Degradation of Work in the Twentieth Century
* Book: Harry Braverman. Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century (1974)
URL =
Discussion
John Bellamy Foster:
"Nowhere in the 1960s, in fact, was the real nature of the labor process systematically addressed, either on the left or in bourgeois social science.9 It was simply assumed that more advanced technology, which was seen as a fait accompli, enhanced the skill of workers while threatening ever higher unemployment. Discussions of alienation, influenced by Marx, saw the relentless mechanization and automation of production as posing “a catastrophe of the human essence,” in the words of Herbert Marcuse.10 Yet, detailed, meaningful critiques of the labor process under monopoly capitalism were missing.
In his foreword to Harry Braverman’s Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century (1974), Sweezy was to highlight this shortcoming of Monopoly Capital with respect to the labor process, while seeing Braverman’s work as filling this enormous gap. “I want to make it clear,” he wrote,
that the reason Baran and I did not ourselves attempt in any way to fill this gap was not only the approach we adopted. A more fundamental reason was that we lacked the necessary qualifications. A genius like Marx could analyze the labor process under capitalism without having been immediately involved in it, and do so with unmatched brilliance and insight. For lesser mortals, direct experience is a sine qua non, as the dismal record of various academic “experts” and “authorities” in this area so eloquently testifies. Baran and I lacked this crucially important direct experience, and if we had ventured into the subject we would in all probability have been taken in by many of the myths and fallacies so energetically promoted by capitalism’s ideologists. There is, after all, no subject on which it is so important (for capitalism) that the truth should be hidden. As evidence of this gullibility, I will cite only one instance—our swallowing whole the myth of a tremendous decline during the last half century of the percentage of the labor force which is unskilled (see Monopoly Capital, p. 267).
In contrast, Braverman had a wealth of experience in the monopoly-capitalist labor process and was able to combine this with an extraordinarily deep understanding of Marx’s treatment of the working day in Capital, plus an examination of the entire history of modern management and the development of labor-saving machinery.12 Yet, while Braverman’s Labor and Monopoly Capital served to fill the gap left in Baran and Sweezy’s Monopoly Capital, Braverman at the same time took the description of the Scientific-Technical Revolution developed in Sweezy’s monograph, together with the general analysis of Monopoly Capital, as the historically specific basis of his own analysis.13 Fifty years after the publication of Labor and Monopoly Capital, the work thus remains the crucial entry point for the critical analysis of the labor process in our time, particularly with respect to the current AI-based automation.
Marx, Braverman, and the Collective Worker
Braverman’s basic argument in Labor and Monopoly Capital is now fairly well-known. Relying on nineteenth-century management theory, in particular the work of Babbage and Marx, he was able to extend the analysis of the labor process by throwing light on the role of scientific management introduced in twentieth-century monopoly capitalism by Fredrick Winslow Taylor and others. Babbage, nineteenth-century management theorist Andrew Ure, Marx, and Taylor had all seen the pre-mechanized division of labor as primary, and as the basis for the development of machine capitalism. Thus, the logic of an increasingly detailed division of labor, as depicted in Adam Smith’s famous pin example, could be viewed as antecedent and logically prior to the introduction of machinery.14 In Babbage’s case, Smith’s pin example was reconfigured to account for the economics of both manufacturing (the early factory system under cooperation) and modern industry (or machinofacture). The logic of the capitalist division of labor set the stage for Babbage’s designs of early calculating computers, aimed at the progressive development of the detailed division of labor as a means for promoting surplus value. Hence, there was a direct connection in the emerging theory of management of the nineteenth-century Industrial Revolution between the detailed division of labor, automation, and the development of the computer.15
It was Braverman, following Marx’s lead, who brought what came to be known as the “Babbage principle” back into the contemporary discussion of the labor process in the context of late twentieth-century monopoly capitalism, referring to it as “the general law of the capitalist division of labor.” According to this principle (now often divided into two parts), the division of labor in capitalist conditions was about determining (1) the least amount of labor necessary for each individual task, broken down into its smallest components, thus (2) generating an economy in labor costs, since each individual task could be assigned the cheapest amount of labor necessary for its fulfillment.16
Babbage had explained the benefits of the division of labor in terms of assigning the less demanding tasks (then seen as requiring less muscular effort as well as less skill) to cheaper female or child labor, as opposed to more expensive adult male labor, traditionally artisan labor. “By dividing the work to be performed into different processes each requiring different degrees of skill or force,” he wrote, the owner “can purchase exactly the precise quantity of both which is necessary for each process.”18 “The whole tendency of manufacturing industry,” according to Ure, was, if not bound to supersede human labor altogether, at least a means with which “to diminish its cost by substituting the industry of women and children for that of men, or that of ordinary labourers for trained artisans.”
