This essay was originally written in early 2021 as an initial attempt to synthesize thoughts derived from my daily working life as an engineer with some influential concepts picked up from various theorists and writers. It is a little sprawling and uses ideas already well-established elsewhere as a staging ground to discuss the particularities of engineering labor. This was never directly published before now, however a significantly condensed version was published as The Present and Future of Engineers under the "Thinking About Communism" rubric for Field Notes, the political portion of The Brooklyn Rail edited by Paul Mattick Jr. The condensed version is largely the same in focus as this full-sized piece, however it also includes some new concepts on the role of ideology among engineers, the nature of engineering in communism, and an aside on software engineering. If you are interested in deeper elaboration and extensive usage of examples, read on! If you want most of the same concepts but without the high word count, just go check out the Rail piece. Special thanks to Phil and Tina for detailed comments on early drafts that improved readability.
Secular Trend Towards Rationalization of Engineering Work
Objective Subjugation
The popularization of scientific management of labor occurred thanks to Taylor1 in his attempts to standardize and thus control the concrete details of manufacturing. While this process is most commonly associated with the proliferation of factory-style labor with many workers performing dull repetitive tasks on an assembly line, this need for capital to standardize, and thus cheapen and accelerate, labor wormed its way into the realm of white collar work where much office work was stripped of its intellectual content and rendered just as tedious and low-paying as the workers out on the manufacturing floor. A data entry worker in an open office differs from an assembler on a production line in very little. Engineering work obviously requires a different kind of expertise than data entry, however aspects of the day-to-day reality of many engineers certainly resemble these Taylorist processes which have had drastic effects on other office dwellers.
The rationalization of office-work follows the same historical trajectory as the rationalization of the work directly involved in commodity production. Underlying both are the imperatives of capital imposing its abstract demands in the concrete reality of the workplace. Historically this first affected clerks and other administrative workers in factories, as the increasing complexity of industrial production necessitated a growth in office staff to coordinate the growingly complicated production process2. Much like the factory workers, their duties were split into multiple different roles then simplified with machinery such that each individual worker can be paid less and produce more. Since then the rationalization bug has bitten office managers world-over, and modern office workers from secretaries to engineers are subjected to the demands of financial efficiency. If the prestigious clerk of early capitalism, with the close trust of the factory owner and diverse skill-set, can be shattered into numerous separate professions, relieved of the need for advanced skills, and be relegated to increasingly precarious and low-paying jobs, there is no reason to think that engineers are immune to the rationalization that virtually every other profession undergoes as a result of the need for firms to reduce costs and boost productivity. Engineers simultaneously carry out the technical implementation of rationalizing measures while also being subject to work rationalization. The degree to which engineering work is tightly rationalized depends on its proximity to the production process itself, and its proximity to the commodity being produced. Proximity in this case refers to the level of abstraction between the labor of the engineer and the process or commodity in question, where being more proximate means being directly involved in the process performance or commodity creation (as in physically doing it themselves), and being less proximate means developing abstract concepts, designs, or models that will be used to guide the concrete aspects of the process or commodity in question.
The production process, as discussed earlier, is where material is augmented by the labor process so that its value increases3. Proximity to this can be described through its two extremes. At one end is the worker who carries out the production process and, through their labor, increases the value of the commodity. At the other extreme (but still within the realm of production with which this essay is concerned) is the manufacturing engineer (almost assuredly an engineering manager) who, at a high level of abstraction, maps out the processes and material flows required to develop a commodity without regard for the specifics of each process. What quantitatively separates the labor involved for each of these extremes is the amount of established planning and information in place that is required for the successful completion of the tasks at hand. For the worker directly manufacturing the commodity, just about everything has been established already. The material they will work on has been decided for them. What they are to do to it has not only been decided, but standardized so that the way they do it is predetermined. Their machine is programmed by someone else to perform the specific task at hand, the worker simply actuates it. Not only is the task and its character predetermined, but the specific amount of this task that this worker must complete per shift is also planned out. At the other extreme, the only pre-planned aspects of the engineer strategizing the production process of the firm as a whole are on the level of company directives as a whole. Their concern is the commodity production process only as far as it achieves the return on invested capital that the finance strategists and company directors as a whole have set as their goals. The only predetermined aspects to the individual at this extreme are the most abstracted aspects of production: the abstraction of concrete processes into monetary terms, into quarterly revenue targets and expenditure forecasts.
