling of personal loyalty to the power-holder.34 Stated thus, Weber's category of traditional authority comprises three separate bases of legitimacy, namely, the sanctity of tradition, convention, and the personal relation of power-holder to power-subject. The basis of legitimacy to which Weber accords most emphasis in his analysis of traditional domination is the sacredness of tradition. According to Weber, a system of domination can be called 'traditional' if legitimacy is claimed for it and believed in on the basis of the sanctity of order and the attendant powers of control as they have been handed down from the past.35 Thus he notes that it is characteristic of patriarchal and patrimonial authority as major sub-types of traditional domination, that the system of inviolable norms which govern the relationship is considered sacred, any infraction of which would result in magical or religious evils.36 The second basis of legitimacy upon which traditional domination rests is convention or tradition itself. In this sense, command and obedience are considered to be legitimate if they are in accord with custom or are 'traditional'. As Bendix notes, the legitimation of traditional domination occurs wherever the 'authorities' claim obedience on the basis of established usage.37 Hence a patriarchal ruler can lose the feeling of obligation among his subjects if he seeks to establish rights for himself which exceed the limits of custom.38 Benn and Peters, in a discussion of traditionalist grounds of authority and obligation, note that for a traditionalist, an authority is legitimate if it is sanctioned by custom.39 It should be noted that convention as a source of legitimacy is distinguishable from the sanctity of convention as a source of obligation. In the first case, the rationale for compliance is that compliance is conventional or prescribed by customary norms. In the second case the rationale for compliance is that the norms which prescribe command and obedience are sacred. Thus Roderick Martin has argued that Weber's category of traditional domination consists of two elements, namely, 'sanctity' or the sacredness of the past, and the worship of established precedent that occurs equally in systems where norms are legal rather than customary.40 The third separate basis of legitimacy in systems of traditional domination is the personal relationship that exists between power-holder and power-subject. This element is identified by Weber as that of personal 'loyalty', and according to him, it constitutes the ground of an obligation of obedience on the part of power-subjects.4l This element is typical of 'patriarchalism' which, together with gerontocracy, forms one of the two major sub-types of traditional domination. Thus Howard Newby has noted that Weber's category of traditional authority involves legitimation both on the basis of traditional norms and on the basis of personal loyalty, and that legitimation on the latter basis is typical of paternal authority relationships in general. 207

11 208 Craig Matheson Closely related to but distinct from the category of traditional domination is the category of charismatic domination. According to Weber, charismatic domination can be said to exist when the governed submit because of their belief in the extraordinary quality of a specific person.43 This belief in and devotion to the extraordinary quality is the basis of the legitimacy of charismatic rule:44 hence the followers of the charismatic leader are duty-bound to obey their leader.45 As well as being a quality of a specific person charisma can also exist in a 'depersonalized' form when it becomes the attribute of a group or norm. In this form, it exists as familial and institutional charisma, and as such, can constitute a basis of legitimacy separate from that of personal charisma. As it can be seen, the concept of charismatic domination outlined here comprises two analytically distinguishable bases of legitimacy. The first of these is the sacredness 'extraordinary quality' of a person, group or set of norms which charisma in its personal, familial and institutional forms represents. The second is the 'personal quality' which charisma in its form as a quality of a specific person represents. The first basis of legitimacy which constitutes charismatic domination, namely, the quality of 'sacredness', is identified by Weber as an 'extraordinary quality' possessed by persons or objects, by virtue of which they are thought to bendowed with 'supernatural, superhuman, or at least specifically exceptional powers or qualities'.46 This usage of the term 'charisma' makes it equivalent in meaning to the concept of the 'sacred' which is employed by other authors. As Weber himselfnoted, both charismatic and traditional domination rest upon belief in the sanctity of something.47 Indeed, traditional domination is the outcome of that process which Weber described as the 'routinization' of charisma, in which charisma ceases to be a quality of a specific person and becomes instead the quality of a set of norms based on precedent.48 Thus we find that in its depersonalized form as both institutional and familial charisma, the quality of charisma is closely associated with traditional domination. In these cases, charisma as a ground of legitimacy is equivalent to the sacred as a source of legitimacy, for the rationale for compliance is that a set of norms possess an extraordinary quality. Charismatic domination for Weber is however, not distinguished simply by the fact that it rests upon belief in the sacred, for otherwise it would be indistinguishable from traditional domination. Rather, what especially distinguishes charismatic domination for Weber is the fact that the extraordinary quality in question is, in the form of personal charisma, believed to be the quality of a specific person. Charismatic domination in this form is therefore wholly distinct from the other types of domination, for it rests exclusively upon the charisma of an individual person, as distinct from that of social norms. As Bendix notes, the feature that distinguishes charismatic

