Benedict Clouette

In 2013, you wrote an essay entitled “On the Phenomenon of Bullshit Jobs: A Work Rant,” which was widely circulated on social media when it was published. It seemed to give credence to an intuition that many people had—that their jobs were, to use your term, “bullshit jobs.” You recently adapted the essay into a book, Bullshit Jobs: A Theory (2018). What is a bullshit job?

David Graeber

A bullshit job is one where even the person doing the job can’t justify its existence, where that person is fairly confident that, were their job to disappear, it would make no difference at all—or might even make the world slightly better. Really, it’s not huge news that a lot of people think that their work is pointless. All I’m suggesting is that we take these intuitions seriously.

BC

After the essay was published, many people wrote to you to confide their own experiences. What were the stories from those exchanges that were most surprising to you, or that were the most consequential in shaping the theory that you later developed in the book?

DG

One thing that surprised me was how often people were willing to admit that their jobs didn’t need to exist or shouldn’t exist—even people in well-paid positions. They would say things like, “It’s true. I’m a corporate lawyer, and actually, I contribute nothing to society. I’m miserable all the time.” You might imagine that corporate lawyers drink the Kool-Aid and think that they’re doing something useful. But no; not only are they aware of the uselessness of their jobs, they’re completely unhappy.

Another thing that struck me was how many people who sit in government offices and do terrible things—like clerks whose job it is to expel homeless people from a shelter if they lack two forms of ID—people who we might ask ourselves, “how do they live with themselves,” actually can’t. Many are keenly aware that what they do is evil. They really wish they could do something about it, or could do something else for work, and they feel trapped.

BC

In the book, you describe a blind spot of early scientific management: while it focused on the efficiency of industrial workers, it never questioned the productivity of the managerial class. The architec­tural spaces associated with Taylorism emphasize the supervision of workers, allowing managers to directly observe the assembly line or typing pool from a central location. But offices today are often organized differently. How do you think that the transformation of spaces of work—from the early 20th-century Taylorist model, to the open floor plan of the postwar cor­poration, to the more informal, flexible, and collaborative work environments of today—might relate to the emergence of bullshit jobs?

DG

One way in which the organization of office space has paralleled industrial organization is the use of surveillance. In my exchanges with people about their jobs, sometimes they would say, “You’re really lucky when you get a place where your computer is facing the wall and there is no camera on you, and nobody can see what you’re doing. . . . If you only have an hour of work a day, you can make cat memes or play computer games, and it’s okay.” But most people are clustered in offices that were designed so that everybody is staring at each other and therefore must keep pretending to work all the time. Surveillance and the organization of space are supposed to ensure that people are busy, but it often doesn’t matter what, in particular, they’re busy doing.

You might imagine that these workers have everything it takes to be happy: they’re getting paid to do nothing, so why don’t they learn Italian, write a play, or volunteer? But that’s quite rare. Typically, their time is broken up and organized in such a way as to make it impossible, or at least extremely difficult, for them to repurpose it. One of the ironies of the bullshit job is that the use of space to encourage work actually has little effect on productivity—in fact, I suspect it often has a negative effect on productivity—but nonetheless destroys time by making the experience of sitting around almost intolerable. This is one of the main components of the violence of a bullshit job.

BC

As an anthropologist, how would you describe the ethnographic scene of the bullshit workplace? Is there a texture, a way that it looks, or a sensibility of the space?

DG

It’s cubicle life, where in theory you’re alone but you’re actually never alone. On the one hand they leave you with nothing to do, but on the other hand they never leave you alone. You’re neither engaged with a group nor left to do your own thing. Spatial organizations like the cubicle are conducive to a particular kind of isolation: they’re set apart but still open enough that someone could be peeking in at any time. In a way you could say it’s the spatial equivalent of boredom. Boredom is not just having nothing to do; if you really had nothing to do, you could relax, let your mind wander; boredom is typically an industrial phenomenon, when you have to keep your attention focused even though there’s nothing interesting to focus on, like a production line worker who has to do the same thing over and over. In a similar way, here you’re both in a constant social relation but that social relation has no real social content.