The amount of rationalization inherent to a job in the manufacturing process is not only proportional to the amount of predetermined information required for the execution of tasks, but is directly a result thereof. A manufacturing engineer performing the highest levels of planning does not have a strongly rationalized job because there is very little about that job that can be rationalized. The content of this individual’s work cannot be strictly controlled, even if it is predictable. Their ability to do their job well hinges on their capability for abstract thought and planning company directives based on the extremely complex world of customer needs, supplier capabilities, the ebbs and flows of that particular industry, and the general state of the global economy. Rationalizing such an individual’s work would be impossible, as the majority of their job involves converting unknown factors into concrete decisions. The laborer operating a machine on the production line, however, can be reduced to little more than a machine themself as a result of rationalization. Of course, these two examples are extremes and engineers reside almost entirely in between them. The level to which an engineer is proximate to the actual production process determines the level of rationalization. This can even vary on a task-by-task basis, as some labor performed by a single engineer is more proximate than other labor. A process engineer developing a single production cell will have their job performance evaluated by how well the cell metrics (throughput, error rate, etc) are conducive to the broader manufacturing plan. A manufacturing engineer, who at a small company may build the first-article version of a complex assembly themselves for validation and documentation purposes, is significantly less subject to the strict productivity standards a non-engineer assembler would be, however the process of documenting the process and ensuring that the design meets manufacturability standards is itself part of the rationalization process. If this same company is larger in size, the engineering labor might well be divided such that junior manufacturing engineers perform the most tedious of the documentation tasks in a repetitive fashion where the engineering tasks that require more expertise are given to other groups in the manufacturing engineering department. A test engineer responsible for evaluating product performance may be held to looser deadlines when thoroughly evaluating a representative product sample across a wide range of use-conditions, as extra engineering time (money) spent here to attain good test results will save money down the line as the product is manufactured in larger numbers. If the nature of the product requires a test of every single unit produced, and if the number of units produced is sufficiently high, then the engineering labor costs cannot be thinned out over a large number of commodities, and thus it becomes increasingly crucial to speed up this testing process to the point where the engineer is likely to be replaced with a technician (who is paid less than an engineer) on a standardized testing setup itself designed by an engineer to improve testing throughput and consistency.
Ultimately, the level of rationalization of manufacturing engineering labor is determined by how much rationalization can be gotten away with before the tasks become unperformable, which is inversely linked to the level of abstraction from the production process. On the design side of engineering there is a similar dynamic at play wherein the amount of prefigured information and decisions dictates the style of rationalization undertaken.
The most readily visible form of rationalization present in the labor of design engineers is the division of labor. A single engineer only designs an entire product from scratch if the product is very simple, the company has a very small number of design engineers in its employ, or the nature of the design work requires simply combining various prefigured options that themselves have already been designed. In the latter case the engineer is not even truly designing the entire product themself but rather performing a simplified actuation of a standardized process in a sort of mirror image of the assembly line worker slotting together prefabricated components into a larger assembly based off of specific work instructions. In most situations, however, one product represents the congealing of much design work from engineers of various disciplines and across various firms, especially as the complexity of the product grows. As the complexity of a design increases, it becomes less and less probable that there is a single engineer with a detailed understanding of every subsystem of the overall assembly. In the design of a complex machine, such as a semiconductor chip bonder or a CNC brake press for sheet metal, it may take a small team of mechanical engineers to design all the physical components, another team of electrical engineers to develop the circuit boards and wiring, and yet another team of computer and software engineers to create the layers of software that bring the machine seemingly to life. Within these teams tasks may be divided such that the work of any one engineer is greatly simplified. In some mechanical design contexts where the creation of the 3D models and manufacturing drawings is not intimately linked to the design process, these tasks are instead performed by a draftsperson who is typically paid less than an engineer and not required to perform actual design4. In the electrical realm the simple and repetitive design of cable harnesses and basic circuit board assemblies may be handed to a (lower-paid) junior engineer. In software it is not uncommon to see less-glamorous debugging and testing tasks performed by junior engineers or technicians, leaving code writing to more skilled engineers, and broader system software architecture to even more skilled and experienced engineers. This division of labor where simpler tasks are given to those with less skill and experience and more complex tasks are assigned to those with higher abilities is business common sense, as salary tends to correlate with experience and expertise.
