In December 2017 the federal government implemented a requirement that all truckers buy, install, and use electronic logging devices (ELDs): digital systems that capture data about truckers’ activities, particularly their work hours, intended to keep them from driving for more time than federal regulations allow. When truckers meet the time limits, they are supposed to stop and rest—sometimes for as long as thirty-four hours—before they can legally drive again.
These devices are intended to address one of the most important and pervasive problems in trucking: fatigue. Truckers are notoriously overworked and underslept, a problem that can have deadly consequences. For decades truckers have been subject to federal “hours-of-service” regulations that limit the number of hours they can drive each day and each week before taking long breaks. These rules have been in place since the 1930s—and have been enforced, until recently, by requiring truckers to keep track of their hours using paper-and-pencil logbooks, which are subject to inspection by law enforcement at roadside or at weigh stations. But because truckers are economically incentivized to drive for as long as possible, owing in large part to the fact that they are paid by miles driven, it’s an open secret in the industry that paper logs are frequently falsified—so much so that classic trucker anthems even make allusions to these regulatory “swindle sheets.”
The federal government’s solution to the fatigue problem has not been to restructure trucker pay to reduce truckers’ incentives to overwork. Rather, it has turned to digital monitoring. ELDs create a digital record of truckers’ activities and are intended to be more “tamperproof ” and less falsifiable than paper logs (though, as we shall see, they are not completely so). And crucially, these devices facilitate surveillance not only by the government, but also by trucking firms—serving as a technological backbone that scaffolds a great deal of real-time data collection about truckers’ bodies and behaviors, and fundamentally changing the nature of the trucking workplace. Many truckers interpreted the introduction of ELDs as a matter of utter shortsightedness and disrespect. Truckers are strong, vigorous, essential workers on whom the economy depends, as pandemic supply chain issues have made all too clear over the last few years. Yet in their view, they are being hemmed in by a ridiculous device that prevents them from doing their jobs—and we will all pay for it in the long run.
Truckers’ work is being drastically affected by this proliferation of surveillance technologies. These technologies are part of an emerging regime of digital enforcement—the use of technology to enforce rules, both legal regulations and organizational directives, more “perfectly” than might otherwise be possible. In trucking, digital enforcement confronts the existing social order of the industry: it upends the occupational autonomy truckers have traditionally held, reconfigures information flows within trucking firms, alters how truckers and law enforcement officers interact with one another, and creates new sites of contestation and resistance. The economic realities of trucking have long depended on truckers’ discretion, including flexible recordkeeping routines and the ability to direct their own work in the face of unpredictable and often inhospitable conditions. But digital enforcement doesn’t address these realities; it papers over them.
The social and economic consequences are serious—and portend changes in how we interpret what rules mean and how they function in contemporary workplaces.
The rationale behind the ELD mandate is that “electronic logs take the noncompliance issues off the table” (as one trucking executive put it), making it much less feasible for truckers to break the hours-of-service rules. We might, of course, accept this goal as normatively desirable and relatively uncontroversial. If we have rules in place for safety reasons, at which we have arrived through legitimate political processes, and those rules are easily broken, why shouldn’t we enforce them more consistently if we can do so? Aren’t rules rules?
In fact, it’s not so simple—rules aren’t always rules. Rules are shaped by social, cultural, and economic realities, and are almost never as simple as they might seem on paper. As a simple example, consider how you’d feel if you were ticketed by a police officer for driving sixty-six miles per hour when the speed limit is sixty-five. Officially, you’ve broken the rule—but the violation is so trivial, and the rule is so commonly broken, that to-the-letter enforcement would very likely strike us as unfair and unreasonable. (Indeed, the policing of minor speed violations is often merely a pretext for police to stop and investigate drivers, particularly Black drivers, for other reasons.)
