Cornelius Castoriadis

On the Content of Socialism

From the Critique of Bureaucracy to the Idea of the Proletariat's Autonomy


Published: Originally as "Sur le contenu du socialisme, III: La Lutte des ouvriers contre l'organisation de l'entreprise capitaliste," S. ou B., 23 January 1958. Reprinted in EMO 2, pp. 9-88.
Translation: David Ames Curtis
Transcription: Class Against Class
HTML-markup: Jonas Holmgren

Table of Contents:


We have tried to show[1] that socialism is nothing other than people's conscious self-organization of their own lives in all domains; that it signifies, therefore, the management of production by the producers themselves on the scale of the workplace as well as on that of the economy as a whole; that it implies the abolition of every ruling apparatus separated from society; that it has to bring about a profound modification of technology and of the very meaning of work as people's primordial activity and, conjointly, an overthrow of all the values toward which capitalist society implicitly or explicitly is oriented.

This elaboration allows us in the first place to unmask the mystifications that have been built up for many long years around the notion of socialism. It allows us to understand first of all what socialism is not. Cast in this light, Russia, China, and the "popular democracies" show their true face as exploitative class societies. With respect to the present discussion at least, the fact that bureaucrats have taken the place of private employers appears to be of absolutely no consequence.

But it allows us to say much more. Only by beginning with this notion of socialism can we comprehend and analyse the crisis of contemporary society. Going beyond the superficial spheres of the market, consumption, and "politics," we can see now that this crisis is directly tied to the most deep-seated trait of capitalism: the alienation of man in his fundamental activity, productive activity. Insofar as this alienation creates a permanent conflict at every stage and in all sectors of social life, there is a crisis of exploitative society. This conflict is expressed in two forms: both as the workers' struggle against alienation and against its conditions, and as people's absence from society, their passivity, discouragement, retreat, and isolation. In both cases, beyond a certain point this conflict leads to the overt crisis of the established society: when people's struggle against alienation reaches a certain intensity it becomes revolution. But when their abstention from society goes beyond a certain limit, the system collapses, as the evolution of Poland's economy and society in 1955 and 1956 clearly shows.[2] Oscillating between these two limits, there unfolds daily life in modern societies. These societies succeed in functioning only in spite of their own norms, inasmuch as there is struggle against alienation and inasmuch as this struggle does not go beyond a certain level. Such societies therefore are based on a fundamentally irrational premise.

In resuming our analysis of the crisis of capitalism, we start therefore with an explicit notion of the content of socialism. This notion is the privileged centre, the focal point that permits us to organize all perspectives and to see everything again with new eyes. Without it, everything becomes chaos, fragmentary statements, naive relativism, mere empirical sociology.

But this is not an a priori notion. The proletariat's struggle against alienation and its conditions can take place and develop only by setting forth - be it in the shape of real relations between people, be it in the shape of demands, aspirations, and programs - forms and contents of a socialist nature. Consequently, the positive notion of socialism is only the historical product of preceding developments, and in the very first place, of the activity, the struggles, and the mode of living of the proletariat in modern society. It is the provisional systematisation of the points of view that the history of the proletariat offers, of its most everyday gestures as well as its most ambitious actions.

In a shop, the workers set things up among themselves in order to make the maximum amount in bonuses as well as to produce less than the norms. In Budapest, they battle against Russian tanks, organize themselves into councils, and lay claim to the management of the factories. In the United States, they insist on stopping the assembly line twice a day so they can have a cup of coffee. At the Breguet factories in Paris last spring, they went on strike and called for the abolition of most of the categories into which they are divided by management. For more than a century they have gotten killed crying, "To live working or to die fighting." In the "socialist" factories of the Russian bureaucracy, they force wages to be levelled out, despite the bitter complaints Khrushchev and his clique make in their speeches. With different degrees of development and various levels of clarification, all of these manifestations and, figuratively speaking, half of all the everyday actions of hundreds of millions of workers in all the enterprises of the world express this struggle for the instauration of new relations among people and with their work. And these manifestations and actions are comprehensible only in terms of a socialist perspective.

We must understand fully the dialectical unity the following diverse moments constitute: the analysis as well as the critique of capitalism, and the positive definition of the content of socialism as well as the interpretation of the proletariat's history. No critique, not even an analysis of the crisis of capitalism, is possible outside of a socialist perspective. Indeed, such a critique could not be based

Upon nothing at all - unless it be upon an ethics, which twenty-five centuries of philosophy have succeeded neither in founding nor even in defining. Every critique presupposes that something other than what it criticizes is possible and preferable. Every critique of capitalism therefore presupposes socialism.

Inversely, this notion of socialism cannot be merely the positive, flip side of this critique; the circle would then run the risk of becoming completely Utopian.

The positive content of socialism can be derived only from the real history, from the real life of the class that is tending toward its realization. Here we have its ultimate source.

But neither does it mean that the conception of socialism is the passive and complete reflection of the history of the proletariat. It is based just as much upon a choice, which is merely the expression of a revolutionary political attitude. This choice is not arbitrary, for there is here no rational alternative. The alternative would be simply/the conclusion that history is only a "tale told by an idiot, full of sound and fury, signifying nothing," and that it can only remain so. It is only in terms of a revolutionary politics that the history of the proletariat can be the source of this politics.

For someone with a different attitude, this history is merely the source of statistics and monographs of anything at all and ultimately of nothing at all. Indeed, neither the critique of capitalism nor the positive definition of socialism, neither the interpretation of the history of the proletariat nor a revolutionary politics is possible outside of a theory. The socialist elements that the proletariat continually produces have to be extrapolated and generalized into the total project that is socialism, otherwise they are meaningless. The analysis and critique of class society have to be systematized, otherwise they have no portent of truth. Both are impossible without a theoretical labour in the proper sense, without an effort to rationalize that which is simply given.

This rationalization involves its own risks and contradictions. As theory it is obliged to begin with the logical and epistemological structures of present-day culture - which are in no way neutral forms, independent of their content, but which express rather, in an antagonistic and contradictory fashion, the attitudes, behaviours, and visions of its subject and object, which have their dialectical equivalents in the social relations of capitalism. Revolutionary theory therefore constantly runs the risk of falling under the influence of the dominant ideology in forms that are much more subtle and deep-seated, much more hidden and dangerous than the "direct" ideological influence traditionally denounced in opportunism, for example. Marxism has not escaped this fate, as we already have shown,[3] and we will provide still more such examples.

It is only by returning each time to the source, by confronting the results of theory with the real meaning of the proletariat's life and history, that we can revolutionize our very methods of thought, which are inherited from class society, and can construct through successive upheavals a socialist theory. Only by assimilating all these points of view and their profound unity can we advance.

We begin our analysis of the crisis of capitalism with an analysis of the contradictions of the capitalist enterprise. The concepts and methods thus acquired in this primordial domain, the domain of production, will allow us then to generalize our investigation and subject the various social spheres and finally all of the social as such to this examination.


The Contradictions in the Capitalist Organization of the Enterprise

For the traditional view, which is still quite widespread today, the contradictions and irrationality of capitalism exist and actively manifest themselves on the overall level of the economy, but affect the capitalist enterprise only by ricochet. Leaving aside the conditions imposed upon it by its integration into an irrational and anarchic market, the enterprise is the place where efficiency and capitalist rationalization reign supreme. Under penalty of death, competition obliges capitalism to pursue the maximum result with the minimum of means.

For is this not the very goal of economics, the definition of its rationality? In order to arrive at this goal, it puts "science in the service of production" to an ever-increasing extent, and it rationalizes the labour process through the intermediary of engineers and technicians, those embodiments of operant rationality. It is absurd that these enterprises manufacture armaments, absurd that periodic crises make them work below capacity - but there is nothing to criticize in their organization. The rationality of this organization is the basis on which socialist society will be built once the anarchy of the market is eliminated and other goals - the satisfaction of needs rather than the maximization of profits - are assigned to production.

Lenin subscribed to this view absolutely. As for Marx himself, there was no basic difference. For him, the enterprise certainly is not pure rationalization. Or, more precisely, this type of rationalization contains a profound contradiction. It develops by enslaving living labour to dead labour, it signifies that the products of man's activity dominate man, and therefore it gives rise to a kind of oppression, a kind of mutilation that increases without ever stopping.

But it is a contradiction that is, if we can call it that, "philosophical," abstract, and this is so in two senses. First of all, it affects man's fate in production, and not production itself. The permanent mutilation of the producer, his transformation into a "fragment of a man" does not impede capitalist rationalization. It is merely its subjective side. Rationalization is exactly symmetrical to dehumanisation. The same step carries both of them forward. To rationalize production means to ignore and even to deliberately crush people's habits, desires, needs, and tendencies insofar as these are opposed to the logic of productive efficiency, while ruthlessly subjecting all aspects of labour to the imperatives of achieving the maximum result with the minimum of means. Necessarily, therefore, man becomes the means of this end that is production.

As a result, this contradiction remains "philosophical" and abstract also in a second sense: Without mincing words, it is because we cannot do anything about it. This situation is the inexorable result of a phase of technical development and ultimately even of the very nature of economics, "the reign of necessity." This is alienation in the Hegelian sense: Man has to lose himself first in order to find himself again - and to find himself again on another plane, after having gone through purgatory. It is the reduction of the workday that will allow there to be a socialist organization of society, and it is the abolition of the waste-fullness of the capitalist market that will make man free - outside of production.[4]

In fact, as we shall see, this philosophical conception is the real contradiction of capitalism, and the source of the crisis in the most down-to-earth and material sense of this term. In its most microscopic as well as its most large-scale aspects, the crisis of capitalism directly expresses the following fact: that man's situation and status qua producer under capitalism are contradictory and ultimately absurd. The capitalist rationalization of the relations of production is only rationalization in appearance. This huge pyramid of means ought to acquire its meaning from its ultimate end. But having become a goal in itself and detached from everything else, the ultimate end of increasing production for its own sake is absolutely irrational. Production is one of man's means, man is not one of the means of production. The irrationality of this ultimate end determines from one end to the other the entire capitalist production process; whatever rationality it might contain in the domain of means, when it is put in the service of an irrational end, it becomes irrational itself. But the principal one of these means is man himself. To make man completely a means of production signifies the transformation of the subject into an object, it signifies treating him as a thing in the domain of production. Whence we have a second irrational aspect, another concrete contradiction, insofar as this transformation of people into things, this reification, is in conflict with the very development of production, which is indeed the essence of capitalism and which cannot take place without also developing people.

But what thus appears as an objective and impersonal contradiction acquires its historical meaning only through its transformation into human and social conflict. It is the producers' permanent struggle against their reification that transforms what otherwise would remain a mere opposition between concepts into a crisis rending the entire organization of society. There is no crisis of capitalism resulting from the operation of "objective laws" or dialectical contradictions. Such a crisis exists only insofar as people revolt against the established rules. This revolt, inversely, begins as a revolt against the concrete conditions of production; it is therefore at this level that we must seek both the origin of and the model for the general crisis of the system.


The Hour of Work

The contradiction of capitalism appears from the outset in the simple question of how capital and the worker relate to each other: What is an hour of work?

Through the labour contract the worker sells his labour power to the enterprise. But what is this labour power? Does the worker sell his "time"? But what is this "time"? The worker, of course, does not sell his mere presence. During a period when the workers were struggling to reduce the workday from twelve or fourteen hours, Marx asked, what is a workday? This meant: How many hours are there in a workday? But there is an even profounder question: What is an hour of work? In other words, how much work is there in an hour? The labour contract can define the daily duration of work and the hourly wage - and therefore what the capitalist owes to the worker for an hour of work. But how much work does the worker owe to the capitalist for an hour? It is impossible to say. It is upon this sand that capitalist relations of production are built.

In the past, the mode and pace of work were set in an almost immutable fashion by natural conditions and inherited techniques, habit, and custom. Today, natural conditions and techniques are in a constant state of upheaval so as to accelerate production. The worker, however, has lost all interest in working except as it helps him to win his bread. Inevitably, therefore, he resists this attempt to accelerate his work pace. The content of an hour of work, the actual labour the worker has to furnish in an hour, thus becomes the object of a permanent conflict.