The design side of engineering can also be viewed in the same terms of predetermined information and decisions, much like the manufacturing side can. The simplest of design tasks, such as the design of a basic cable harness, are simple by virtue of many of their parameters having already been decided throughout the design of the other components to be interfaced with. If one is designing a cable harness to connect several electrical components to a power supply, the designer already knows the types of terminals needed for each cable based off of the component specifications, what type of wire to select based off of current ratings, and how long to make each cable based off the known physical layout that has already been designed. The design of the harness becomes trivial busywork that in the case of a smaller company may be quickly designed by the same engineer who designed or selected all the components the harness joins. If there are many cable harnesses to be designed though, the ghost of Taylor possesses the department manager and maneuvers them into piling all cable harness design onto one unfortunate engineer to whom the other engineers simply send a standardized list of electrical parameters the harness needs to meet. This engineer, of course, can be paid less for the decreased amount of skill required to do the repetitive drudgery of these designs. Similarly, one engineer may be tasked with the important but tortuous task of error correction on engineering design drawings (the 2D drawings that specify the pertinent details to the fabricator, such as component dimensions or location of components in an assembly). In a company where many such drawings are produced, it is inevitable that small mistakes are made, which pile up. Creating these drawings often requires extreme attention to detail and it is easy to produce errors. A rationalized design engineering department would record the errors as they are noticed (either by that same department at a later date, or by a different department entirely) and sort them by importance. Critical errors may be fixed immediately, but if most are minor to medium importance, they may be saved such that they can be corrected all at once so as to not have to frequently ask various engineers to stop their work to fix a minor issue. Once the pile is sufficiently large, one unfortunate engineer must spend hours, days, or maybe weeks going through and fixing these minor errors. Some may be a quick fix, such as a typo, some may take more time, like changing every screw in an assembly (and its corresponding tapped hole) from imperial sizes to metric sizes. Some may take even longer, such as restructuring the hierarchy of parent and child subassemblies in a bill of materials and having to rebuild the corresponding parametric models and drawings linked to them. Needless to say, this work does not often require a lot of creativity, and is generally tedious and frustrating. Much like the cable harnessing, the majority of the information necessary to conduct the task has already been determined, the only remaining portion is the repetitive and dull labor of the engineer.
Subjective Results of Rationalization of Engineering Labor
Insofar as engineering labor occupies a unique and pivotal point in commodity production, one that carries out the abstract demands of capital on both the physical world and on other workers, it is also subject to these same tendencies. The historical erosion of craft production was often opposed (by craftspeople) on the grounds that their expertise and skilled labor were a source of social standing and personal pride5. It is not uncommon to hear similar sentiments echoed today from engineers whose labor is increasingly rationalized. Engineering is intellectually-driven and benefits from its practitioners cultivating as much expertise as they can as to ensure their particular scope of work is best serving the process as a whole. To the average engineer, this “process as a whole” is synonymous with their employer, or maybe the temporary partnership formed between their employer and customers. This occludes the actual totality that engineers serve, which is that of capital and capitalism. For many engineers the effects of enacting rationalizing measures on their work, be it overly-granular division of labor, repetitive task work, under-utilization of abilities, intellectual-understimulation, or lack of knowledge transfer between departments, serve as major sources of frustration and dissatisfaction.
