Copyright © 1996 R.J. Kilcullen
'GM' refers to H.H. Gerth and C. Wright Mills (trans. and ed.), From Max Weber (New York, 1946) (H/33/.W36).
'SEO' refers to Max Weber, The Theory of Social and Economic Organization, tr. Henderson and Parsons (New York, 1947) ((HB/175/.W364).
'ES' refers to Max Weber, Economy and Society, ed. G. Roth and C. Wittich (New York, 1968) (HM/57/.W342).
'Beetham' refers to David Beetham, Max Weber and the Theory of Modern Politics (London, 1974) (JA/76/.B37).
In this lecture I want to look at what Weber says about bureaucracy, in G and M, p. 196 ff, and in SEO, p. 329 ff.
First, something about the word. 'Bureau' (French, borrowed into German) is a desk, or by extension an office (as in 'I will be at the office tomorrow'; 'I work at the Bureau of Statistics'). 'Bureaucracy' is rule conducted from a desk or office, i.e. by the preparation and dispatch of written documents - or, these days, their electronic equivalent. In the office are kept records of communications sent and received, the files or archives, consulted in preparing new ones. This kind of rule is of course not found in the ancient classifications of kinds of government: monarchy, aristocracy, democracy - and bureaucracy? In fact it does not belong in such a classification. It is a servant of government, a means by which a monarchy, aristocracy, democracy, or other form of government, rules. Those who invented the word wanted to suggest that the servant was trying to become the master. Weber is of course aware of this tendency; in fact he attacked the pretensions of the Prussian bureaucracy to be an objective and neutral servant of society, above politics, and emphasized that every bureaucracy has interests of its own, and connections with other social strata (especially among the upper classes); see Beetham, chapter 3. But formally and in theory the bureaucracy is merely a means, and this is largely true also in practice: someone must provide policy direction and back the bureaucrat up (if necessary) with force. 'At the top of a bureaucratic organization, there is necessarily an element which is at least not purely bureaucratic', SEO, p. 335, to give policy direction.
In the middle ages the most effective kings ruled from horseback: they travelled round the country, armed, accompanied by armed men, and enforced their will. They were prepared if necessary to enforce their will on their armed companions by personal combat, though their prestige was such that this was seldom necessary. The king was accompanied also by 'clerks', i.e. clergy, who could read and write, who took along a chest containing records and writing materials; the modern bureaucracy developed from this. In modern countries the ruler does not have to fight in person, or travel round much; he or she rules by sending messages, through a bureau. The messages are usually acted on mainly because of the government's moral authority or prestige (a 'status' phenomenon), but also because they can be backed by force, by a 'staff' of police or soldiers. As Weber points out (e.g. SEO, pp. 330-1), armies have been bureaucratized. Napoleon had to watch his battle from horseback, but the modern general receives and sends messages. Napoleon had a 'staff', officers who galloped off with written messages, the modern army has a 'general staff'; the Prussian general staff was in Weber's time regarded with pride one of the key institutions of the German empire - it was in Weber's terms a bureaucracy. As he also points out, not only government services but also political parties, churches, educational institutions, and private businesses, and many other institutions have bureaucracies. That is, they all have a professional staff for keeping records and sending communications which will be regarded, at least by other staff of the same institution, as authoritative directions. Bureaucracies are found in ancient Egypt, ancient Rome, in the middle ages (notably the bureaucracy that served the pope). Bureaucracy is a pervasive feature of modern societies, ever growing in importance, Weber believed.
Weber sets out an 'ideal type' (see last lecture) for bureaucracy, characterised by an elaborate hierarchical division of labour directed by explicit rules impersonally applied, staffed by full-time, life-time, professionals, who do not in any sense own the 'means of administration', or their jobs, or the sources of their funds, and live off a salary, not from income derived directly from the performance of their job. These are all features found in the public service, in the offices of private firms, in universities, and so on.
