What is bureaucracy? by Claude Lefort

From Telos #22

Although the concept of bureaucracy has fallen into the common domain of political sociology, theory of history, and public opinion, and has been sanctified to the success it has today, it has nevertheless remained so imprecise that it is still meaningful to question the identity of the phenomena it claims to describe. At first one is astonished at the diversity or ambiguity of the responses. But this is only a first impression. Bureaucracy appears as a phenomenon that everyone talks about, feels and experiences, but which resists conceptualization. Thus, rather than immediately attempting to provide a new definition or description, we will measure the difficulties encountered by theory, assume that they have a meaning, and from the very beginning critically examine what both motivates and perpetuates these difficulties.

Outline of The Problem of Bureaucracy

in his Critique of Hegel's Philosophy of Right, Marx draws attention to the specific nature of the social stratum in charge of the administration of public affairs. To corporations dedicated to particular activities and attached to particular interests, this stratum appears to represent a universal interest. We will follow the development of the theory of the state in Marx's later works, then in Lenin's Stale and Revolution, and its application to post-revolutionary Russian society by Trotsky, together with an examination of the role that the bureaucracy plays as a stratum inextricably bound to the structure of a class society. From this viewpoint, the bureaucracy is neither a class nor a stratum. It is a result of the division of society into classes and class struggles, since its function is to secure the acceptance of the rules of an order (an order undoubtedly connected with relations of production, but in need of being formulated in universal terms and maintained by force). Bureaucracy is "normally" at the service of a dominant class since the administration of public affairs within the framework of a given regime always assumes the preservation of its status. But since it is not simply a section of this class, when the balance of social forces permits it, it can run counter to some of its interests, thus acquiring a relative autonomy. The limits of power are always determined by the configuration of social relations. In short, bureaucracy is a special body in society because its function is such that it supports the established structure and its disappearance would mean the end of bourgeois domination. (Marx said that the Commune's first revolutionary measure was the suppression of the bureaucracy through the lowering of functionaries' salaries to that of the average worker.) Since it is not a key to social stratification, its role in the society is ascribed by the real historical agents-classes in struggle.

The viewpoint changes as soon as one observes the growth of the stratum devoted to administrative tasks in the various sectors of civil society. Thus, it is tempting to look for criteria defining a type of social organization that recognizes the similarities between the bureaucracies of the state, industry, party, unions, etc. Comparison encourages research into the conditions for the emergence of bureaucracies in order to define a type which would pull together various characteristics.

From this viewpoint, very close to Weber's thesis, the bureaucracy appears again as one particular mode of organization corresponding to a more or less extended sector within society. In other words, the social dynamic doesn't seem to be affected by the development of bureaucracies. The mode of production, class relations and political regimes can be studied without reference to a phenomenon designating only a certain type of organization.

A qualitative change in the theory of bureaucracy takes place when it is used to refer to a new class considered to be the dominant class in one or several countries, or even seen as destined to displace the bourgeoisie all over the world. This is suggested by the evolution of the Russian regime after the rise of Stalin, with the disappearance of the old proprietors and the liquidation of the organs of workers' power along with a considerable extension of the Communist Party bureaucracy and the state, which took over the direct administration of society. Similarly, social transformations connected with the development of monopolistic concentration in large industrial societies (notably in the United States) also generate reflection on the development of a bureaucratic class. This necessitates a change in the theory since, because of its role in economic and cultural life, the bureaucracy is now understood as a stratum able to displace the traditional representatives of the bourgeoisie, thus monopolizing power.

Finally, we believe that a completely different conceptualization is required if the phenomenon of bureaucratization is seen as a progressive erosion of the old distinctions linked to private property. Bureaucratization here refers to a process seeking to impose a homogeneous social form on all levels of work-at the managerial as well as the executive level-such that the general stability of employment, hierarchy of salaries and functions promotion rules, division of responsibilities and structure of authority, result in the creation of a single highly differentiated ladder of socio-economic statuses. This last thesis refers to a social dynamic in bureaucracy, and lends it a goal of its own, the realization of which engenders an upheaval of all of society's traditional structures. If this is what the problem of bureaucracy boils down to, it is important to examine each of these theses and explore their contradictions.

The Marxist Critique of State Bureaucracy

As in Hegel, the Marxist account of bureaucracy is conditioned by a theory of history. In fact, when Marx criticizes Hegel's Philosophy of Right, his own theory is still in gestation. Yet, the philosophical viewpoint still takes absolute precedence, and it is remarkable that Marx could sketch out a description of bureaucracy.

According to Marx, Hegel's error consists in having accepted the bureaucracy's self-image. It claims to embody the general interest, and Hegel decides that it does so. Marx argues that the general interest is actually reduced to the bureaucracy's interest which requires the permanence of particular spheres, i.e., the corporations and the estates, in order to appear as an imaginary universal. The bureaucracy assigns its own goals to the state. It maintains the social division in order to confirm and justify its own status as a particular and privileged body in society. As real activities take place in civil society, the bureaucracy is itself condemned to formalism since it is completely occupied with preserving the frameworks in which its activities are carried out and in legitimating them. This critique reveals a series of empirical traits of bureaucracy whose relevance remains concealed to those who cling to appearances First it is the reign of incompetence Marx writes: "The highest point entrusts the understanding of particulars to the lower echelons, whereas these on the other hand credit the highest with an understanding in regard to the universal and thus they deceive one another." But this incompetence is rooted in the system The bureaucracy is a circle from which no one can escape Finally it lives for the secret the hierarchy guards the mysteries of the state and acts as a closed corporation with respect to the outside world. Furthermore, it engenders a cult of authority which is "the principle of its knowledge and being," while "the deification of authority is its mentality."

Finally, it is exposed to a "crass materialism." The bureaucrat makes the goal of the state his own private goal: "a pursuit of higher positions, the building of a career." Marx also shows that this materialism is accompanied by a similarly crass spiritualism: the bureaucracy wants to do all, and, in the absence of a real function, it is condemned to an unrelenting activity of selfjustification.2 Marx's analysis applies to nineteenth century Germany, i.e., to a backward society. Its relevance, however, is not thereby diminished. When he elaborated his theory of the state as an instrument at the service of the dominant class, through the study of a nation where bourgeois development had erased particularism and destroyed the corporations (the France of Louis Napoleon Bonaparte), Marx kept the idea already developed against Hegel: that the state bureaucracy is essentially a parasitic body. Thus, in dealing with Bonaparte's regime, he writes: "This executive power, with its enormous bureaucratic and military organization, with its ingenious state machinery, embracing wide strata, with a host of officials numbering half a million besides an army of another half million, this appalling parasitic body, which enmeshes the body of French society like a net and chokes all its pores, sprang up in the days of the absolute monarchy, with the decay of the feudal system, which it helped to hasten."3 In Marx's eyes. the Paris Commune's most revolutionary measure was to have installed cheap government and to have suppressed the privileges and hierarchy characteristic of state bureaucracy. In State and Revolution, Lenin only reiterates what Marx had said on these points. The bureaucracy and the standing army considered as two typical institutions of the state are seen as parasites engendered by the internal contradiction which tears this society apart, but parasites which block its vital pores." To be sure, he clarifies this notion of parasitism, and points out that the recruitment of the bureaucracy from the middle and lower strata detaches part of their members from the rest of the people and allies them to the dominant class. Furthermore, the state bureaucracy is the "stake" in a permanent battle between the large parties fighting over the administrative domains. Particularly during a change of regime, these parties tend to appropriate a substantial part of the booty for their clients.

What is the relevance of the Marxist analysis and what difficulties does it encounter? In the first place, taken as an empirical phenomenon, it presents the state bureaucracy in a light which continues to clarify it today as it did a century ago. It is a critique that resembles common opinion but gives it its reasons. It is still the case that bureaucracy is a circle out of which no one can escape, that subordinates rely on their superiors to take the initiative and to resolve difficulties, while the superiors expect their subordinates to solve particular problems which elude the level of generality where they have been conceived. This solidarity in incompetence goes quite far in tying the employee, situated on the bottom of the ladder, to the system of which he is a part. As a result, it is impossible for him to denounce this system without simultaneously denouncing the vanity of his own function, from which he derives his own material existence. Similarly, bureaucrats seek the highest positions and work itself is subordinated to the gaining or maintenance of personal status, such that the bureaucracy appears as an immense network of personal relations. Actually, relations of dependence displace the objective relations outlined by the division of labour, while internal struggles are superimposed on the formal hierarchy and constantly tend to remodel it according to their exigencies. Today, more than ever, the distribution of the most important positions between the large parties appears as a division of booty whenever there is a change in regime. These observations are worth stressing. Such traits are well-known, but what is not explained is why they are not investigated: Marx, and Lenin after him, gave an account. Even if they were wrong, that should not be an excuse for not considering it. But in recognizing its relevance, it is not sufficient to stop at a superficial account of bureaucracy which retains only its official image. In this regard, Marxism preserves a freshness of approach which fares well when contrasted with the vision of certain contemporary sociologists. As already indicated, Marx only sketched out a description which was subsequently smothered by a theory. From this comes the treatment of state bureaucracy as a general category without any attempt to explain its functioning. If the bureaucracy includes all of its members, it still remains stratified (it is, in its essence, stratification) and all of its members do not participate in it in the same way. What is the location of the bureaucrat's power? Why does the bureaucracy always grow in size? Does the very life of the bureaucratic organism include a principle of proliferation? Clearly, state bureaucracies are usually staffed by middle-class elements. By becoming bureaucrats, do they remain part of their class? Do they change their mentality? Do they become sensitive to new interests? Marxism does not answer these questions: its conception of society as completely regulated by the class struggle does not encourage a study of bureaucracy for its own sake.

