Let’s just get this out of the way: All jobs are bullshit jobs. Even if you’re a public defender or work for Médecins Sans Frontières, insofar as your labor is determined by a system of abstract compulsion—insofar, that is, as it exists within capitalism—it’s bullshit. You know this.

In his new book, Bullshit Jobs: A Theory, David Graeber is interested in a particular variety of bullshit and work. In 2013, the anthropologist and anarchist (he hates to be called “the anarchist anthropologist”) published an essay slamming the proliferation of “pointless jobs” that seem to exist “just for the sake of keeping us all working.” The response was tremendous: It turns out that many people have jobs that they believe require them to do nothing of value (or to do nothing whatsoever while trying to appear to be doing something).

Graeber sifted through the responses and solicited additional input on Twitter in a quest to categorize the “five basic types of bullshit jobs” and document the absurdist travails of those who hold them. From such data, he constructed a working definition of the subject at hand:

[A] bullshit job is a form of paid employment that is so completely pointless, unnecessary, or pernicious that even the employee cannot justify its existence even though, as part of the conditions of employment, the employee feels obliged to pretend that this is not the case.

Graeber distinguishes these bullshit jobs from “shit jobs,” which serve a purpose but suck. Which is not to say that bullshit jobs don’t suck as well, but they suck precisely because they don’t serve a purpose. Much of the stress they produce—the “spiritual violence,” as Graeber terms it—results from the contortionist maneuvers that employees are forced to perform in order to pretend to be working when they have nothing to do. And as Graeber notes, this sense of purposelessness is widespread: To give just two examples, 37 percent of the UK respondents to a poll on the subject, and 40 percent of the Dutch ones, insisted that their work is utterly useless.

In 1930, John Maynard Keynes predicted that, by the end of the century, technology would have become so far advanced that developed economies would have a 15-hour workweek. So how did we get to our current state, almost two decades into the 21st century? It turns out that Keynes was only half right—technology has advanced spectacularly, but we are far from a 15-hour workweek. Keynes thought that the developed economies would adjust to a growth in productivity by decreasing workers’ hours. Instead, capital absorbed those gains but did not free up the now-superfluous human labor—a tendency that Karl Marx noticed long ago.

For Marx, this pattern is intrinsic to capital, whose constant expansion of its own value requires the reproduction of existing social relations. For Graeber, however, this pattern has less to do with capital’s prerogatives than with human agency; the problem “clearly isn’t economic: it’s moral and political,” he writes. Yet it would be truer to say that the problem is not merely economic, but also moral and political, and even truer to relate these spheres to one another, a point that Graeber himself makes later: “[E]very day it’s more difficult to tell the difference between what can be considered ‘economic’ and what is ‘political.’” But despite a muddled sense of causes and effects, Graeber’s book offers us an engaging—albeit at the same time tremendously disheartening—portrait of labor in 21st-century capitalism.

In his previous books, especially 2009’s Direct Action: An Ethnography and 2011’s Debt: The First 5,000 Years, Graeber’s ear for anecdote lent his activism the air of folktale. Debt’s opening vignette, for example, set at a garden party at Westminster Abbey, offers a charming little parable about our tacit beliefs and assumptions. At the party, Graeber suggests to an attorney he meets that the developing world’s debt should be abolished. “But,” she objects, “they’d borrowed the money! Surely one has to pay one’s debts.”

In Bullshit Jobs, Graeber similarly employs anecdote in order to illustrate just how much insanity we take for granted. Liberally drawing from the respondents to his original essay, he recounts stories that read like Philip K. Dick at his least plausible. Some are sad, others infuriating, and many are both. A number verge on the absurd: One woman’s job was to go around demanding IDs and proof of income from temporarily sheltered homeless people so that “the temporary homeless unit could claim back [the] housing benefit.” If homeless people couldn’t provide the necessary paperwork—as often happened—their caseworkers would kick them out. In another instance, a “subcontractor of a subcontractor of a subcontractor for the German military” describes driving for hours and filling out pages of paperwork simply to prevent a soldier from carrying his computer about 16 and a half feet down a hallway to his new office.

Most of the stories involve jobs that are also nightmarish in their unrelenting tedium. My favorite is the museum guard whose job was to protect an empty room, apparently to make sure no one started a fire in it. To ensure his vigilance, he was forbidden to read a book or even look at his phone.

All of these jobs sound terrible, but are they also bullshit? The people who have to do them think so. But Graeber’s reliance on subjective impressions of whether work produces value is the book’s major weakness. He brings up Marx’s distinction between productive and unproductive labor—between workers who produce surplus value and those who do not—simply to brush it off. And it is telling that he focuses on “information work” and what he calls “salaried paper pushers.” While he claims that these kinds of positions, rather than “waiters, barbers, salesclerks and the like,” account for “the bulk” of service jobs added to the economy since 1990, the US Bureau of Labor Statistics could have set him straight on this score. As Jason E. Smith pointed out in his review of Graeber’s book in The Brooklyn Rail, the bureau’s table of “occupations with the most job growth” actually does include waiters and retail salespeople, not to mention nurses, customer-service representatives, janitors, health aides, fast-food workers, cooks, and construction workers. Most service workers, in other words, are indeed providing valuable services—caring for others and feeding people.

