In 2009, Monthly Review published an open letter from three labor activists saying that their union — UNITE HERE — had developed “a cynical and manipulative system of control” designed to create “a cult-like relationship of dependency between staff and their supervisors.” This system, known as pink sheeting, first came to public attention during a prolonged and ugly divorce between the union’s two halves, UNITE and HERE. While the Monthly Review authors called the UNITE leadership’s decision to publicize the existence of pink sheeting a “sensational” and “not sincere” attempt to discredit HERE, they also said the practice itself was seriously troubling.

Under the pink sheet regime, a lead UNITE HERE organizer would persuade workers and lower-level organizers to divulge personal information, often about past traumas. The lead organizer would frame this as a private disclosure, a gesture of trust and connection. But if the lower-level organizer ever second-guessed a command from on high, the lead organizer would use that personal information to pressure the subordinate into carrying out orders.

Hotel worker turned UNITE HERE organizer Julia Rivera told the New York Times that her supervisors had pushed her to reveal details of how she was sexually abused as a child. Then “her supervisors ordered her to recount her tale of abuse again and again to workers they were trying to unionize at Tampa International Airport, convinced that Ms. Rivera’s story would move them, making them more likely to join the union.”

The authors of the Monthly Review letter wrote: “This command and control-style union culture is endemic to much of the movement.” And indeed, while pink sheeting may have been limited to UNITE HERE, the mentality that spawned it is not. Take, for example, the actions of SEIU earlier that year against its own members: In a May 2009 Labor Notes essay, Union of Union Representatives (UUR) president Malcolm Harris announced that his organization had filed unfair labor practice charges against SEIU for trying to bust its own staff’s union.

The “reorganization” within SEIU that led to the termination of seventy-five UUR members, wrote Harris, “reveals SEIU’s cynical view of organizers and organizing. While it lays off experienced and dedicated International organizers, it is outsourcing our work to non-union contractors and hiring dozens of temporary staff for health care reform efforts.”

These are the stories that are known, the ones that have attracted broader attention. But in many other cases, labor unions have managed to keep shoddy treatment of their own workers quiet. Organizers and other employees, perhaps fearing that they will be blacklisted from the movement or inadvertently help to discredit it, only talk among themselves about the wrongful terminations, low pay, tyrannical bosses, and outright union-busting they face.

The stories I’ve heard are shocking: people being ordered to “organize” around ancillary political issues that have nothing to do with their real job descriptions, an acquaintance who found her position summarily liquidated and herself blacklisted due to an employer’s personal vendetta, countless examples of union-busting.

The problem here is not any particular malice on the part of union management. The problem is that they feel the same pressures as any large employer: the constant need to maximize efficiency and workforce flexibility while minimizing labor costs. Whereas a corporation feels that pressure most acutely from competitors and shareholders, unions experience it in the form of political reaction and corporate opposition to unionization. Despite these differences, the outcomes are remarkably similar: managers in both sectors constantly look for new ways to extract productivity from their labor forces in exchange for stagnant — or diminishing — pay.

But while labor leaders and corporate bosses face analogous pressures, their situations are not identical. Union bureaucracy is exceptional because the ideology that helps to sustain its sometimes exploitative labor practices is also exceptional. Pro-labor sentiment demands that union employees tolerate their own exploitation as a necessary condition of working to free others from exploitation.

Granted, some other employers use related modes of ideological conditioning to acclimate their workers to shoddy labor conditions. For example, Capitol Hill staffers accept remarkably low pay and long hours in exchange for the prestige of traversing the halls of power. A young creative professional — a designer, say — might tolerate low pay and no benefits because he loves the casual-cool ethos and high social capital of his workplace and industry.

We might be tempted to say these examples are basically identical to the case of the union staffer, but we can’t do that if we’re to take the ideology of unionism seriously. If promoting workplace democracy is a fundamentally noble goal, then those who make sacrifices in order to achieve that goal are not suckers. We should admire, not pity, the idealistic college graduate who gives up a shot at a high-paying private-sector career in order to become an organizer. We should also assume that she is not stupid, that she’s aware of the great personal and financial hardships organizers endure. In that case, the line between exploitation and laudable sacrifice becomes a lot fuzzier.

To look at it from the employer side: When a private employer insists that you must accept precarity as the price of a “fun” job and the ability to wear a hoodie to work, we can dismiss this as obviously ridiculous. Not only is the inverse relationship between the employer’s ability to provide security and the “fun-ness” of the work environment rather dubious, but meeting the basic subsistence needs of workers is a greater moral imperative than providing free pizza on Fridays. But labor unions don’t offer their staff an ideology of fun and free pizza; they claim, with some moral force, that their staff is involved in an emancipatory project of monumental importance. Furthermore, unlike most other employers, they can plausibly claim that any effort to give, say, regular hours to their employees would undermine their ability to satisfy their other moral imperative.