The larger a company is, the more strictly demarcated its division of labor tends to be both between engineers and non-engineers, but also between different engineering job types. A rigid division between engineering duties (e.g. electrical design vs mechanical design, or process engineering vs quality control engineering) ensures that engineering time is spent in ways that management has strong control over, which is necessary for the completion of large projects involving many people. This division of labor, however, simultaneously undermines a corporation’s ability to extract the highest quality labor from its engineers. It is very rare for an engineer to only need to understand a small band of knowledge to do their job properly. The overwhelming majority of engineers strongly benefit from familiarity with the other engineering duties involved in the production of the commodity in question, especially those duties adjacent to theirs in the production process. A research engineer/scientist must have a sufficient understanding of the practical needs of the field in order to ensure that their research and findings are useful and applicable. A design engineer must understand enough about the manufacturing processes and application of their design to ensure that it is cost efficient to manufacture and can be utilized as intended by the end user. Likewise, the manufacturing engineer and the applications engineer cannot do their jobs properly if they do not understand the design intent of the commodity they work with. A manufacturing engineer must ensure that the fabrication they oversee is capable of yielding commodities that work as intended, and the applications engineer cannot best develop a product application for the customer if they do not have a full understanding of the capabilities and limits of the design. The best way for these engineers to understand the pertinent details of each other’s work is to be directly involved with each other’s work such that a strong intuitive understanding can be developed. Taken to its logical conclusion, this would dissolve the distinction between different types of engineers and different job duties. In reality this only happens at incredibly small companies (and even then everybody still retains a specialty based off of their abilities and interests), but the tension between empowering engineers with the ability to attain necessary knowledge vs retaining a centralized system of control to ensure financial goals are met is something managers in every engineering department struggle with. Allowing engineers too much freedom and autonomy makes it difficult for management to control the character and timeline of what is produced, however chaining everybody to their cubicle and requiring all communication to pass through management will quickly kill both the effectiveness and morale of engineers. A manager who is good at their job is capable of balancing this tension, however the reality is that division of labor tends to make it difficult for engineers to interact meaningfully with other departments, especially at larger companies.
The frustration over a strict division of labor tends to manifest in engineers and technicians who believe that their working life is being made more difficult by the shortcomings of other departments who lack the knowledge or understanding needed to make everybody’s life easier. Technicians become frustrated when asked to fabricate a design that was clearly not designed with the fabrication process in mind. An electrical chassis designed with no consideration for assembly might not have sufficient space for the technician to reach inside to connect a cable, or enough space to fit a socket wrench to fasten a grounding lug. A machine frame designed with a heavy cantilevered component may require two people to install it, requiring the technician to pull a colleague away from their work, when in reality a different location or fastening method could have been chosen that only requires one individual for assembly. Perhaps the documentation created by the design engineer or manufacturing engineer to instruct a technician is overly vague on the orientation of a certain component with respect to other components, requiring time spent to attain clarification. An engineer, upon learning how frustrated technicians are with their design, may be perplexed as the purported difficulties strike them as having obvious solutions, or that anybody who understands the design intent will automatically be able to infer important information from the documentation. For both the technicians and engineers a lack of understanding of each other’s work leads to this difficulty. Engineers lacking an understanding of the hands-on details of technician work may simply not consider many details that can add time and frustration to the fabrication process. This is compounded by the assumption that what is obvious to an engineer intimately familiar with the design intent, is likely not obvious to a technician who is generally asked to follow instructions without needing an understanding of the larger picture. Between engineers, similar frustrations are common. A manufacturing engineer, due to not sufficiently understanding the original design intent or the way the product is used by the customer, may inadvertently introduce unacceptable levels of variation into the fabricated item by not denoting that feature as having a critical function and defining a tight tolerance for it that the technicians must adhere to. The quality control engineer, if also not familiar with the end-use, may not think to check that function and thus not flag them for improvement, much to the annoyance of the applications engineer who must account for the inconsistent behavior of the end product in their work. Likewise, the quality control engineer may overly fixate on a specific metric that they erroneously believe to be important, causing significant frustration to the manufacturing engineer, meanwhile the applications engineer is blissfully unaware of this struggle as the feature in question has little impact on their work. Meanwhile, if the original designer is not given the opportunity to understand these difficulties, they might feel maligned at hearing other departments criticize their work when, to the designer, everything seemed to be perfectly in order. This interdepartmental frustration is not even limited to the technical portions of commodity production, and can include the work necessary for the capital to valorize itself in the form of the commodity being sold. It is incredibly common for engineers to have anecdotes of ways that the marketing department or a salesperson promised something to a customer that is absurdly difficult or impossible for the engineers to design and bring to fruition. It is obvious that all of these interdepartmental difficulties can be harmful to the bottom line. Poorly designed products, manufacturing difficulties, and cynical employees can absolutely destroy sales and, thus, profit. Capitalism, as it tends to do, creates small-scale tendencies that in isolation may be profitable but absolutely undermine the greater systemic processes as a result. The division of technical labor (and labor in general) under capitalism is no exception.