Now, in the capitalist universe there exists no rational criterion that would allow this conflict to be resolved. Whether the worker "loafs" or dies of exhaustion over his machine is neither "logical" nor "illogical." The relation of forces between workers and capital alone can determine the pace of work under given conditions. Every real solution therefore represents a compromise, a truce based on the relation of forces existing at that moment. By its very essence such a truce is temporary. The relation of forces changes. Even if it does not change, the technical situation will be modified. The compromise that was arrived at so arduously starting from a given configuration of machinery, a particular type of manufacturing process, etc., collapses; in the new situation the previous set of norms no longer makes any sense. And thus conflict begins anew.

Nevertheless, in order to overcome this conflict as well as to be able to plan production in the enterprise, capitalism is obliged to search for an "objective," "rational" basis for setting production norms. The essential element of this planning process is to be found in the labour time devoted to each operation. Insofar as production has not been completely automated, each unit of time always boils down in the last analysis to "human time," in other words, to the output actually obtained where living labour continues to make itself felt. This truth remains concealed from the production engineers insofar as "depreciation on equipment," for example, can appear (when the factory is not completely integrated) as an autonomous and irreducible element in cost computations. This, however, is only an optical illusion that is due to the fact that under the present structural setup, the engineer is obliged to take the part for the whole. The cost of equipment depreciation is nothing but the labour of workers who manufacture it or repair it (machinists). Hence, one cannot calculate, for example, a machine's "optimal running speed" - which balances the labour cost of the worker utilizing it against the cost of "depreciation on equipment" - unless the actual efficiency of these machinists also is taken into account.

We will return later to this question, which is decisive for the "rationality" of capitalist production. It suffices to point out here two things. First, the inability to consider the entire production process beyond the accidental boundaries of the particular enterprise destroys at its base all pretension to "rationality" on the part of the capitalist organization; one is obliged to consider as irreducible givens what are in reality a part of the problem to be resolved. Second, even on the scale of the individual enterprise, the capitalist management inevitably remains, as will be seen, at least partially in the dark about the real output and efficiency of different types of labour. This state of ignorance therefore also makes it impossible for this type of management to plan production in a rational manner.

Taylorism and all the methods of "scientific management" that flow from it claim either directly or indirectly to furnish such an "objective" basis. Postulating that there is only "one best way" to accomplish each operation, they try to establish this "one best way"[5] and to make it the criterion for how much output the worker should furnish. This "one best way" is to be discovered by breaking down each operation into a series of movements, the duration of which is to be measured, and by choosing, among the various types of movements carried out by various workers, the most "economic" ones. By adding together these "elementary times,"[6] one is supposed to be able to determine the normal amount of time required for the total operation. For each type of operation, we then would be able to tell the actual amount of labour an hour of clock time contains and thereby overcome the conflict over output. Ideally, this method ought to allow us even to eliminate supervisors, insofar as the latter are used to make sure that the workers furnish the maximum amount of labour possible: Paid according to the ratio between their output and the norm, workers would supervise themselves. One part of the conflicts over wages finally could be eliminated, since the actual wage would depend henceforth on the worker himself.

In fact, this method runs aground. Taylorism and the "scientific management" movement have resolved certain problems;[7] they have created many others and on the whole they have not permitted capitalism to get beyond its daily crisis in the area of production. Because of the bankruptcy of "scientific" rationalization, capitalism is constantly obliged to return to the empiricism of coercion pure and simple, and thereby to aggravate the conflict inherent in its mode of production, to heighten its anarchy, and to multiply its wastefulness. First of all there is an insurmountable gap between the postulates of the theoretical conception and the essential characteristics of the real situation in which this conception tries to assert itself. The "one best way" has no relation to the concrete reality of production. Its definition presupposes the existence of ideal conditions, conditions that are extremely far removed from the actual conditions the worker faces, such as problems relating to the quality of equipment and raw materials, the need to establish an uninterrupted flow of supplies, etc. - in short, it presupposes the complete elimination of all the "accidents" that often interrupt the course of production or give rise to unforeseen problems.[8]

But in particular, there are flaws immanent in the theoretical conception itself. From the physiological point of view, work is an expenditure of effort multiplied by a duration of time. This duration is measurable, but the expenditure of effort is not (it involves a muscular component a component of attentiveness, an intellectual component, etc.). "Time studies" can take into account only the duration of time. As for the rest, they confine themselves to "the subjective judgments or interpretations of the engineer responsible for the measurement of the practical calculations and this deprives the results of any scientific value."[9] But work is not only a physiological function; it is a total activity of the person who accomplishes it. The idea that there is "one best way" for each operation ignores the basic fact that each working individual can have and does have his manner of adapting himself to the job and of adapting it to himself. What appears to the scientific manager as an absurd, time-wasting movement has its own logic in the personal psychosomatic makeup of the worker in question which leads him to follow his own "best way" to complete a given operation.

The worker tends to resolve the problems his work poses for him in a manner that corresponds to his overall way of being. His gestures are not like a set of toy blocks where one could pull out one cube and replace it with a "better" one while leaving everything else in place. A gesture that is apparently "more rational" and "more economical" can be much more difficult for some particular worker than the way of doing things that he has invented for himself and that thereby expresses his organic adaptation to this hands-on struggle with machinery and materials that constitutes the work process. Such a gesture is carried out more rapidly because another one is carried out more slowly; merely adding together the minimum amounts of time used by different workers is a glaring absurdity, but applying a standard "normal performance rate" to all the successive phases of an operation carried out by the same worker is an even greater one. The worker's entire set of gestures is not a garment that might be replaced with another. A human being cannot spend two-thirds of his waking life carrying out movements that are foreign to him and that correspond to nothing within him. Tacking "rational" gestures onto the worker in this way is not simply inhuman' in actuality, it is impossible, it never can be fully realized. Indeed, even for the gestures that workers make up themselves, and even for each worker taken individually, there is no "one best way"; experience shows that the same worker alternatively uses several methods of carrying out the same job, if only to relieve the monotony of his work. [10]


Critique of the Theoretical Critique

The idea that labor is only a succession of elementary movements of a measurable duration, that this duration of time is their sole significant feature, makes sense only if we accept the following postulate: The worker in the capitalist factory should be completely transformed into an appendage of the machinery. As with a machine, one determines the movements that are "rational" and those that are not; one retains the former and eliminates the latter. As with a machine, the total time to complete an operation is merely the sum of the "elementary times" of the movements into which one can, mechanically, resolve this operation. Like the machine, the worker does not have and should not have any personal traits; more precisely, as with the machine, the worker's "personal traits" are considered and treated as irrational accidents to be eliminated.[11]

The theoretical critique of Taylorism, in particular as it is conducted by modern industrial sociologists[12] lies essentially in showing that this view is absurd, that man is not a machine, that Taylor was mechanistic, etc. But this is only a half truth. The whole truth is that the reality of modern production, where hundreds of millions of individuals spend their lives in enterprises dispersed all over the world, is precisely this very "absurdity." Taylor, from this point of view, did not invent anything at all; he merely systematized and brought to its logical conclusion what has always been the logic of capitalist organisation that is to say, the capitalist logic of organisation. What is astonishing is not. that mechanistic" and absurd ideas were able to germinate in the heads of the Ideologues and organizational managers of industry. These ideas merely give expression to the peculiar reality of capitalism. The astonishing thing is that the sphere of production, capitalism almost has succeeded in transforming man into an appendage of the machinery, that the reality of modern production is only this very endeavour renewed each day, each instant. This endeavour fails only to the exact extent that in the sphere of production people refuse to be treated as machines. Every critique of the inhuman character of capitalist production that does not take as its point of departure the practical critique of this inhumanity that the workers themselves bring to bear in the sphere of production through their daily struggle against capitalist methods ultimately is merely literary moralizing.


The Workers' Practical Critique

The root of the failure of "scientific management'; methods is the bitter opposition that the workers have shown from the very outset. And of course, the first manifestation of this resistance is the permanent struggle that sets workers against the time-study men. It is on the terrain of this struggle that in every factory the workers immediately realize a spontaneous association. For obvious reasons, the actions that are the expression of this spontaneous association are little known, but their import and universality become clear once we listen to an author who is familiar with what goes on inside a factory.[13]

The first outcome of this resistance obviously is that all semblance of "objective" justification for such "elementary times" is destroyed. The conflict between workers and management is transposed onto the plane of determining these time periods. This process of determination presupposes a certain degree of collaboration on the workers' part. The latter refuse to do so. Management might have been able to dispense with this collaboration if its techniques were unchanging; in that case, little by little it would have been able to set down for good norms representing the maximum amount of output that can be extorted from the worker under a given set of conditions. These techniques, however, are constantly changing; norms have to be reset, and conflict begins anew.

Speaking of an enterprise in which there is a methods department that brings up to date" the times allotted to workers, a right-thinking author writes:

Surveys are constantly being brought up to date to take account of:

a) rapid technical development: improvements in processes and in the machinery manufactured.

b) the large number of operations.

The allotted time is frequently revised and should normally be agreed upon by the workers. Experience shows that this is not the case and that these revisions are the cause of frequent disputes capable of leading to local strike action.[14]

As the norms cannot be put into effect or even established without at least a certain degree of grudging acceptance on the part of the workers, and as the latter do not cooperate, the exploiters' first counter response is to establish them with the collaboration of a minority that they buy off. Here is the ultimate meaning of Stakhanovism: It is to establish monstrously exaggerated norms based on the output of certain individuals who are given a privileged position and who are placed under conditions that bear no relation to the current conditions of the actual production process.[15] A twofold result thus is aimed at: (1) to create within the proletariat a privileged stratum that is a direct support for the exploiters and that is helpful in dissolving working-class solidarity precisely on the terrain of their resistance to increases in output; and (2) to utilize the norms thus established, if not as such, at least in order to shorten the times allotted for the mass of production workers. But Stakhanovism is not the invention of Stalin; its true father is Taylor. In his first "experiment," at Bethlehem Steel Company, after a "scientific" motion study was conducted, Taylor set a norm four times higher than the average output theretofore achieved, and he "proved" three years later with a specially chosen Dutch worker that this norm "could have been realized." Nevertheless, when one tried to extend this system to seventy-five other workers on the gang after having taught them the "rational" method of working, it was discovered that only one worker in eight could keep up with the norm.

Consequently, the problem was posed anew, for norms established based upon the output of a few "rate-busters" or a few Stakhanovites cannot be extended to the rest of the workers. The Russian bureaucracy's ultimate abandonment of Stakhanovism is the glaring admission of the bankruptcy of this method.

In fact, management's real counter response - which at the same time wipes out all of Taylorism's scientific pretensions and closes the discussion from this standpoint - is that it itself sabotages every "rational" employment of scientific management methods and reverts to arbitrarily imposing norms, backed up with coercion. Each year, hundreds and thousands of books and articles appear on the topic of "scientific management," "time studies," etc.; hundreds and thousands of individuals are "trained" to apply these methods. Simplifying the issue but remaining faithful to the essence of the actual situation, we can state that all this is an enormous masquerade that has nothing to do with the setting of norms as it is practiced in a real industrial setting. The objective basis for establishing these norms essentially comes from fraud, spying, and assorted types of pressure.

Workers who think of the time-study men as policemen refer not only to the content but to the methods of their "work" as well. In the Renault factories, the setting of norms often occurs in the following fashion: Unknown to the workers, a new time-study man is sent to walk around the shops and to note while passing by unnoticed the amounts of time required for various operations (one can easily imagine the true value of the "times" noted in this way). With the aid of these "times," the time-study man mixes up a concoction - the new "norm" - which he then will haggle over with the supervisor of the shop in question. The final norm is the outcome of this process of haggling. One or two weeks later, a ritual performance is enacted in the shop: The time-study man comes to time the workers, starts his stopwatch, bustles about, pronounces some cabalistic words, and then disappears. Finally, the result is proclaimed - which had been decided upon in advance.[16]

In another factory, in September, 1954, the Methods Department timed all the operations carried out in the assembly shop; the time-study engineer, questioned by the head of the workshop and a delegate, stated that he was carrying out a revision of the operating methods shown on the chart. ... On December 29, 1954, new times, representing an average reduction of about 20% in allotted time, were notified to the shop delegates. ... The workers concerned stopped work; the arguments put forward by their delegates were as follows:

1. The delegates and the workers had been misinformed about the purpose of the time-study operations. ...[17]

If management's agents are forced to hide like thieves in management's own shops, we can definitely say that all discussion about "rationalizing" efficiency and norms is nothing but mystificatory drivel. In fact, in such a situation, norms express merely management's Diktat - the enforcement of which depends on the workers' capacity for resistance.

