This article originally appeared in the Full Stop Quarterly’s special issue on graduate student organizing. You can support the quarterly by subscribing to our Patreon

Since the installation of a Trump-appointee majority on the NLRB, the organizing landscape for graduate workers has looked bleak. Though there have been notable successes even in the wake of Trump’s election—Harvard has recognized and begun to bargain with its graduate student union, while Tufts and Brandeis have reached deals on contracts that have afforded graduate workers meaningful gains—the majority of graduate student unionization campaigns have stalled. Administrations have united in their refusal to recognize the legitimacy of the elections that have taken place, even in the face of overwhelming majority support among graduate workers. While the 2016 Columbia decision grants graduate workers at private institutions the right to organize and bargain collectively, that right remains profoundly precarious: Functionally, any actual petition by any graduate student campaign is likely to face appeal up to the national Board and a reversal of the Columbia decision that would effectively end graduate worker organizing for a union at private universities for the foreseeable future, just as the Brown University decision did in 2004.

Though the NLRB process governs labor organizing in the private sector, and though graduate workers are, at least for now, considered employees under labor law, we face an unusual predicament: How can you make use of a right that you technically have, but which the administrative agency tasked with its protection promises to threaten? Graduate workers at both Brown University and Georgetown University, like workers at Cornell and NYU before them, have engineered a workaround by reaching negotiated agreements with their universities that provide for a mutually agreed-upon election outside of the NLRB and recognition process.

We are two such workers at Brown University whose involvement in the unionization effort on this campus spans several years. We have found that organizing in such an environment has its benefits and its drawbacks. We write in the hope that our campaign, which on November 19 won our election 576 to 394, can serve as one example of how labor campaigns can move forward when the institutions designed to protect workers rights have failed. We are thinking here of a variety of alternative paths (each organizing drive looks different!), but these paths share in common a recognition that the power of the union comes from its members, over and above the increasingly unreliable protection of the law. Because a union’s leverage always ultimately derives from its members’ capacity to take action collectively, campaigns that cut out the NLRB to bring the employer directly to the table can use many of the same tactics as garden-variety union organizing drives: public petitions, media pressure, slowdowns and strikes. However, there are some differences. Without the protection of the Board, drives require more imagination, more organization, a greater commitment to maintaining member mobilization, and often much more time to pressure the employer for recognition independent of the law’s recognition. What we advocate isn’t easy, but for many workers seeking to access the benefits of unionization, it may be the only option—and for graduate workers, at least for now, it already is.

Since the New Deal, the NLRA has governed the ability of most private sector workers to engage in union activity, access the benefits of collective bargaining, strike, settle contracts, and have formally- mediated relationships with management. Indeed, the NLRA itself during the Great Depression was passed in an attempt to manage labor strife at a time of profound economic and social disruption. Through the Act, the government established itself as the final arbiter of disputes between capital and workers, in an attempt to prevent these disputes from turning violent, or to their greater fear, harming the nation’s capacity for production and economic growth. Still, labor’s history in the US is one of civil disobedience. (There is a reason why the tactics of civil disobedience so resemble labor tactics.) Even following the Act’s passage many workers have taken collective actions to improve their working conditions without an expectation that the law would protect them and at times with reasonable fears that they may face criminal charges for their tactics, including mass pickets that bar an entrance.[1] At the same time, the Act has led to safer and more peaceful organizing conditions for many workers.

Yet the contradictions imbedded in the Act and its application have also long been manifest. Some of these contradictions center on the question of constitutional rights, particularly First Amendment rights and their interpretation. This has been especially true in cases when a set of rights might be perceived as conflicting with another set of rights, such as the right to work and the right to strike. Since the Act’s beginnings, determining whose rights it protects has been a political question and a political balancing act, rather than a genuine reflection upon which human powers, abilities, and possibilities ought to be constitutionally protected. This fact has led some scholars to question whether continuing primarily to appeal to the Board and to the courts to protect our freedoms as workers and union members by demanding that they enshrine these freedoms as constitutionally protected rights is really the best means of organizing for workplace representation.[2] It has led others to suggest that unions would be better protected without the NLRA.[3]

The effectiveness of the NLRA to protect labor has been linked to interpretations of First Amendment Law, which, as Laura Weinrib has documented, has become increasingly unfavorable to labor since the Reagan administration. Weinrib writes of this trend,