BC

Even if it’s not your boss. I wonder whether the dark side of teamwork and collaboration, with its affected casualness about hierarchies, is that in such a workplace everyone is monitoring everyone.

DG

Right, or that they could be. It sometimes happens that the lines are so clearly drawn in a workplace that all the workers know it’s bullshit, talk to each other about it, and are in solidarity. However, it’s far more common that everybody asks themselves: “Does my supervisor know? Probably. Can I tell her? No. Or, I can, kind of, if it’s a really bad day—maybe I can say something, but there’s a tacit understanding that I’d better not go too far. How about my fellow workers? It’s pretty clear that they know, but that we shouldn’t discuss it. And then again, it’s possi­ble that they actually don’t know. Or that if they get mad at me for some reason they’ll pretend to have just discovered it.” Many degrees of sus­picion and doubt are introduced. Anybody could be an ally or an agent of the other side, and you never know who will turn on you when there’s an opportunity for a raise or when somebody actually wants your bullshit job.

BC

Why do you think it’s so difficult for companies to acknowledge the existence of bullshit jobs, when doing so might make the work more tolerable and workers less anxious?

DG

In the book, there’s a story of an Egyptian engineer who trained to solve advanced technical problems for a career in developing new technologies. After school, he gets a job at a public enterprise in Cairo, where he gradually realizes he’s really just there in case the air conditioners break. But, of course, his supervisors don’t say, “Look, you’re just here in case something goes wrong. In the meantime, just try to stay out of people’s way and don’t be more than 10 minutes away at any given time. Other than that, do what you like.” Letting him do what he liked when he wasn’t needed would have seemed immoral to them. Instead, he has to complete all these rituals of checks and forms, which fill up a full day. Eventually, he figures out which exercises he doesn’t need to do—which would go unnoticed if left undone—and he wins half his time back. But he finds that he can’t bring himself to repurpose that time for what he was trained for, so he ends up becoming an amateur film critic. Which is fine, but he had all this scientific training, so why can’t he use it? Why can’t he be allowed to use his idle time for his own projects? If he comes up with an invention, it might even help the company. It could be like Bell Labs. But oddly it seems a matter of principle in most enterprises not to let that happen.

BC

What you’re suggesting is similar to Google’s long-standing policy, now abandoned, of allowing engineers to spend 20 percent of their time working on their own projects, which other tech companies have tried to replicate. But those policies are never extended to other classes of workers. Do notions of class affect which companies allow which workers to direct their own time?

DG

The ways that companies treat people who are actually making, moving, maintaining, and fixing things and the ways that they treat people who are thinking about how to do those things are completely different. For most blue-collar jobs, there is increased Taylorization: if you’re delivering packages, for example, management is all over you, moni­tor­ing your efficiency and speed. Whereas, for someone in an office job at the same company, it’s the opposite—there are endless numbers of people doing nothing, and it seems to be okay.

BC

Many of the workers featured in the book have job descriptions that foreground cognitive skills: communication, coordination, information processing. A common perception is that the workplace for that type of labor is becoming more dispersed or diffuse, both spatially and temporally—more mobile and flexible. No one has a fixed desk; everything is in the cloud, on their laptop, and on their phone. Do you think this diffusion can be liberating for workers? Or, conversely, is work now more pervasive, leaving an imprint on all aspects of life? Is there a correlation between bullshit jobs and a workplace that is both everywhere and nowhere?

DG

In my research for the book, people sometimes asked: “Why can’t I do this from home? Why am I coming to the office and pretending to work for eight hours a day?” One reason is that if you were working from home, you’d be judged on what you were or were not doing; since no one is monitoring how many hours you’re spending sitting in front of the screen, on any given day, they can only measure output. It’s a paradox. You can have sanity and freedom, but maybe not enough to pay your bills, or you can make what’s effec­tively a ritual sacrifice of your time so as to prove yourself worthy of a middle-­class lifestyle.

The pervasion of jobs into life is a related phenomenon, but it’s a complex relation. If there’s a task that any employee would want to do for any reason other than the money, the first impulse of executives is to try to get someone else to do it for free, usually through online, work-at-home services. For example, companies can get non-native French speakers to translate French texts for free, or almost free, because these amateur translators get to practice their French. Work that has any pleasurable element, educational benefit, or other redeeming value gets outsourced, so that people do it at home, but they’re paid extremely little or nothing.