A dissolution of capitalism and the historically-specific forms of division of labor that accompany it have the potential to alleviate a lot of these tensions. As discussed earlier in this essay, after the dissolution of capital’s profit-oriented logic there would be no imperative to simplify manufacturing work and strip expertise from those who naturally accumulate it in the course of their labor. Likewise, for those who participate in labor that can be considered “engineering” in a socialist/communist society, the elimination of the antagonism between profit and everything else would mean that the tension between centralized control over attaining planned goals and the need for those performing technical labor to understand each other’s work can be mediated along lines of human need and holistically-rational decision-making, rather than the brutally authoritarian fever-dream logic of endless profit.
The frustrations arising from division of labor are likely to play a large role in disillusioning engineers with capitalism to the point where they choose to become involved with a real movement capable and willing to overturn capitalism and install socialism/communism. Capital’s tendency to rationalize all work performed in its name does not and will not spare engineers much of the same fate the majority of the world proletariat already faces, although the level of intensity of course varies. Performing repetitive work with much of the creativity stripped out of it for stagnant wages is not a recipe for a docile and complacent workforce. Given that getting into engineering in the first place requires an actual desire to do so and at least 4 years of rigorous study, there is a level of pride inherent to the common subjectivity in engineers that directly runs afoul of capitalism’s rationalizing tendencies. Engineers typically want to perform fulfilling work that satisfies some important purpose (however “important” is defined to them). Not every engineer wants to go save the world, but the satisfaction of performing meaningful work is implicitly promised through shared cultural understandings of what highly-educated work entails. Big-name companies working on high-profile products often use this as a psychological tool to squeeze long hours and low pay out of their engineers. “You’re not in this for the money right? You’re in this for the pride of making rockets/futuristic computers/fancy automobiles/popular software/important infrastructure!” they say to their workers. Unfortunately it can be an effective tool for the capitalist class. There exists a possibility that such rhetoric could be deployed by an anti-capitalist movement in a similar fashion. Appeals to the desire to do meaningful labor that is personally fulfilling and important for society have potential to be effective if used properly. Engineers are simply people, after all. As people, we are just as capable of being ideologically chained to capitalism and its self-implied permanence as we are of viscerally feeling its abjecting tendencies and declaring that we want a better society that works for everyone, not just for the inhuman needs of capital.