Almost nothing is changed in this situation when the trade unions intervene. In theory, the trade unions' line is that they are "opposed to all modifications of the norms and speed of production, unless these modifications are justified by improvements in the equipment or changes in the manufacturing processes." In reality, management constantly modifies its equipment and its manufacturing processes precisely in order to accelerate the work pace. Hence we see that the trade unions end up being opposed to modifications of norms in all cases ... except, it turns out, when it is really important. How indeed can it be judged whether or not some particular equipment change or alteration of the manufacturing process "justifies" a change in the norms? Management constantly relies upon this inability to make a judgment in order to cut down on any "slack in time," and it does so under the pretext of "technical modifications" that are in fact fictional. An American worker put it this way: "They'll tear a machine to pieces to change something to cut a price."[18]

Once the norm is set, one's problems are far from being over. Management is assured of the quantity of the workers' output but not its quality. Except for the simplest of jobs, this is a decisive question. Rushed by norms that are difficult to adhere to, the worker naturally will have a tendency to make up for it on the quality of his work. Quality control over manufactured parts becomes a new source of conflicts.[19] On the other hand, products cannot be manufactured without greater or lesser depreciation on the equipment - and generally, it is easier to increase output by depreciating one's equipment to an abnormally high degree. Management's only response lies in setting up additional supervisory controls - whence there arise additional conflicts.[20]

Indeed, the problem of effective output remains completely open. We will see how workers succeed in emptying a set of norms of its content and even in turning it against management.


The Collective Reality of Production and the Individualized Organization of the Capitalist Enterprise

In an abstract form, the contradiction of capitalism appears at the outset in production's molecular element: the individual worker's work hour. The content of the work hour has directly opposite meanings for capital and for the worker. For the former, its meaning is that of maximum output; for the latter, it is the output corresponding to the amount of effort he thinks is fair.

But in modern production the individual worker is an abstraction. To a degree which was unknown under other historical forms of production, capitalist production is a collective form of production. Not only in society, but in the factory and in each shop, the jobs performed by one person are dependent upon the jobs performed by everyone else. This dependence takes on more and more direct forms as its scope continually widens and as it comes to cover all aspects of the operations of production. No longer is it merely the case that a worker cannot carry out some operation on some components if unfinished components are not provided at the required speed; the worker must also be provided with tools, power, "services" (tool setups, stock management, etc.). Furthermore, every aspect of an operation is directly interdependent with every aspect of all the operations preceding it as well as with those that will follow. Indeed, on a production line and, still more, on an assembly line, individual rhythms and gestures are only the materialization of a total rhythm that pre-exists them, controls them, and gives them a meaning. The true subject of modern production is not the individual; it is, to various degrees, a collectivity of workers.

Now, capitalism simultaneously develops this collective reality of modern production to the extreme and, in its mode of organization, fiercely repudiates it. At the same time that it absorbs individuals into ever-larger enterprises, assigning them jobs whose interdependence increases every day, capitalism claims to be Concerned only with, and wants to be concerned only with, the individual worker. This is not just some contradiction on the level of ideas - although that too exists and manifests itself in a thousand ways. It is a real contradiction. Capitalism is perpetually trying to retransform the producers into a cloud of individual dust particles lacking any organic tie among themselves, yet management clusters this cloud of dust together at convenient spots on the mechanical Moloch, according to the "logic" of this total machine. Capitalist "rationalisation" begins by being, and remains to the end, a meticulous regulation of the relationship between the individual worker and the machine or the segment of the total machine on which he works. This, as we have seen, is in keeping with the very essence of capitalist production. Work is reduced here to a series of meaningless gestures going on at a frantic pace, during the course of which the worker's exploitation and alienation unremittingly tend to increase. For the workers, this work is a kind of forced labour to which they put up both individual and collective resistance. As a counter response to this resistance, capitalism has at its disposal only economic and mechanical forms of coercion. Payment in terms of achieved output is supposed to furnish the worker with motivations capable of making him accept this inhuman situation. But this payment has meaning only in relation to the individual worker, whose gestures have been taken apart and timed, whose work has been defined, measured, monitored, etc.

Thus, this method comes into violent conflict with the reality of collectivised and socialized production. Dissolving the organic ties between the individual and his group and transforming the producers into an anonymous mass of proletarians, capitalism is destroying the social groups that preceded it, the corporation or the village. Grouped into enterprises, these proletarians cannot live and coexist without resocializing themselves, at a different level; they are resocialized under the new conditions created by the situation in which they are placed within the capitalist world and which, by becoming resocialized, they transform. In the factory, capitalism is constantly trying to reduce them to mechanical and economic molecules, to isolate them, to make them gravitate around the total machine under the hypothesis that they obey only the dictates of economic motivation, this Newtonian law of the capitalist universe. And each time, these attempts are shattered when confronted with the perpetually renewed process through which individuals are socialized in the world of production - a process upon which capitalism itself is constantly obliged to rely.

The spontaneous constitution of elementary collective units within the framework imposed by capitalism is the first aspect this process of socializing the workers takes on. These elementary groups[21] constitute a firm's basic social units. Capitalism clusters individuals together within a team or a shop, pretending to keep them isolated from each other and linking them solely through the intermediary of production processes. In fact, as soon as workers are brought together to do a job, social relations are established among them, a collective attitude toward the job, supervisors, management, and other workers develops. The first facet of this socialization process on the level of the elementary group is to be found in the fact that the workers who make it up spontaneously lend to organize themselves, to cooperate with each other, and to deal with the problems raised by the work they have in common and with their relationships to the rest of the factory and to management. Just as an individual, when confronted with a job, organizes himself - half-consciously, half-unconsciously - in order to carry it out, so, on a different level, a number of workers, when confronted with a job, will tend to organize themselves - half-consciously, half-unconsciously - in order to carry it through, to give some order to the relations among the individual jobs of its members, and to make it into a whole corresponding to the goal in question. It is to this type of organization that elementary groups correspond. Elementary groups of workers include a varying, but generally small, number of persons. These groups are based on the direct and permanent contacts established among their members and on the interdependent character of the jobs these people perform. Workers in a workshop may form one or many elementary groups, depending upon the size of the shop, the nature and degree of unity of the jobs they carry out, but also as a function of other factors related to personal, ideological, and other kinds of attraction and repulsion. Often, but not necessarily, elementary groups coincide with the "crews" designated in the official organization of the shop.[22] They are the living nuclei of productive activity - as elementary groups of another type are the living nuclei of all social activities at different levels. Within them we find already manifested the workers' self-managerial attitude, their tendency to organize themselves in order to resolve the problems raised by their work and by their relations with the rest of society.


Elementary Groups and Industrial Sociology

Bourgeois academic sociology has brought to light the fact that in reality modern production relies for the most part on this spontaneous association of workers into elementary groups, or more exactly on the self-transformation of fortuitous assemblages of individuals into organic collectivities.[23] Undoubtedly, modern industrial sociology has made a decisive contribution to the recognition of the fundamental importance of this phenomenon, and concurrently, to the critique of the capitalist organization of human relations in production, starting out from this point of view. This contribution is totally undermined, however, by the general outlook of its authors just as the critique of the capitalist enterprise that follows therefrom only results in a Utopian and impotent reformism.

The perspective through which industrial sociologists most of the time view elementary groups is "psychologistic." Like all human beings, workers tend to become socialized, to enter into reciprocal relationships, to form "bands." Their motivation to work is constituted starting from their belonging to a "band" and not starting from economic considerations. The "work ethic" depends on this feeling of belonging, on the ties that unite the individual and his group. The fundamental flaw of the capitalist organization of production is that it ignores these phenomena. From its own point of view management is wrong to arbitrarily transfer workers, to assign a new trainee to a given crew without worrying about the relationships that might arise between him and others, and more generally, to be unaware of the reality belonging to the elementary group. This regrettable lack of awareness is to be attributed to the erroneous theoretical conceptions (those that Mayo[24]) encapsulates under the name of the "rabble hypothesis" and that we prefer to designate henceforth in this text by the term "molecular hypothesis"[25] that have predominated for some time now. The critique of this conception ought to lead production managers to change their attitude toward the problem of human relations in the enterprise, thus allowing actual conflicts and wastefulness to be eliminated.

The paternalistic and idealistic character of these solutions, their thoroughly Utopian content, and their laborious naiveté are obvious. Management's theoretical conceptions do not determine the relations between management and the workers in the capitalist firm. These conceptions merely give abstract expression to the inescapable necessities management faces qua external management and qua exploitative management. The molecular hypothesis is a necessary product of capitalism and will disappear only when it does. From the practical point of view, when faced with the anarchy that characterizes both the capitalist enterprise and its relations with the market (or with the "plan"), management has other, more pressing matters than to be bothered with its employees' personal feelings toward each other. At the very most, a new bureaucratic department responsible for "human relations" may be created within the managerial apparatus. If it takes its role seriously, this department will be in permanent conflict with the exigencies of the "production" managers, and it will be reduced thereby to a decorative role; otherwise, it will put its "sociological" and "psychoanalytical" techniques at the disposal of the factory's system of coercion.[26]

But the main point lies elsewhere. The workers' spontaneous association in elementary groups does not express the tendency of individuals to form groups in general. It is simultaneously a regrouping for the purposes of production and a regrouping for the purposes of struggle. It is because they have to resolve among themselves the problems involved in organizing their work (whose various aspects are mutually interrelated) that workers necessarily form elementary collectivities not mentioned on the organizational chart of any enterprise. It is because their situation in production creates among them a community of interests, attitudes, and objectives irremediably opposed to those of management that, at the most elementary level, workers spontaneously associate together to resist, to defend themselves, and to struggle.

To invite management to recognize these elementary groups means to invite it to commit suicide.[27] For these groups are constituted from the start against management, not only because they struggle to make their interests prevail in irremediable opposition to its interests, but also because the very foundation of their existence, their primary objective, is the management [gestion] of their own activity. The group tends to organize the activity of its members, to define the norms relating to how much they should exert themselves and how they should behave. All this signifies a radical challenge to the very existence of a separate management [direction]. The inability of "elementary group" sociologists to recognize clearly the consequences of this state of affairs constitutes the main stumbling block for this type of sociology.[28]


The Informal Organization of the Enterprise

This challenge indeed goes far beyond the bounds of the elementary group. On the one hand, these groups tend to put themselves in contact with each other; on the other hand and more generally, contacts and relationships are established between individuals and groups throughout the enterprise, alongside and in opposition to the official organization. Along with modern industrial sociology, we are learning that the enterprise has a double structure and leads, so to speak, a double life. There is, on the one hand, its formal organization, the one represented on organizational charts, the one whose ruling summits proceed along the lines of these charts in order to allocate and define the work of each person, to keep informed, to send orders, or to assign responsibilities. To this formal organization there is opposed in reality the informal organization, whose activities are carried out and supported by individuals and groups at all levels of the hierarchical pyramid according to the requirements of their work, the imperatives of productive efficiency, and the necessities of their struggle against exploitation.[29] Correlatively, there is what indeed might be called the formal production process and the real production process. The first includes what ought to happen in the enterprise according to the plans, diagrams, regulations, methods for transmitting information, etc., established by management. The second is the one that actually is enacted. It often bears little relation to the first.

The failure of the individualist type of capitalist organization therefore goes far beyond the elementary group. Cooperation tends to be carried out alongside and in opposition to this type of organization. But what is most important is that this opposition is not the opposition of "theory" and "practice," of "beautiful schemes on paper" and "reality." It has a social content, a content having to do with struggle. The formal organization of the factory coincides as a matter of fact with the bureaucratic managerial apparatus's system of organization. Its nodal points, its articulations are those of this apparatus. For in the official diagram of the enterprise, the whole enterprise is "contained" in its managerial apparatus; people exist only as provinces of power for those in charge. Beginning with the summit of what is properly called "management" (president-CEO in the firms of Western countries, the factory director in the Russian factory) and passing through the various offices, departments, and technical services of the enterprise, the bureaucratic managerial apparatus terminates with the shop foremen, supervisors, and team leaders. Formally, it even completely encompasses the executants - who in the official diagram are only clusters around each foreman or team leader.