The past decades have witnessed a marked shift in First Amendment Law from the protection of disfavored minorities against state suppression to the insulation of industrial interests against government regulation. That effort has overlapped almost perfectly with the decline in labor’s power.[4]

It is not surprising that the effectiveness of the NLRA to actually protect labor has been dependent on changing judicial interpretation of rights-based claims. Consider that when the NLRA was first passed, businesses actually continued to ignore their legal obligation to bargain because they anticipated that the courts would find that the Act infringed upon their property rights.[5] This practice of employers ignoring the law, or at least waiting until the law suits them before following it, is certainly familiar to graduate workers at private universities. As mentioned, we have seen employers delay in negotiating because they hope the Trump NLRB will restore the pre-Columbia order, once again denying private university graduate students the right to unionize as employees. What we learn from taking into account the history of the NLRA and its application is that employers have long relied upon the political instability of workers’ freedoms as guaranteed constitutional rights.

Even putting aside the legal and philosophical problems inherent in the NLRA framework, serious practical questions remain about its continued usefulness as a path to union recognition across the contemporary workplace. The problems with the current environment around NLRB elections center on two important issues: first, the speediness—or lack thereof—with which the Board resolves election-related disputes, and second, the perception that the Board has become unduly politicized, acting, during Democratic presidential administrations, only in the interests of labor unions, and during Republican ones, only in the interests of employers.

A union election can be either stipulated or contested. In a stipulated election, workers seeking union representation voluntarily agree with the employer on the conduct of a representation election. In a contested election, the employer and the union cannot come to agreement on issues like whether the threshold of authorization card signers has been met by the union, or whether the employees seeking representation are eligible under the law. In this instance, the regional Board must first resolve any disputes existing between the parties before an election can be ordered. In general, most union representation elections are stipulated, but most graduate worker union elections have been contested, since the status of graduate workers as eligible employees covered under the NLRA has changed so often[6]. Frequently, the board can take weeks or even months to resolve these disputes and order an election, leaving employers valuable time to begin an anti-union campaign in the workplace. Jeffrey M. Hirsch, considering a 2016 NLRB rules change designed to speed up elections, makes the case that NLRB delays in ordering “vigorously contested” elections have been “indefensible”:

Over the last decade, the median time between the filing of the election petition and the actual vote has ranged from about 37–39 days. However, there are sharp differences between the time to conduct stipulated elections, which occur in a median of 36–39 days, and contested elections, which take 59–70 days. Much of the difference between stipulated and contested elections result from the Board’s need to resolve preelection disputes.[7]

In every instance, a lengthy lead-up to an election favors the interests of the employer, who can take advantage of the lag time to make the case against unionization to employees in the unit. Graduate unionization drives have faced fierce persuasion campaigns waged by university administrations who, though they have not typically followed the lead of other private sector employers in resorting to harassment, interrogation, threats, discipline, and dismissal of employees who support unionization[8], have taken the opportunity to oppose unionization on two fronts: campaigning to change the hearts and minds of union-friendly graduate students on the one hand while using their legal teams against graduate student worker eligibility on the other.

Universities appeal to the NLRB to gain relief from the obligation to organize elections, and to respect the results of such elections should the union win, because they are certain to encounter an administrative body that, even when staffed with labor-friendly members, has failed to adapt to the changing circumstances of work and employment, and even when acting with the best of intentions, has struggled to adapt an octogenarian labor law regime to the realities of the contemporary workplace.[9] One such workplace is the contemporary university, which has come to rely disproportionately on the work of adjunct, contracted, and other precariously-employed instructors to perform teaching and eliminated tenure-track employment, particularly in the humanities and social sciences. This is assuming a board staffed by members who value freely-chosen collective bargaining. Given the current composition of the Board, graduate student workers have no reason to believe that any petition they file would be decided in their favor: the best-case scenario, in these instances, is a delayed election that gives the university ample time to campaign against the union and preserves the prerogative of management to challenge the results after the fact, while the worst-case scenario gives the national Board the opportunity to revoke graduate student employees’ access to collective bargaining under the NLRA framework.

Since the right to organize and bargain collectively is guaranteed by an uncertain precedent and administered by an enforcement agency eager to overturn its own precedent, we might be forgiven for asking whether—and where—such a right exists at all. Unless, of course, the right to organize, the right to bargain collectively, and yes, the right to strike, comes from something other than a determination of eligibility under the NLRA.