Nonetheless, we should be able to draw up some general principles that help to distinguish exploitation from necessary sacrifice. After all, employees in plenty of other professions, like nursing, find ways to balance worker solidarity with their profession’s unique ethical obligations. It is one thing to say that a worker has professional obligations; it’s quite another to say that he must abandon all of his concerns about workplace conditions and bow to the limitless authority of the boss.

So while some level of personal sacrifice on the part of union organizers is not only inevitable but reasonable, that can’t possibly justify rendering them powerless over their own workplace conditions. To argue otherwise would be to argue against unionism itself. Even the labor movement’s right-wing enemies acknowledge that it would be radical to call for the abolition of teachers unions on the sole basis that teachers serve a higher calling. Any attempt to distinguish union organizing from those professions that deserve collective decision-making is bound to be incoherent.

The only lucid arguments for tolerating union staff abuse can be extended to the workforce as a whole. The logic of organizer exploitation is the logic of worker exploitation in general, and therefore corrosive to the very idea of unionism.

Consider the “shit happens” defense, a less ideologically charged cousin of the appeal to shared sacrifice. This line of reasoning excuses abusive practices in a union on the grounds that all organizers and other staff have had similar experiences. In other words, the person making the complaint should not think she’s special. Low or no pay, long hours, sadistic bosses, and so forth just come with the territory, and everyone should learn to accept that.

It shouldn’t be too hard to guess how unions respond to that argument when it comes from bosses in workplaces they seek to organize. Yet I’ve heard senior union staff readily apply it to their own work environments, seemingly without any pangs of cognitive dissonance. In one case, when told that an intern at a labor organization had frequently received his weekly stipend several weeks late, one of his superiors just shrugged, said “shit happens,” and pointed out that plenty of interns work for no pay at all.

Of course, this particular intern had been promised a modest weekly stipend, and had taken the job expecting to receive it. Imagine how this same labor organization would likely react if it were organizing workers whose boss routinely violated prior agreements and justified these violations by pointing out how much worse labor conditions elsewhere happened to be. The special exception that this intern’s supervisor made for his own staff may not be malicious, but it is both damaging and incoherent.

A slightly more plausible argument could be made that granting union staff any collective representation would lead to unsustainable internal politics. After all, there would be nothing to prevent the new staff union from making unreasonable demands, raiding members’ dues to grant themselves lavish salaries, overriding strategic concerns out of pure self-interest, and so on. Union leadership might argue that giving the staff control over the bureaucratic wing of the union would result in corruption, stagnation, or self-destruction.

Such a claim is superficially reasonable, but again, does not apply only to organized labor. In recent years, the American right has marshalled similar arguments against teachers unions. In Wisconsin, for example, Governor Scott Walker and his allies insisted that they would not be able to balance future state budgets as long as the state’s public school teachers were able to collectively bargain for their wages and benefits. The subtext of the argument was that selfish, intransigent teachers would demand obscene pay at the expense of everyone else who lived in the state.

To unionists, that was unthinkable. For one thing, the fight to save collective bargaining rights took place amid a budget debate during which the teachers had already made significant concessions.

The Right’s depiction of teachers as psychotically rapacious was just sort of bizarre. Teachers are the people who we trust to safeguard our children’s physical safety and mental development; if, in reality, most teachers are interested merely in fattening their wallets at the expense of the school district, the taxpayer, and their own students, then the fact that they have the same workplace rights as everyone else is not the problem. No one who fits that psychological profile should be allowed anywhere near a school, and if we had any reason to believe that it was an accurate description of most public school teachers, then enrolling America’s children in school could be reasonably considered a reckless act of child endangerment.

Thankfully, we can dismiss that scenario as absurd. But if we feel comfortable trusting the majority of teachers with the wellbeing of our children for a large chunk of the day, then surely we can grant them the same discretion when they bargain for contracts. If anything, their ability to do harm is much more limited in contract negotiations than during the school day.

The same logic applies to organizers. It is difficult to imagine the person who becomes an organizer just for the money. Unions so often get away with mistreating organizers in large part because of the sincere passion and conviction that brought those organizers into the profession. Just as it is impossible to imagine an entire union of teachers dedicated to bleeding the taxpayers dry at all costs, it is equally hard to imagine an entire staff of organizers whose only aim is to extort money from their union. If they do not care deeply about furthering the cause of unionism, they’re unlikely to be organizers; and if they do care deeply, then they’re unlikely to make collective decisions that they know will financially ruin the union.