12 Weber and the classification offorms of legitimacw domination from other types of domination for Weber is the fact that the exercise of authority is bound up entirely with a concrete person and his distinctive qualities.49 People do not obey by virtue of tradition or statute, according to Weber, but because they believe in the existence of the charisma of the individual.50 Hence the process of the routinization of charisma is one in which rules in some form always come to govern, so that the power-holder no longer rules by virtue of purely personal qualities.5l Contrariwise, the emergence of charismatic domination is a process that occurs when the permanent institutionalized structures of domination are no longer adequate and that has a 'revolutionary' character in so far as it challenges the established order. As Weber notes, charismatic domination is the direct antithesis of both traditional and rational-legal domination in so far as the latter are forms of action which are bound to every day routine.52 Thus we may conclude that the category of charismatic domination comprises two analytically distinguishable bases of legitimacy, namely, the sanctity or extraordinary quality of persons, groups or norms, and the extraordinary quality of an individual person. The former is an instance of authority whose source of legitimacy lies in the sacred. The latter is an instance of authority whose source of legitimacy lies in the personal quality of an individual; the quality in this case being that of the extraordinary or super-human. This twosided nature of charismatic domination as both authority legitimized by the sacred and authority legitimized by an extraordinary quality of the person, has been recognized by a number of commentators. Bendix, on the basis of Bierstedt's distinction between leadership as personal domination and authority as institutionalized domination, distinguishes charismatic leadership and charismatic domination.53 Whereas the former is domination by virtue of the extraordinary qualities of a person, the latter is domination by virtue of 'established authority', or in Bierstedt's terms, institutionalized power. Dennis Wrong argues that we cannot adequately define charismatic authority in terms of the 'sacred' since, given the fact that Weber spoke of the sacredness of tradition as a basis of legitimacy, this.would make charismatic authority indistinguishable from traditional domination. According to Wrong, the essence of charismatic authority lies in the fact that it is a form of personal authority, or that type of authority in which subjects obey out of a regard for the personal qualities of the power-holder.54 Benn and Peters similarly argue that Weber's category of charismatic domination revresents an instance of legitimation on the basis of personal factors.5 The third type of domination identified by Weber was that which he termed 'rational' or 'legal'. According to Weber, legal authority exists when rulers base their claim to leitimacy upon a belief in the 'legality' of rationally established rules. 6 Rules are said to be legal when they have been enacted in accordance with a rationally enacted 209

13 210 Craig Matheson or interpreted constitution. Bendix expresses the relation between legality and legitimacy thus: laws are legitimate if they have been enacted, and the enactment is legitimate if it has occurred in conformity with the laws prescribing the procedures to be followed.57 From the laws which govern the authority relationship rulers derive their right of command and subordinates derive their corresponding duty of obedience. Command and obedience in such circumstances constitute forms of obedience towards a norm rather than an arbitrary freedom, favour or privilege, or an expression of devotion to the person of the authority holder.58 Legal authority therefore differs from charismatic authority, where command and obedience occur not in accordance with norms but in accordance with the personal beliefs and expectations of superior and subordinate concerning the legitimacy of command and obedience. Legal authority also differs from traditional authority, for whereas the former rests upon belief in the legality of rationally established norms, the latter rests upon a belief in the sanctity of norms based on precedent. From this outline of the nature of legal domination we can identify two separate bases of legitimacy upon which this category rests. The first of these is convention, or the legal norms which govern the relationship of domination and submission. Command and obedience are carried out in such circumstances, in the belief that they are legitimate because they are prescribed by the law. As Weber notes, submission in the case of legal authority is based upon 'duty of oflice' which is fixed by rationally established norms.59 The corresponding right of command is one that is grounded in legal rules, and not, as in the case of charismatic domination, in the person of the powerholder.60 As Benn and Peters note, when Weber talks of an authority such as legal-rational or traditional authority, he means that there exists a right of command which is bestowed by a set pattern of rules, wheth#r these are enacted or customary.6l A second sense in which convention is the source of legitimacy is present in Weber's definition of legal authority in so far as he sees the legitimacy of laws which prescribe domination and submission as deriving from their enactment in accordance with legally or constitutionally defined procedures for such enactment. Weber argues that the legitimation for establishing legal rules in the case of legal domination derives from a rationally enacted or interpreted 'constitution'.62 As Bendix notes, laws are legitimate in the case of legal domination if they have been enacted in accordance with the laws prescribing the procedures to be followed.63 In this case, command and obedience are legitimized indirectly, by way of their prescription by norms which are considered to be legitimate. This is similar to the situation that exists in the case of traditional domination, where command and obedience are legitimized by virtue of the fact that they are prescribed by norms which are considered to be sacred. Here, part of the sanctity of tradition lies in