Let me comment on these points, starting with the 'economic' features. There have in history been governments whose members made no distinction in resources, income, expenditure, etc. between public and private. Weber calls these 'patrimonial' (from the Roman law term for property that can be bought or sold). In Europe in the middle ages, for example, 'jurisdiction' was often as much a piece of property as a building or a horse. A kingdom might change hands as part of a marriage settlement. This was not true of jurisdiction and property in the Church, which did distinguish the prelate's private property from that of his church, and did not allow jurisdiction to be inherited or transferred as property; it forbade 'simony' (buying and selling office in the church), and enforced celibacy to keep church office and property from falling into the patrimony of families. Weber also speaks of 'prebends' or 'benefices', meaning an office to which is attached some income-yielding property, e.g. a farm, or tithes, or tax-gathering rights, from which the office holder lives. The modern bureaucrat does not have a prebend, but is paid a salary. He is not allowed to charge fees for himself (if fees are charged they belong to the government, firm, etc.), or to accept gifts. The ideal lying behind this is that if the official has any source of income apart from his salary he will not reliably follow the rules. Reliable following of the official rules is one of the highest values in a bureaucracy.
The modern bureaucrat does not own his job (SEO, p. 332). Some governments have sold offices, to raise money. This was true, for example, of judicial positions in 18th century France, of commissions in the army and navy in most European countries into the 19th century. The vested rights of office holders were an obstacle to reorganization, an impediment to efficiency; so they were bought out, or expropriated with compensation.
Bureaucrats do not own the 'means of administration' - the computers, the furniture, the files, etc. Weber suggests a parallel with capitalist productive enterprise (GM, pp. 81-2). Similarly, in modern armies the soldier does not own his weapons, whereas in ancient armies he did (GM, pp. 221-2). For example, in ancient Rome when the army was called together the 'classes' were expected to come equipped to a certain standard at their own expense - 'classification' was a form of taxation. Soldiers were expected to bring money to buy food from the locals (when they did not take what they wanted by force); they got no pay or provisions. In modern educational institutions teachers do not own what they use (in medieval universities originally they did, and in fact each 'master' owned a school which was a private business enterprise). 'The bureaucratization of... the universities is a function of the increasing demand for material means of management... Through the concentration of such means in the hands of the privileged head of the institute, the mass of researchers and docents (lecturers) are separated from their "means of production" in the same way as capitalist enterprise has separated the workers from theirs', GM, pp. 223-4. In the modern army, public service, private firm, the equipment is provided by the organization partly because this is more efficient now that it is so elaborate and expensive.
The modern bureaucrat is a full-time, life-time professional; this requires a sufficient salary and job security, because otherwise people will not stay in the job full time for life. Unless they do, the organization will not be efficient. It takes time and experience to learn the job, not so much because it is difficult to perform the particular task, but because it all has to be coordinated. An elaborate division of labour requires stability of staff. Because of the nature of bureaucratic work, and also perhaps because of the importance of training and coordination in the job, the bureaucracy wants educated recruits. Their education will be attested by some certificate (partly just to prove they have been educated, but also perhaps because a bureaucracy likes to work with clear impersonal criteria). Weber speaks of 'credentialism', the preoccupation evident in modern societies with formal educational qualifications. All these things - credentials, fixed salary, tenure, stability of staffing, Weber incorporates into his ideal type. They are all required, he believes, for the efficient functioning of an administrative machine.
Another feature is the impersonal application of general rules, both to the outsiders the organization deals with, and to its own staff. The Taxation Commissioner's staff impersonally, objectively, apply the rules to the taxpayer, and their own duties and rights within the organization are defined by rules applied to them impersonally by their superiors. In Weber's mind this is the most important feature of bureaucracy. It underlies the features we have been commenting on up to this point: bureaucrats do not own their equipment or their job, and receive a fixed salary etc., because these things ensure reliable rule-following. In ESO he treats of bureaucracy under the heading of Types of Legitimate Authority. There are three types: rational, traditional and charismatic. Charismatic authority is regarded as legitimate, and works, because followers are personally devoted to the 'gifted' leader. Traditional authority is regarded as legitimate because everyone has always obeyed whoever was in the leader's position, and no one thinks of disputing his authority. Rational authority is the 'rule of law': it exists in a community in which there is a moral attitude of respect for the law as such, or because the law has been arrived at in a way that is regarded as legitimate. Rulers are recognised and obeyed if they can show a warrant in the law. Bureaucracy obviously exists within such a framework: even in the bureaucracy of a private firm, subordinates want to be assured that orders are properly authorised. Bureaucracy is the most efficient way of implementing the rule of law: the legal rules are recorded, studied, and applied in a carefully considered and reliable way to individual cases.