“In the mythology of capitalism,” Braverman wrote,
the Babbage principle is presented as an effort to “preserve scarce skills” by putting qualified workers to tasks which “only they can perform,” and not wasting “social resources.” It is presented as a response to “shortages” of skilled workers or technically trained people, whose time is best used “efficiently” for the advantage of “society.” But however much this principle may manifest itself at times in the form of a response to the scarcity of skilled labor…this apology is on the whole false. The capitalist mode of production systematically destroys all-round skills where they exist, and brings into being skills and occupations that correspond to its needs. Technical capacities are henceforth distributed on a strict “need to know” basis. The generalized distribution of knowledge of the productive process among all its participants becomes, from this point on, not merely “unnecessary,” but a positive barrier to the functioning of the capitalist mode of production.
With the advance of the detailed division of labor, as Marx argued in his critique of capitalist production, machinery could be introduced to replace labor altogether, generating what was potentially automatic production, while throwing masses of workers into the relative surplus population, or reserve army of labor, thus decreasing labor costs across the board. The worker where still present was reduced to an appendage of the machine. This whole tendency was evident, as Marx pointed out, in the fact that the vast majority of workers in the textile industry at the heart of the Industrial Revolution in England were women and children, who were superexploited, receiving only a small fraction of the wage of the male artisanal workers that they had replaced, which was not enough for subsistence. All of this fed the development of machine industry and the further exploitation of workers, whose conditions—whether their wages were high or low—placed them at an increasing disadvantage in relation to the enormous productive apparatus that their collective labor had generated, and which was imposed on them like a deadweight to enhance both their exploitation and their displacement by machines.21
Still, in order to develop the division of labor further, it was necessary to break down the resistance of the workers with the aid of science as a direct power within production. This enabled what Marx called the real, as opposed to merely formal, subsumption of the worker within the capitalist production process. As Matteo Pasquinelli states in The Eye of the Master: A Social History of Artificial Intelligence: “Marx was clear: the genesis of technology is an emergent process driven by the division of labour,” while the implementation of the Babbage principle pointed all the way to automation and the dominance of the machine as the means for the enhanced exploitation of labor.
The incorporation of science, personified by what Sweezy was to call “the collective scientist,” as itself a new emergent power within capitalist production, was only actually possible with the economies of scale and the extension of the market associated with the growth of the giant corporation of monopoly capitalism. Simple management carried out by the owner and a handful of overseers in small-firm freely competitive capitalism would no longer suffice to maintain profitability under the new conditions of the giant, multidivisional corporation following the massive merger waves of the late nineteenth and early twentieth centuries.23
The new approach to management was best captured by Taylor; so much so that scientific management and Taylorism became synonymous terms. Taylorism was summarized by Braverman in terms of three distinct principles:
(1) “disassociation of the labor process from the skills of the workers,”
(2) “separation of conception from execution,” and
(3) “use of this monopoly over knowledge to control each step of the labor process and its mode of execution.”
Although Taylor claimed wage increases were integral to the system, at least in the early stages of the employment of scientific management in a given industry, the overall object was to reduce employers’ unit labor costs. “Taylor,” Braverman wrote, “understood the Babbage principle better than anyone of his time, and it was always uppermost in his calculations.… In his early book, Shop Management [1903], he said frankly that the ‘full possibilities’ of his system [of scientific management] ‘will not have been realized until almost all of the machines in the shop are run by men of smaller caliber and attainments, and who are therefore cheaper than those required in the old system.'” Taylor’s own distinctive contribution was to articulate a full-scale managerial imperative for increased job control, to be implemented primarily through deskilling. Hence, within Taylorism, Braverman maintained, “lies a theory which is nothing less than the explicit verbalization of the capitalist mode of production.”
The full contradictory logic of the capitalist mode of production and the possibilities for a revolutionary socialist response were, for Braverman, only brought out with mechanization and automation, including the introduction of AI (a more advanced form of automation) within monopoly-capitalist production. Here Braverman’s analysis relied fundamentally on Marx’s concept of the collective worker, which Marx used as a category to encompass the totality of the detailed devision of labor, the hierarchy of labor, and the incorporation of labor knowledge into machines. Even in the context of higher levels of mechanization associated with the deskilling and displacing of workers, the labor process, according to Marx, remained organically, and in terms of labor value as its basis, essentially the same.