14 Weber and the classification offorms of legitimacw the fact that it is indeed 'tradition' or precedent, that is, handed down from the past. Tradition possesses this quality of persistence through time because it is not consciously evaluated or changed by succeeding generations. As a result of this non-rational reliance upon precedent and resultant orthodoxy, tradition comes to be regarded as being of suprahuman origin and therefore 'sacred' in nature. Hence, interwoven with the justification of traditional norms on the basis of their sanctity is the justification of these norms on the basis of their 'traditionality'. The norms are right because they are traditional. Similarly, legal norms can be justified on the basis of their 'rationality', or the fact that they have been consciously enacted. This is the other principle of legitimacy upon which Weber's category of legal domination rests. As MacIver, following John Burgess, points out, there are basically two ways in which a rationale for obedience to the law can be given. The first of these refers to the legitimacy of the source of the law, that is, the law-making authority. The second refers to the intrinsic merit of the law itself; to the rationality of its content.64 In the case of the category of legal domination, the former justification is present in so far as laws are legitimized on the basis of the legality of their enactment. Laws are legitimate because they have been enacted by a legally constituted authority. The latter justification is present in the case of legal domination in so far as laws are legitimized on the basis oftheir 'enactment', or the fact that they have been rationally created. Thus Weber distinguishes traditional and legal domination in terms of a contrast between claims to legitimacy that are based upon traditional grounds and rational grounds respectively.65 This is because legal domination, in contrast to traditional domination, rests upon rules that, as Weber notes, have been 'rationally established'.66 As Martin Albrow observes, the very conceptions of legitimacy which sustain lebsal authority are rational, since legal norms are intrinsically rational.6 We may conclude then, that Weber's category of legal domination comprises two analytically distinguishable bases of legitimacy, namely, convention and the rationality of law. The former is present in so far as power-holders claim legitimacy upon the basis of 'legality', whether this concerns command and obedience as such or legal norms which prescribe command and obedience. The latter is present in so far as power-holders claim legitimacy on the basis that laws which prescribe command and obedience have been 'enacted', that is, rationally established. Thus we have demonstrated that Weber's three principles of legitimation each comprise a number of different principles of legitimation. There exist not three principles of legitimation but five. Does this mean then, that Weber's classification of principles of legitimation is mistaken? The answer is no, for Weber's three principles of legitimacy, while not constituting three analytically 211

15 212 Craig Matheson distinguishable types, do form three sets of principles of legitimation which are commonly found together in conjunction with distinct forms of command-obedience relationship. If we take the three principles of legitimation or rationales for attributing legitimacy which underlie the category of traditional domination, namely, the sanctity of tradition, personal relationships, and convention, then we find that these three principles are commonly encountered together in those situations where command-obedience relationships are regulated by customary norms. This situation is implied by the very notion of legitimation on the basis of customary conventions and the sanctity of tradition. We have previously noted the way in which the mode of origin of tradition favours its sacralization, and thus we tend to find that legitimation on the basis of customary conventions and on the basis of the sanctity of tradition are both present. Legitimation on the basis of personal relationships will also tend to be present in such situations because the nature of custom as a set of discrete unsystematized precedents leaves room for the exercise of personal discretion and thus for the exercise of personal favouritism in command-obedience relationships. As Weber noted, there is an element of arbitrary power present in traditional domination owing to the fact that traditional limits upon such demands are ill defined.68 Also present in traditional authority relationships is obedience motivated by a sense of obligation that is based on the personal relation of the power-subject to the power-holder. Such forms of obedience are more likely to be present in cases where command and obedience are governed by customary norms, since these leave the conditions of obedience ill-defined. Hence we find that the organization of administration under systems of legal domination is, in contrast to that found under systems of traditional domination, one in which obedience to a command is purely impersonal.69 Thus we may conclude that the principles of legitimacy which underlie the category of traditional domination identified by Weber form an identifiable class of such principles which are found in association with command-obedience relationships that are regulated by traditional norms. Indeed, in his analysis of the structure of traditional domination it is to the latter feature of the relationship of command and obedience that Weber accords the most attention. The features of systems of administration under traditional domination that are outlined by Weber, such as the lack of clearly defined spheres of competence, the arbitrary nature of the exercise of power and of the delegation of authority, and recruitment on the basis of personal loyalty,70 can all be derived from the nature of traditional domination as a command-obedience relationship which is regulated by customary norms. In the case of charismatic domination, the two principles of legitimacy upon which it is based are, as in the case of traditional domination, ones that tend to occur in conjunction. Charisma as an

