On June 20, 2011, the US Supreme Court put an end to the largest civil rights class action lawsuit in American history: Betty Dukes v. Wal-Mart Stores, Inc., a case that pitted over 1.5 million female Wal-Mart workers against the country’s largest private employer. Suing on behalf of all women who had worked at Wal-Mart between 1998 and 2011, the plaintiffs in Dukes accused the retail giant of discriminating on the basis of sex in pay and promotions, in violation of Title VII of the Civil Rights Act of 1964. Their case was historic — and slow moving. By the time it reached Washington, Dukes had been working its way through the federal courts for ten years. With it came ten years of accumulated resources: seven teams of legal counsel, over two dozen amicus briefs, and hundreds of pages of data backing the plaintiffs’ claims.
The women’s case was convincing, to say the least. Study after study revealed that in the thirteen-year window under investigation Wal-Mart paid its male employees more than its female employees in every one of the store’s forty-one retail regions. Worse, the wage gap at Wal-Mart actually increased every year after 1997, even though women who worked at the chain generally stuck around longer than men and received better performance ratings. Women made up roughly 72 percent of Wal-Mart’s hourly employees, yet only 34 percent of management. Those in management saw the gap between their salaries and those of their male colleagues widen over time, the plaintiffs’ expert statistician Richard Drogan noted in his report — even “for men and women hired into the same jobs at the same time.” Bringing these numbers to life, Dukes offered damning anecdotes from the plaintiffs: women who were told by their managers to “blow the cobwebs off their makeup and doll up” if they ever wanted to advance in the company; or reminded that “God made Adam first, so women will always be second to men”; or asked if they wouldn’t rather be at home, raising their kids, than seeking out promotions — all while their male colleagues’ fatter paychecks were justified by the assertion that men were “breadwinners.” Male favoritism at Wal-Mart could be subtle and insidious, but it was just as often overt and absurd: Sue Brown, a former schoolteacher with a master’s degree, had been working as a full-time sales associate in the electronics department of a Kentucky Wal-Mart for several years when she discovered that a teenage boy, whom she herself had trained, made more money than she did.
Women who challenged this trend often suffered reprisals. After joining Wal-Mart as a cashier in 1994, the first named plaintiff in the case, an African-American woman named Betty Dukes, was demoted and subjected to other disciplinary actions for identifying discrimination through the company’s “Open Door” hotline — a 1-800 number for internal complaints eventually known among Wal-Mart employees as “1-800-YOU’RE-FIRED.” Dukes filed a sexual discrimination claim in 2001; in 2004, on the fiftieth anniversary of Brown v. Board, a US district court in California granted Dukes and six other named plaintiffs certification to pursue their case as a class, representing all of Wal-Mart’s female employees. Seven years later, all evidence gathered by the plaintiffs’ lawyers made the Wal-Mart situation clear. As Ruth Bader Ginsburg confirmed in her dissenting opinion for the Court: “gender bias suffused Wal-Mart’s company culture.”
But evidence, however clear, isn’t always enough to carry a case. Dukes had a stake in both legal and social matters that went beyond the women’s claims and Wal-Mart’s employment practices. Columnists used Dukes to rhetorically lament the state of labor politics: Were class actions really the best means of collective bargaining? Where, after all, were the unions? The Court, meanwhile, faced more specific legal problems. The class’s enormous scale — a consequence, of course, of the enormity of Wal-Mart — raised philosophical and technical questions about class actions that promised to have broader political consequences. Two questions especially shaped their deliberations. First: could plaintiffs in a class action reasonably ask for both “injunctive relief” — the sort of court-ordered mandate that forces an institution like Wal-Mart to resolve a systemic issue like sex discrimination — and ask for back pay (a.k.a. monetary relief, a.k.a. damages)? And second: if certified classes could grow to represent groups numbering in the millions, was there such thing as a class that would fail, on size alone, to meet the class action requirement of “commonality”? Could a class, in other words, be too big to win?
When the time came to answer the legal issues Dukes raised, in June 2011, the Court ruled unanimously on the first procedural question at hand: the matter of classification vis-à-vis back pay. Class actions fall under Rule 23 of Civil Procedure, which specifies, among other things, what kinds of relief classes can seek. In general, classes seeking injunctive or declaratory relief file as a “b(2)” class — which the women in Dukes did — whereas classes seeking monetary relief file as a “b(3)” class — as most well-known consumer class actions do. As a civil rights case suing under Title VII, Dukes was squarely b(2). But the Court ruled that the women’s additional demand for back pay — which would amount to millions, perhaps even billions of dollars in withheld wages to women across the country — could only belong in a b(3) claim. This misclassification was enough to compromise the women’s class status, sending them back to square one.