But there’s an even more important point here. It’s not just that some amount of rule-breaking, like the behavior of our barely speeding driver, is considered unobjectionable. The more fundamental point is that we often depend on rule-breaking to make the world function. When former New York City Mayor Bill de Blasio proposed a zero-tolerance policy for jaywalking, economists were quick to point out that if anti-jaywalking rules were followed to the letter, pedestrian commute times would increase, and the city’s social and economic life would suffer for it. We rely on this sort of routine rule-breaking to make society function efficiently.
Perhaps no phenomenon illustrates the point more clearly than the “work-to-rule” labor action, in which unions exert pressure on management by following every rule in the handbook to the letter. Doing so slows organizational functions to a halt—but because employees are officially following the rules, it is difficult to discipline them. The fact that working “to rule” functions as a resistance tactic demonstrates that work practices typically do not fully accord with the rules—indeed, organizations rely on the fact that they don’t.
Examples abound—particularly in the workplace. Organizational theorists have long observed that firms, in response to pressures for institutional conformity, often adopt certain formal rules and structures to achieve legitimacy, while decoupling those rules from practices that allow them to function efficiently. Alvin Gouldner’s seminal ethnography of a gypsum mine, Patterns of Institutional Bureaucracy (1954), describes workplace rules that were in neither managers’ nor workers’ interests to enforce. These rules were enforced only on rare occasions—for instance, a smoking ban was enforced only when insurance inspectors came by—while most of the time managers “looked the other way” at employee rule-breaking.
This “mock bureaucracy” effectively gave supervisors a bargaining chip they used to maintain friendly relationships with workers. More recently, Michel Anteby described similar practices in an aeronautical factory, where managers tolerated and even encouraged employees making small souvenirs, called “homers,” for new retirees, using company time and materials. Though homer-making was officially forbidden, the fact that this rule was so often broken served a similar function as Gouldner’s smoking ban: managers could ultimately exert more control over workers by seeming to be “on their side” when it came to rule-breaking, and workers viewed the practice as a source of pride and occupational identity.
To be sure, nonenforcement of rules, in the workplace or more generally, isn’t necessarily a good thing. Many rules are in place for a good reason—to protect workers and others in vulnerable positions, or as a check on the powerful. Selective enforcement of laws can be the basis of arbitrary or discriminatory treatment. And we may or may not think that the managers in Gouldner’s gypsum mine and Anteby’s aeronautical plant were ultimately doing workers any favors when it came to smoking and homer-making: “looking the other way” for certain rule violations gives managers leverage to pressure workers in other ways, and managers can “play favorites” by enforcing rules in some situations (or against some workers) and not others.
But we should not focus too strongly on determining whether strict, to-the-letter enforcement of rules (or nonenforcement on the other hand) is altogether “good” or “bad”—the world is far too complex to make such a sweeping statement. It is simply crucial to remember how much of social life, in the workplace and more broadly, relies on the “gap” between rules on the books and practices on the ground. The “wiggle room” around rules is a site for strategic negotiation, for economic functioning, for relationship management—for both good and ill. So when we decide to more strictly enforce rules using technology, without accounting for what has been happening in the gap, we may well disrupt the social order of a particular context in important and unforeseen ways.
Despite this fact, the notion of “more perfect” enforcement via digital technology is a common refrain, often motivated by the idea that technology can help us to close the gap between rule and practice—transforming society into a more consistent and just version of itself. As David Friedberg, then CEO of data analytics firm Climate Corporation, described it in 2014: “Technology is the empowerment of more truth, and fewer things taken on faith.” By using technology, the logic goes, we can ascertain what people really do, instead of what they say they do; we can catch and deter cheaters and liars; we can generate knowledge where before we had only hunches and secrets; we can make people follow the rules.
Sometimes, digital enforcement happens through attempts to prevent violation, making rules more difficult to break—using code to make it more onerous (or even impossible) to deviate from an imposed rule. For example, digital rights management technology makes it (nearly) impossible to violate copyright law. If these technologies work as “perfectly” as intended, rule violation is completely impaired, and violation becomes practically impossible (or at least much more difficult).