Today the state is the largest capitalist and the largest investor. In addition to what it administers directly through fiscal and economic policy, it tends to direct investments on a national scale. Although the state is a battlefield between large political parties which include representatives of private capital, and its policies are often the result of countervailing social forces, tile struggle between these groups is not the same as that which unfolds in civil society. When joined to the requirements of public administration, the division of interests creates a space for its own decision-making -a space which grows and develops as thhe process whereby the state increasingly drains larger amounts of capital and takes over an increasing number of tasks previously left to private initiative. The defence of the established order which guarantees the position of the rulers over those who are ruled, creates and recreates everyday the foundations of this sovereignty. In this perspective, the previous conception of state bureaucracy cannot be held any longer. In particular, the concept of parasitism seems inadequate, or at least inaccurate: why does the bureaucratic mode of organization as such multiply parasites? The thesis that, on the whole, bureaucracy is a parasitic phenomenon threads its way into Marxist theory. Actually, the bureaucracy is necessary in the context of capitalist society. In order to be effective, the critique must be located at the same level as that of capitalistic organization. If this is the case, does it not seem as if there is a dialectic of domination in modern society whereby a social stratum meant to plan and improve conditions of domination grows in proportion as industrial work invades all sectors of social life, and that the life of the masses must be subordinated to it? When all is said and done, does this process of bureaucratization, so visible in the framework of the state, also obtain within what the young Marx called civil society?

The Bureaucracy as a Type of Organisation

Let us provisionally skip these questions in order to deal with another perspective which uncovers the multiplicity of bureaucracies in modern society and draws attention to their common function and similarity. Here Max Weber is the starting point. He lists certain traits he considers typical of modern bureaucracies: (I) The duties of functionaries are officially fixed by laws, rules, or administrative dispositions; (2) The functions are hierarchical and integrated into a system of command such that at all levels lower authorities are controlled by higher authorities; (3) Administrative activity is spelled out in written documents; (4) These functions require a professional apprenticeship; (5) The work of functionaries demands complete devotion to the office; (6) Access to the profession is at the same time access to a particular technology, jurisprudence commercial science, administrative science, etc.

From this analysis some conclusions can be drawn concerning the bureaucrat. His office appears to him as the exercise of a profession to which a determinate ensemble of knowledge is attached. Moreover, it is neither de facto nor de jure the source of fees any more than it is the object of a contract in terms of which the employee sells his labour power. The particular nature of the office implies that in exchange for certain material guarantees (the assurance of a suitable standard of living), the functionary contracts a specific duty of fidelity to the office; he is in the service of an objective and impersonal goal, not of a person. This goal is inherent in the enterprise to which he is attached-state, commune, party or capitalist enterprise. Secondly, the bureaucrat enjoys a certain prestige with those he dominates. This prestige is usually guaranteed by a special status which confers on him certain rights sanctified by rules. Thirdly, the functionary is normally appointed by a superior authority. If it is true that there are certain bureaucracies whose members are elected, the pure type requires the principle of appointment. Hierarchical discipline is undermined when the functionary derives his power from the approval of electors, i.e., from below and not from above . Fourthly, the stability of employment is normally assured, even though a right of possession over the office is never recognized. Fifthly, the bureaucrat normally receives a remuneration in the form of a salary determined by the nature of employment and, possibly, by seniority. Sixthly, parallel to the hierarchical order of the bureaucracy a hierarchy of salaries is established: the majority of functionaries desire that promotions be made as mechanically as possible. Max Weber also indicates the role of certain factors in whose absence the bureaucracy would not completely develop. For example, its structure is not definitively established until the natural economy has been eliminated, i.e., until capitalism dominates society. Furthermore, the emergence of democracy allows the substitution of an administration of anonymous functionaries, detached from every particular social milieu and devoted to tasks of universal significance, for the traditional administration by notables, provided with local authority. Finally, Weber goes so far as to identify the movement of bureaucratization with the process of capitalist rationalization. More than the quantitative development of administrative tasks, what appears decisive is their qualitative change, the necessity for a large enterprise, whatever its nature, to envisage its activities from a strictly technical viewpoint and to obtain a predictability or a calculability of results as exact as possible. Bureaucracy in this sense is the social form most adequate to the capitalist organization of production and to a society based on it. The elimination of personal relations, the subordination of all activities to the application of a norm linked to an objective goal, makes it into a model of economic rationality established by industrial capitalism. Here, from a technical viewpoint, Max Weber does not hesitate to judge modern bureaucracy as superior to all other forms of organization.

It does not follow, however, that the development of bureaucracies must affect the nature of a political and economic regime, no matter how necessary they might seem once certain conditions are fulfilled. On the contrary, Weber claims that the numerical importance of this form of organization does not in any way determine its relation to power. The proof is that the state bureaucracy accommodates itself to diverse regimes - as demonstrated by France, where the state bureaucracy has remained remarkably stable. The proof lies also in the fact that during war, the bureaucratic staff of a conquered country is used by the foreign power, and continues to carry out its administrative tasks. In principle, bureaucracy is indifferent to the interests and values of a political system, i.e., it is an organ at the service of rulers located somewhere between the rulers and those who are ruled.

These analyses do not reveal their full meaning until they are located within a certain methodological perspective. Weber sees bureaucracy only as a type of social organization. Actually. bureaucracies do not necessarily realize their pure form: certain empirical conditions are needed in order that the various characteristics be simultaneously present. But once defined, the type makes the impure forms intelligible. Even when Weber states that the process of bureaucratization and of capitalist rationalization are closely tied together, this could be misleading. Historical explanations are something different from the determination of a social type. Thus, the method partly determines its results. If bureaucracy is seen as essentially neutral in relation to an economic and social system, and it appears as having no historical goal, it is because Weber sees it from a purely formal viewpoint. as a type of organization, and not as a specific social stratum which, in establishing a certain set of relations between its members, generates its own history. Consequently, Weber cannot deal with "state socialism" without prejudice. According to him, bureaucracy can adjust more easily to state socialism than it can to bourgeois democracy. Yet, the history of state socialism is alien to that of bureaucracy. Strangely enough, Weber's conclusions on this point are similar to those of some Marxists, although inspired by different principles. In the eyes of these Marxists, state bureaucracy is alien to the social dialectic obtaining at the level of relations of production. For Weber a sequence of events can be reconstituted to make sense out of the development of state socialism. Although favored by these events, however, bureaucratization does not generate them. Yet, certain historical developments can be adduced to refute this thesis even more easily than Marx's account which is concerned with an empirical description. In the system resulting from the Russian revolution, which Weber calls "state socialism" (an expression which need not be criticised here), bureaucracy is not actually alien to power. The future state leaders come out of it: Stalin made a career in the party bureaucracy. For a long time he sought the highest position before obtaining it; he added to his functions of party secretary those of the state bureaucracy before becoming the master of power. Just because his rule had a charismatic character does not mean that he was independent of the bureaucracy: the latter was the permanent foundation of his power. While charisma can disappear or change its character with the death of the dictator, the new power will reconstitute itself on the basis of the bureaucracy. The political battles concerning the direction of the state take place in the upper reaches of the bureaucracy.

Extended to the limit, the state bureaucracy comes to take over final political and economic decisions, i.e., it becomes the focus of a new system. Had Weber accepted this, he would not have formulated his definition of the bureaucratic type as he did. Given the nature of his thought. he refused to regard bureaucracy has having a dynamic goal of its own. Thus, he was unable to investigate its constitutive traits, i.e., how it is rooted in social being and increases its power. The enumeration of criteria can be useful, but as long as what holds them together remains un-investigated, the phenomenon described remains indeterminate. It matters little if one adds or subtracts a criterion: nothing allows one to decide if, in the absence of certain selected traits, a social complex is or is not bureaucratic. In order to decide, it is necessary to grasp the source of bureaucratisation.

The above does not apply only to Weber, but to every attempt at a formal definition. Thus, Alain Touraine writes: "I call bureaucracy a system or organization where status and roles, laws and duties, conditions of access to a position, controls and sanctions, are defined by their location in a hierarchical line and thus by a certain delegation of authority. These two characteristics assume a third; that the fundamental decisions are not taken within the bureaucratic organization. the latter being only a system of transmission and of execution." This definition, obviously inspired by Weber-although more concise-can readily find many applications. It is easy to agree with Touraine when he claims that a ministry is a bureaucratic organization. The same cannot be said when he adds that an industrial enterprise is only partially so. If only the first characteristic of bureaucracy is found here, how can one claim that the enterprise is a partial bureaucracy? Does he not mean that a system of organization functioning according to fixed rules and in an impersonal manner already entails bureaucratization? If, on the other hand, it is admitted that the delegation of authority is decisive, and that workers do not participate in decisions, does it make sense to speak of a "bureaucratization of work"? This ambiguity grows when Michel Crozier, in elaborating on Touraine's definition, decides that "the Western workers in general, and French workers in particular, have already largely entered the channels of bureaucracy. He tells us that "delegation of authority is not necessary for participation in a bureaucratic system." It is characterized primarily by the existence of hierarchy. Does this mean that it is possible to participate in a system without possessing authority? In such a case, however, the problem would only be displaced, because relations within the bureaucratic system, between the authority and the executive sector dedicated to the manufacturing tasks and subjected to external authority, remain to be defined: the problem of knowing what role the relations of authority play in the constitution of the bureaucracy would remain. If it is necessary to admit, on the other hand, that a bureaucratic system on the whole does not necessarily locate these relations and that it is essentially characterized by the existence of a hierarchy. it is still necessary to determine the meaning of a bureaucratic hierarchy. The notion is vague enough to be applicable to very different structures: nothing is more hierarchical, for example, than the court of an hereditary monarchy. What is, then, the basis of hierarchy in bureaucracy? What justifies a vertical classification of functions and roles? The question is always reintroduced to evaluate the import of this or that criterion functioning in the conception of bureaucracy.

If Weber enumerated certain precise characteristics of bureaucracy without wishing to privilege any one of them which could have designated another social reality, it is because he had a strong feeling of its specificity. What is interesting in his analysis is what he links to this feeling, i.e., the multiplicity of bureaucratic forms of organization in modern society. Even if he fails, he at least forces us to confront his examples and types, and to come up with a new account.