Built Accordingly
Engineers play a critical role in the maintenance of capitalist social relations. Our cornerstone position with respect to commodity production, the very backbone of capitalism, coupled with high levels of technical training not easily attainable by most other workers, results in an undeniably privileged status for engineers in terms of pay, benefits, and social prestige compared to the rest of the workforce. Work, no matter how well paid, is still work though, and every worker will have grievances against their job. Frustration over the usual suspects, such as mandatory (and often unpaid) overtime, or depressed salaries compared to average market rates or costs of living, can conceivably serve as a focal point for labor organizing among engineers. This is especially true if conditions are such that simply switching employers does not solve the issue. This is a double-edged sword, however, as labor organizing for better treatment under capitalism often poses little challenge to the capitalist order itself, especially as formal unions are just as dependent on the perpetuation of capitalism for their own existence as a corporation is. A formal dues-collecting union necessarily depends on the continuation of business in order to maintain its own existence, and is vulnerable to politics that pose zero fundamental threat to the reigning capitalist order. Unionization of engineers is particularly dangerous in this regard, as it would not be difficult for a minority union of workers who hold a near-monopoly on technical expertise to exploit their strong position to further entrench capitalism’s oppressive character to their own benefit at the expense of the larger workforce, rather than militating for the dissolution of capitalism. The formal unionization of engineers as a method to secure better working conditions for engineers could very well result in further exploitation of “unskilled” labor by amplifying the rationalization of their work. Even if engineers and non-engineer employees unite on a per-company basis, this can only serve to make engineers and their colleagues willingly complicit in the inherently predatory global supply chain, which relies on the imperialism of strong states to secure raw materials and cheap foreign labor that are the fundamental building blocks of the high-tech industries in which engineers are employed. While engineers share plenty in common with the rest of the workforce and are fundamentally proletarian, the echoes of pre-capitalist artisanal mastery over our work leave us not quite as dispossessed from the means of production as the typical worker, whose comparative lack of formal technical training makes them significantly more replaceable and, as automation is increasingly deployed in global commodity production, increasingly superfluous. Engineers would be just as much a liability to the dissolution of capitalism as we would be a critical component. This is hardly unique to engineers however, as the historical upending of capitalism to be replaced by socialism/communism would require that just about every component of capitalist society negate the innate kernel of capitalist logic within in favor of a truly human mode of production.
Engineering is a crucial component to the present mode of production, and would be similarly crucial to any communist project. The historical origins of modern engineering are multiple, with elements of pre-capitalist artisanal work, proletarian factory labor, and bourgeois scientific management all playing important roles in shaping the lived experience of the modern engineer. There is no easy answer to how engineers are to be meaningfully included into a movement aiming to transform the mode of production (which is to say, a movement that does not yet exist in any coherent form). What is clear, however, is that any answer absolutely must account for the role engineers play as both the intellectual linchpins of production while also being subject to its domination. Those of us who have accumulated the technical expertise necessary to enable the endless churning of things from factories have done so as part of a vast machine built on an absurd amount of suffering. This suffering cannot be undone, but can potentially be redeemed if this expertise is leveraged towards building a society where planetary domination by the alien logic of capital is consigned to the dustbins of history. Those of us who design the world’s commodities and bring thousands of years of accumulated knowledge to bear in flooding the world with manufactured goods can repurpose this knowledge towards communist society where human welfare is the defining logic of production. In doing so we would be destroying the very foundation of our own privileged status among the present productive status quo. Like the self-destruction of the entire proletariat, this should be embraced rather than resisted. The destruction of capitalism would not mean engineers must sink to the level of precarity and dispossession currently inflicted upon the laborers whose expertise we have stripped. Instead, that which makes our work so noteworthy, the opportunity for creativity, the ability to create, the means to systematically solve problems and bring concepts to fruition, would be shared by all those capable of doing it. In making the wonders of engineering accessible to all through the abolition of capital, we would also eliminate the horrid aspects of engineering tied to our present need to work for a wage. This cannot happen without a broader movement of workers and dispossessed proletarians for whom the capitalist order is no longer acceptable. While that which sets us apart from the rest of the proletariat is noteworthy, it is what we all have in common in the face of capital’s totalizing domination that is important.
A better world is possible, but it must be built accordingly.
Footnotes
- Braverman, Harry. Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century, Monthly Review Press, 1998. p. 60
- ibid, p. 203-245
- C-M+P-C’ in Capital value schemas
- The line between design and documentation of said design is very blurry. Drafters and CAD modelers inherently make design choices when bringing a more-abstracted engineering design to the less-abstracted model or drawing state, however these implicit design choices are subject to review by the engineer for compatibility with the original engineering design intent.
- Braverman, Harry. Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century, Monthly Review Press, 1998. p. 91-94
Leave a Reply