Richard Locke, our Provost at Brown, has written of the NLRB,

Like it or not, the NLRB is the government agency that has governed private sector labor relations in the United States for more than 50 years. It is an important institution that has withstood many presidential transitions. While we can challenge an outcome of a decision using legal means, this is not an entity to disregard simply because we do not agree with the politics of the current occupant in the White House.[10]

What this claim in effect means is that Brown views graduate students as workers when the NLRB does but not when the NLRB does not, regardless of whether the NLRB is composed of members who would like for labor to die or not. But does the fact that an agency has lasted for decades by itself really mean that we need to continue turning to it and empowering it, especially if it has not served us well and promises to serve us even worse? What should we do if this entity that was designed to facilitate workers unionizing has become largely a tool to prevent us from doing so? As workers and organizers, one thing we can do is re-conceptualize what it means to successfully organize. When workers cannot count on the NLRB to protect them in a timely and fair fashion, they can still compel employers to respond to demands and to voluntarily bargain by running pressure campaigns, which, as Joe Crump notes, “[focus] on employers and [make] them pay for operating nonunion.”[11] At Brown, we employed some of these pressure tactics to come to a pre-election agreement that would allow for us to have our election outside of the NLRB. For example, in emails, Facebook posts, posters, protests, and letters, we called upon Brown to not align itself with the Trump administration by insisting upon an NLRB election. Brown benefits from its reputation as a liberal and inclusive university, and so these acts and communications of ours forced Brown to choose between its preference for remaining union-free and its desire to maintain a reputation that is beneficial for sustaining the university.

Of course, our successful negotiation of a stipulated election administered by the American Arbitration Association, not the NLRB, was in part the result of fortuitous circumstances. We negotiated with a University administration which, it appears, wished to avoid a long and polarizing fight over unionization. We organized in a political environment that saw academic communities, for better or worse, beginning to circle the wagons against perceived political threats such as the 2017 GOP tax bill, which included an unprecedented university endowment tax, as well as a possible tax on graduate tuition remission as income. But, perhaps more importantly, we were able to turn the legal limbo of graduate student workers’ right to organize to our advantage: In 2004, unionization efforts by graduate workers at Brown were vigorously opposed by an administration led by then-Provost Robert Zimmer (now President at the University of Chicago, which has set an industry standard for busting academic unions). The University took its opposition to the national Labor Board, leading to a decision which reclassified graduate workers as students and stymied graduate organizing for over a decade, until Columbia. This time around, Brown University administrators were not eager to reprise this precedent- setting role.

Graduate student employment resembles traditional employment relationships in important ways: Employed graduate students generally work for a single employer, over a fixed period of time, and are directly paid by that employer in exchange for research or teaching performed. In other ways, however, graduate student employment resembles less traditional work arrangements, in which hours, duties, and even the structure of supervision remains highly variable, and the terms of employment depend greatly on the needs of departments, programs, schools, and the university itself. In this way, graduate workers have much in common with millions of workers across the United States, who labor in increasingly casualized ways, and for whom the terms and conditions of employment are never quite clear. Such workers, like graduate students, have enjoyed only very inconsistent access to the rights and protections afforded by American labor law. Yet, increasingly, they are finding ways to fight for their rights at work, and even sometimes to win.

The legacy of such nonstandard (at least in the United States) approaches to unionism is long and even its recent history would be impossible to summarize in this article. Suffice it to say that organizing strategies have begun to show signs of catching up to the fractured landscape of work and employment. From high-profile, national movements like Fight for 15, which has won important gains across the country, to local struggles among rideshare drivers in Seattle and Los Angeles, as well as cab drivers in New York City, new coalitions of unionized and nonunion workers are seeking justice for employees who do not enjoy the sometimes dubious protections offered by the NLRA. Most recently, Somali workers in Minnesota, organizing without a union but with the help of a worker center, brought Amazon to the table to negotiate over discipline and working conditions[12]. Amazon had never before agreed to negotiate as a result of organized worker pressure. This unprecedented victory supports our continued belief that, even when the options seem limited, organizing works.

If the American labor movement is to survive the challenges posed by the Janus decision, on top of the already-outdated legal framework governing union activity, not to mention the challenges posed by the rise of the nationalist far right at home and abroad, it will have to find new and innovative ways to organize. This is not a new insight but an established truism within the labor movement. It is the reason that unions have poured so many resources into new organizing, fighting not merely to retain existing members but to make inroads into new industries, new communities, and new types of work.