Let us return to state bureaucracy, in order to ask what stratum of functionaries Weber dealt with. His definition surely applies to ministerial personnel, or at least to those functionaries whose duties carry certain responsibilities and whose 'office' entails a loyalty to the goal of the enterprise. It is a professional formation having specialized knowledge in relation to subordinates assigned to purely executive tasks and whose labour-time is rigorously checked. But, strictly speaking, does this definition apply to all the functionaries who hold 'office'? Can one say, for instance, that according to Weber's framework, secondary school teachers are part of the French bureaucracy? The professor's personal position corresponds to Weber's characterization of the bureaucrat. Only on one point is the definition inadequate: participation in a system of authority. Access to a certain position or a level in the hierarchy does not give him power over subordinates. Similarly, his relation to his superiors is special. Obviously, he is subjected to administrative power.His lot depends on decisions taken at the managerial level. Yet, he largely escapes this power; the content of his activity is only very partially determined. His professional activity has its own goal. It is not justified in terms of a transformation of the object-which cannot be confused with the objective goal immanent in the ministerial enterprise. Finally, and above all, the secondary teacher is not in the process of making a career out of a job. He can hope for a change from one grade of seniority to another by the most rapid route. But, unlike bureaucrats, he does not seek a new function which will carry with it a higher social status, expanded responsibilities, and increased power over subordinates. The secondary school teacher remains largely an isolated individual. Undoubtedly, his activity is social, since it necessarily brings him into contact with a public, but it is not socialized. The division of labour can oblige him to specialize in a branch of teaching and thus to relate his activity to those of other teachers, without, however, generating a unity of production. rn short, if we try to apply the concept of bureaucracy in the way Weber himself did (neglecting the value judgements implied in his description), we are led to exclude certain levels of functionaries from the framework of bureaucracy while also reforming his system of interpretation.

Since Weber did not integrate French high school teachers into his type, it follows that most of the characteristics, which he considers typical and which apply to our example, acquire an import only in certain precise cases. On the other hand, the absence of certain traits makes it difficult to speak of bureaucracy. In the first place, we see a connection between a certain hierarchy and a system of authority (of command-subordination, according to Weber), such that progression in the hierarchy corresponds to amassing a higher status, new responsibilities, and more power. In the second place, Weber's idea that bureaucracy expects its members to identify with the undertaking appears at first glance to have only an apologetic function, but proves to have some sociological content. Such an identification assumes some professional activity linked to a role, itself determined through relation to other roles within the enterprise.

The bureaucracy expects a subordinate to say "the Ministry" or "the Service" instead of "I," and by this act of identification, this person exists as a bureaucrat. But this act has no meaning for those whose work renders them strictly anonymous or for those who are individualized to the point where work as such becomes a sufficient justification of existence. In other words, what Weber calls the identification with an office is something other than professional consciousness. The latter finds its end in the act of production; the former, in the occupation of an office. This professional consciousness calls for a behaviour conforming to the interest of the bureaucracy, in response to the expectations of hierarchical superiors-a behaviour proper for each member of the bureaucracy in a similar situation. Thus, the activity of bureaucrats has two characteristics: it is technical and bureaucratic. It can lose the first, not the second. For example, the intense circulation of reports or of memos in offices serves only to express the necessity of each manifesting his function to others, and the bureaucracy functions only by virtue of a mutual recognition, constantly renewed according to determined ceremony. The volume of paper internally consumed by an administration allows one to measure its coefficient of bureaucratic integration. Stripped of ~l malevolent intentions, this shows that the bureaucracy can only act by constantly reflecting its activity in the mirror of its constitution. Finally, because of the place it gives to the system of command-subordination, Weber's analysis presupposes the existence of a geographical unity-a spatial framework determined by bureaucratic activities. Of course, all the members of a bureaucracy are not necessarily assembled in the same place, but their relations, the discipline that unites them, the control of each by the others, tends to circumscribe a specific world of offices. A second example mentioned by Weber, that of the industrial enterprise, will allow us to test out his ideas and to specify ours. In the first place, we are again led to ask whether bureaucracy is only an organ of transmission and execution. Although an industrial enterprise is never autonomous and its functioning must take into account the interests of a financing capital on whom it depends, or the directives of a ministry if it is a question of a nationalized economy, the fact remains that management has a considerable power of decision. For the totality of decisions is not the action of an individual. Whatever the personality of the general manager, the power of decision is necessarily distributed among the different services and is concretized at the heart of each service only through a more or less collective participation in the solution of problems. To ask whether the direction is or is not distinct from the bureaucracy, is to pose a false problem. In every organization in which hierarchy results in delineating a function of direction -this transcends all those subordinatedd to it. Yet, the fact remains that, if the power that it formally holds was actually composite, i.e., if the decisions which fall to it by virtue of officially fixed allotment had actually been partially elaborated at various lower levels, it is still part of the framework which it dominates.

On the other hand, as with state bureaucracy, the most important thing concerning the bureaucracy of an enterprise is its boundaries. Who are the bureaucrats? Who can be assimilated into the bureaucracy? Finally, who definitely falls outside this category? Clearly, for Weber, the definition of the capitalist enterprise as a bureaucratic organization does not specify which sector, within the enterprise, could be designated as bureaucratic (although he goes so far as to claim that the capitalist enterprise offers an unequalled model of bureaucratic organization). To maintain, with Crozier, that workers are part of a bureaucracy as soon as they are placed with engineers and directors under a single hierarchical ladder, would have seemed extravagant to Weber, not because some of his criteria would have been contradicted, but because the position of a social group cannot be established only by considering its juridical status. The fact that some workers find their work assimilated to that of functionaries, says nothing about the specific nature of their work, or of their relations with other social strata within a given enterprise. The question of establishing the real situation of workers is not automatically answered if the enterprise is nationalized, the stability of work is guaranteed, or the workers are integrated with the cadres in the same hierarchical system-although these conditions could have important effects. In the industrial enterprise, the mass of workers is confined to purely operational tasks. The ordering of workshops, the number and distribution of positions, the rhythm of production, the duration and the intensity of work-all is prescribed by an administration functioning at a distance from the place of production and constitutes an alien and closed world with respect to it.

On the other hand, is it possible to consider all those who work in offices as the bureaucracy? First of all, technical services should not be confused with services of administration and exploitation. Although both share certain common norms of organization, it remains that the social relations are different by virtue of the different work performed. In short, the relations of authority and the connections with the enterprise are not similar. In technical services the engineers and the technicians, the draftsmen, have a relative autonomy by virtue of their professional knowledge. Control over labour can only be effective if the boss has a technical competence at least equal to that of his subordinates, i.e., his control must be considered a technically superior function. Social control could be practically non-existent where the work requirements were of fixed duration, sufficing to establish a normal rhythm of output. Moreover, the technicians' autonomy is also measured by their ability to move from one enterprise to another by virtue of their knowledge. Generally speaking, the position of technicians depends more on the work performed than on his place in the social organization of the enterprise.

The functioning of administrative services is something else. Here, at the bottom of the ladder, we find unskilled employees whose professional competence is rudimentary or non-existent. Between them and the general management of the enterprise, the hierarchy of positions is a power hierarchy. The relations of dependence become determining and having a function defines one against a higher level, whether it is that of the departmental supervisor, a boss, or a director. Here the double nature of employment reappears. It answers to a professional activity and expresses an established social order in which the enterprise finds its concrete existence. In fact, from the top to the bottom of the ladder, relations are such that they always serve to reiterate the authoritarian structure of administration. This does not mean, however, that those located at the bottom of the ladder participate in the bureaucracy in the same way as those at the middle or upper levels. In certain respects, the employees are like workers, deprived of any authority. They are often paid less than certain hourly labourers. Thus, their work cannot be described as an 'office' and we cannot assume that they could identify with the goals of the enterprise. Nevertheless, they are not alien to the bureaucracy: they are the dependents . They often enter into the enterprise only when provided with references certifying their "good character." They cannot advance unless they prove their aptitude to obey commands: they live in the hope of moving to a higher status. Thus, the situation of the employee is ambiguous. He is not integrated into the bureaucratic system. lie only endures it. Yet, everything tends to make him adhere to it, and he does so effectively when he accepts his superiors' ideal: promotion. Moreover, he is even less able to detach himself from the bureaucratic milieu since his work is determined by the social organization of the enterprise, and in extracting the resources that assure his subsistence, he perceives it as being as necessary as the organization itself.

Bureaucracy thus overflows the active core of middle and upper level functionaries tied to administrative and exploitative tasks: it is a hierarchy which plunges its roots even into the productive sector, where supervisors and foremen control the work of labourers. These functionaries hold real authority. Not only do they hold positions with official duties defined by a certain division of labour and submit to a certain discipline, but their function makes them participate in the power of management and leads them to identify with the enterprise as such. To say that they identify with it does not mean that they necessarily have a correct idea of the enterprise's interests nor even that they are led to place this above their own interests. In their eyes, the horizons of the enterprise are absolutely confused with those of their employment. They see the social order immanent in the enterprise as both natural and sacred, their own function as something other than a mere source of remuneration or of professional activity, but as the backbone of a system which needs their co-operation to subsist and expand.

To possess a status apparently differentiating his position from that of mere labourers, to enjoy a prestige generating others' respect, to obtain a remuneration and material advantages assuring a privileged condition of existence, to belong to a milieu from which authority flows, where subordination is the other side of a command, with opportunities for promotion-all these are the traits of the bureaucrat.

Finally, the bureaucracy of the enterprise exemplifies the mystification implicit in a purely formal description. The latter assumes that bureaucratic organization is identical with the rational organization of the enterprise, insofar as it is technically required by production itself. For, as soon as we seek to locate the strictly bureaucratic sector and to emphasize a specific type of conduct, we discover a dialectic of socialization different from the dialectic of the division of labor. This does not mean that we can determine what an adequate social organization of the enterprise would be like at a given stage of the division of labor, since this depends on historical conditions resulting from technical evolution and class struggle, but rather, that bureaucratic organization has its own ends which cannot be deduced from the necessities imposed by the organization of production. Once it is recognized that, in addition to the manufacturing and the technical sectors, every large enterprise must deal with tasks pertaining to the administration of personnel, to the sale of products and to the purchase of primary materials and machines, to the determination of production costs, etc., it does not follow naturally that the specialized services function as they do in the real framework of the modern capitalist factory. The requirements of planning, coordination, and information do not necessarily create a determined social order. This order is instituted by virtue of a social activity. From this viewpoint, it is essential to grasp how the bureaucracy creates its order. The more activities are fragmented, services diversified, specialized and partitioned, the more numerous the structural levels and the delegations of authority at each level, the more co-ordination and control sectors multiply because of this dispersion. Thus, the bureaucracy prospers. The status of a bureaucrat is measured by the number of secretaries and employees who depend on him, by the number of telephones and machines at his service, more generally, by the authority allocated to his domain of organization. As soon as conditions allow, he seeks to expand his sphere of influence and to preserve it. This tendency engenders the formation of cliques and hidden wars between departments which is stimulated by their separation. Each department is quick to blame others for errors or delays in carrying out a program. But at the same time, since this tendency responds to a common aspiration, it works itself out. The more the bureaucrats multiply, the more complicated the system of personal dependence becomes, the more the bureaucracy as a whole becomes a rich and differentiated milieu. As this process intensifies, bureaucrats derive a growing sense of their own objectivity. The bureaucracy loves bureaucrats as much as bureaucrats love the bureaucracy.