The managerial apparatus pretends to be the only organization in the enterprise, the sole source of all order and of any kind of order. In fact, it creates as much disorder as order and more conflicts than it is capable of resolving. Facing it is the enterprise's informal organization, which includes the elementary groups of workers, various modes of lateral connections [liaison transversale] among these groups, similar associations among individuals in the managerial apparatus, and lots of isolated individuals at various levels who in extreme cases only have among themselves the relationships that the official diagram assumes they have. These two organizations, however, are truncated. The formal organization is riddled with holes by the base, it never succeeds in actually encompassing the immense mass of executants. The informal organization is thwarted by the heights; beyond the elementary groups of executants, it actually includes the individuals formally belonging to the managerial apparatus only when this apparatus starts to grow to enormous proportions, when the division of labour is pushed even further and is accompanied by further collectivisation, and, finally, when the work of the lower echelons of the managerial apparatus is transformed into merely another form of executant work, thus creating even within this apparatus a category of executants that struggles against the summits.[30]

The formal organization, therefore, is not a facade; in its reality it coincides with the managerial stratum. The informal organization is not an excrescence appearing in the interstices of the formal organization; it tends to represent a different mode of operation of the enterprise, centred around the real situation of the executants. The direction, the dynamic, and the outlook of the two organizations are entirely opposite - and opposed on a social terrain that ultimately coincides with that of the struggle between directors and executants.

For a struggle takes place between these two modes of organization, which is in all respects permanent and which ends up becoming identical with the enterprise's two social poles. This is what industrial sociologists, who usually just criticize the formal schema as absurd, too often forget. This situation is analogous to the one we discussed apropos of Taylorism, and the shortcomings of a purely theoretical critique are the same here. The managerial apparatus is constantly struggling to impose its scheme of organization; the absurdity of this schema is not theoretical, it is the reality of capitalism. What is astonishing is not the theoretical absurdity of the schema but the fact that capitalism almost succeeds in transforming people into points on an organizational chart. It fails only to the exact extent that people struggle against this transformation.

This struggle begins at the level of the elementary group, but it extends throughout the entire enterprise through the very need to produce and to defend against management; ultimately, it embraces the entire mass of executants. Its extension is founded on several successive moments. The position of each elementary group is essentially identical to that of the others; each of these groups inevitably is led to cooperate with the rest of the enterprise;[31] and ultimately they all tend to merge in a class, the class of executants, defined by a community of situation, function, interests, attitude, mentality. Now, industrial sociology denies deep down this class perspective that verbally it accepts. It speaks of elementary groups as a universal phenomenon; but while it is willing to compare them with each other, it refuses to add them together. Nevertheless, it does more than just add them together since it recognizes in them the subject matter and a the same time the principle of the enterprise's informal organization. But it keeps these two moments - the identity of elementary groups throughout the enterprise and their cooperation - separate and does not venture to ask itself why there is a passage from one to the other. It therefore renders itself incapable of seeing the polarization of the enterprise between directors and executants and the struggle that sets them against each other, all the more so as it includes under the rubric of informal organization phenomena whose significations are radically different, such as when it compares the tendency of the executants to form their own type of organization to the formation of cliques and clans within the ruling bureaucracy. This actual refusal to place the firm's problems within a class perspective (and the process of class formation can be seen most vividly through an analysis of the enterprise) makes it sink into theoretical abstraction as well as get lost in "practical solutions," the Utopian character of which is based precisely on the imaginary suppression of the reality of classes.

We must add that Marxism admits of an abstraction that is almost symmetrical to the preceding one insofar as it has limited itself to immediately positing the concept of class and to directly opposing the proletariat and capitalism while neglecting the basic articulations within the enterprise and among the human groups within the enterprise. It thus has prevented itself from seeing the proletariat's vital process of class formation, of self-creation as the outcome of a permanent struggle that begins within production. It also has prevented itself from relating the proletariat's organizational problems in capitalist society to this process. And finally, insofar as the primary content of this struggle is the workers' tendency to manage their own work, it has prevented itself from posing workers' management as the central feature of the socialist program and from drawing from it all the possible implications. To the abstract concept of the proletariat corresponds the abstract concept of socialism as nationalization and planning, whose sole concrete content ultimately is revealed to be the totalitarian dictatorship of the representatives of this abstraction - of the bureaucratic party.


The Contradictions Proper to Management's Bureaucratic Apparatus

To achieve its own ends, the capitalist organization of production is obliged to pursue the fragmentation of production tasks and the atomisation of the producers ad infinitum. With respect to the end in view - the total subjugation of people - this process culminates in a double failure and leads to tremendous waste. At the same time, however, it gives rise very sharply to a second problem: that of how to recompose these operations of production into a whole. Individual jobs, supposedly defined, measured, monitored, etc., have to be integrated anew into a unified whole [ensemble], outside of which they are meaningless. Now, this reintegration can be accomplished in the capitalist factory only by the same authority following the same method of decomposition that "preceded" it, by a managerial apparatus separated from the producers that aims at subjecting them to capital's requirements and that treats them to this end as things, as fragments of the mechanical universe that are comparable to all others. Logically and technically, reintegration is only the flip side of decomposition; neither one can be carried out or have any meaning without the other. Economically and socially, the realization of the goals pursued during the phase of decomposition is impossible if these goals do not also predominate over the process of reintegration: The ground taken from the producers during the phase of decomposition could not be given back to them during the phase of reintegration without putting back into question the very structure of the relations of exploitation.[32]

As a consequence, the managerial apparatus will try to resolve the problem of reintegrating jobs itself, thereby denying deep down the collective character of production that it is obliged to grant on a formal level. For the managerial apparatus, the collectivity of workers is not a collectivity but a collection. Their labour is not a social process whose every part is in a constantly changing interdependence with all the others and with the whole, and whose every moment perpetually contains the seeds of something new; it is a sum of parts that someone from the outside can decompose and recompose at will, like a game of blocks, and that can change only insofar as something else is introduced into it. For it is only upon this condition that the command post of this collective activity could be transposed outside this activity with no repercussions. It is only upon this condition that exactly what one has put into its parts could be rediscovered in the whole, without losses or gains.

The managerial apparatus thus is obliged to take everything upon itself. In theory, all acts of production have to be doubled ideally and a priori within the bureaucratic apparatus; everything that involves a decision has to be worked out in advance - or after the fact - outside the operations of production themselves. Execution has to become pure execution, and symmetrically, management has to become absolute and perfect. Of course, such a situation never can be realized; but the "organizational" activity of the managerial apparatus is dominated by the necessary pursuit of this chimera, which puts it up against insoluble contradictions.

First of all, the very concept of a perfect, separate management is contradictory. A perfect, separate management is possible only if its complementary pole, a perfect, separate execution, also is possible. Now, perfect, separate execution is nonsensical. As human activity - as activity that cannot be conferred upon automated machinery - execution necessarily involves the element of self-direction; it is not and never can be execution pure and simple. Man is not and cannot be a perfect, separate executant, and this singular attempt to make him one creates in him both a situation and reactions that produce the opposite effect. This contrary situation is established because the suppression of the faculties of and capacities for self-direction (which are indispensable for tasks of "execution") are precisely what make him a bad executant. And these contrary reactions are created because man always tends in one fashion or another to take on the direction of his own activity and he revolts against this expropriation of his self-directing activity to which he is subjected. During the historical stages that preceded capitalism, this contradiction remained abstract and merely potential, basically because the form and content of productive activities were fixed once and for all. But capitalist production, which is in constant upheaval is continually obliged to call upon the human faculties of its executants in order to function. In this way the contradiction becomes an active and actual one, since the way the system functions leads it to affirm two things at once: "The worker should confine himself to the pure and simple execution of the tasks prescribed to him"; and, "The worker should bring about the end in view whatever the real conditions and available means and no matter how far these depart from theoretical conditions and means."

This gap cannot be bridged. Perfect, separate management can be conceived of only as the organ promulgating the perfect plan, which obviously cannot exist. Such a perfect plan would imply that management has absolute foresight and exhaustive information, both of which are impossible in themselves, two times impossible for a separate management, and three times impossible for a management that exploits the producers. Of course, modern industry tends to "rationalize" the set of conditions, means, and objects of production, and this rationalization is presented as the elimination of chance, of the unforeseen, and as the creation of standardized conditions for the production process as a whole. Under such conditions, it ought to be possible, after a period of trial and error and through successive approximations, to reach a "point of rest," after which production finally could unfold according to plan. But this would imply that from this moment on the conditions, methods, instruments, and objects of production were unalterably fixed. Now, the very essence of modern industry is perpetual change. From a large-scale point of view, one stage of technical development hardly has arrived at a level of "consolidation" when a new stage comes crashing onto the scene. From a small-scale point of view - which is just as important in the everyday life of the factory - "consolidation" is never achieved; "small" changes continually are being made in the materials, the machinery, the objects manufactured, and the ways people and machines are arranged (and these changes are precisely the expression of this process of "rationalization"). Thus, the plan has to be perpetually modified, and there never is time perfectly to adapt it to the unfolding of the production process.

Indeed, "standardization" remains an ideal norm that is never realized, for both social and "natural" reasons. Everything used at any given stage of the production process already is the result of previous industrial labour. In theory, this result, this product - whether we are talking about raw materials or a machine or a detachable part - is supposed to conform to a rigorous definition, to precise specifications of size, shape, quality, and so on within set margins of tolerance. It suffices that any one of these material or ideal components not correspond in reality to its theoretical definition for the plan not to be able to be put into effect as is; this does not mean, of course, that production collapses or even that there is necessarily any significant damage - but it implies that only the vital intervention of real people can serve as a substitute for some now out-of-date directive and can adapt on the spot the available means - which are different from the theoretical ones - to the end in view.

That all the components of any job are the result of a previous job signifies that as soon as the actual results of this job deviate at a given stage from the "theoretical" results, this gap has repercussions in one fashion or another upon the subsequent stages of the manufacturing process. Now, gaps of this kind are absolutely unavoidable in capitalist production, not only because the exploited executant is not interested in the result of his work and therefore often turns in "made up" results (which go along with a whole gamut of means for struggling [against the factory's "inspectors"), but also because the compartmentalized executant does not know and by definition should not know what is important and what is not important in what he is doing. All specifications that are set for him by the production directives he receives seem to be of equal importance (with allowed margins of tolerance). In fact they are not, either in the absolute or from the point of view of possibly making up for some gap without difficulties arising at a subsequent stage in the production process. Inasmuch as the executant, pressed by time restrictions, cannot handle everything at once, he will take shortcuts at random. For its part, the planning department cannot establish which aspects are truly important and which ones are not: On the one hand, it does not itself know which ones are important, for the establishment of such a hierarchy results from actual practice within an industrial setting from which it is, by definition, separated; on the other hand, its role is to present all directives as equally and absolutely important. Thus, by rendering an intelligent execution of tasks impossible, the methods of a separate managerial stratum themselves lead toward their own defeat.[33]

Similarly, there is always an unforeseen "natural" element, even under the conditions of large-scale modern industry. Even materials manufactured under the best possible conditions present specific, unanticipated problems that must be compensated for in an equally unforeseen manner as they are worked upon. Even electronic computers, which are manufactured not under industrial conditions but under laboratory conditions, break down or go haywire for unknown reasons.[34] At each new stage, modern industry stretches to the limit its exploitation of the possibilities of knowledge and of matter; during each new period, it tends to work at the edge of the known and the feasible. This continuous displacement of its frontiers signifies that it can never comfortably remain within the regions it has already fully explored. A new territory has hardly been opened up when it must already be exploited under the conditions of mass production. Its means expand at a dizzying rate - but so do its objectives and manufacturing requirements. Instruments become finer and finer and more and more precise - but at the same time the limits of tolerance become narrower and narrower. In the past, the "unforeseen," the "irrational," and the "accidental" consisted of a cleft in the steel bar; today it can lie in infinitesimal irregularities in the chemical composition of molecules. It is not the degree of matter's resistance to man that is diminishing, it is the line on which this resistance becomes effective that is being displaced - so that the gap between theory and reality can always be filled in only by practice, only by man's simultaneously rational and concrete intervention. But this practice itself is constantly being elevated to a higher level, and it presupposes that the individual's ever more highly developed capabilities - which are absolutely incompatible with the role of a pure and simple executant - will be put to work. These are the reasons why the reality of production always deviates in a more or less appreciable manner from the plan and from production directives - and why this gap can be filled only by means of the practice, the invention, the creativity of the mass of executants. Each time that a new manufacturing process is introduced or a new product model is to be manufactured, and after the factory's various departments and engineers have spent months or years developing and "perfecting" the process or product in question, weeks or months will pass before production begins to flow in a somewhat satisfactory manner. Car drivers know that when a factory "launches" a new model, the cars produced during the first few months generally have serious defects.[35] And yet, their "prototype" had been tested for years, they had driven it in the Sahara and in Greenland, etc. But the time that has elapsed between the debut of the new manufactured product and the rolling-out of nearly satisfactory copies is the time needed for the mass of the factory's executants as a whole to give concrete form to initial manufacturing directives under real work conditions, to fill in the holes in the production plan, to resolve unforeseen problems, to adapt the manufacturing process to their own needs in their defence against exploitation (for example, to "make do" with the blueprint "specs" they are given), etc. Equilibrium between the production plan, the real state of the factory from the viewpoint of what is possible within the manufacturing process and the workers' struggle against exploitation thus is attained - until a new modification is introduced.