But even more common than tools of prevention are tools of detection—technologies that function not by making rule-violating behavior more difficult to execute, but by creating a comprehensive account of our behaviors. These are surveillance technologies. For example, body-worn cameras don’t make it impossible for a police officer to use unauthorized force against a civilian, but are intended to make the officer more accountable should they do so (though they accomplish this goal with questionable effectiveness). These technologies may work by deterring sanctioned behaviors—knowing that one is being observed can incentivize rule-following—or because they enable enforcers to more swiftly detect and punish rule-breaking.
Perhaps nowhere do we see this trend more clearly than in the workplace, where surveillance over workers’ behaviors has become a favored method for compelling compliance with the aims of management. This practice has deep roots—but contemporary workplace surveillance has some new features, too.
We often anticipate the “future of work” in either dreamy or dystopian terms. The phrase has been widely adopted by technologists and commentators, either to describe a paradisiacal ideal—in which people have much greater autonomy and flexibility to do work in ways that suit them while affording them ample time for leisure—or to describe a dark alternative: a future in which workers have ever-diminishing social and economic power and in which their every move and thought is overseen, predicted, and optimized by management, human or algorithmic. Both visions, though, are united by the assumption that the future of work (whatever it looks like) will happen, well, in the future.
In other domains, the way we talk about technology tends to be more focused on what is occurring now or in the very near term, but when it comes to work, we maintain some temporal distance, at least in our discourse, from these changes. In reality the “future of work” is not some distant or discrete mode of social organization so unlike the one we have today; the management practices of tomorrow are, in many ways, not particularly different from the management practices of the past. They’re built on the same foundations—motivating efficiency, minimizing loss, optimizing processes, improving productivity. And one of the most common strategies for achieving these goals, then and now, is increased oversight over the activities of workers.
So what is new about today’s workplace surveillance? Is this not just more of the same, driven by the same organizational goals that have always motivated managerial oversight—even if the specific technologies that are used to do so have changed form in one way or another? Some workplace monitoring is old wine in a new bottle, a contemporary instantiation of the manager with a clipboard looming above the factory floor. This is not to say that these practices don’t deserve scrutiny or critique—but we should be precise about what, if anything, is new here.
In fact, there are at least four subtle but important dynamics that distinguish contemporary workplace surveillance from what’s come before, and that are especially visible in the story of trucking.
First, contemporary technologies facilitate surveillance in new kinds of workplaces. Geographically distributed and mobile workers, for example, have historically maintained more independence from oversight than workers centralized in non-mobile workplaces, like factories, call centers, and office buildings—but location tracking, sensor technology, and wireless networking have changed that. Porous boundaries between home and work also facilitate surveillance in new places. For example, the growth of work-from-home arrangements during the COVID-19 pandemic has led to greater use of tracking software to monitor workers’ keystrokes, locations, and web traffic—as well as video capture of the kitchen tables and living rooms in which their work now takes place.
Second, new kinds of data also come to the fore. As sensor technologies become cheaper and easier to deploy, and workplace surveillance capability is more frequently embedded in software by default, employers are well positioned to capture more and more fine-grained data about workers’ movements and activities. Wearable technologies, like those used in Amazon’s warehouses, monitor and evaluate workers’ speed with much more precision than was previously possible—including the number and length of their bathroom breaks. Employers increasingly monitor and analyze datapoints like workers’ social media posts, phone calls, and attendance at meetings; Microsoft faced pushback in 2020 when it built “productivity scoring” into its widely used Office 365 product, which gave managers access to “73 pieces of granular data about worker behavior” like email and chat frequency. And as we’ll discuss, biometric data is also becoming more commonly collected in the workplace—from authentication mechanisms like fingerprint and retinal scans to behavioral data about workers’ attention and fatigue.
Third, these new data streams fuel new kinds of analysis that impact how workers are managed. In some contexts, managerial decisions are implemented through opaque algorithmic systems that can create acute information asymmetries between workers and firms—like Uber’s use of algorithms to apportion rides and determine rates without making those rules transparent to its drivers. Other analyses are predictive, designed to forecast which workers are likely to be most productive, how many workers to staff at a given time to meet demand, or which worker is likely to make a sale to a particular customer.