The consequences of this situation can appear paradoxical. Weber is right in claiming that the capitalist enterprise offers bureaucracy a privileged framework for development, that the latter finds a motive for its organization in the process of economic rationalization: the need for rigorous calculability and of a predictability favoring the emergence of a special stratum of administrators and imposing on them a certain kind of structure. Yet, this stratum elaborates its conduct, actively intervenes in the structuring, and, located in historically created conditions, develops while following its own interests. Thus one can see what is behind the mask of law and impersonality, the proliferation of unproductive functions, the play of personal relations, and the folly of authority.

Our third example will provide a kind of counter proof, since it presents a bureaucracy which is apparently extremely different from what we have just seen: the mass party. It is not surprising that Weber also refers to this example. Weber did not fail to observe that there is a close connection between the party and state bureaucracy since he had witnessed the emergence of a state bureaucracy in Russia under the Communist Party. Yet, one wonders why this did not lead him to revise his definition of bureaucratic organization. Actually it is not sufficient to claim that the mass party is led by a body of "professionals" in order to associate them with state functionaries or with managers. Most of Weber's criteria do not apply to them. In the first place, if one considers the organization of the party, it becomes obvious that the bureaucracy is not only an organ for executing and transmitting orders: the management becomes part of the Politbureau or a general secretary emerges from the bureaucracy. It matters little that an individual or a handful of individuals holds all real power. They have obtained it only by rising through the hierarchy of the party and keep it only because they are supported by a stratum of bureaucrats who direct party activities according to their directives, justify their decisions and apply them, while ousting all opposition. If this stratum falls apart, the power of the leaders dissipates. In the second place, the functions of bureaucrats are well fixed by rules, but they do not form a whole as in the structure of a state administration or an enterprise. There are no strict rules regulating the passage from one position to another; there is no hierarchy of salaries. The bureaucrats do not enjoy a special, officially defined status distinguishing them from the rank and file. Access to the highest positions does not depend on technical knowledge linked to a profession, and if the principle of nominating leaders by the main organs is recognized, it coexists with a principle of election, since these organs themselves are assemblies composed of delegates elected by the rank and file. Finally, it is not even necessary to be remunerated by the party in order to have an important central function in the hierarchy. This particular characteristic of the party bureaucracy follows from the position that it occupies in society as a whole. Its function is not defined by the division of labor. Rather, it is an institution based on voluntary participation which attempts to influence power-either to participate in it, or to capture it by rallying a mass of individuals around a program of demands. That a group of professionals is formed in the party in the process of coordinating its activities changes nothing in the formal definition and ascribes to this sector characteristics apparently very different from those found in industrial enterprises.

If so, how can one speak of the mass party as a bureaucratic institution? This question leads us closer to what we have sought to formulate since the beginning of our analysis: what is the social nature of the bureaucracy? If we characterize mass parties as bureaucratic institutions, it is not because we can define the parties by criteria equally applicable to industrial enterprises. Things are more complicated. In the party, we distinguish a specific sector where functions are hierarchical by virtue of participation in power; where decisions are taken in the absence of any control from below, where respon-sibilities are distributed in an authoritarian way, where organizational discipline detracts from the free examination of decisions, where a conti-nuity of roles, actions and persons is institutionalized, thus making the ruling minority practically permanent. In other words, in the party bureau-cracy appears as the antithesis of democracy. But this does not make much sense until we understand how the bureaucratic organization is constituted. Its genesis is all the more comprehensible when it does not immediately depend on economic conditions. As previously mentioned, the party is based on voluntary participation motivated by an ideological agreement on a program. This entails no particular form of Organization. The technical organizational requirement comes about only when the party attracts large masses. But the coordination of the activities of small, local sections, of assuring the best propaganda, of properly managing the assets gathered among the militants, does not necessitate any specific social milieu. It is as a result of choice that this milieu turns out to be bureaucratic. Choice here need not mean that individuals deliberately decide to create a bureaucratic organization. It only means that a certain behavior becomes predominant, with certain requirements coming to take absolute precedence, others fading. Since the party adherence is a function of voluntary participation based on the shared acceptance of some ideas, it would seem to follow that the maintenance of this participation and agreement is essential to the life of the organization. Since the party claims to articulate a collective will, and presents itself as a locus of cooperation, it would seem to lose its raison d'�tre if it used coercion with its members. Furthermore, formally it could not do so since the members are not dependent on the party for their livelihood. Yet, the party must operate within society as a whole as a coherent force, maintaining continuity of action, permanently binding those who participate in it, and finding a structure which guarantees its unity, independently of the uncertain participation of its militants.

Now, if the existence of the mass party generates this alternative, bureaucracy comes about by giving the latter considerations absolute primacy over the former and it does so in a way that makes its existence increasingly more necessary and its choices irreversible. From the very beginning, bureaucrats come into being as those whose work safeguards the party's existence and unity, while their activity in the party makes them indispensable. But this activity is peculiar. This becomes clear as soon as one compares it with the activity of ordinary militants: it is based on the very institution. It is what is usually called an organization activity. But the term is imprecise because it hides essential features: that it is always a question of directing the militants' work in a way that reinforces the party's existence and power. The organization's fundamental aspect is the multiplication of party organs: the more cells and sections there are, the more the life of the institution is differentiated, the more is its power materialized, the more the leaders appointed to be in charge of coordinating each sector. Thus, the efficiency of bureaucratic work is measured by the leaders' ability to preserve and extend the field of activity that they organize. This measurement, however, can be formulated in objective communicable terms if one considers only the formal aspect of the bureaucrat's activity. This is what gives rise to the fetishism of the agenda at party meetings, festivals or commemorations. This is why what is called activism-a feverish and vain agitation~has become routine. The number and diversity of the ceremonies from which the institution derives its daily justification goes hand in hand with the proliferation of bureaucrats. If they are entirely at the party's service, they become professionals, although they need not be that in order to act as such. It is only necessary that their activity be precisely specified, that their aim be mainly party preservation and that it be carried out according to the leaders' instructions, which makes it seem as a form of employment. On the whole, bureaucracy is this milieu for which the party structure is both necessary, sacred, and irremovable. But this milieu generates its own structure: in identifying with the goals justifying the party's existence, it makes the party-to paraphrase Marx-its private property: it sees itself as necessary, sacred and irremovable. The defense of the party is the bureaucracy's self-defense. But this implies a particular interpretation of the party's goals which results in the distortion of its original vocation. In fact, the party cannot intervene directly in the social struggle as it should according to its principles, nor can it be the locus of ideological discussion without running the risk of self-transformation or even self-destruction. Thus, the bureaucratic group feels threatened as soon as a principle of change is introduced in the party: it is naturally conservative. This conservatism inspires all inter-bureaucratic relations: the cult of authority, the will to control all activities, the value of prestige around the functions of responsibility, etc. All these are too well known traits to require further elaboration, In the last analysis, the bureaucracy's behavior has its logic. The party, in fact, is not a purely artificial organism, born out of ideological motivations. It exists as a mass organization within society as a whole. Not only does it seek power, but presently it penetrates, in various degrees, all sectors of society. This penetration assures it the allegiance of an important part of its militants who are employed in services where the party controls recruitment, either directly, or through a friendly union. Although it can appear as an incomplete bureaucracy if seen as an isolated institution, the party reveals certain material determinations of bureaucratic stability, when considered within society as a whole.

Of course, the examples which have been chosen and purposely borrowed from Weber present common traits. Most of all, however, they allow us to deal with the phenomenon in a certain way. In our eyes, the bureaucracy is a group which makes a certain mode of organization prevail, develops under certain conditions, expands along with certain states of the economy and technology only by virtue of a social activity. To attempt to grasp bureaucracy without focusing on a type of specific behavior is condemned to failure from the very beginning. Bureaucracy exists only through bureaucrats and their common aim to form a milieu apart from those whom they dominate, to participate in socialized power, and to interdefine each other in a hierarchy which guarantees them either material status or prestige.

To stress the phenomena of social behavior is not to reduce bureaucracy to a sum of similar actions. The activity of the isolated individual is unintelligible: it becomes meaningful only when placed in the context of a group. In fact, the bureaucracy comes about in an immediate socialization of activities and behavior. Here the group is not a category of activity or of socio-economic status; it is a concrete milieu where each draws his own identity. It is here that we can locate the link between bureaucracy and mass institutions. It is in ministries, unions, parties and industrial enterprises that the bureaucracy finds its adequate form because of the structural unity, the interconnection of the tasks, the number of jobs, the proximity of men within each sector, the perspective offered by a growing institutional development, the volume of capital engaged, etc. All of this defines a field of social power. It follows that the bureaucrats' identification with their enterprises is a natural mediation in consciousness whereby a group acquires its own identity. But this identification must not conceal the fact that in reality the bureaucracy does not have its destiny strictly defined by the technical structure of the mass institution. It also makes its own destiny. As the agent of a particular stratification, it multiplies positions and services, partitions various activities, generates artificial controls and coordinations, and reduces an ever growing mass of workers into merely mechanical functions in order to exercise its authority at every level.