Management, of course, is "conscious" in general of these gaps between the production plan and what really goes on in the factory, and in principle it is supposed to fill them in itself. In practice, this obviously is not achievable: If each time something went wrong it was necessary to stop everything and ask for instructions back up the hierarchical chain of command, the factory would accomplish only a small portion of its production goals. Let it be said in passing that just because management is forced to tolerate the indispensable initiatives of the executants does not make the latter's role any easier. The managerial apparatus is both jealous of its prerogatives and completely fearful of its responsibilities; as much as it can, it will avoid tackling a question unless it is "covered," but it will harshly reproach its subordinates for having done so themselves. If the initiative succeeds, it will merely grumble, and then will try above all to grab the credit itself; if the initiative fails, it will deal with them severely.[36] For the executant, the ideal attitude is for him to take initiatives that are really effective while making it seem like he is following all the official directives - though this is not always easy. The factory thus comes to constitute in places a double world - where people make it seem like they are doing one thing while doing another.

Both the foresight required for planning and the need for ongoing readjustment of the plan to a constantly evolving reality pose the problem of how to obtain information about what is going on in production. This problem quickly becomes insoluble for a bureaucratic managerial apparatus. The ultimate source of all information is the executants who are constantly engaged in the battle for production. Now, these people do not collaborate in the process; not only do they not necessarily inform management about the situation, but very often they are led into a tacit conspiracy to hide the real situation from management. The managerial apparatus can react to this only by creating special organs for obtaining information - which quickly run up against the same difficulty, since they too live to obtain original information from the outside. The conspiracy surrounding, the obtainment of information indeed is not limited to executants. The managerial apparatus itself participates in it. In fact, this is an essential aspect of the activity of its members. They make up the results of their own activity or the activity of the sector for which they are responsible. Their fate, the fate of their clan or their department depends upon it.[37]

Obtaining information, however, is not simply the gathering of "facts." It already is their choice, but it is also and much more their elaboration, the disentangling of the relationships and perspectives that tie facts together. This is impossible outside a conceptual framework, therefore outside a set of organized ideas, therefore outside a theory (even if it remains unconscious). Consequently, all information the managerial apparatus may have at its disposal is undermined by its theory of society - or of industrial reality. This is plainly apparent when we consider the bureaucratic apparatus that runs the entire society - the State or bureaucratic party. To run society presupposes that one knows it, and to know society signifies that one has an adequate theoretical conception of it. But today's leaders can try to grasp social reality only by subordinating it to absurd schemata. The same is true of their ideologists. Sometimes these ideologists plan out the operations of society, using the functioning of a mechanism as their model; at other times, when disheartened by the failure of this absurd attempt at comparison, they take refuge in irrationalism, the accidental and the arbitrary. We will encounter these problems again later.

The ruling apparatus of the enterprise is faced with the same questions and the same impossible options. The reality it needs to know is the reality of production. The latter is first and last a human reality. The most important facts are those that concern the situation, the activity, and the fate of people in the production process. Obviously, it is impossible to know these facts from the outside. Moreover, management does not bother itself very much about them. To the extent that it is obliged to worry about them, however, it can do so only by considering them as external facts, by transforming them into mechanical entities capable of being observed - in short, by destroying their very nature. In management's eyes, consequently, the worker either does not exist at all or else he exists only as a system of nerves and muscles capable of carrying out a certain quantity of gestures - gestures that can be increased in proportion to the amount of money he is promised. This entirely imaginary view of the worker is the basis for the "knowledge" of the reality of production that management possesses. In the manager's very gaze is incorporated, through a process of construction, the negation of the inherent [propre] reality of the object he claims to be looking at, for recognition of this inherent reality would imply, conversely, that the manager denies himself qua manager.

This situation hardly is modified at all when the crude old methods and the schema of "molecules irresistibly attracted by money" are abandoned in favour of more modern conceptions and the discoveries of industrial sociology. Only the nature of the "laws" supposed to rule people and their relations changes; the basic attitude remains the same. It no longer is assumed that the worker is capable of murdering his buddy and killing himself at his job for a few extra pennies - it now is assumed, quite to the contrary, that he is essentially determined by a "group solidarity." But in both cases, it is merely a matter of management's knowledge about the workers, and this knowledge is supposed to allow management to utilize them better for purposes of production. Group solidarity in its turn has become the new external motive determining the worker's acts; knowing the motive and acting upon it, one can bring the worker to do what is wanted of him. Management's situation still remains that of the engineer charged with laying out and ordering the assembly and operation of the parts of the human mechanism that make up the enterprise and of which he knows the laws. That the author of these laws is no longer Bentham, but Freud or Elton Mayo, changes nothing. And we need hardly add that it is still impossible to know industrial reality. Mired in this perspective and utilized toward these ends, psychology, psychoanalysis, and sociology are emptied of their content and transformed into their opposite.[38] That the group, for example, is not for its members an external motive, that it is the unity of self-determination creating and recreating itself, that thereby it sooner or later can only set itself against every kind of external management that tries to impose itself on this group - these truths can be of no service to management, for they challenge its very foundations. Management can possess only the theory of its own practice, i.e., of its social existence.

But contradictions that are just as insoluble tear apart the managerial apparatus, independently, so to speak, of its permanent struggle against the executants. A series of factors, all of which derive in the last analysis from the tendency to confine labourers to more and more limited tasks of execution, leads to an extraordinary proliferation within the managerial apparatus itself. Taking on itself a constantly increasing number of tasks, the managerial apparatus can exist only as an enormous collective organ. In a large enterprise, the individuals employed in offices and departments already constitute in themselves a sizable enterprise.[39] This collective organ itself undergoes a twofold division of labour within its own ranks. On the one hand, the managerial apparatus is subdivided into "specialized branches" - the various "services" in the enterprise's offices. On the other hand, within this apparatus as a whole and within each of these "services," the division between directors and executants inevitably is instaurated anew. By this very fact, all the aforementioned conflicts reappear within the managerial apparatus.

The organization of work within the managerial apparatus obviously can occur only under the same forms of "rationalization" as were applied to production proper: subdivision and compartmentalization of tasks, transformation of individuals into a mass of anonymous and interchangeable executants, etc. It engenders the same consequences in both places. In order to tame the workers' struggle, management thus ends up introducing the class struggle into its own ranks. Condemned to a compartmentalized job, deprived of all meaningful skills, reduced to salaries comparable to those paid to workers, deprived (in statistical terms) of any real chance of advancement, the vast majority of employees in the managerial apparatus now have trouble distinguishing themselves from their fellow workers on the shop floor; at bottom, only illusions that are being increasingly undermined by their real situation are capable of keeping them separate from the workers.[40] Independent of this process that unifies the various strata of executants in the enterprise, the principal result of the appearance of this mass of executants within the managerial apparatus is that management no longer has even itself at its own disposal; even if they are not in solidarity with the workers, vis-à-vis their work the lower strata of nonproduction employees have the same attitude as production workers.

On the other hand, the unavoidable fragmentation of the managerial apparatus into a series of specialized services inevitably creates a problem of reuniting the activities, methods, and viewpoints of these services. Each of them tends to champion its own viewpoint at the expense of the others, for this is the sole means by which it can assert its importance and enlarge its position within the apparatus. Now, the summit of the managerial apparatus, which is charged with resolving these conflicts, does not in general have any rational criterion for doing so. To do this, indeed, it would have to be able to take on itself all opposing points of view; i.e., it would have to in fact "duplicate" all the costly services that have been set up so laboriously. This is in fact the solution to which a number of managers are led: They surround themselves with an exclusive personal team, a sort of private and clandestine general staff.[41] Management thus is obliged to instaurate its own informal organization in opposition to the formal one it has already set up. However, it is obvious not only that these two solutions refute each other (either the clandestine general staff is useless or else it proves how useless a good part of the official departments are) but also that their juxtaposition can only be the source of new conflicts. And ultimately, top management does not run anything at all; it is reduced to arbitrating between opposing viewpoints and it does this in a truly arbitrary fashion, for it knows hardly anything about the problems in question. Logically speaking, its sole foundation now is merely that whatever decision it makes, even an arbitrary or absurd one, is more valuable than the total absence of decision making.[42]

The absence of rational criteria capable of aiding in the resolution of conflicts between opposing points of view that arise unavoidably within management's bureaucratic apparatus is combined with another phenomenon of capital importance: the absence of rational criteria concerning the placement of individuals within this apparatus. These two factors are at the root of the traits that are characteristic of every modern bureaucratic apparatus: the struggle of all against all for "advancement," the formation of cliques and clans that dominate in a hidden [occulte] fashion the "official" life of the apparatus, and the transformation of objective options into stakes in the struggle between cliques and clans.

We must fully understand the meaning of this analysis of the contradictions of bureaucratic management. We are not comparing the latter to a perfect management in order to draw out the failings it exhibits in relation to such an imaginary standard. There is no perfect management, whatever the social structure (even if it be the organized collectivity of producers), and such a comparison would be completely meaningless. From every standpoint we have examined, a human management would encounter problems as well as difficulties as to how to solve such problems. The preceding discussion has no bearing on the possibility of eliminating these problems. It shows rather that the structure and the nature of the present form of management, which is a bureaucratic form of management external to the activities it is supposed to direct, make its problems insoluble, or at best, prevent its problems from being "resolved" except at the price of enormous wastefulness and perpetual crises.

Perfect foresight will never exist. And it need not exist for production to be organized rationally. The present structure, however, is implicitly based on the hypothesis that such foresight exists, and that management possesses it. Since in theory the producers are incapable of carrying out "on the job" the permanent readjustment of the plan to reality, this adjustment must be carried out a priori and once and for all by management. By virtue of this, the "production plan" - of the enterprise or of the entire economy - acquires an absolute value. Since the permanent process of making adjustments between foresight - without which there is no rational action - and reality is upset by the fact that managers are radically separated from executants, balance can be re-established in each instance only by fits and starts, and through specific, belated, spasmodic interventions.

The problem of obtaining adequate information will always exist. But the present structure renders the problem literally insoluble, for its very existence drives the whole of society to conspire to mask reality. The problem of making individuals adequate for the functions they fulfil will exist for a long time to come. But, by arranging these functions along a hierarchical pyramid, by tying not only the economic fate of the individual but also his total situation and ultimately his sense of self-worth to his success in a desperate and absurd struggle against everyone else, the present structure destroys all possibility of a rational solution. Human society will always be faced with options that are not geometrical problems admitting of a single, unique solution at the end of one rigorously defined path. But the present structure either fails to pose these problems explicitly or resolves them in terms of factors that are external to their content.

Now, unless there is a radical overthrow of the present structure, this separate type of management is inevitable. The activities of thousands of individuals and elementary groups have to be coordinated in one fashion or another. The "universal" point of view of the enterprise's operation has to prevail over the "particular" viewpoints of the workers or of their groups. Ultimately, then, a particular group of managers has to take it upon themselves to impose this "universal" viewpoint upon the totality of producers. From then on, conflict is inevitable.