Finally, contemporary workplace surveillance can blur boundaries between the workplace and other spheres of life, creating new kinds of entanglements across previously disparate domains. Surveillance of work-from-home environments can facilitate data collection about family, friends, and living situations. Managers often keep tabs on workers’ online activities on social media platforms. Workplace wellness programs can facilitate employers’ collection of data about worker health, stoking concerns about discrimination. And “bring-your-own-device” policies, in which an employer’s software is installed on a worker’s own personal phone or computer, can further muddle distinctions between home and work and create additional data privacy and security concerns.
All of these dynamics are at play in trucking. For decades, the mobile, isolated nature of truckers’ labor provided a buffer against managerial oversight, giving truckers much more freedom and autonomy in their day-to-day work than other blue-collar workers. But this has changed drastically, owing to the proliferation of digital monitoring technologies like the ELD: the road no longer affords drivers the independence they once had. Indeed, surveillance in trucking involves the collection of new kinds of fine-grained data about truckers’ behaviors, bodies, braking patterns, even brainwaves. This data collection supports new forms of analysis: firms can compare truckers’ performance against one another and predict what they are likely to do in the future—giving them much more visibility into, and control over, truckers’ work than ever before. And trucker surveillance involves deep entanglements among the interests of many different actors in different social spheres—not only the trucker, the government, and the firm, but also truckers’ families, insurers, third-party companies seeking to make money from the data, and the public at large. As such, truckers may be canaries in the coal mine: investigating digital surveillance and rule enforcement in this industry can give us important clues about how these dynamics may function in other contexts, both within and outside the workplace.
Trucking is a job, but it is more than that. Trucks are understood by their drivers both as workplaces, relatively free of meddlesome bureaucratic oversight, and as homes, in which they live, eat, and sleep for days or even weeks at a time, and in which their privacy is sacrosanct. And similarly, understanding trucking as the mere activity of driving a truck is only one facet of what trucking means to those who call themselves truckers. Trucking work is bound up with cultural constructs of manhood and virility, performed through displays of physical and mental stamina. The industry is about 94 percent male. The “asphalt cowboy” has been an iconic figure in trucking for decades; workers who strain against authority in traditional employment settings may self-select into trucking as an occupation in which the day-to-day routines of work have, traditionally, been largely self-directed.
In short, trucking is an identity: an enactment of masculinity, a form of economic provision, and an extension of sexuality. Sociologists have observed that for working-class men, “bodily capacities are their economic asset”—the ability to push their bodies up to and past the limit is how they maintain economic autonomy. The ELD reduces this autonomy and impugns the self-knowledge on which it relies. To a far greater extent than in many other workplaces in which digital monitoring has proliferated, in trucking such monitoring clashes acutely with truckers’ collective and individual self-definition and occupational identities, built up over decades.
The ELD, then, operates simultaneously as a legal, economic, and cultural object. To regulators, it’s a legal creation—the product of federal regulations designed to enforce compliance with rules, and a strategy to address safety problems that plague the industry (though with mixed evidence of effectiveness). To firms, it’s primarily an economic tool to align workers’ behaviors with organizational aims such as maximizing fuel efficiency and minimizing out-of-route driving. (Firms, meanwhile, are major stakeholders in the formulation of the legal rules, since they too have interests in how much their drivers—and their competitors’ drivers—may legally drive.) And to truckers, the ELD is a cultural object that challenges the value of their “road knowledge” and occupational identity, as well as long-held industry norms for getting work done independently.
This complex interplay of legal rules, socioeconomic organization, cultural norms, and technical capabilities reveals the interaction of multiple domains in shaping surveillance technologies. Only by examining these dimensions together can we understand how and why digital enforcement works—or doesn’t.