It’s the same for open-source software. Large companies use code that people develop on their own time because it’s fun. Often the code isn’t compatible with other software, so companies pay the same people, during the day, to do the boring work of making everything compatible. The basis of the work that invades daily life often isn’t even being paid.

BC

A current tendency in workplace design is to include spaces for leisure or rest—various derivatives of the Silicon Valley clichés like the Ping-Pong table, the daily happy hour, or the room with big pillows. The association of that type of office with the high valuations of tech companies has made it desirable, even for companies that aren’t based on “innovation” or “creativity,” as a model for new workplaces. Do you see the weird, coercive sociability of these environments as a consolation for the meaninglessness of the work?

DG

This cult of creativity is part of the culture of bullshit jobs. In my book The Utopia of Rules: On Technology, Stupidity, and the Secret Joys of Bureaucracy (2015), I discuss a series of managerial documents about creativity and innovation that might as well have been designed to strangle any possibility of creativity in the cradle. The rituals you mention are part of that phenomenon. In my research for Bullshit Jobs, I spoke with a consultant who gave seminars on creativity. Occasionally she would try to tailor her seminar to the particular business that hired her—to throw in some new ideas—but every time she tried, her company told her, effectively, that introducing any element of creativity into her creativity seminars was strictly forbidden.

BC

In The Utopia of Rules and Bullshit Jobs, you point out that, while the rhetoric of the free market always assumes that the inefficiency of bu­reau­cracies is a phenomenon of public services, it’s actually the opposite: with every move toward deregulation, many new layers of administration and paperwork are often produced. What do you make of companies like Uber or Airbnb, where the value of the company is premised on the elimination of management and the “disruption” of highly regulated industries? Do they follow the same logic as other forms of deregulation, or are they something new and different?

DG

The general bullshit jobs model is that you Taylorize the workers—in this case, the drivers—by breaking down their work into small, repetitive, closely evaluated tasks, increasing creativity to increase profits, while creating endless layers of unnecessary management in the offices, which soak a lot of those profits up. Uber eliminates the management level at the border between the workers and the executive management of the company. As I was writing the book, most of Uber’s executives were forced to resign due to a scandal, and it apparently made no difference for the day-to-day operations of the company. So, clearly, there are unnecessary jobs at Uber. Both of these companies traffic in services that challenge the boundaries between what might be called the traditional productive economy and the caring economy.

Ironically, the introduction of digital technologies into fields like nursing or teaching often has the effect of making them less, not more, productive. This is confirmed by Federal Reserve statistics, by the way: while productivity has been skyrocketing in, say, manufacturing, it’s been declining in health and education. This is due in part to the fact that digi­tization in those fields means that people have to spend so much time translating qualities into quantities, to translate what they’re doing into terms that a computer can understand. So teachers spend less time teaching and nurses spend less time nursing because they’re doing paperwork. So when it comes to care work, digitization clearly decreases productivity, putting a squeeze on profits, and therefore companies try to cut wages.

BC 

What does the rise of bull­shit jobs imply for the design of the workplace?

DG

Rather than spaces for forced sociability, there’s a need for spaces where people can run away and hide. Give them ways to defeat monitoring, so that they can be measured by their actual output rather than by what they appear to be doing at any given time.

BC 

You’re trained as an anthro­pologist. What do you think designers and architects can learn from the techniques of anthropologists?

DG 

Anthropology is based on a fairly simple, but in some sense radical, idea: Why not just listen to what people say and try to take them seri­ously? Everybody has at least one idea that you or I never would have thought of about how to improve things—how to do things in a more sensible fashion. The problem isn’t that people don’t have ideas, but that almost all of those people spend their whole lives being told to shut up. They’re very aware, but there are all these mechanisms that subtly silence them. Through my research, I gave them a space in which they could talk about this stuff, with no ramifica­tions in their actual places of work. You should assume that people know what is happening around them and that they have ideas, even if they look like they’re just sitting around.