Graeber’s picture of a Dickensian bureaucratization run amok has other problems as well. He correctly notes that our economic system has undergone profound transformations since the 1970s, with declining manufacturing and wages and a rising service and financial sector. But according to Graeber, these changes mean that the existing system isn’t exactly capitalism anymore, but rather a kind of “managerial feudalism”—one that involves “hierarchies,” “class loyalty,” and “moral envy”—that is the result of political will rather than structural determination. However, while one can surely find such attributes at work in the global economy, the system remains capitalist: predicated on the extraction of profit from the labor of others (even when such profit is mediated by financial markets). “Class loyalty” and “moral envy” are the products of such a system.

In his work, the Marxist theorist Moishe Postone (who died earlier this year) explored “the domination of people by time” under capitalism in ways that bolster some of Graeber’s claims. Postone’s discussion of the shift from the “variable” time of the Middle Ages, which was determined by the different kinds of human activity, to the clock time of the modern period, an invariable standard that dictates the workday, parallels Graeber’s own.

Yet there are important differences as well. In his discussion of value, Graeber (like some Marxists, it must be said) attributes to Marx a “labor theory of value” akin to David Ricardo’s, according to which the value of a given commodity is equal to the amount of labor that went into its production. But the point of Marx’s theory, as Postone makes clear, is precisely to refute this: Value for Marx is not a market mechanism focused on exchange relations, but a social mediation. It is that which compels workers to reproduce capitalism. For Marx, capitalism’s alienated social relations are not by-products of capital’s expansion of value—they are how capital gets valorized.

Also, capitalism—like feudalism—is a system of domination. But in comparison, feudal domination was overt and easy to comprehend—no peasant had to wonder what he was working for. This is why Marx’s theory can indeed help to explain the situation that Graeber rightly decries: because it is a theory about how social relations get reproduced, including those that seem irrational and unnecessary.

Despite Graeber’s focus on surface phenomena like hierarchy and envy, he is correct to conclude that the only thing keeping capitalism going is our refusal to stop it in its tracks through collective action. One of his respondents, whom he calls Lilian, captures the pathos of our continued submission to our own domination: “I get most of the meaning in my life from my job,” she writes. But the “meaning” of most jobs is meaninglessness itself.

Here we find Graeber exploring what is perhaps his true subject: not jobs that seem unnecessary, but the unnecessary compulsion of wage labor. In a free society—one in which your time and work are your own rather than commodities—Lilian’s sentiment would not necessarily be pathological. Work doesn’t need to be drudgery; we can find meaning in our jobs. But a society based on the production of value is by definition unfree, since we don’t really have a choice about whether to participate in it, and because work often becomes merely a tedious means of survival.

We have all experienced the truth of this. After college, I worked briefly as a temp doing data entry for a corporate law firm. I sat in a windowless room with a bunch of other temps, all of us squeezed together at a long table like students in a computer lab. We earned a little over the minimum wage. As often as I could, I would shirk my duties and surf the then-nascent Web. I had my spreadsheets minimized in a corner, ready to click should a paralegal come in to pick up something from the printer. But it was a fellow temp, who sat on my left, who objected to my wretched rebellion. “You’re not getting paid to surf the Web,” he informed me. I was just trying to reclaim a little of my time from those who were stealing it. And it wasn’t even a very effective protest, since I still had to sit in that depressing room and fill out enough spreadsheets to keep from getting fired. But my co-worker was simply expressing an assumption so commonplace that it hardly ever needs to be articulated: Your time does not belong to you.

Some of the first factories in London went bankrupt because laborers refused to work all day, every day. To the factory owners, this proved the workers were indolent loafers, so they reduced wages to the point that workers were forced to put in even more hours to survive. But this was really doing the workers a favor, the owners insisted, because otherwise they’d just get drunk and lie about. “Productive activity,” as André Gorz noted, began to be “cut off from its meaning, its motivations and its object and became simply a means of earning a wage.” Now we’ve all internalized this view of work.

Graeber doesn’t mention a project I recently learned about from Franco Berardi’s Futurability: The Age of Impotence and the Horizon of Possibility, but it represents the ne plus ultra of bullshit work. Berardi reprints an article—one of the most terrifying things I’ve ever read—that describes Candelia, a job-training center in France:

Sabine de Buyzer, working in the accounting department, leaned into her computer and scanned a row of numbers. Candelia was doing well. Its revenue that week was outpacing expenses, even counting taxes and salaries. “We have to be profitable,” Ms. de Buyzer said. “Everyone’s working all out to make sure we succeed.”

This was a sentiment any boss would like to hear, but in this case the entire business is fake. So are Candelia’s customers and suppliers, from the companies ordering the furniture to the trucking operators that make deliveries. Even the bank where Candelia gets its loans is not real.

The wages are imaginary, too. Nothing is produced in this “job” except the illusion of waged labor, but de Buyzer “welcomes the regular routine.” France has more than 100 of these “staged companies.”

This is the world we’ve inherited—one in which we reflexively inquire of strangers, “What do you do?” which means, of course, “How do you earn a living?” And this is so even when there’s no social need for everyone to be working all the time. Bullshit jobs are only one idiotic facet of this larger decoupling of work from meaningful activity. If the problem were managers and bureaucracy, then we would simply need to eliminate them. But if the problem is capitalism, then we need to change the world. The familiar slogan of Occupy Wall Street and the global justice movement of the early 2000s, both of which Graeber was involved in, was “Another world is possible.” We’re told this is idealistic and naive. But it’s not bullshit.