First of all, for each group of workers, the imperatives arising out of this "universal" standpoint of the management take the form of an arbitrarily imposed external law. Its justification cannot even be known, and by this very fact it therefore appears to be completely irrational. But management's "universal" point of view is in fact another particular point of view; this viewpoint, which is partial in both senses of the word [partial et partiel], is the viewpoint of a particular stratum that has access to only a part of reality, that lives a life apart from actual production, and that has its own interests to put forward. Inversely, the "particular" point of view of groups of producers is in fact a universal point of view. The point of view of each elementary group is found again in all the others. The norms arising within them are identical. The interests they try to advance are the same. Management endeavours to think about the actual reality of production. The producers are this actual reality itself. Taken in their totality, they embrace the totality of aspects of the activity of the enterprise - in fact, they are this totality.

But are they really? Can they, across the many shops and offices of the enterprise, actually form an organic unity? Are they not all riveted to specific places on the total machine of the workplace? Is not each of them deprived of a view of everything else and incapable of connecting with the overall living totality of the enterprise? An analysis can show their mutual identity, and it can combine them. But can they themselves become united? Only the analysis of working-class struggles can furnish an answer to these questions.


The Working-Class Struggle against Alienation

The capitalist organization of production is profoundly contradictory. Capitalist management claims it deals only with the individual worker, whereas in fact production is carried out by the collectivity of workers. It claims to reduce the worker to a limited and determined set of tasks, but it is obliged at the same time to rely upon the universal capacities he develops both as a function of and in opposition to the situation in which he is placed. By exhaustively defining in advance the methods by which these tasks are to be executed, it claims to remove from them every element involving managerial duties. But as such, an exhaustive definition always is impossible. Production can be carried out only insofar as the worker himself organizes his work and goes beyond his theoretical role of pure and simple executant.

The conflicts that result from this situation culminate in a veritable anarchy of production in each enterprise. But they create at the same time a contradictory situation and a contradictory attitude in the workers themselves. The conditions in which they are placed impel them to organize their production work in the most effective manner, to upgrade the machinery, to invent new processes, etc. The way capitalism organizes production obliges them to do so, for when something goes wrong it is the workers who pay (and who cannot defend themselves merely by pointing out that the factory is badly organized). On the other hand, however, as soon as they manifest themselves, the workers' organization and creativity are combated by the managerial apparatus. In any case, these qualities are continually being disrupted and butchered by this apparatus. Indeed, under present conditions, improvements in the organization and methods of production initiated by workers essentially profit capital, which often then seizes hold of them and turns them against the workers. The workers know it and consequently they restrict their participation in production, both consciously as well as unconsciously. They restrict their output; they keep their ideas to themselves; they make use of improvements on their individual machines that they carefully hide from the foremen; they organize among themselves to carry out their work, all the while keeping up a facade of respect for the official way they are supposed to organize their work - and so on.[43]

This contradictory attitude on the part of the workers signifies that the insurmountable conflict that tears through capitalist society is transposed into the heart of the proletariat itself, into the behaviour of the individual worker as well as into the attitudes of the working class. It would be entirely wrong to represent the proletariat as a full positivity, like some kind of class that already bears within itself the solution to all problems and that an enemy class and a form of social organization that remains foreign alone prevent it from achieving such solutions. That would be both a demagogic mystification and a poor, superficial theory. Capitalism would not be able to continue to exist if the crisis it is undergoing did not have repercussions within the proletariat itself. The oppression, the exploitation, and the alienation created by capitalism express themselves in the working class through contradictions that till now it has not succeeded in overcoming. The positivity of the working class comes from the fact that it does not remain simply torn by these contradictions, but constantly struggles to overcome them and that, at the most diverse levels, the meaning [contenu] of this struggle is the autonomous organization of the working class, workers' management of production, and, ultimately, the reorganization of society.

Bureaucrats - and sometimes even revolutionary militants deformed by a narrow "Marxism" they have outgrown but have not been able to shed - do not want to see in the proletariat's struggles anything but a tendency toward improving its standard of living, or at best a struggle "against exploitation." But the proletariat's struggle is not and cannot be simply a struggle "against" exploitation; it necessarily tends to be a struggle for a new organization of the relations of production. These are only two aspects of the same thing, for the root of exploitation is the present organization of the relations of production. The worker can be exploited, i.e., the fruits of his labour can be expropriated from him, only insofar as the direction of his labour is expropriated from him. And the struggle against exploitation quickly places before him the problem of management. This always is true on the shop floor and periodically on the level of the factory and of society as a whole.

Usually one fixes one's eyes on the "historical" moments of proletarian action (revolutions and general strikes) or, at the very least, on what can be called its explicit organization and activity (trade unions, parties, big strikes). But these actions and organizations can be comprehended only as moments of a permanent process of action and organization that finds its origin in the depths of everyday life in the workplace and that can sustain itself and remain adequate to its intentions only on the condition that it continually returns to these depths. Under the title of implicit struggle we include this everyday activity and organization, the capital importance of which must henceforth be given full recognition. It is implicit in the proletariat's existence, in its very condition as being proletarian. The informal or elementary organization of workers is only one aspect of this struggle. Organization is only one logical moment of the process of struggle - and the same is true of action. Struggle includes action, organization, and the setting of objectives. Our purpose is much more general than the analysis of informal organization since it also includes both informal actions and informal objectives. This implicit struggle is only the flip side, one could say, of the proletariat's everyday work. Work in the capitalist enterprise does not occur without struggle. This situation follows directly from an organization of work based upon the opposition between directors and executants.

Thus, the capitalist organization of work tends to rely upon the definition of work norms. Workers struggle against these norms. In this struggle, only a "defence against exploitation" can be seen. But in fact, it contains infinitely more: Precisely because he is trying to defend himself against exploitation, the worker is obliged to demand the right to determine his own work pace and to refuse to be treated like a thing.

Once a norm is defined, problems are far from being settled. It is only the boundaries of a battlefield that have just been defined. In this battle, the battle over actual output, the workers are led to organize themselves, to invent new means of acting, and to define objectives. Nothing is given to them in advance; everything has to be created and conquered in the midst of struggle.

The dynamic of the sequence of objectives, organization, and means of action, is plain to see. The workers aim for the maximum amount of pay for "an honest day's work." This maximum has meaning only as a collective maximum - in other words, every attempt to reach a maximum amount of pay for an individual quickly is revealed to be illusory and ultimately is turned against the individual who made the attempt. The achievement of this initial objective implies the pursuit of the greatest possible amount of freedom within the given framework of the capitalist enterprise. It equally implies the pursuit of the maximum amount of real efficiency in production - an indispensable condition for achieving labour savings. The workers thereby are led to struggle against the entire set of methods for organizing production along capitalist lines. They are led equally to organize themselves in an "elementary" or "informal" fashion under forms that capitalism constantly breaks up and that they continually recreate.

We are not saying that the workers always or even most of the time achieve these objectives. In the last analysis, they cannot achieve them without smashing the capitalist organization of the enterprise - which is impossible without at the same time smashing the capitalist organization of society. Setbacks and defeats are inevitable phases in this process. But as long as the capitalist organization is there, the struggle will always be reborn from its ashes and will be led both by its own dynamic and by the objective dynamic of capitalist society to widen and deepen. This is the meaning of this struggle that we have been trying to bring out.

Neither are we saying that this meaning is simple, a state of grace automatically investing the working-class condition, a socialist apriority innate to proletarians. The proletariat is not socialist - it becomes so, or more exactly, it makes itself socialist. And, long before it came to appear as socialist by organizing itself into trade unions and parties with this name, it makes appear the embryonic elements of a new form of social organization, of a new type of behaviour and of a new human way of thinking, in its everyday life and in its daily struggle within the capitalist enterprise. It is upon this terrain that we will now begin to analyse the dynamic and the signification of working-class struggles.


The Struggle over Output

The tendency of workers to regulate their own work pace to the greatest extent possible - by combating management's norms, and then by "bending" these norms with all the means at their disposal - appears to management as "restricting output" or "restricting production." Faced with such curtailment, the classical "rational" counter response is "output-based wages" or "piece-rate wages."[44] The worker thus will be driven, "in his own interest," to increase output to the maximum. In doing so, he also will, incidentally, provide indications of what levels of output can be attained - which will make it possible to revise the norms downward when the time comes.

Industrial sociologists (mainly the Elton Mayo school) have criticized this method as "mechanistic" because it postulates that the worker is an "economic man" whose sole motive is getting the maximum amount of earnings whereas in reality other motives play a much more important role. This critique starts from a correct idea in order to come up with a false conclusion. It gets at the capitalist system as a whole, but falls far short of the problem that concerns us. Workers certainly are not "economic men." They behave exactly like "economic men," however, toward management. They pay management back in its own coin.

First of all, workers generally do not go for the efficiency bait, for experience teaches them that after a short period of receiving bonus pay a draconian reduction in the norms will supervene.[45] Next, they discover ways to get an increase in wages without a real or apparent increase in output.

In small- or medium-scale production with individual bonuses, the means used by workers are practically unstoppable. Taking as an example the shop described by an American author,[46] these means can be set forth as follows.

1. To avoid having the norms revised after output increases, the workers never show (which does not mean that they never attain) results surpassing 145-150 percent of the norm.

2. On the "gravy jobs," which represent nearly half the jobs done in the shop and which are defined by the possibility of going far beyond normal output, when the workers cannot "fix" the actual output so as not to appear to exceed these set maximums, they "loaf," either literally or figuratively. The resulting wastefulness is estimated by the author with the help of some long and involved, but quite conservative, calculations at around 40 percent of the workers' time - and that, in his opinion, is an "underestimation."

3. On the "stinkers," which represent the other half of the shop's jobs and which are defined by the fact that it is impossible to get a substantial pay bonus no matter how much effort is made (the cut-off point seems to be, in the case analysed by Roy, in the neighbourhood of 120 percent of the norm), the workers generally "goldbrick" and fall back on the base rate (the hourly rate determined in collective bargaining, whatever the output actually achieved). There is, nevertheless, an important exception: If the "stinker" in question comes in large lots or is a job that must be done often, there begins a relentless struggle with the time-study men to revise the norms.[47] The wastefulness brought about in such a case is, according to the author, comparable to that of the previous case.

4. The very existence of these two types of jobs (as well as other minor jobs paid by the hour: machinery setups, jobs for which norms have not yet been established, "reworking" defective pieces) gives the workers ample opportunities to increase their pay without their apparent output going beyond the "normal" rate. Thus, if a worker has a "gravy job" for four hours, during which he could work at 200 percent of the norm, and a "stinker" for four hours, during which he will not be able to work at the norm, he can choose between three options.

He can (a) follow management's formal rules, in which case he will make a twelve-hour wage (4 x 2 + 4 x 1) - with the certainty that a few days later the time allotted for the "gravy job" will be reduced. He can (b) hold back on the gravy job to 150 percent; he then will make a ten-hour wage (4 x 1.5 + 4 x 1). Last, he can (c) work at 200 percent of the norm on the "gravy job" and at 100 percent on the other one, but report that the first job was carried out in 5'/3 hours and the second in 2% hours. It then will appear that the worker had worked at 150 percent of the norm in both cases, he will make a twelve-hour wage, the maximum amount of production will be carried out - and there will be no danger of the time allotments being reduced.[48]

The worker can obtain a similar result by changing the apparent allocation of his time between the "gravy jobs" and jobs paid by the hour (with the difference that in this case he increases his pay without increasing production).

5. For the workers to be able to realize these possibilities, most of the work rules established by management have to be broken. In fact, the whole system of capitalist "rationalization" of labour is struck down by it; management loses the ability to determine the breakdown of the workers' hours between various jobs, and ultimately all its accounting procedures and calculations of profitability are utterly ruined. Therefore, management has to react and it can do so only by instaurating additional "controls." If these controls are "effective," they lead the workers toward solution (b) as described in 4 - namely, restriction of output, and hence wastefulness.