Bureaucracy or Class

At this point, we can examine the thesis that the bureaucracy is a class. Undoubtedly, there is a ruling class in the USSR. Those who persist in denying it do so by reiterating quotations from Marx according to which the abolition of private property entails the disappearance of the ruling class, without seeing that at a deeper level a class opposition has been reintroduced in the relations of production. Here, the ownership of the means of production is no longer decisive. What determines the proletariat as an exploited class is its exclusion from the administration of production and its reduction to merely mechanical functions. What determines the position of a ruling class facing the proletariat is the fact that all decisions concerning economic life (i.e., concerning the volume and distribution of investments, wages, intensity and duration of labor, etc), are made by a particular social stratum. What is relevant here is not to discuss the class nature of the USSR, but to emphasize that bureaucracy cannot be seen as a class without analyzing its dynamics within the context of traditional capitalist society and the mass institutions where bureaucracies develop. To merely define it there as a parasitical organ, or as a simple economic category, is to overlook ho'," through its specific behavior, it creates a power base , and how it uses circumstances to consolidate and grow. On the other hand, to recognize its historicity and to establish the horizons of its activity is to locate a world which it has made in its image and where it is the ruling class. In the last analysis, the genesis of the bureaucracy in Russia is intelligible only if it is related to the social type which, in different forms, obtains in all modern countries.

But this observation concerning the conditions leading to the formation of a ruling class after the Russian revolution applies only to a special case where the bureaucracy has built its power through a specifically social activity. If it is claimed that today this class is what it is only because of its function in production, planning and the nationalization which guarantees its material basis, then it becomes difficult to claim that it results from a political bureaucracy whose earlier versions were not concerned with the extraction of surplus value within the context of modern industry, but with the concentration of authority in the hands of a ruling minority, the exclusion of the masses from decision-making and from the information pertinent to these decisions, the hierarchization of functions and the differentiation of wages. the rigorous division of competence's; in short, the scientific organization of inequality such that it becomes the principle of a new form of class oppression. Certainly, the party bureaucracy has not artificially created a whole new world. Yet, it would be inadequate to simply say that it has been served by circumstance. The new type of class domination was prepared by the destruction of the political and economic powers of the old owners, the state's taking over large sectors of production,the existence of an already concentrated industry with a modern administration, and the example of large industrial capitalist countries with a growing fusion of capital and the state. But this domination forced its way through only with a party which, by means of ideology, terror, and privilege, melted elements torn from all the classes of the old Russian society into the same mould.

It is inadequate, however, to point out the existence of a privileged class in the USSR, or even to examine its own genesis, in order to comprehend what bureaucracy actually is within the whole society that it dominates. An analysis limited to exploitation within relations of production altogether misses the nature of the bureaucratic class. Such an analysis would locate the privileged strata. But the factory managers and the planners are not the only members of the ruling class and all those who are privileged are not necessarily part of these groups. As in the industrial enterprise, a mere foreman, as opposed to an engineer, can be considered a bureaucrat because he has authority and he identifies with management against workers, Similarly, on a social level, some union or political functionaries can be considered members of the bureaucracy while some technicians, although earning higher salaries, are not members of the ruling class and do not share their values or lifestyle. The social nature of the bureaucracy cannot be deduced from its economic function. In order to be understood, it must be observed. In the absence of an observation, the question dealt with here prevents a schematic conception of history. Undoubtedly, in the USSR, as in Western countries, there is more than one class facing the industrial and agricultural proletariat. The bureaucracy is not composed either by the ensemble of the working class nor simply by some thousands or tens of thousands of leaders supported by the political police: one can only define it by pointing to the solidarity which unites its members and crystalizes them in the exercise of domination.

It is possible, however, to indicate certain traits of this class both by examining its constitution as well as by extrapolating from the testimonies of observers, or of political leaders aware of the difficulties that the regime must confront. Here two remarks are in order. First of all, the bureaucracy involves a mode of social participation different from that of the bourgeoisie. Bureaucrats do not derive private power from a professional activity which allows them to develop as a ruling class. They do not have a common interest which could generate a power to manage society in their name. They are immediately members of their class, and their personal attributes are a function of this connection: they are what they are only by virtue of their dependence from the state power which grounds and maintains the social hierarchy, i.e., political power and economic power are merged within the bureaucratic class. To participate in the appropriation of surplus value for them is the same thing as participating in the system of domination. What this means is that the bureaucracy is the privileged terrain of totalitarianism, i.e., of a regime where all social activities are measured by the same criterion of validity dictated by state power. Here pluralism of systems of behavior and of values immediately constitutes a menace not only to the ruling minority but to the ruling class itself whose integration depends entirely on its submission to the established power. Secondly, in spite of the reinforced tendency to make a single authority prevail at all levels, as already indicated, the bureaucracy cannot avoid conflicts which not only contrapose groups against each other, but also contrapose bureaucracies against each other. If the above is correct, bureaucracies exist full-fledged within mass institutions: in parties, unions, in various branches of production and various cultural sectors.In each of these contexts they attempt to grow and monopolize an increasing part of social capital in order to expand in as broad a field as possible. There is no pre-established harmony within the bureaucracy, and the unity of the class does not 'naturally' prevail: it involves a constant activity of unification. The rivalry of bureaucratic apparatuses reinforced by the struggle of inter-bureaucratic cliques is only managed by the intervention of a political principle at all the levels of social life. But the party which applies this principle is itself the broadest and most complete bureaucracy. If class unity is inconceivable without it since its mediation "politicizes" all of society so that the state tends to merge with civil society, its presence and its natural tendency to control and subordinate everything to its own power generates the sharpest tension within the ruling class. Thus, the bureaucratic system is unceasingly torn by internal conflicts, certainly different but not any less dreadful than those typical of bourgeois regimes.

To maintain that the bureaucracy is the ruling class in the USSR does not settle the question of its status in large industrialized Western nations. From one viewpoint, the formation of a bureaucratic class seems to be an extension of bureaucratic organizations: they blossom within mass institutions because technological developments make human activities increasingly more interdependent and impose a socialization of administrative tasks parallel to that of production. From another viewpoint, this class requires such a political integration and subordination to state power that it cannot operate without instituting a system of total domination. Yet, these two viewpoints are not incompatible; they allow us to see bureaucracies as a type of social behavior whose success or failure is not preordained but is a function of a complex of historical conditions. Bureaucratic organizations have an affinity for regimes where the definitive elimination of private property assures the broadest possible development and their integration within a new class structure.

Similarly, rooted in bourgeois society and fettered in their development by their natural conservatism, as well as by the profits they derive from the established mode of production, they prove incapable of doing more than invading bourgeois society, i.e., incapable of transforming the system of power. In other words, nothing warrants the claim that, in the absence of a radical social upheaval which would sweep away old regimes (as happened in Russia by workers' and peasants' revolution, and in the peoples' democracies by war), bureaucratic organizations would naturally overcome their division and become integrated within a new state apparatus as parts of a ruling class. Furthermore, bureaucracies exhibit an indeterminacy which is the source of the difficulties encountered by theory. The bureaucracy is not a class as long as it is not the ruling class, and when it becomes so, it remains essentially dependent on a political activity of unification.

To maintain that bureaucrats are already a class within all of society, would mean that they are distinguishable because of their particular interests, values, or lifestyle. Actually, they are different only in terms of their aggregation and by how they gain their status as members of a collectivity. Surely, this trait is important. The interrelations of bureaucrats within each institution correspond to a specific social model and outline a new global structure. But so long as this structure is not realized, the bureaucracy does not constitute a separate world: bourgeois society assimilates it. It is inadequate to point out that high state functionaries are members of administrative councils, or that important groups derive part of their income from the stocks that they own, since this phenomenon of embourgeoisification is comparable to a similar phenomenon of aristocratization of the bourgeoisie, who, during certain epochs, rushed to buy land and nobility titles. What is important is that the difference in the appropriation of wealth is not linked to production relations, while in the context of society as a whole the various bureaucracies split along traditional lines thus remaining heterogeneous and unaware of their identity-at least in the absence of a social crisis. Moreover, polycentrism, which is part of the essence of bureaucracies meant to crystalize into particular institutions prevents the development of class unity.

From another viewpoint, the bureaucracy retains a principle of indeterminacy even when this unity is attained: it does not exist apart from a social form of power. It is not an economic category but comes about by participating in a system of domination. Thus, there is a great temptation to deny that the bureaucracy is a class where it is seen to rule or, in specific social contexts, where it multiplies within bourgeois societies. If, on the contrary, it is claimed that it is a ruling class in the USSR, there is a tendency to neglect or underemphasize its basic constitution, the change in the function of the political in bureaucratic society, the heterogeneity of organizations, the intra-bureaucratic battles, and the differences of integration of the various strata within the class. Most of all, this class could be seen as a general model in the process of realization throughout, as if bourgeois society must naturally turn into a bureaucratic society because of capital concentration. Economic rationalization and bureaucratization are then associated, and the latter is seen as the adequate expression of the former, forgetting that rationalization obtains within a regime based on exploitation and that bureaucratization is part of a system of domination. By stressing the phenomenon of bureaucratic parasitism, it is possible to ignore that the bureaucracy simultaneously penetrates social life and poses itself as an end: it responds to technical needs but also subordinates them to power imperatives.