Why does Weber regard the rule of law as 'rational'? One possible answer is suggested by his statement that 'any given legal norm may be established... on grounds of expediency or rational values or both, with a claim to obedience', SEO, p. 329. 'Expediency' is, in Weber's thinking, one of the two main forms of rationality, and 'rational values' is the other. So he is saying that law may be rational in either or both of those ways, and (therefore?) claim obedience. Insofar as the law is rational, obedience is rational, and the rule of law is rational.
In other places he emphasises the rationality of bureaucracy in precisely the first of those two senses. So let me explain the two senses more carefully. He distinguishes the 'zweckrationell' from the 'wertrationell', the 'goal-rational' and the 'value-rational' (SEO, p. 115). 'Zweck' means end, purpose, goal. Goal-rational behaviour is whatever course of conduct is well-adapted as a means to one's ends, whatever they may be; i.e. it is economic efficiency from the actor's point of view - given that these are my goals, and these are the resources available to me, what is the effective way of achieving these goals? The Nazi 'final solution' might be said to be rational if it really was an efficient solution to what its proponents saw as a problem, whether they were right to see it as a problem or not. And very often Weber writes as if the intelligent choice of means is all that rationality can be. But from time to time he says that the rationality of actions is not always determined by their effectiveness in furthering goals, but sometimes by some other sort of relation to values that are not goals, and that goals and other values also can be rational or irrational. For example, to tell a lie may be an effective means of furthering one's goals, but it may violate a moral value, a value that truth-telling serves in some sense other than as a means to achieve a goal; and truthfulness is not a goal, but a 'value' of some other sort (we also 'value' ultimate goals). So occasionally he distinguishes between 'goal-rationality' - effectiveness in serving one's goals whatever they are, rational or irrational - and 'value-rationality', the rationality of goals (and not merely as means to some ulterior goal) and other values, and of actions in their relation (otherwise than as means) to some value. But only occasionally: often he treats rationality as synonymous with efficiency. And it is in this sense, I think, that he means that bureaucracy is rational in the following: 'Experience tends universally to show that the purely bureaucratic type of administrative organization... is... capable of attaining the highest degree of efficiency, and is in this sense formally the most rational known means of carrying out imperative control over human beings. It is superior to any other form in precision, in stability, in the stringency of its discipline, and in its reliability. It thus makes possible a particularly high degree of calculability of results for the heads of organization and for those acting in relation to it. It is finally superior both in intensive efficiency and in the scope of its operations, and is formally capable of application to all kinds of administrative tasks', ESO, p. 337.
Weber is thus not one of those who regard bureaucracy as synonymous with inefficiency: quite the reverse, it is the supremely efficient way of conducting administration. This is why is has been adopted by capitalistic firms, and in every institution. An institution served by a bureaucracy will out-perform its competitors, and prevail in the struggle for survival: bureaucracy has spread and continues to spread because of its survival value for social institutions. 'When those subject to bureaucratic control seek to escape the influence of the existing bureaucratic apparatus, this is normally possible only by creating an organization of their own which is equally subject to the process of bureaucratization', GM, p. 338 - because they can't beat a bureaucracy except with the aid of another one. (This is the theme of the book on Political Parties by Weber's protege Roberto Michels; his book shows how the Marxist Social Democratic Party, despite its belief in internal democracy, had become thoroughly bureaucratized and undemocratic. Later Trotsky explained Stalinism as a 'bureaucratic deformation' of Marxism.) Just as Adam Smith's pin makers who divide their labour will make more pins and sell them more cheaply than their old-fashioned competitors, and will drive them out of the market, so an army with a general staff, a government with a bureaucracy, a pope with a chancery, a firm with an efficient office, will prevail over their competitors.