Marx’s analysis of the collective worker in Capital transcended his discussion of the general intellect in the Grundrisse, written around a decade earlier. In what came to be known as the “Fragment on Machines” in the Grundrisse, the “general intellect” was incorporated into machines, leading to the apparent elimination of labor—and even of labor value—in production with the growth of automation.26 Braverman himself was to refer in Labor and Monopoly Capital to Marx’s statement in the “Fragment on Machines,” where Marx had written: “The production process has ceased to be a labour process in the sense of a process dominated by labour as its governing agency.”27 The “Fragment on Machines” has sometimes been used erroneously in recent discussions to argue that Marx saw the labor theory of value as being progressively displaced by machine production and automation.28 Yet, this has been refuted by analyses of how Marx’s later concept of the collective worker came to demystify the entire process of mechanization and automation, demonstrating both the continuing centrality of labor and of the labor theory of value.29
Braverman’s approach to the seeming contradiction associated with the subsumption of the labor process to the machine was to focus precisely on Marx’s concept of the “collective worker,” not only as accounting for labor’s everlasting centrality to production, but also pointing to new revolutionary possibilities. In the collective worker, labor as a whole was seen by Braverman, like Marx, as materialized within an organic process, encompassing the hierarchy of labor and mechanization.
Commenting on automation and the collective worker in Capital in response to Ure, Marx had written:
Dr. Ure, the Pindar of the automatic factory, describes it, on the one hand, as “combined co-operation of many orders of work-people, adult and young, in tending with assiduous skill a system of productive machines continuously impelled by a central power” (the prime mover); and on the other hand as “a vast automaton composed of various mechanical and intellectual organs, acting in uninterrupted concert for the production of a common object, all of them being subordinate to a self-regulated moving force.” These two descriptions are far from being identical. In one, the combined collective worker appears as the dominant subject [übergreifendes Subjekt], and the mechanical automaton as the object; in the other, the automaton itself is the subject, and the workers are merely conscious organs, co-ordinated with the unconscious organs of the automaton, and together with the latter subordinated to the central moving force. The first description [related to the collective worker in general] is applicable to every possible employment of machinery on a large scale, the second is characteristic of its use by capital, and therefore of the modern factory system. Ure therefore prefers to present the central machine from which the motion comes as not only an automaton but an autocrat. “In these spacious halls, the benignant power of steam summons around him his myriads of willing menials.”
In this contradictory posing of the implications of automation by Ure, the first description, corresponding, as Marx suggested, to the phenomenon of the collective worker in general, is consistent with the development of socialist production. The second corresponds to the myth of the machine itself, endowed with a general intellect, and in which labor is either totally absent or reduced to an abject, brainless state. For Ure, “when capital enlists science into her service, the refractory hand of labor will always be taught docility.”31 For Marx, in contrast, the revolutionary response was to enlist science on behalf of the collective worker in such a way as to enhance free social development.
What was to emerge as the culmination of Braverman’s own analysis, building on that of Marx’s Capital, was the development of a revolutionary approach to the division of labor, mechanization, automation, and AI, in which the collective worker was at least potentially the active subject of social labor. Such a view was strongly opposed to the more machine-fetishized characterizations—the preferred view of Ure and Taylor—of a “vast automaton composed of various mechanical and intellectual organs” and functioning as the insurmountable autocrat of production, with the workers reduced to mere appendages."
In Conclusion: The Collective Worker, AI, and the Reunification of Production
"In Braverman’s critique, modern technology, including automation and AI in the digital age, ultimately represented a powerful tendency to reunify a labor process that had been degraded by the capitalist division of labor. Significantly, all of the tasks used by Smith in his pin example at the beginning of The Wealth of Nations were now united in a single machine, allowing for the reunification of the labor process itself. Yet, capitalism in its monopoly stage, in which the exploitation of labor and the valorization process were still rooted in the Babbage principle, constantly sought to use higher levels of mechanization and automation to reinstitute what was now an increasingly archaic division of labor. As Braverman declared, “The re-unified process in which the execution of all the steps is built into the working mechanism of a single machine would seem now to render it suitable for a collective of associated producers, none of whom need to spend all of their lives at any single function and all of whom can participate in the engineering, design, improvement, repair, and operation of these ever more productive machines.” However, these possibilities technically open to the collective worker as a result of developments in the forces of production are thwarted by the social relations of production of monopoly capitalism. “Thus the capitalist mode of production enforces upon new processes devised by technology an ever deeper division of labor no matter how many possibilities for the opposite are opened by machinery.”