Of course, the idea that union staff would have any policy-setting abilities is enough to make some people uncomfortable. This leads us to the most persuasive argument against giving collective bargaining rights to organizers and other staff: the union is supposed to be an agent of the membership’s will, and no one else’s. True unionists tend to (rightfully) view staff-run unions as oligarchies, disconnected from the concerns of their members and little better than other hierarchical private entities.

In the event that union staff formed their own bargaining unit, they would be forming an institutional counterweight to the uninhibited control of their members — or at least to their members’ elected representatives. Therefore, any union with a staff union is at greater risk of becoming a staff-run union. Some of the risk could be contained by clearly defining the extent of the staff’s bargaining power — for example, by banning them from bargaining over anything but wages, hours, benefits and grievance procedures — but such regulations never entirely work. They often turn previously explicit demands into implicit demands. (This was the case in the recent Chicago teachers’ strike, which was nominally about wages and teacher evaluations but fundamentally had quite a lot to do with charter schools and resources for students.)

Of course, objections of a similar kind are often made to the existence of public-sector unions. Public sector employees, the argument goes, are meant to carry out the wishes of the democratic polis (or at least, again, of its elected representatives). They are not supposed to shape the policy of their own departments by going against the will or interests of the people. The key difference here is that public-sector employees, unlike union staff, are expected to have some political input anyway; they are both employees and citizens (and therefore, in a distant sense, employers), whereas union staff are not themselves members of the unions they serve.

Another important distinction lies not between civil servants and union organizers, but between ideal theory and gritty reality. Yes, in an ideal democracy, the people would be the absolute sovereign. And yes, in an ideal union, so would the members. But there is no such thing as a neutral institution, and the institutions created by both governments and unions to serve their own interests will inevitably have agendas of their own. No one truly believes that government departments are capable of perfectly divining majoritarian consensus and carrying it out without the slightest deviation. Few people would even desire that state of affairs, given that government employees often know more about what they’re doing than the average voter. We acknowledge that, in the real world, we need to grant government employees some level of discretion.

Similarly, though organizers should strive to represent their members as best as they can, we must acknowledge that union staff will sometimes pursue their own interests, or imperfectly carry out the wishes of the membership. We must also acknowledge that strategic researchers or labor attorneys may know more about their jobs than the unions that employ them. The only way to entirely eliminate the tension between staff and membership would be to have unions without staff, where all of the essential functions are carried out by volunteer members. Even then, however, different volunteer members would have different beliefs, levels of knowledge, and interests — much like citizens who are also public employees. Besides, modern unions need staff. Private employers have developed exceedingly sophisticated union-busting tactics over the past century, and beating them often requires a dedicated team of communicators, researchers, lawyers, organizers, and statisticians, among others.

We could no more do away with professional union staff than with modern government bureaucracy. As a result, some level of tension between contradictory interests is inevitable. But no serious unionist would suggest that this tension is reason enough to do away with public-sector unions. We can certainly conceive of a situation in which the collective will of a public-sector union goes against the collective will of the public at large, but employees of the public are still employees, and it is a fundamental tenet of unionism that a workplace is unjust when its workers have no recourse to democratic decision-making. A state-owned workplace is a workplace like any other.

If we accept that reasoning, then it is difficult to imagine the grounds on which we might argue that a union’s bureaucracy is not a workplace like any other. If the tension between workplace democracy and the will of the stakeholders is tolerable in a government, it should surely be tolerable in a union. In fact, a union is easier to consider in simple employer-employee terms than a state. Whereas public employees are also citizens who elect their own employers, the power relations between elected union leaders and union staff are far more easily conceptualized using the traditional boss-worker bifurcation.

It seems that labor leaders cannot escape from unionism’s categorical imperative: Any abuse, exploitation, or union-busting is a fundamental betrayal of the union ideal. A union president who tries to prevent his staff from organizing might acknowledge this truth but counter that it’s not his problem. In the real world, he might say, the leadership and senior staff are here to serve the interests of the members. Sometimes those interests will directly contradict the lofty principles of unionism.

That’s all well and good, but it doesn’t sound much like the attitude of someone who considers himself part of a movement. In a time when social unionism seems like the last best chance for organized labor, that kind of atomized thinking is retrograde at best, and suicidal at worst. If workplace democracy is to survive and grow over the next few decades, it requires the cooperation of a diverse coalition in its favor. Firm principles unify such coalitions. Unions abandon them at their peril.