What do workers control when they control the means of production?
I’ve been reading Jean-Louis Cohen’s Building a New World: Amerikanizm in Soviet Architecture, and one of the striking features of the early chapters is the enthusiastic reception of Taylorism in Russia. That Taylorism was popular among reforming liberals in the pre-revolutionary period is not surprising, as Taylorism is a feature of advanced industrialism these reformers wanted to introduce to backwards Russia. The enthusiasm of many Bolsheviks is more puzzling, though: Taylorism seems like a specifically capitalist form of industrialization, a subordination of the worker to the factory or, as Lenin initially evaluated it, “sweating in strict accordance with all the precepts of science” (quoted in Cohen, 105). Lenin later revised his opinion, however, claiming to have identified a positive side to Taylorism in which “the Taylor system – without its initiators knowing or wishing it – is preparing the time when the proletariat will take over all social production and appoint its own workers’ committees for the purpose of properly distributing and rationalising all social labour” (105). Taylorism, that is, holds out the possibility of a rational organisation of labour, and while this is currently being employed in the service of capitalism, it would find its full realisation in the truly rational organisation of labour that would be the result of replacing the anarchy of capitalism with workers’ control.
This bad dialectics of attempting to dividing the good and bad parts of the current situation is exemplified even more strongly by Trotsky, who wrote that Taylorism “is the most sophisticated form of exploitation of the workforce, as well as the most brutal…. But on the other hand, it is a system allowing a discerning expense of the human force participating in production” (108). The correct dialectical procedure, rather than attempting to sort out the good from the bad, attempts to understand how they are internally related: how the apparently good is a form of appearance of the bad. In the case of Taylorism specifically, approaching the phenomenon in this way would allow us to see how the rational organisation of labour in Taylorism is necessarily connected to capitalist control over labour (not something re-usable outside of capitalism); and so might help us imagine what a non-capitalist form of control over production might look like.
We can think of Taylorism as having two broad themes: the rational control of work, and the material control of workers. The rational control of work is the observation of workers and the formulation of alternative, more efficient movements, which workers are trained to perform. The classic Taylorist example is the chronocyclograph, a mechanism whereby lightbulbs were attached to workers’ bodies and they were filmed in their work, to produce a detailed record of movements that could be examined for any inefficiencies (although it is a technically inaccurate to call this “Taylorist”, as it was developed by Frank and Lillian Gilbraith, and rejected by Taylor himself). The material control of workers’ bodies involves constructing the work environment in such a way that workers are compelled or encouraged to move in the most efficient ways; the classic example would be designs for factory floorplans based on studies of the time it took workers to move themselves and materials between different stages of the labour process. These two aspects of Taylorism aren’t rigorously distinct – both involve a rational element (the scientific manager determining the optimum form of work) and a material element (the worker doing the work), but it’s helpful to keep in mind that the rational element can appear more or less external to (or contrarily, more or less embodied in) the material element, because the movement between the two is central to how Marx understood the development of the labour process under capitalism.
Marx distinguished between two forms of industrialization, manufacture and mechanization. Manufacture arises when a number of workers are brought together as part of a single enterprise. At first, the capitalist merely reaps the benefits of economies of scale, but the concentration of workers also allows for an alteration of the way work is carried out. When the capitalist considers the group of workers he has brought together, he sees “a productive mechanism whose organs are human beings” (Capital, 457); and because production is carried out by this entire mechanism, the division of labour can be intensified such that, instead of each worker producing a finished product, they instead carry out an increasingly small and specialised part of the production process.
Manufacture modifies the labour process by subdividing it into ever smaller and more repetitive parts. However, although this rationalises the earlier craft methods, of production, it does not fundamentally alter them. Because “handicraft remains the basis” of manufacture, it “excludes a really scientific division of the production process, since every partial process undergone by the product must be capable of being done by hand, and of forming a separate handicraft” (458). This changes with the introduction of large-scale machinery, which does not just change how the work is divided up, but fundamentally changes what work is done. Modern machinery embodies the process of previously undertaken by the worker within its own mechanism. Marx gives the example of the slide rest, which “replaces not some particular tool, but the hand itself, which produces a given form by holding and guiding the cutting tool” (507). This means that the activity of workers in mechanized industry is of a fundamentally different kind to the work done in manufacture: “along with the tool, the skill of the worker in handling it passes over to the machine” (545), and while “in handicraft and manufacture the worker makes use of the tool, in factories the tool makes use of him” (548).