The study of bureaucracy, and the discussion that it calls forth, become fruitful only if these simplifications are rejected. Then the true questions can be asked, and advances toward their solution can be made on the condition that the following principles are observed: (I) Attention must be paid to the various bureaucracies instead of immediately swallowing up this image in a concept which can then be handled with such an ease that it deprives bureaucracy of all content; (2) Bureaucracy must be seen as a social formation, as a system of meaningful behavior, and not merely as a system of formal organization. This implies a historical definition of the phenomenon as a human enterprise with its own goals; (3) Special emphasis must be placed on the relations of the bureaucracy with other social strata and particularly among various bureaucratic groups within a given institution; (4) From the social nature of the bureaucracy (its sociality) no deduction should be made concerning a future based on a whole series of historical conditions which are extensions of established structures and events; (5) The question concerning the class nature of bureaucracy must be posed. The answer must avoid a comparison between the bourgeoisie and the bureaucracy. There must be a description of the specific participation of bureaucracy in society as a whole, along with the connection of its political, economic, and cultural determinations instead of relying on an a priori definition (having alleged universal significance whereas it is actually an abstraction from the nineteenth-century bourgeoisie) of the essential and the accidental features of a class; (6) In studying a particular bureaucracy, the self-image of top bureaucrats must not be uncritically accepted. The whole bureaucratic milieu must be considered in order to define the bureaucratic mentality and behavior, by relying on the workers who are most directly affected by the bureaucrats and who, as a result, cannot be easily misled: those people whom the bureaucrats dominate.4

Postscript (1970)

The above text deals with a theme which seems distant from present concerns. For young readers, who are often the ones most engaged in political struggle, this distance is even greater, since they cannot remember the proper context. Those today who are between twenty and thirty years old, have not experienced the overwhelming influence of the Communist Party during Stalin's life. The 1953 workers' insurrection in East Berlin, for some, the first to shake up the image of socialism as it existed in the popular democracies, Krushchev's de-Stalinization begun at the Twentieth Congress, the Hungarian and Polish uprisings-these are merely part of one's personal pre-history, unable to enter into lived experience, thus permitting one to assimilate them. These young people do not remember how progressive intellectuals rallied around the Stalinist banner, returning for the audacity of an independent gesture a redoubled fidelity-a time when the Left was almost entirely limited to Trotskyists, some branches growing into three or four tiny "groupuscles," or when the Communists labelled the Trotskyists fascists, treating them accordingly when they had the chance, with the result that the leftist press did not give them even the slightest mention.

Does this mean that the 1950s can awaken only the interest of the historian and that what we wrote twelve or twenty years ago has only docu-mentary value? Does it mean that, in order to comprehend the present and to attenspt to set the landmarks of change, it is necessary to look at Czechoslovakia rather than to Hungary, to study Brezhnev's last speech rather than Krushchev's Report, to challenge the Sartre who supports La Cause du Peuple rather than the one who, in 1952, gave an apology for Communist politics in France?

The questions posed at that time have not become obsolete and, in spite of changes which occasionally modify the practice and mentality of social actors, or the interpretation of ideologies, a considerable part of the historical context remains, along with the same choices and conflicts.

The distinction between what constitutes the present and the past, of what belongs to new horizons and what is lost in the distance is subtler than we are tempted to believe when considering only the generation gap or in pointing to some major signs of change-which can effectively designate the novelty of a conjuncture while hiding the continuity of the traits of a socio-historical structure. The past is not really past until it ceases to haunt us and until we have become free to rediscover it out of curiosity. But so long as the images and the words continue to fill our thoughts and excite our passions, at a distance from men and events that we have not experienced, they fully participate in the present, whether they serve to destroy, or whether we need them to preserve the framework of our life.

Thus, perhaps Bolshevism, or its Trotskyist variant, and the history of the Russian revolution no longer have any 'real' efficacy. Maybe they only provide fighting symbols for leftist militants whose goals escape them, or an identification to an imaginary community of revolutionaries, in the absence of which their opposition to the regimes in power disappears. Maybe the concept of revolution itself now passes through unprecedented paths. Maybe the USSR Communist leaders themselves need not only Stalin's ghost, but the Bolshevik legend as well, in order to successfully carry out the prosaic tasks, impossible to enumerate, of a new ruling class. What is certain is that today, the glorification of heroes and the repetition of old speeches, always accompany action and mobilize faith.

After twenty years the sources of inspiration have not spoiled. At least for a fraction of the new generations in Western societies, they are much more alive than for their elders. And in Eastern countries, the same references support both the opposition as well as the politics of the masters of power. It follows that the temptation to dissipate certain illusions by examining the great revolutionary politics of the past, to reveal what has been hidden (most often to protect its imitation) is even more necessary in the present than when they had such a great importance for those of us in Socialisme ou Barbarie. I discovered how Trotsky, who for so long we considered the guarantor of the revolutionary attitude, combined the fetishism of the party, the fetishism of the state's 'socialist bases,' and the repression of worker oppositions. Thus, we speak for a tiny number. The circle has expanded and has also intensified the equivocal nature of militancy where the will for emancipation joins the narrow subjection to tradition and the taste for the sacred. Of course, the critique of Bolshevism and Trotskyism has gained ground: the documents of the Workers' Opposition in Russia are better known, we can read Voline, Archinov and Pannekoek while the Kronstadt revolt sometimes has the value of an archetype. But it is too easy to believe that it is enough to substitute a 'good' tradition for a 'bad' one. Too often we are satisfied with changing the symbols without renouncing the authority imputed to the pure image of a founder. Even those who see how the party separates itself from the exploited strata, thus creating the kernel of a new social formation, end up by transferring to the class as such the sacredness hitherto invested in an institution or in men. Thus, questions which emerged when there were militants in the Communist Party, and which burst their system of beliefs, are suddenly extinguished under this new certitude that evil is intrinsic to organizations, while obstinately refusing to look for the conditions of their genesis in the history of the proletariat.

Thought could well free itself of certain images. What prevents it is the relationship we entertain with the representation of the past. It is the mythic function which we force it to play in order to assure ourselves of a truth already given which will not betray us, in order to finally exorcise the indetermination which is reborn ceaselessly in our living history.

In vain one relies on the movement which separates us from our old beliefs. Of course, we have managed to destroy some illusions. But the soil on which these illusions grew nourishes other germs. When we taste the bitter ecstasy of overturning our first theses, it is perhaps then that we remain the most captive of their principles. In any case, so many desires are invested on the political level that the progress of knowledge displaces its own boundaries instead of suppressing them, and each time new doors open before us, we must assume that elsewhere others are closed.

We can easily see limits in others. We are struck by their inability before a troublesome event to draw the inevitable conclusions that we have long since reached. Thus, only recently there were the militant Communist intellectuals, indignant at the Russian intervention in Czechoslovakia. For once, they condemned USSR policies and even labdled them imperialist. But it was to denounce a "tragic error" and to proclaim that a socialist state cannot act like a great power, without disavowing its principles. Their audacity was great. They rose against a hitherto unchallengeable authority. They exposed themselves to exclusion from their party. Yet, they never asked themselves whether it makes any sense to speak of socialism in the case of a state that oppresses its neighbors on economic, political, military, and cultural levels. They defended the "democratic" demands of Czech Communists and they criticized the governing group in the USSR, along with their servile managers in the other countries. They did so, however, only in order to oppose liberalism to authoritarianism, innovative methods to conservatism, as if the conflicts were not rooted in social relations and the political police terror was an accidental trait of the workers' state-the effect of a bad interpretation of revolutionary strategy or, even better, the sign of the ambitions of an intolerant clique. They deplored an error, but were quick to limit it to the case of Czechoslovakia. These ardent defenders of national communism, who still applaud the entrance of Russian tanks into Budapest, shamelessly maintain that the Hungarian insurrection was the work of reactionaries and American agents. Besides, while condemning Moscow, has it not been pointed out by others that opposition to the USSR carries the germs of counter-revolution? flow could they explain that a regime of popular democracy was able to last twenty years almost entirely withdrawn from exchange with capitalist countries and fused into the socialist bloc while preserving a bourgeoisie so strong as to endanger it at the leaders' first error? While supporting Dubcek, they were worrying about the consequences of reforms. They based all their arguments on the defense of soesalism, whileb they saw present everywhere under Brezhnev, Novotny, Dubcek or Husak. The Counter-revolution was seen sprouting everywhere so that they would not dare take a step in one direction without immediately beating a retreat. But one should also consider the position of certain non-party leftists: they also surrounded themselves with strange hesitations. The Czechs' taste for liberty raises their suspicion. What in bourgeois society they consider the most precious acquisition-however fragile, insufficient, and falsified in practice-they are ready to brand as a sign of corruption in Czechoslovakia. They themselves evoke anti-socialist forces, forgetting that they do not believe in the reality of socialism of the peoples' democracies. Thus, they remain caught in the representations which they thought they had discarded. So powerful and so widespread is the idea that the world has been divided into two camps since the Russian revolution, that they take it up again, in spite of all they have learned about exploitation and oppression in the USSR. They seemed to be certain of the fact that the abolition of private property ends in the fusion of Capital and State. But then they mechanically repeated that all that American imperialism profits from is reactionary, and that relations between East and West ultimately decide the revolutionary significance of an event. Thus the whole casuistry elaborated by Sartre and other progressive intellectuals at the time of the Hungarian insurrection has not lost its effectiveness. Its terms are disjointed, the interpreters are ideologically displaced, but the essence of the old position is preserved, the captive imaginary has not really been released.

Undoubtedly, it is difficult to discover in one's own mind the forces which draw it backwards. At least with time I have acquired a certain power, and if I still conceal a part of what guides my judgments, I feel less disarmed before old writings. It is useless to pretend a modesty which has no place when an author must stand aside before the questions of knowledge; the essays which I have written after I left the Trotskyist party are, in my opinion, better equipped to explain the phenomenon of bureaucracy than most of the analyses presently circulating under the label of the Revolution. What is interesting about them is that they were guided by the desire to apply to the labor movement or to the forces which claim to be part of it, the principles of analysis that Marxism had elaborated in the critique of bourgeois society. To be sure, this was not the result of an intellectual decision. It was the experience of militancy, lasting several years, that taught me to scrutinize the strange logic whereby a group (weak, numerically and, because of the inefficiency of its actions, relatively free of political and economic constraints), reintroduced the rules, practices, and inter-personal relations typical of the organizations which they wished to fight, reweaving the same kind of social fabric, cultivating the principles of frag-menting sectors of activity, segregating information, making its existence an end in itself and finally, presenting a cloudy account closed to reflection. Such an experience indicates some of the reasons why Trotskyism, in spite of all its critiques against Communist Parties, did not really succeed in distinguishing itself from them. Although it formulated different objectives in its program and it insisted on the decisive function of mass mobilizations, the social relations that it instituted were ordered according to a similar model. This was best shown in its inability to confront the essential question of a sociological definition of Stalinism, i.e., of inquiring into its social basis. At best, Trotskyists reproduced Lenin's account of the degeneration of social democracy in terms of the emergence of a labor aristocracy. Ordinarily, they stuck to the pure and simple denunciation of leading groups judged opportunist or incompetent, and associated their prestige to the Russian revolution, in the conviction that the isolation of the socialist state left the revolutionary enterprise in shambles, thus favoring the temporary advent of a bureaucratic caste. But their failure was symbolic, for the same conceptions ultimately pervaded all the analyses of the non- Stalinist Marxist Left. Of course, this Left distinguished itself from the Trotskyists on many issues, beginning with their insistence on remaining in the immediate vicinity of the Communist Party. But this Left was Trotskyist in its ignorance, by virtue of its double conviction that party policies were explainable in terms of methodological errors or in terms of a deformed representation of the revolutionary task, and that these policies were the result of unexpected "accidents" after the October revolution, i.e., disturbances in the 'normal' development of socialism.