These controls, however, quickly become ineffective. If the inspectors remain in their offices, they basically can inspect nothing at all. This is the case with the time-study men, who are used in fact, according to Roy's phrase, as the true "hatchet men" of upper management: Though they are merciless against machine operators whom they find breaking the rules and get them dismissed immediately, these time-study men described by Roy appear only very rarely on the shop floor. If they are stationed in the shop, they cannot resist the continuous pressure of the operators for long.[49] Such is the case with the "time checkers" who are supposed to record the time at the beginning and the end of each job specifically to prevent any "fixing" of real output. Quite soon these time checkers themselves ask the operators, "When do you want me to check you?" In fact, not only production workers but all "service" employees who are in direct and continuous contact with them ("time checkers," tool-crib attendants, stock-chasers, setup men, inspectors, and ultimately even foremen) continually cooperate to a greater or lesser degree to break management's rules (which in their eyes, and objectively, are absurd) and to allow the workers to "figure the angles." "Figuring the angles" would be impossible without this constant cooperation involving all the parts of the managerial apparatus that are in ongoing contact with the producers.

Not being able to trust its human representatives, management is obliged once again to fall back on impersonal and abstract regulations. It introduces new regulations aimed at making the transgression of its rules "objectively impossible." But the objective observance of these new regulations of necessity depends in turn upon human control: Their effectiveness presupposes that the problem they are called upon to resolve is already resolved. From this standpoint, additional regulations are made in vain, for workers in cooperation with the lower strata of the "auxiliary services" quickly succeed in circumventing them.

But there is more: Most of the time these regulations introduce an additional degree of wastefulness and anarchy. The operators and the service employees are obliged by this very fact to devote part of their efforts not only to circumventing the regulations but to compensating for its irrational effects.

Thus, in the factory described by Donald Roy, in order to keep the machine operators from "figuring the angles" (allocating the apparent distribution of their time between different jobs as it suits them), management appoints "time checkers." In fact, the latter become the operators' allies and are turned against management. At a certain point, management decides to react and to make a "ruling" aimed at making the operators' "make-out angles" "objectively impossible." The "ruling" in question forbids the operators from keeping their tools and other auxiliary means of production (the "setups") next to their machines after a given job is finished as well as from getting what they need from the tool-crib attendants "in advance" (these two practices obviously being necessary to do any other work than what they are supposed to be doing). Tool orders in triplicate are used to guarantee adequate monitoring. At the end of each shift, the work-order card and all tool setups have to be turned in to the tool-crib attendants, whether the job is finished or not. The setup work then has to be started all over again by the next shift.

The rule's effects - which indeed have been foreseen by experienced workers - are not long in coming: a considerable increase in the tool-crib attendants' workload resulting both from increased paperwork and from the need to reassemble and re-sort the requested tools after each shift (up until then, the machine operators and setup men served themselves from the tool crib); also, there is a considerable loss of time for the workers and long lines begin to form at the tool crib. But management's desired result is not achieved: The triplicate forms ire filled out and exchanged each time - but the tool-crib attendants continue to supply the operators in advance with their tools.

Faced with this situation, management, four months later, modifies its first rule with a second one. To avoid long lines forming in front of the crib the shifts no longer are obliged to turn in their work-order cards and tools at the end of their workday, but tools can be furnished from then on only upon an order in duplicate from the "time checkers." At the same time, the inspectors have to countersign the time a job ends before a new work order can be obtained (this is done to permit a cross-check of the times marked by the "time checkers" and the inspectors).

Nevertheless, the second rule also results merely in increased paperwork for the tool-crib attendants. The setup men, who are allowed to go into the tool crib, pick up setups ahead of time for the operators. The inspectors quickly fall in step and "countersign" the time cards as requested by the operators. The shop gets back into a routine again, under slightly different procedures - and with a notable increase in the production of pink, white, and blue paper.

Management does not let itself get discouraged. It publishes a third "ruling" officially forbidding anyone from going into the tool crib except the tool-crib employees and two superintendents. The order, signed by Faulkner, the director of the factory, is posted on the tool-crib door.

An old machine operator, Hank, predicts that the new order "won't last out the week," and a setup man explains why its effects will be tough on the grinders and crib attendants, because setup men and foremen have been doing much of the [tool] grinding and have made it easier for them by coming in to help themselves to tools, jigs, etc.

A new line forms in front of the crib as a result of the third rule. The foremen are furious, they yell at the crib attendants and warn them that they will make out allowance cards charging them for every minute of time the workers are delayed because they do not have their tools. The boys who are standing in line at the crib window growl or wisecrack about the crib attendants.

Then Jonesy, the most conscientious and most efficient of the crib attendants, declares that he has "had enough" and lets foremen and setup men back into the crib again. The notes taken the same evening by D. Roy are worth citing verbatim.

Just ten days after the new order was promulgated, the sun began to break through the dark clouds of managerial efficiency. Hank's prediction was off by four days. ... Johnny (setup man) and others seemed to be going in and out of the crib again, almost at will. ... When I asked Walt (crib attendant) for some jaws to fit the chuck I had found, he said: "We've got lots of jaws back here, but I wouldn't know what to look for. You'd better get the setup man to come back here and find you some." Walt said to me: "I break the rules here, but not too much - just within reason to keep the boys on production. Faulkner's order still hangs at eye level on the crib door. ...

"And so much for Faulkner's order!" The "fix" was "on" again, and operators and their service-group allies conducted business as usual for the remaining weeks of the writer's employment.

The dialectic of this situation can be summed up easily in a certain number of moments of universal import. The essential element in production costs is human labour (in any case, the sole element upon which management can or thinks it can continually act: the others depend on factors that for the most part are beyond its control). Management seeks to reduce its costs by trying to obtain maximum output with minimum pay. The workers want to get maximum pay by providing what they consider a fair amount of output. Whence the fundamental conflict over the content of the work hour.

Management tries to overcome this conflict through' 'rationalization," through a strict definition of the amount of effort to be provided by the workers, tying their pay to the amount of production attained. This "rationalization" only makes the initial conflict grow and blossom into a number of specific conflicts: over the setting of norms, the concrete application of such norms, the quality of tools and machinery and their depreciation, the application of regulations aimed at organizing work from management's viewpoint.

The initial conflict, far from being overcome, is broadened at the same time as it is deepened, for management's successive counter responses force the workers to put all aspects of the organization of labour into question. At the same time, the overhead costs of capitalist management are considerably increased: voluntary restriction of output on the part of the workers, time taken up merely struggling against norms and regulations, multiplication of auxiliary services and in particular "supervisory" services that in each instance have to be rechecked by others, etc.[A*]



[1] In CS II.

[2] See RPB, in SB 1, pp. 286-310 [T/E: see "The Proletarian Revolution against the Bureaucracy," this volume, starting with the section entitled "Working-Class Resistance: Ultimate Cause of the Failure of the 'Plan,'" and ending with the first half of "The Political Evolution of De-Stalinization"].

[3] Concerning the problem of remunerating labour in a socialist society: CS I, pp. 12-15 [T/E: reprinted in CS, pp. 83-87, and included in PSW 1 as the second section of CS I, "Marxism and the Idea of the Autonomy of the Proletariat"]; apropos of the very nature of work and of the "reduction of the workday" as a solution to the problem of alienation: CS II, pp. 14-22 [T/E: reprinted in CS, pp. 123-37, and included in this volume as section 4 of CS II, "Socialism Is the Transformation of Work"].

[4] See the critique of this conception in CS II, pp. 14-22 [T/E: see preceding note].

[5] T/E: Castoriadis uses the phrase "une seule bonne méthode" followed by the English phrase "the one best way" within quotation marks and in parentheses.

[6] With the addition of various other factors, like the percentages allotted for "taking account of unforeseen possibilities" - which in fact can be assessed only empirically and arbitrarily and which thereby ruin the alleged "rationality" of the rest.

[7] We are talking about scientific management insofar as it applies to the problems of output by human beings. As production engineers, the Taylorists were able to play a positive role in a host of domains concerning the material rationalization of production - and sometimes also the rationalization of human motion by making known to others the most economical methods, as picked up from individual workers.

[8] Thus a strike breaks out in an enterprise following an average 20 percent reduction in time allowances in the assembly shop. Among other issues, the shop stewards brought up the fact that "components were now supplied in bulk, whereas previously they had been sorted and laid out in a carrier; moreover, frequent stoppages were caused by bad supply arrangements at assembly points, which penalized workers paid on an output basis" (R. J. Jouffret, "Description of Two Cases in Which Human Relations in Industry Were Impaired by the Efficient Use of Time Study in Determining Production Bonuses," in Human Relations in Industry [Paris: European Productivity Agency, 1956], p. 202). Such situations exist everywhere.

[9] Jouffret, ibid., p. 201. The times noted are adjusted to the "normal (performance) rates" and "rest coefficients," which can be based only upon the time-study engineers' estimations.

[10] Here we have one of the "findings" of the famous Hawthorne factory experiments conducted in the United States from 1924 to 1927 under the direction of Elton Mayo: "It was found that the more intelligent the girl, the greater was the number of variations (in her movements)." J. A. C. Brown, The Social Psychology of Industry (London: Penguin, 1956), p. 72.

[11] The "objective-scientific" measurability of labor-time aimed at by Taylorism "extends right into the workers' 'soul': even his psychological attributes are separated from his total personality and placed in opposition to it so as to facilitate their integration into specialized rational systems and their reduction to statistically viable concepts. ... In consequence of the rationalization of the work-processes, the human qualities and idiosyncrasies appear increasingly as mere sources of errors (G. Lukács, History and Class Consciousness [Cambridge: MIT Press, 1971], p. 88.)

[12] See the summary of this critique in J. A. C. Brown, The Social Psychology of Industry, chapters 1 and 3. Speaking of Taylorism, Alain Touraine writes (L'Evolution du travail ouvrier aux usines Renault [Paris: CNRS, 1955], p. 115): "Since Taylor, personnel administrators have striven to stop (the workers) from 'loafing,' but Taylor's pseudoscientific and purely coercive methods today are condemned; the importance of human relations, of communications, of informal organization, i.e., of social adjustment [T/E: Touraine places the English phrase 'social adjustment' in parentheses and in italics, following the phrase 'integration sociale'] of the worker into the enterprise, has become the principal theme of American Personnel Management." [T/E: "Personnel Management" appears in English.] But what value is there in condemning Taylor when it is well known that the great majority of French businesses pay workers on an output basis, using time-motion studies (R. J- Jouffret, "Description," p. 200)? In fact, as we shall see, management has responded to the bankruptcy of Taylorism with more and not with less coercion. As for "human relations," we will come to it later.

[13] The first person to experience this struggle obviously was Taylor himself. Speaking of the first years of his career, when he himself applied his method in factories, he wrote, "I was a great deal older than I am now, what with the worry, meanness, and contemptibleness of the whole damn thing. It's a horrid life for any man to live not being able to look any workman in the face without seeing hostility there, and a feeling that every man around you is your virtual enemy" (cited by J. A. C. Brown, 'The Social Psychology of Industry, p. 14). See a description of the workers' attitude toward time-study men in Georges Vivier, "La Vie en usine," Socialisme ou Barbarie, 12 (August 1953), pp. 38 and 40, Daniel Mothe, "L'Usine et la gestion ouvière," ibid., 22 (July 1957), pp. 90-92 [partially reproduced in Journal d'un ouvrier (Paris: Minuit, 1959]; Paul Romano, "L'Ouvrier americain," ibid., 2 (May 1949), pp. 84-85 [T/E: "Life in the Factory," in Romano and Stone, The American Worker (1947; reprinted, Detroit, Bewick Editions, 1972), p. 9]: "When the time-study men are about, the worker will find a multitude of reasons for shutting the machine down." The systematic slowdown of work performed in front of the time-study men is a universal rule. When time studies are done, the workers switch to lower speeds and slower "feeds" than the ones they will use later on; "operators deemed it necessary to embellish the timing performance with movements ... that could be dropped instanter with the departure of the time-study man" (Donald Roy, "Efficiency and 'The Fix,"' American Journal of Sociology, 60 [November 1954], pp. 255-66).

[14] R. J. Jouffret, "Description," p. 201. The idea that the workers "should normally" accept revisions in the hotted times is all the more astonishing since the author himself shows that the revision that provoked the conflict ended up stealing from the workers at least 10 percent of their time and since he concludes his study by saying that in this firm "the lack of confidence felt by the workers in the procedure of the Methods Department proved to be largely justified as a result of the joint survey subsequent to the dispute."