As Marx himself recognized in his conception of the collective worker, and as Braverman was to highlight in the context of monopoly capitalism, the new technological possibilities for human freedom, in which human beings potentially are the subjects of production, are turned against them. The worker becomes a mere commodified object in a world where capital management uses new machine technology to reinforce the detailed division of labor, treating the ever more “intelligent” machine as itself the subject of production. In Braverman’s terms, Marx’s collective worker was itself degraded under monopoly capitalism. “While production has become collective and the individual worker has been incorporated into the collective body of workers, this is a body the brain of which has been lobotomized, or worse, removed entirely. Its very brain has been separated from its body, having been appropriated by modern management as a means of controlling and cheapening labor power and labor processes.”
But if Ure’s notion of collective labor as reduced to a machine logic was clearly present under monopoly capitalism, Marx’s collective worker, combined with Sweezy’s collective scientist, stood for the new revolutionary possibilities that emerged as machines became more automated, incorporating knowledge of the labor process developed over the course of human history. With more extended education of workers in science and engineering through polytechnic schools made possible by increased productivity, this could lead to the reunification and enhancement of human labor and creativity. Ironically, the more that this became feasible, the more the capitalist education system was itself degraded, keeping workers under the domination of the Babbage principle, which depended on the devaluation of the knowledge of the worker.
Hence, in monopoly-capitalist society, education is increasingly subjected to the same logic as the detailed division of labor. The imperative of the system in this respect was clear from the start. As Frank Gilbreth, one of the founders of scientific management, wrote: “Training a worker means merely enabling him to carry out the directions of his work schedule. Once he can do this, his training is over, whatever his age.”34 This principle, coupled with the degradation of work, lies behind the intensive degradation of education in public schools in the United States and elsewhere. Science, culture, history, and critical thinking are being systematically removed or deemphasized at the K–12 levels, which are increasingly devoted, particularly in the early grades, to a reductive process enforced by standardized testing. It is as if the system has finally found the means to take full advantage of the classical-liberal political economist Adam Ferguson’s adage, “Ignorance is the mother of industry,” emphasizing that workers are more productive from the standpoint of capital the more mindless they are.35 The digitalization of education, rather than expanding knowledge and creativity, is leading to the opposite: relentless standardization. The goal seems to be to convert the larger portion of the population to what C. Wright Mills called “cheerful robots.”36 With the rise of large-scale language models, coupled with the growth of generative AI capable of incorporating masses of data inputs and artificially synthesizing information in “neural networks” in accordance with predetermined algorithms, university students are increasingly being encouraged to use these technologies as a mechanical substitute for actual learning.37 Rather than a collective worker or a collective scientist, the emphasis is on AI as a collective machine intelligence.
Behind this, in the hidden abode of production, lies the continued degradation of human labor. Google hired one hundred thousand temporary and contract workers to scan books at a rapid pace in time with a rhythm-regulated soundtrack as part of its plan to digitalize all of the world’s books (estimated to be 130 million unique volumes). Although the project has been largely abandoned, it was viewed as a mechanism for the development of generative AI.38 The rise in the number of temporary and contract workers, constituting precarious labor, are the hidden realities of the digital/AI era, obscured by the mystique of “cloud computing.” New platform jobs employ millions of contract workers. Online surveys of the national workforce by business groups such as the McKinsey Global Institute “indicate [that] between 25 and 35 percent of workers” in the United States have “engaged in non-standard or gig work on a supplementary or primary basis in the preceding month. As of 2024, that means at least 41 million people in the United States are engaged in some form of gig [or platform] work,” usually as contingent workers. Although countless jobs are threatened by AI—the estimates of which vary greatly—work is not so much being displaced overall as being made more contingent and precarious.
Yet, there are opposing tendencies to this seemingly inexorable degradation of work. New revolutionary struggles aimed at “the reconstitution of society at large” inevitably emerge, as Marx famously observed, where expanding human potential, associated with the development of productive forces, is fettered by the social relations of production.40 Today’s class struggles over the labor process are not directed against new digital technology or AI, but against the reduction of human beings themselves to mere algorithms. The collective worker as the embodiment of the general intellect can only control the conditions of production for the benefit of society as a whole under a developed socialism, or an egalitarian and sustainable system of human development."