16 Weber and the classification offorms of legitimacy 213 'extraordinary quality' is, together with 'leadership', one of the two personal qualities which are most often regarded as forming entitlements to rule. The reason for this is that in times of trouble when permanent institutional structures of domination do not suffice, people turn to the person who possesses extraordinary gifts. Such a leader frequently appears in the guise of a messiah or saviour. Thus Weber's category of charismatic domination as charismatic 'leadership', or domination by virtue of extraordinary personal qualities, constitutes a distinct type of domination by virtue of personal qualities that is to be frequently encountered throughout history. The two principles of legitimacy which underlie the third category of legal domination, namely, convention and rational enactment, are also principles which tend to occur in conjunction. This is because they both presuppose the existence of rationally established norms which prescribe command and obedience. As Weber notes, the existence of legal authority requires the prior existence of a legal code consisting of a system of abstract rules which regulates the command-obedience relationship, so that both acts of obedience and of command constitute forms of adherence to a norm rather than acts of personal devotion or an arbitrary freedom.7l Thus we may conclude that the five principles of legitimacy which underlie Weber's three types of domination tend to form three sets of principles because they occur in conjunction with specific types of command-obedience relations. In the case of traditional domination, this command-obedience relation is one that is regulated by customary norms. In the case of charismatic leadership, it is one that occurs in times of trouble when permanent structures no longer suflice, so that people turn to a 'leader' who possesses extraordinary qualities. In the case of legal domination, this command-obedience relation is one that is regulated by legal norms. Given the fact that it is the principles of legitimacy which vary according to the type of command-obedience relation, then Weber's principle that according to the type of legitimacy claimed the structure of domination will vary, can be reversed. Weber distinguishes the type of obedience encountered, the kind of administrative staff and the mode of exercising authority, as aspects of the structure of domination.72 The type of obedience encountered will indeed vary according to the kind of legitimacy claimed, since the type of obedience forms part of the very definition of authority as legitimate. The kind of administrative staff and the mode of exercising authority however, vary not so much according to the type of legitimacy claimed but more according to the type and degree of institutionalization of the authority relationship. Only in the case of charismatic leadership can it be said that the kind of administrative staff and the mode of exercis-ing authority vary according to the type of legitimacy claimed, for here it is the belief of ruler and subordinate in the legitimacy of command and obedience

17 214 Craig Matheson that is the sine gzua non of charismatic leadership. Such a connection of the belief in legitimacy to the command-obedience relationship does not exist in the case of the two other types of domination, for here, the particular relationship of command and obedience can exist in the absence of any belief in legitimacy on the part of superiors and subordinates. The mode of exercising authority in the case of traditional or legal domination for example, can exist in the absence of a belief in legitimacy, for the conditioning factor in these cases is the regulation of the command-obedience relationship by particular types of norms. The same is also true with respect to the kind of administrative staff characteristic of each. The difference between personal retainers and vassals on the one hand, and bureaucratic oflicials on the other, as the type of administrative staff characteristic of traditional and legal authority respectively, is due to the greater degree of institutionalization of the authority relationship that occurs with the transition from customary to enacted law. It would be perfectly possible for a bureaucratic administrative staff of the type outlined by Weber to exist even in situations where no belief in legitimacy was present, as long as the command-obedience relation concerned was regulated by legal norms. Given that this is the case, we cannot say that the principle that the type of legitimation will vary according to the type of command-obedience relationship, means that in all cases where a particular type of command-obedience relationship is found so too will the corresponding type of legitimation. All that we can say is that the existence of a particular type of commandobedience relation is a necessary but not a sufficient condition for the existence of certain types of legitimation. It is only in so far as this is the case that Weber's argument that the structure of domination varies in accordance with the type of legitimacy claimed is correct, for the existence of certain types of legitimation presupposes the existence of certain forms of command-obedience relationships. Craig Matheson Campbell, ACT, Australia NOTES 1. D. H. Wrong, Power Its Forrns, Religion, Harmondsworth, Middlesex, Bases, and Uses, Oxford, Basil Blackwell, Penguin, 1973, p , p S. I. Benn and R. S. Peters, Social 2. Ibid., pp Principles and the Democratic State, London, 3. G. Balandier, Political Anthropology, Allen & Unwin, trans. A. M. Sheridan Smith, New York, 6. Ibid., p Pantheon, 1970, p Ibid., pp. 27-8, P. L. Berger, The Social Reality of 8. Berger, op. cit., p. 41.