[15] See "Stakhanovisme et mouchardage dans les usines tchécoslovaques," by V. W. in Socialisme ou Barbarie, 3 (July 1949), pp. 82-87, and Guillaume's short report, "La Destakhanovisation en Pologne," ibid., 19 (July 1956), pp. 144-45.

[16] Testimony gathered by us from factory workers.

[17] R. J. Jouffret, "Description," p. 201-2.

[18] Donald Roy, "Quota Restriction and Goldbricking in a Machine Shop," American Journal of Sociology, 57 (March 1952), pp. 427-42. It should be noted that the entire analysis of the "Hawthorne experiment" made by the Elton Mayo school is based on the assumption that workers in the shops studied had no "rational reason" for restricting their output and that it therefore was necessary to find "nonlogical" motives for their behaviour. Roy remarks in this regard: "John Mills, onetime research engineer in telephony and for five years engaged in personnel work for Bell Telephone Company, has recently indicated the possibility that there were factors in the bank-wiring room situation which the Mayo group failed to detect: 'Reward is supposed to be in direct proportion to production. Well, I remember the first time I ever got behind that fiction. I was visiting the Western Electric Company, which had a reputation of never cutting a piece rate. It never did; if some manufacturing process was found to pay more than seemed right for the class of labour employed on it - if, in other words, the rate-setters had misjudged - that particular part was referred to the engineers for redesign, and then a new rate was set on the new part. Workers, in other words, were paid as a class, they were supposed to make about so much a week with their best efforts and, of course, less for less competent efforts' (The Engineer in Society [New York: Van Nosirand, 1946], p. 93)." (Quoted by Roy, "Quota Restriction," p. 431.) Let us add that the Mayo research group literally lived in the shop in question for five years and that it claimed to be studying reality without any pre-established theoretical schema, without any "preconceived ideas." This is what allowed them to rediscover in reality their unconscious ideas (for example, that management is always logical, and that, if the workers oppose management, it can only be for "nonlogical" reasons) and to ignore facts as massive as those mentioned by Mills.

[19] On conflicts over quality control, see Mothe's article, "L'Usine et la gestion ouvrière," in 5. ou B., 22 (July 1957), particularly p. 103. "To succeed in 'earning a living' (i.e., in not exceeding your time allotments), one has to cut corners on quality, eliminate an operation here and there. In the factory, this currently is called 'sabotage'" (G. Vivier, Socialisme ou Barbarie, 14 [April 1954], p. 57). This cutting of corners is the "streamlining" [T/E: the word appears italicized and in English in the original] of American factory parlance; cf. Roy, "Efficiency and the 'Fix,'" p. 257. On the contradictions, the resort to empirical methods, and the proliferation of piecework-related supervisory services, see Touraine, L'Evolution, pp. 169-70. Touraine concludes that ultimately "the unwieldiness of supervisory controls poses the question of returning to self-control," i.e., quality control over pieces by the semiskilled workers who manufacture them. It is not difficult to see that such an apparently minuscule change is impossible without a total overthrow of the structure of the factory, of wages, of the relations between the worker and his work.

[20] Roy, "Efficiency and the 'Fix.'"

[21] These are what Anglo-Saxon sociologists call "informal groups" or "primary groups." [T/E: In the original, Castoriadis gives the French translation of these two phrases. We have retained throughout his phrase, "elementary groups," to distinguish his analysis from that of these "Anglo-Saxon sociologists."]

[22] We shall see later that the divergence between the workers' spontaneous organization and the factory's official organization is, from a certain point of view, the condensed expression of all the conflicts and of all the contradictions of the capitalist enterprise.

[23] The study of elementary groups goes back to Charles H. Cooley (Human Nature and the Social Order [1902; reprinted, New Brunswick, N.J.: Transaction Books, 1983]). Its application to industrial sociology is tied to the works of Elton Mayo and his school. See, in particular, Elton Mayo, The Social Problems of an Industrial Civilization (1945; reprinted, Salem, N.H.: Ayer, 1977).

[24] Mayo, Social Problems, Chapter 2, "The Rabble Hypothesis and Its Corollary, the State Absolute."

[25] T/E: Castoriadis uses the English phrase "rabble hypothesis" in italics, followed by the French phrase "postulat de la horde." What we have translated as "the molecular hypothesis" is what he calls the "postulat moléculaire."

[26] Remark by Philippe Guillaume.

[27] Unless, once again, such "recognition" [reconnoitre] means inviting management to utilize its "acquaintances [connaissances]" in such groups in order to worm its way into them, the better to combat them. Contemporary American literature and cinema offer many examples of this type of utilization: Thus in the film Blackboard Jungle, an elementary group is broken up by discrediting the "ringleader" in the eyes of its members.

[28] We are thinking in particular of Mayo, but the same can be said of all of industrial sociology. Thus Brown, in his excellent synthesis of industrial sociology already cited, persistently recapitulates the criticisms developed by several writers in this regard against Mayo and emphasizes that elementary groups have their own logic, in no way "inferior" to management's logic, but he remains unable to get himself out of the contradiction as thus stated. And for good reason, for the only way out is workers' management - obviously an "unscientific" idea for a sociologist.

[29] See the extraordinarily vivid description of this informal organization in the Renault factories by Mothe, "L'Usine et la gestion ouvrière," in particular pp. 81-90, 101-2, and 106-10.

[30] An informal organization also exists, of course, at higher echelons in the management apparatus - but, as will be seen later, it obeys another type of logic than that of an informal organization of executants.

[31] See a description of this kind of cooperation in Mothe's "L'Usine," as well as the long quotations from Roy that we provide later.

[32] Of course, it is not a matter here of separate time periods, but of simultaneous facets, of logical moments in the process of organizing production.

[33] See in this regard Mothe's long exposition in "L'Usine"; likewise those of Vivier (Socialisme ou Barbarie, 12 [August 1953], pp. 46-47, 14 [April 1954], pp. 56-57) and of Paul Romano (ibid., 2 [May 1949], pp. 89-91 [T/E: 1972 American edition, pp. 12-14].)

[34] Cf. N. Wiener, Cybernetics (New York: Wiley, 1948), pp. 172-73.

[35] "After each model change, the supervisors frenetically run through the factory trying to get the plans and machinery which have been studied for months in the offices to work normally. At this moment the foreman is boss; he puts the workers where he wants, he breaks up old groups, he asserts his authority. It is the moment of greatest disorganization in the factory. For precisely this reason few Detroit autoworkers will buy a new car immediately after the model changes. They leave this lemon to people who don't work in a factory and therefore don't know any better. It is only when the workers are able to reestablish a certain amount of order in production that things go smoothly. The foreman has been put in charge of a group of workers and he is told what he should make them do. The organization he brings about is always bad. The assembly line goes too quickly or else there is only a single man where there should be two. The workers explain that to him, but he has his orders and cannot make any changes based on what the workers say. The men therefore are obliged to take the situation in hand themselves. They screw up the work so that the assembly line has to be stopped. Finally, after this situation has gone on for some time, management wises up, production is adjusted, and the cars produced are worth the price of purchase" (The American Civilization, monotyped text produced by the American group from Detroit, Correspondence, p. 47; [T/E: despite a long search, no copy of this text has been found; we therefore have retranslated Castoriadis's French back into English.])

[36] See Mothe, "L'Usine," p. 88.

[37] See RPB, in SB 1, pp. 279-81 [T/E: see "The Proletarian Revolution against the Bureaucracy," this volume, the third unnumbered subsection of the section entitled "Bureaucratic Planning"].

[38] For example, every form of psychoanalysis worthy of the name is based on the idea that the freedom of the subject is at one and the same time the end and the means of the therapeutic process - and every utilization of psychoanalysis by industrial sociology is based on the manipulation of the subject, both as means and as ultimate end.

[39] In the Renault factories, the percentage of "monthly salaried workers" went from 6.5 of the total in 1919 to 11.7 in 1930, 17.8 in 1937, and 20.2 in January 1954 (Touraine, L'Evolution, pp 164-65). On the development of offices in American industry, see C. Wright Mills, White Collar (New York: Oxford University Press, 1956), pp. 65-70.

[40] In this regard, the analysis of the attitude of these strata, as furnished by C. Wright Mills in the final chapters of his White Collar, has the following shortcomings:[1] It mixes disparate categories of "white-collar proletarians" whose situations and outlooks differ fundamentally; and[2] it does not take into account the dynamic of their situation. In particular, illusions about "status" will not outlive for long the real conditions that once had nourished them. The phenomenon of the industrialization of office work obviously is of decisive importance in this regard. Cf. R. Berthier's excellent analysis, "Une Experience d'organisation ouvriere," in 5. ou B., 20 (December 1956), pp. 6 ff.

[41] At an entirely different level, this phenomenon of "duplicating" the bureaucratic structure that blankets all of society with a more exclusive managing organ, the Party (which unsuccessfully tries to be the authoritative seat of reunification and thereby also tends to render the State's entire bureaucratic apparatus useless) has been brought to light by Claude Lefort, starting off from the speeches of the Twentieth Congress of the CPSU. See, in 5. ou B., 19 (July 1956), his article "Le Totalitarisme sans Staline," in particular pp. 45 ff. [now in Elements, pp. 166 ff.; T/E: 1979 ed., pp. 203 ff.]. Let us add that in duplicating the structure of the State bureaucracy, the Party is obliged to reproduce it within its own ranks, creating specialized commissions, etc. That is to say, this is no solution to the problem, by near or by far.

[42] On the necessary incompetence of managers within the present system, see C. Wright Mills, The Power Elite (New York: Oxford University Press, 1956), especially pp. 138-46 as concerns managers of industry, pp. 205-24 as concerns military leaders, and the final chapter of the book.

[43] See the articles by Romano, Vivier, and Mothe already cited. Noting the relatively small number of "suggestions" from workers that are aimed at improving production, Touraine writes: "How is this relative failure to be explained? In the first place by remembrance of the past. The worker, used to seeing his suggestions and his initiatives turned back against him when the time-study men are called in, abandons his former mistrust only slowly" (L'Evolulion, p. 121). "To abandon slowly" is a euphemism: The figures cited by Touraine refer to the period 1945-47. What has happened since then has not prompted the workers to abandon their mistrust. Quite the contrary.

[44] The types, formulas, and names for "wages based on output" are innumerable. But as far as we are concerned here, only the general meaning [contenu] of these formulas matters: The worker's wage is, within ample limits, a function of the quantity of production provided.

[45] One of the workers in the shop where Roy worked said to him, "Don't you know that if I turned in $1.50 an hour on these pump bodies tonight, the whole God-damned Methods Department would be down here tomorrow! And they'd retime this job so quick it would make your head swim! And when they'd retime it, they'd cut the price in half?"

[46] Roy, in his articles cited earlier.

[47] Roy describes at length an epic struggle in such a case between the four best workers in the shop and the time-study men, a struggle that lasted nine months and only came to an end when the workers won. This outcome makes one think - just as Mothe's remarks ("L'Usine," pp. 91-92) do - that the great majority of jobs are "stinkers" at the outset and that it is the workers' struggle against the time allotments that progressively transforms them into "gravy jobs."

[48] This third option, very likely applied as soon as the conditions for it are given, corresponds exactly to the concept of "maximization of profits in the long run" recently discovered by bourgeois economists as the principle that ought to guide the decision making of capitalist entrepreneurs.

[49] Let us recall that the stomach ulcer is the occupational illness of the foreman.



[A*] This text - of which the first part, a sort of programmatic introduction, was published in July 1955 in S. ou B., no. 17, and whose second part was devoted to a discussion of the problems of a socialist society, in issue no. 22 (July 1957) - continued with an analysis of the proletariat's political struggles, a critique of the overall organization of capitalist society, and an analysis of the crisis of contemporary culture. Events (May 1958, the scission within the S. ou B. group) interrupted its elaboration and publication. Parts of the first draft have been used in the writing of PO I, MRCM/MCR, and MTR/MRT. [T/E: The first two texts are included in this volume; the third is to be found in IIS.]


Last updated on: 6.14.2009