This development has a couple of consequences for thinking about the limitations of a “socialist Taylorism”. One is the dialectic of increasing abstraction. In craft production, where one worker undertakes every part of the transformation of raw material into a finished object, we have a direct production of a concrete use value. In manufacture, on the other hand, each worker only takes on a small part of the process, and the concrete product only emerges from that total process. From one point of view, this is an abstraction – the production process moves from a concrete individual worker to an omnicompetent “collective worker”. This collective worker abstracts away from the individual characteristics and flaws of any particular worker, and instead “possesses all the qualities necessary for production in an equal degree of excellence” (469). From the point of view of the individual worker, however, work becomes increasingly, and punishingly, concrete, as the labour process is subdivided into smaller and more specialised tasks. “Since manufacture carries the social separation of labour much further, and also by its peculiar division attacks the individual at the root of his life, it is the first system to provide the materials for industrial pathology” (484). In mechanized industry, however, the distinguishing concrete characteristics of work become embodied in the machinery itself, and the actions of the worker become increasingly abstract: “in place of the hierarchy of specialised workers that characterises manufacture, there appears in the automatic factory a tendency to equalise and reduce to an identical level every kind of work” (545).
Alongside this increasing abstraction of work comes a change in the way the power of the capitalist appears in the labour process. In manufacture, where the final product is produced by the cooperation of numerous specialised workers, there is a need for external coordination of these worker: “the interconnection between the various labourers confronts them, in the realm of ideas, as a plan drawn up by the capitalist and, in practice, as his authority, as the powerful will of a being outside them” (450). As manufacture is increasingly refined, the command of the capitalist becomes increasingly indirect, expressed in the plan for the coordination of different workers, until, with the development of mechanized industry, the despotism of capital is fully embodied in the machinery itself. In mechanised industry, “the automaton itself is the subject, and the workers are merely conscious organs coordinated with the unconscious organs of the automation, and together with the latter subordinated to the central moving force” (544).
The increasing embodiment of capitalist control in the matter of the labour process itself creates a problem for the idea of a socialist Taylorism. If the capitalist’s authority is external to the labour process, it seems plausible that the workers could wrest that authority from him, and replace it with a democratic authority, informed by a rational science of labour organisation descended from Taylor. It’s harder to imagine what democratic control of the means of production would look like without this external power to take over. I large-scale, mechanized, industry, power is embedded in machines, processes, and architecture; what does it mean to democratize that? It would have to be a different form of democracy from the liberal model, in which political power is externalised to a sovereign in order that the actions of that sovereign can be exercised in accordance with the will of the people. Instead, it would be a materialist democracy, not an externalisation but an embedding of democracy into machines and organisations. This is a hint of the immense effort of the imagination that will be required to develop the new forms of communist life.
Two thoughts on how this might pan out, one positive, one negative. On the positive side, Marx’s account of the increasing abstraction of work describes a tendency, not a completed reality. I’ve never worked on an assembly line myself, but reports from those who do suggest that the work is less evacuated of concrete content than capitalists would wish, and still requires a measure of human skill and knowledge. This kind of practical knowledge (which we can differentiate from the technocratic knowledge of Taylorism) may provide the grounds for a democratization that builds, or re-builds, the materials of production to reflect and strengthen the collective decisions of the workers (this perhaps has something in common with the constructivist idea that socialist objects would be “connected as co-workers with human practice”).
More depressingly, though, it seems like capitalism is continuing to innovate in the sphere of materialising control. The app-mediated gig economy places increasingly complicated mechanisms of control in technologies and processes which are increasingly distant from where physical labour is performed. Again, this technology cannot do without workers’ practical knowledge, as much as that might be the goal; but it does provide new opportunities to constrain and capture that knowledge in an immaterial infrastructure, fortified in data centres spread across the world. How would we democratise that?