The 2004 Brown University decision effectively spelled the death of graduate worker organizing at private universities on all but a handful of campuses. Only the prospect of a labor-friendly Board stocked with Obama appointees prompted successful campaigns at places like NYU and Columbia. We have no reason to believe that the Columbia precedent will survive the next several years (though we obviously hope that it does!). The graduate worker unionization movement, then, has two urgent and related tasks: the first is to stake a claim for labor rights that does not rely on the decisions of a highly politicized and increasingly illegitimate arm of the administrative state. This will require not only changing the tactics and the strategies we employ, but possibly adopting a more militant attitude towards the establishment and protection of collective bargaining, with or without the NLRB. We cannot ask union members to patiently wait out the results of legal battles to secure them unionization rights: increasingly, we have to ask them to take the fight directly to their employers, by making direct use of their collective power. The second is the immediate necessity to connect the experiences of graduate workers as workers to those of workers across the economy, and to forge new bonds of solidarity that can help the union movement grow and continue to do what it has done in the best moments of its history: fight for a better and fairer future for all of us who labor. We do not claim to have found the solution to either of these challenges. But our experience of organizing has helped to point us in the right direction, and we are excited to continue to find the way together.

[1] On the continued role of criminal law in deterring strikes and criminalizing workers, see Ahmed White, “The Crime of Staging an Effective Strike and the Enduring Role of Criminal Law in Modern Labor Relations” in The Journal of Labor and Society, Vol 11 (2008).
[2] See, for example, for a critique of unions’ growing “dependency on the state,” Ahmed White, The Last Great Strike: Little Steel, CIO, and the Struggle for Labor Rights in New Deal America, (Oakland: University of California Press, 2016).
[3] See, for example, Harry Bernstein, “Creativity Needed to Stem Unions’ Decline,” in LA Times, (1989), and Julius G. Getman, “The NLRB: What Went Wrong and Should We Try to Fix It,” in Emory Law Journal, 64, Special Issue (2015).
[4] Laura Weinrib, “The Right to Work and the Right to Strike” in University of Chicago Legal Forum 2017, no. 20 (2018), 519.
[5] See Weinrib, 524.
[6] Since 1970, the Board has ruled at least four times on the status of graduate student workers as employees covered by the NLRB, each time reversing its previous decision: in Leland Stanford Junior University (1974), which held that graduate students were primarily students, not workers; New York University (2000), which reversed the previous decision; Brown University (2004), which overturned the 2000 decision, and, most recently, Columbia University (2016), which overturned Brown University. Sheldon D. Pollack and Daniel V. Johns provide a useful history of early NLRB case law on graduate student workers in their article “Graduate Students, Unions, and Brown University,” The Labor Lawyer no. 241 (2004).
[7] Jeffrey M. Hirsch, “NLRB Elections: Ambush or Anticlimax?” in Emory Law Journal vol. 64, special issue, 2015.
[8] See Kate Bronfenbrenner, “No Holds Barred: The Intensification of Employer Opposition to Organizing” (Econ. Policy Inst., Briefing Paper No. 235, 2009)
[9] For a fuller accounting of the ongoing crisis of legitimacy confronting the NLRB, see Wilma B. Liebman, “Decline and Disenchantment: Reflections on the Aging of the National Labor Relations Board,” Berkeley Journal of Employment and Labor Law 569 (2007). The challenges outlined by Liebman in the article have only compounded in the eleven years that have passed since publication. Liebman chaired the NLRB from 2009-2011, repeatedly calling for updates to the country’s labor law framework to further encourage collective bargaining.
[10] Richard Locke, “We should respect federal ruling on graduate student unionization,” letter to the editor, in The Brown Daily Herald, 2018, http://www.
[11] Joe Crump, “The Pressure is On: Organizing without the NLRB” in Labor Research Review, 1, no. 18 (1991), 38.
[12] workers-minnesota.html

Dennis M. Hogan is a PhD candidate in Comparative Literature at Brown University. He has been organizing with SUGSE since 2015.

Hilary Rasch is a PhD student in English at Brown University. Most of her writing and teaching focus on the intersections of literature, activism, and healthcare in twentieth-century North America. She loves cats and hates prisons.

Become a Patron!

This post may contain affiliate links.