In Against Security (2012), his critique of the post-9/11 security apparatus, sociologist Harvey Molotch draws a useful distinction between two different views of how the social world works: its “apparent order” and its “actual order.” All too often, Molotch argues, we impose top-down solutions to social problems based on how we think the world functions (or at least, how we think it should function)—while disregarding, deliberately or out of ignorance, how the world actually functions. In other words, we impose apparent order to the detriment of actual order. As Molotch explains, it is
best to replicate or at least build on people’s complexly tacit and mutual means of doing the world rather than trying to invent a world that does not exist. Too often the temptation is to reform through rules and official procedures that can sidestep those usual ways of doing things. But mundane life . . . should be interfered with only on pain of screwing things up in a big way. To create apparent order, you kill the actual order. . . . you efface the tacit mechanisms and social work-arounds that people use to get things done.
Molotch’s concern manifests in spades when we use digital systems to enforce rules. In trucking, the imposition of a digital enforcement regime wholly fails to account for numerous elements of the actual order of the industry—the political, logistical, and economic realities with which the industry and its workers must contend. These include: the fact that truckers’ economic well-being has long depended on widespread noncompliance with federal regulations; the logistical realities of unpaid detention time; the dearth of safe available truck parking and the lack of political will to build more; the unpredictable contingencies of the public highway, from weather to accidents to traffic congestion; the work pressures, sometimes coercive, from employers; cultural norms around machismo and stamina, which manifest themselves both in economic provision for family and in resistance to authority; and the occupational pride and deep-seated identity founded in a long tradition of workplace autonomy, just to name a few.
Of course, the “actual order” in today’s trucking industry is also far from economically equitable, socially just, or without need of serious reform. It exploits workers and risks public safety, and it’s no wonder there is so much driver turnover in the industry. An incredible amount of economic activity rests precariously on a fundamentally unsustainable system in which workers are uncompensated for far too much of their dangerous and difficult labor.
As a result, we shouldn’t conclude there is no possible place whatsoever for technologies like ELDs in trucking. If integrated alongside other meaningful economic reforms, such technologies might well be one component of a healthy, reorganized trucking industry. But by using digital surveillance to enforce rules, we focus our attention on an apparent order that allows us to ignore the real problems in the industry, as well as their deeper economic, social, and political causes. According to the apparent order envisioned by ELD advocates, the fundamental problem in trucking is that truckers cannot be trusted to reliably report how much they work, and the solution to that problem is to make it more difficult for them to fudge the numbers. But under the actual order, the problem in trucking is that drivers are incentivized to work themselves well beyond healthy limits—sometimes to death.
The ELD doesn’t solve this problem, or even attempt to do so. It doesn’t change the fundamentals of the industry—its pay structure, its uncompensated time, its danger, its lack of worker protections. At best, it prevents some of the very worst-of-the-worst carriers from pushing drivers too far, but it remains both possible (and sometimes encouraged!) for drivers to exploit the limitations of electronic monitoring. In short, it’s a failure to change the actual order that leads drivers, under the gun of companies’ profit motives, to compensate for the ELD by speeding to reach their destinations on time, making them (and all of us) less safe on the highway. And it’s a failure to change the actual order that leads technologists and trucking firms to embrace the promise of autonomous trucks—even though these technologies are nowhere near safe enough to deploy on their own, and end up leading to halfway solutions that imperil safety and intrude on drivers’ bodies.
There are solutions to these problems, but they’re not technological solutions. On their own, technology fixes are, at best, Band-Aids—superficial ways of covering up a problem that don’t address root causes. More meaningful reform in trucking would require a ground-up rethinking of how the industry is structured economically—most importantly, fundamental changes to the legal framework that dictates how workers are paid and the safety of their working conditions. So long as trucking is treated as a job that churns through workers, these problems won’t be solved.
Editors’ Note: This essay is adapted from the author’s new book Data Driven: Truckers, Technology, and the New Workplace Surveillance, published by Princeton University Press. Copyright © 2023 by Karen Levy. Reprinted by permission.