It is not by chance that, in a polemic with Sartre, I developed an argument which, although it was seen as "Trotskyist," was largely directed against Trotskyism. I discovered that progressives and Trotskyists could not help but meet as soon as they eliminated some social phenomena from Marx's critique. Marx had stressed the divergence between ideology and praxis. Moreover, he knew how to do a critique of economic, political, religious, or philosophical ideologies as a privileged means of unveiling contradictions operating at the level of praxis. For the progressives and Trotskyists. this route was lost as soon as what they were dealing with was no longer the bourgeois class or the Western capitalist system. They limited their critique of Communist Parties and of the social strata from which they drew their force to the level of ideas. They attacked these ideas head-on as if they were without depth, self-sufficient, and did not occlude social relations. Yet, Marx distinguished historical from sociological analysis. His study of capitalism, so rich in references to events, focused on the specific logic of a system and on the articulation of oppositions which develop once the division of capital and labor is carried out on a large scale. Aware of the need to describe the capitalists' actions, the correctives they needed, the resistances that they awakened, and thus to outline certain sequences of an empirical genesis, he nevertheless sought to decipher in apparently contingent facts the signs of a necessity which was not the conscious product of peoples' activities, but was imposed by their ignorance and often even at the cost of their immediate interests. Our epigones, on the other hand, cling to the level of historical development. When analyzing the USSR, they can only grasp the chain of events invoking the consequences of the civil war, the revolutionary defeats in Europe, or the capitalist blockade. It would have been scandalous for them to admit that the course followed by the past revolutionary regime was inevitable (a limited hypothesis having only heuristic value) and that the social system that emerged had properties which had to be studied in themselves. Furthermore, the powerlessness to detach themselves from an explanation in terms of events coincided with the powerlessness to discover, beneath the representations and institutional forms, the social relations which support them. Convinced that state ownership of the means of production and the institution of the plan were the result of the revolution, they located in them the bases of socialism without ever asking how these institutions modified the relations established in the process of production, their real function in the socio-economic system, and the oppositions in which they were embedded.

No doubt, we would not have been able to base the critique of workers' organizations (emerging from our experience in a small and militant party) and their concomitant mode of representation (i.e., to measure the reversal of the Marxist problematic) if we had not simultaneous]y learned to recognize the USSR-thanks to Castoriadis' enlightening studies of the traits of a bureaucratic capitalism. The two analyses bolstered each other. But with ideas, it is advisable to move beyond historical descriptions. What is important is that knowledge of a bureaucratic phenomenon involves reflection on the social conditions which give rise to it. As long as these conditions remained hidden and we naively accepted the norms of our milieu, we were unable to give free reign to our questions. We continue to believe that an analysis of the USSR will be fruitful only if it is connected to an analysis of the organizations of the labor movement in Western countries and of their mode of insertion in the capitalist system -just as an analysis of revolutionary undertakings at the turn of the century (in particular, Bolshevism), assumes that we examine the divorce of practice and ideology in these organizations. However, as legitimate as it may seem, the critical movement in our earlier essays today seems to suffer from the obstinate prejudice of remaining in the strict framework of the Marxist interpretation. Fidelity turns into equivocation when it looks for pregiven answers to new questions.

Since we are only interested in fixing the stages and limits of our analysis of bureaucracy, it should be noted that it was conducted in such a way as to leave intact the image of the proletariat as a revolutionary class-as the carrier of universal historical goals. When we saw in the USSR the existence of a ruling class whose power was based on collective ownership of the means of production, believing that the whole economic system was ordered in such a way as to maintain the division between a mass of mere "doers" and a minority monopolizing managerial tasks, we assumed-without even making any explicit hypothesis-that the new class antagonism reproduced the contradiction denounced by Marx in his examination of bourgeois society. We substituted the bureaucracy for the bourgeoisie, although it had come about through a different process. At any rate, the proletariat's position remained unchanged. The only difference was that now it was in a position to discover the true nature of its goals, until then concealed under the necessity of the struggle against private property. Only now could it recognize the basis of socialism in the workers' administration of enterprises and collectivities. We imagined that in contemporary bourgeois societies the process of bureaucratization, which was becoming increasingly evident in spite of the maintenance of old forms of ownership, created for the working class an analogous consciousness of its goals-a consciousness that would not fail to operate during periods of crisis when labor Organizations would be forced to openly uphold the capitalist system. In other words, we assumed that the world proletariat had reached a stage in which the task that Marx assigned it could be carried out. When we attacked the development of bureaucracy in the West as well as in the East, we considered the trans-formations in the industrial mode of production-the concentration of enterprises, the rationalization of tasks due to technological change, and the class struggle, the growing intricacy of productive and organizational functions - as affecting only the structure of the ruling class. Finally, our analysis of the genesis of bureaucracy in the organization of the working class and the institutionalization of its forms of resistance, did not challenge but, on the contrary, made even more evident, the proletariat's vocation to install a society freed from all domination. We believed that one needed a trial of alienation, even in the process of emancipation, for the critique of all alienations to be carried out.

In retrospect it seems that we lacked audacity. We were afraid to admit that the transformation of the mode of social domination could involve a profound modification of the antagonistic terms described by Marx and, consequently, would call for a revision of his model. Even when considering the economic sphere we would have had to inquire into the changes affecting the nature of social labor.

Similarly, the process of bureaucratization results in a tendency towards homogenization of models of action, social relations, and norms. Formerly this tendency was limited to workers' labor within large industry. At present this tendency has expanded to strata of technicians, planning agents, and beyond. Not only has it extended to state administrations on which the productive apparatus depends, production of services and to large scientific laboratories, but also into domains which would seem naturally not amenable to such assimilation-public health, education, juridical institutions, etc. At the same time, the relation of workers to the enterprise has been modified. This relation can no longer be encapsulated in the clauses of the contract analyzed by Marx, but has expanded to encompass a network of obligations covering the workers' social life, e.g., through institutions of social security, housing, education, and leisure. Moreover, the evolution of technology and the rationalization of tasks has the consequence of changing the proportion of skilled and unskilled labor in industry, the productive tasks in the old sense of the term, and organizational tasks. In examining these phenomena, it is futile to maintain that the proletarianization of society spreads, according to the schema outlined by Marx, for the mass of men separated from the means of production do not resemble the image that he had of them. Ultimately, those factors resulting in heterogeneity are no less powerful than the forces of resistance. In short, it is no longer possible to mix together in the same, simple social stratum the most dispossessed, the most exploited and the most frustrated in their creativity. The last are precisely those in whom the capacitiy of knowledge and of intervention in the milieu of labor is most stimulated by their training and the quality of their tasks. But they do not suffer the most from exploitation, which remains the lot of the factory workers, nor do they fail to benefit, sometimes substantially, from the growth of revenues. As for the most dispossessed, those who presently perform unskilled labor, they are not the most exploited in the sense that it is not from their production that capital extracts the maximum surplus value that it needs to reproduce itself. It does not follow from this that the working class has been erased: the specificity of blue collar labor remains, along with the division of manual and mental labor, in spite of modifications-especially in the latter. Nevertheless, one should not reestablish the classic antagonism between technicians and professionals, on one side, and administrators and technocrats on the other. This opposition surely exists, but it does not imply, as Marx believed, a class which is excluded from the process of socialization instituted by capitalism: a class condemned to discover itself as alien to bourgeois society, a class which is not a class, witnessing in its very existence-when it escapes the status of an economic category from which it receives its definition from outside-its vocation for communism.

With the expansion of bureaucracy, several contradictions converge: between leaders and led, between the strata which receive only scraps from economic growth and the strata which ceaselessly increase the size of their advantages, and between a minority in control of the means of knowledge and information, the production and diffusion of representations, and the masses who, in spite of their formation and their increasing importance, are deprived. In addition to these contradictions in the labor process, there is another one, which contraposes collectivities in all sectors of social life and culture against rules which determine behavior in every minor detail and plugs it into the planned circuits of giant organizations. But this opposition spreads in several directions. It mobilizes various modes of Opposition. One of these takes place within the system and is the result of bureaucratic impotence to satisfy recognized needs that are even intensified by the multi-plication of organizational apparatuses. Another mode of opposition translates the desire for collective control of resources while a third places the fringes of the Population-essentially youth-in a position of deviance, makes them into outcasts, tends to destroy symbolic references without which the relation to reality dissolves.

In considering the ambiguous characteristics of this revolt that strikes at the very heart of the system of domination by revealing the mechanisms which guarantee the combined functioning of exploitation, oppression, and ideology while at the same time shaking up all symbolic references to socialization, we can measure the distance which separates us from the world analyzed by Marx. In this world, the proletariat was the outsider and at the same time the carrier of productive forces-itself being the greatest productive force. Thus, it was designated as the revolutionary class.

Presently, the producer is not the outsider. It obtains, rather, in the rejection of the models and norms of industrial society. Strategies ordered according to realizable objectives cannot be based on conflicts between owners, means of production, and workers. Thus, it becomes impossible to make everything converge toward a single revolutionary focus to preserve the image of a society centered around the praxis of a class, to maintain, in paraphrasing Marx, that bureaucracy necessarily tends toward its own destruction by raising against itself as in a single man the mass of the dispossessed. The locations of conflicts are many and the revolutionary demand par excellence of collective self-management is gaining ground.