Bureaucracy is in fact the division of labour applied to administration, and bureaucracy occupies the same place in Weber's account of the development of modern civilization as division of labour in general occupies in Adam Smith's account. For Weber this species of division of labour is more fundamental than the others because it initiates and orders other divisions of labour. Instructions come to the factory floor from the office. Just as Adam Smith saw division of labour in general as the cause of progress toward modern, generically commercial, society, so Weber sees bureaucracy as one of the most important causes of the development of capitalism specifically. He points to many cooperating causes (see Collins), and in The Spirit of Capitalism puts some emphasis on the moral causes - on the factors that made people strive for ever increasing profit, and to use their profits not for consumption but for further investment. But among the causal factors he often mentions the adoption of rational accounting methods: no amount of will to make a profit, or willingness to invest, would have had the desired result if investment and management had not been guided by systematic accounting, carried on of course increasingly by a bureaucracy. Once some began to be systematic others had to follow suit or go under. Labourers were 'separated' from the old-fashioned means of production by the superior effectiveness of production guided by systematic accounting - they could get a better living as employees. Capitalists adopted machinery and other innovations when their bureaucracy analyzing the possibilities of investment found that such innovation would be profitable. In fact a bureaucracy finds its own capitalists. As modern Weberians have pointed out, modern firms are run, not by owners, but by their managers, who often initiate the issuing of shares to raise capital.
But although Weber regards bureaucracy as supremely efficient, he regards its inevitable triumph with distaste. Paralleling the distinction between 'goal-rational' and 'value-rational' (and perhaps the same distinction in other words) is a distinction between 'formal' and 'substantive' rationality. Society is 'formally' rational when things are organized to maximise the attainment of people's goals, whatever they are. But it may be formally rational without being 'substantively' rational, because this organization is inimical to values rationally paramount over the goals actually served. One of these values is personal freedom, to which bureaucracy is inimical. 'The quality which best guarantees promotion [in a bureaucracy] is a measure of pliancy toward the apparatus,... of "convenience" for his superior', ES, p. 1449. Socialism would mean one unified bureaucratic system: at least now there are alternative and competing bureaucracies; see ES, pp. 1402-3, 1453-4, and Beetham, pp. 82-9. So for Weber bureaucracy occupies the place capitalism has for Marx, of the admired enemy, spreading inexorably throughout the world and into every department of life. But Weber foresees no 'death-knell'. Bureaucracy is inescapable.
But Weber does not believe that there is no point in resisting the inevitable. He was himself politically active, in a despairing kind of way - he did not expect to have success, but he went on 'resolutely', like a Stoic. Weber as politician takes his stand on certain values although (as a scientist) he cannot rationally justify them, and takes 'responsibility' for organising action aimed at realising those values although he knows that action may fail.
Weber contrasts the status honour of the bureaucrat with the responsibility of politician; see ES, pp. 1403-4, 1417, 1438. If a bureaucrat's superior gives him a directive he considers wrong he should object, but if the superior insists 'it is his duty and even his honour to carry it out as if it corresponded to his innermost conviction', On the other hand 'the politician must publicly reject the responsibility for political actions that run counter to his convictions and must sacrifice his office to them'. A genuine political leader will be ready to accept responsibility for morally dubious action, since the different parts of our value system are irreconcilably in conflict; GM pp.118-28, 147ff.
'The essence of politics is struggle' (ES, pp. 1415, 1450) to attain power; political leaders must be selected through competitive struggle. They will enter parliament only if that is the way to real power; see ES, pp.1409, 1414, 1420-1, 1450. The real leader's task is not merely to compromise interests as if politics were like a market place, but to take a stand on issues that transcend material interests; see Beetham, pp.222-6, 144-7. A person is more likely to care about such issues, and be willing to sacrifice office to conviction, if he is financially independent - he must live 'for', not 'off' politics; see ES, pp.1427, 1448, and GM pp.84-5.