18 Weber and the classification offorms of legitimacy G. Sjoberg, The Preindustrial City: 39. Benn and Peters, op. cit., p Past and Present, Glencoe, Illinois, Free 40. R. Martin, The Sociology of Power, Press, 1960, p London, Routledge & Kegan Paul, 1980, 10. Benn and Peters, op. cit., p pp Weber, The Theory of Social.., op. 11. Balandier, op. cit., pp cit., pp Benn and Peters, op. cit., p H. Newby, 'Paternalism and 13. Wrong, op. cit., p. 53. Capitalism' in R. Scase, (ed.), Industrial 14. C. J. Friedrich, Limited Government: Society: Class, Cleavage and Control, A Comparison, Englewood ClifEs, NJ., pp Prentice-Hall, 1974, p Weber, 'The Social Psychology of 15. Ibid.,p.110. the World Religions', op. cit., p R. A. Nisbet, The Sociological 44. Ibid. Tradition, New York, Basic Books, 1966, p Friedrich, op. cit., p Wrong, op. cit., pp M. Weber, The Theoy of Social and 47. Weber, 'The Social Psychology of the World Religions', op. cit., p Economic Organization, trans. Talcott 48. Ibid. Parsons, New York, The Free Press, 49. R. Bendix, Max Weber: An Inaclkstual 1964, pp Portrait, London, Methuen, 1966, p R. Bierstedt, 'An Analysis of Social 50. M. Weber, 'Politics as a Vocation', Power' in Bierstedt, Power and Progress: in From Max Weber, op. cit., p. 79. Essays on Sociological Theoy, New York, 51. Ibid. McGraw-Hill, 1974, p Weber, The Theoty of Social.., op. 21. Wrong, op. cit., p Benn and Peters, op. cit., p M. Weber, 'The Social Psychology of the World Religions', in H. H. Gerth and C. W. Mills (eds), From Max Weber: Essays in Sociology, London, Routledge and Kegan Paul, 1947, p Benn and Peters, op. cit., p Sjoberg, op. cit., pp Friedrich, op. cit., p Benn and Peters, op. cit., pp Weber, The Theory of Social.., op. cit., p Ibid., pp cit., p Bendix, op. cit., p Wrong, op. cit., p Benn and Peters, op. cit., p Weber, op. cit., p Bendix, op. cit., p Weber, 'The Social Psychology of the World Religions', op. cit., pp and p Ibid., p Ibid., p Benn and Peters, op. cit., p Weber, op. cit., p Weber, The Theory of Social and Economic Organization, op. cit., p Bendix, op. cit. 29. Ibid., pp R. M. MacIver, The Web of Govern- 30. Ibid., pp ment, New York, Macmillan, 1947, p Ibid., pp Weber, The Theoty of Social.., op. 32. J. G. Merquior, Rousseau and Weber: cit., p Two studies in the theory of legitimacy, 66. Weber, 'The Social Psychology of London, Routledge & Kegan Paul, 1980, the World Religions', op. cit., p p Martin Albrow, Bureaucraey, 33. Weber, The Theory of Social.... London, Pall Mall Press, 1970, p. 63. Op. cit., p Bendix, op. cit., p Ibid., p Weber, The Theoty of Social.., op. 35. Ibid., p cit., p Weber, 'The Social Psychology of 70. Ibid., pp the World Religions', op. cit., p Ibid. 37. Bendix, op. cit., p Ibid., p Ibid., p. 333.