Undoubtedly, all of this applies primarily to Western capitalist societies, but it would be a mistake to imagine that the problems are posed in radically different terms in the popular democracies and in the USSR. Certain indications suggest that these problems are only masked by repression which comes down on any and every opposition. The force of this repression, the visible figure of power, has the effect of crystalizing revolutionary energies as soon as authority vacillates. The insurrections of East Berlin, Poland, Liungary. and Czechoslovakia provide ample evidence. Thus, one can expect in the USSR-in some unforesecable future-a crisis of the regime, whose consequences will have an unheard of impact in Eastern Europe as well as in the Western world. But this eventuality should not lead us to forget the complexity, and indeed the heterogeneity, of conflicts at work in modern industrial society-conflicts for which only the lazy imagination of the little heirs of Leninism can delight in foreseeing the solution in a "good" dictatorship of the proletariat.

A study of the peculiar traits of bureaucratic regimes where nothing remains of bourgeois institutions must inquire farther than I have done in order to discover where the critique of totalitarianism leads. It will not do to refer to the logic of bureaucratic organizations, the new mechanisms whereby the state tends to penetrate all the details of the productive process, in all the representations and relations between people in civil society and culture. Nor is it sufficient to recognize in the party the opposite of what it pretends to be-the pivot of a totalitarian integration. Nor can one point to a fundamental contradiction between control and parasitism. In light of this analysis, it is advisable to carry the critique into the home ground of Marxist theory.

An examination of the Russian regime challenges nothing less than the definition of social reality and, with it, the distinction between base and superstructure. Even if one notes that social relations are generated at the level of production, and that property relations are only their juridical expression-as Castoriadis has shown-one still remains too close to the Marxist problematic. What escapes us is what distinguishes bourgoeis society from bureaucratic society. To be sure, it is possible to point to a pertinent structural trait but it should not be forgotten that it alone cannot characterize it. The very definition of relations of production, reduced to the Opposition of means of production and labor power, remains abstract as long as it does not clarify what it deals with, as long as this relation remains purely within economic space. Rather, it must be acknowledged that it allows this space to come about, that it is at the source of a system of operations (specified in terms of production, exchange and distribution), which are in turn dependent on a specific institutional structure, where modes of power and representation are articulated according to various political and symbolic schemas. In a sense, Marx allows one to think with the concept of mode of production, of a structuring of the social field which locates various features of the economy, of the policy and of the system of representation, as well as their articulations. This structuring, however, assumes and does not generate the referents of the economy, politics, or the symbolic. Indeed, we need only consider the advent of capitalism to note the impact of extra-economic factors. Even when it is characterized by a particular order of economic Operations and regulated by specific mechanisms, the logic of a social system can only be grasped by connecting the network of relations under the triple heading of production, power, and representation. Seen in this fashion, it is certainly possible to distinguish between what comes from the base and what comes from the superstructure. However, this cannot be expressed in terms of the distinction between the economic and the political. It obtains on two levels: it acts on the level of representation where the function of imagination cannot be confused with that of symbols which establish the possibility of social communication and make up the shell of the economic-political field. Otherwise, how is it possible to indicate the originality of the modern bureaucratic system? Plow is it possible to escape the alternative of a bad sociology which sees in it either a variation of industrial Society, or a variation of an a-temporal formation such as Asiatic despotism? To move forward in the analysis, we must ask how, with the destruction of the bourgeois regime, are the articulations of a social field reproduced on all levels, how power relations, and the operations of production and representations combine according to a new model of socialization. If, unable to do so, we preserve, e.g., the classic Marxist conception of the state, or if we dismiss a priori the political or symbolic function, the traits of totalitarianism will always appear accidental.

Such an analysis would have at least one practical consequence. So long as we remain prisoner of the Marxist schema, all signs of oppression-no matter how quick we are to denounce them-turn out to have no Importance. Similarly, as we have noted, democratic demands do not constitute an object for sociological interpretation, if they are seen as expressions of the influence exercised by bourgeois regimes, or a reflection of a-temporal humanist values. A new examination of the social system would have to persuade us that with democracy we are in the midst of a fundamental process of socialization-if we can read beyond the forms to which it is attached in bourgeois regimes.

Lastly, it was by reexamining my views concerning the degeneration of the "workers" parties and unions that I became aware of a critique too faithful to the spirit of Marx. Without doubt it is important to observe the structural homology between "revolutionary" organizations and of the organization of the industrial system that they hope to destroy. Lenin's views in What Is to Be Done? bears witness, in an exemplary and explicit manner, to the transfer of norms from the industrial enterprise (the militarization of labor) to the model of the party. Yet, the problem is not exhausted by invoking the alienation which leads the exploited to reproduce in their own organization the constraints that they suffer in bourgeois society-or which leads them to divest themselves of their ability to direct their emancipation after having been dispossessed of the ability to direct their production. Nor is the problem exhausted by emphasizing the role of an intelligentsia quick to transform into power the superiority that knowledge gives them. These answers are not false, but they leave in darkness the mechanisms which determine the repetition. The adherence to models of authority and hierarchy, the belief in the knowledge of the leader, the tenacious fidelity to a tradition, the attachment to symbols, the fetishism of the institution, do not point only to the inability of the working class to discover its own identity. These phenomena take on the investment of energy-both individual and collective-in the service of a socialization about which Marxists wish to know nothing, although they are actually very good at mobilizing it. Like bourgeois regimes, the bureaucratic regime would crumble if it did not nourish identifications which conceal servitude and antagonisms, and which keep the large majority under the authority of the leaders. By wanting to ignore the import of the 'imaginary,' one only exposes oneself (under the good colors of revolutionary optimism-itself mystifying and mystified) to the maintenance of an exercise in repetition. These remarks on bureaucracy are far from attaining their goal. My hope is that the reader, like myself, finds here the inspiration to continue.

Notes

1. This article originally appeared in Arguments, no. 17 (1960); reprinted in Elements dune Critique de la Bureaucratie (Paris: Droz, 1971). The "Postscript" was written in 1970.English translation by Jean L. Cohen.
2. Karl Marx, The Critique of Hegel's Philosophy of Right (Cambridge University Press, 1970), p. 47.
3. Karl Marx, The Eighteenth Brumaire of Louis Bonaparte (International Publishers, ~969), p. 121. Emphasis added.
4. The analysis of burcaucracy by the Socialisme ou Barbarie group differs from that of Trotsky and Burhoam as follows (this section originally appeared in Arguments, no.4l]une-September, 1957]): "The ideas of Trotsky and Burnham differ qualitatively from those of the Socialisme ou Barbarie group. Trotsky always saw the bureaucracy as a parasitic transitory formation typical of a particular historical juncture-as a fungus grown on the socialist organism that a new revolutionary wave would readily sweep away. He always rejected the idea that it represented a social class and a new social type. The existence of the bureaucracy did not alter the nature of productive relations: the Russian proletariat simply had to chase it away as one does with a bad manager, since there already were socialist relations. Socialisme ou Barbarie denounced Trotsky's formalism by showing the absurdity of a socialist society where the producers are expropriated from all managerial tasks. It substituted the idea of a bureaucratic society for that of a bureaucracy in society, i.e., a society which produced and reproduced itself by separating the producing masses from a social stratum which collectively expropriated surplus value. Such a bureaucratic society was made possible by the rigorous integration of all bureaucratic strata by the state apparatus. Furthermore, Socialisme ou Barbarie also pointed to the function of the official ideology-borrowed from Marxism-Leninism-in the interests of the bureaucracy. Finally, Socialiame ou Barbarie claimed that Trotsky's inability to analyze the bureaucratic phenomenon was linked to his general notion of revolutionary struggle (the absolute pre-eminence of the party) and of socialist society (state centralization) which inadvertently facilitated the advent of a new society of exploitation.

Buraham's merit, on the other hand, is to have pointed out the separation in moslem capitalism between production and ownership, and the formation of a new type of society. But this is where the analogy with Socialisme on Barbarie ends since the latter rejects Burnham on the following points: (1) Burnham considers the factory managers as the members of the new class and the real masters of society. He fails to see the changes in the phenomenon of management which now comes to engulf society as a whole. He merely substitutes the managers for the private capitalists without realizing that the process which places power within the context of the enterprise and in the hands of managers tends to dispossess every sector of its autonomy and subordinate everything to the state apparatus.

Furthermore, he fails to see that the bureaucratic mode of social domination generates new relations between its members whose power no longer flows from their private economic activity. They are able to pose themselves as a separate class by rigorously subordinating themselves to a control organ which guarantees a permanent integration through the police and ideology. Hence, Burnham's acrobatics in trying to explain two otherwise inexplicable phenomena: in the USSR the bureaucracy is staffed by a stratum of political functionaries and the factory mangers, whatever their lisiluence, do not hold power. (2) Burnham believes that the rise of industrial managers is a result of their scientific knowledge, against which the mass of producers are seen as ignorant. Accordingly, the managers are indispensable and without them the factory could not function. Against this view, Socialise ou Barbarie claims that (a) the continuous socialization of work has exploded the managers' old tasks since now the working of the enterprise is guaranteed at all levels by collective organs, and that the existence of a separate managerial apparatus answers the social need of exploitation rather than technological requirements,- (Is) constant conflicts tear factory organization apart, with the social hierarchy destroying cooperation and generating an irreducible irrationality; (c) managers confront these basically insurmountable conflicts daily in the attempt to facilitate cooperation and initiative among producers while keeping them in coercion, isolation and inertia. In other words, Socialisme ou Barbarie explains the existence of bureaucracy in terms of the class struggle and not as a function of technological progress. (3) Against Burnham, Socialisme ou Barbarie sees the contradictions of earlier capitalism to have been transposed and even intensified within the bureaucratic society. Actually, the advent of the bureaucracy is the result of a fundamental historical tendency described by Marx as "the socialization of society." The bureaucracy seeks to coordinate all of the activities that it deals with, have all individuals participate in the social totality by formally denying all class distinctions, while radically contradicting this tendency by its very existence, its system of oppression, its hierarchy, and its fragmentation. It itself pays for this contradiction with an unrelenting internal struggle. This double movement is the reason why bureaucracy exists only in the horizon of communism, and generates the need for its own destruction. In other words, for Socialisme ou Barbarie bureaucracy is a total social phenomenon which is intelligible only in the perspective of modern history and class struggle. The theory of bureaucracy is the theory of revolution."