The second question of “commonality,” however, was more contentious — and it’s curious, given that Dukes was more or less a dead horse after the b(2) error, that the Court continued to kick, making commonality central to their ruling. In a split 5-4 decision delivered by Antonin Scalia, the Court ruled that the class represented in Dukes also failed to meet Rule 23’s commonality requirement — that there be “questions of law or fact” common to the class. Bending the language of the original rule, Scalia argued that to claim “commonality” a class must not only share a common problem, but also a common solution to that problem — one that would compensate all members equally in a single stroke. Because the 1.5 million female Wal-Mart employees were not all denied the same promotion, the same pay raise, or insulted, belittled, or obstructed by the same manager in the same store, their cases could not legitimately be litigated all at once. “Without some glue holding the alleged reasons for all those decisions together,” Scalia said, “it will be impossible to say that examination of all the class members’ claims for relief will produce a common answer to the crucial question, Why was I disfavored.” In agreement were Alito, Roberts, Thomas, and Kennedy. Sonia Sotomayor, Elena Kagan, and Stephen Breyer joined Ginsburg in her dissent.
It is true that not all women employed at Wal-Mart since 1998 faced the same degree of discrimination. It’s also true that by including a request for damages the plaintiffs may have compromised their shot at injunctive and declaratory relief.1 But these were arguments that had little to do with the phrasing of the original rule, and Scalia’s revision of what constituted “commonality” raised more questions than it answered. What, after all, was “glue”? What was “some glue,” and what was glue enough? To those who cared, it seemed all too coincidental that a group of people who have historically been denied recognition as an oppressed, exploited class — women — were being denied that recognition yet again under our major civil rights law, this time on a technicality — articulated by something so vague as “glue.” Wasn’t the answer to the question, Why was I disfavored obvious enough? — Because I am a woman?
What the women in Dukes sought to put on trial was an entire culture of sexism: an unspoken, or at least unofficial ethos that influenced millions of decisions by independent actors across the company. What was interesting about their evidence was that it showed how despite the absence of a central policy of discrimination, sexist practices were operative everywhere at Wal-Mart, from Anchorage to Orlando. Too widespread to be dismissed as the errors of a few “bad apples” in a few “problem regions,” the behavior Dukes reported was the product of something much more expansive and amorphous — something that ultimately painted a picture of sex discrimination as it actually exists today. There are no central policies, no pronouncements in company literature that say, “No Women May Advance, nor Receive Equal Pay for Equal Work.” But that does not mean sexism does not exist, nor that women are given due compensation for their labor.
As it was, men at Wal-Mart didn’t need to conspire to keep women down; they did it perfectly well as individuals. Until Wal-Mart computerized its job application process in the mid-2000s, an employee who wanted a promotion first had to ask her manager how to go about getting one. If for whatever reason he (and usually he was a he) didn’t think she (and usually she was a she) deserved it, he would either avoid her, flat out deny her information about required training programs, or tell her some other step was required first — like relocating, holding her present job for at least a year, or being able to demonstrate some arbitrary “necessary” skill, like lifting fifty pounds of dog food. If a new position opened up and a store manager never posted a flyer advertising the job, a woman seeking a promotion might find that a male coworker — or an out-of-towner chummy with a guy higher up — had taken the job before she even knew it existed. If she did manage to land a higher-level job, she could expect to make anywhere from $2,400 to $139,000 less than her male counterparts per year. After a while, even Wal-Mart’s most faithful female employees — those who had hung around for more than a decade, thought unions were “not for this company,” and truly believed in Wal-Mart’s emphasis on being “a family” — would get wise.
Scalia, however, stuck to the letter of Wal-Mart literature in his opinion for the majority, treating corporate communications with as much dignity and reality as black-letter law. “Wal-Mart’s announced policy forbids sex discrimination,” he said — as if that was that. With nothing written down at the top, how could you ever determine whether the alleged discrimination was systemic?
Ginsburg was quick to point out in her dissent that “‘discretionary employment practices’ can give rise to Title VII claims, not only when such practices are motivated by discriminatory intent, but also when they produce discriminatory results.” Nevertheless, it was frightening how reasonable Scalia’s argument sounded — demonstrating how easily sexism can be reasoned away by well-meaning people. A stereotyping insult, a derogatory joke — these are things women have learned to accept in the workplace, and are encouraged to either take in stride or report as the fault of an individual perpetrator. As a result, too many women find themselves isolated in their experiences of prejudice, left to whittle offenses down in their minds to nothing worth complaining about — to accept each experience as “a fluke,” in the words of one Wal-Mart supervisor in Alabama, who told deli manager Gretchen Adams that there was nothing he could do about a sexist pay discrepancy.
For a moment, the Dukes class action had offered an alternative. Viewed in aggregate, the “frivolous” or “dismissible” offenses women suffered at Wal-Mart could finally be seen for what they really were: oppression. By offering working women redress through the vehicle of the class action, Dukes was not only calling Wal-Mart on its sexist, illegal behavior. It was also giving an unprecedented number of women license to recognize themselves as a class, 1.5 million strong — a collective body that could speak its members’ grievances more powerfully than any one of them could individually. It created, in other words, a sort of class solidarity. As Betty Dukes once told an interviewer, “I found myself standing alone. But I wasn’t standing alone.” By stripping the Dukes plaintiffs of their technical, legal commonality, the Court had undercut something much more important.
Afterward, the media presented several ways to view the ruling. It was a message from the judiciary that labor squabbles weren’t theirs to settle, or a gesture protecting the rights of corporations, or a tactical move to inhibit class actions of any sort. But the case could also be seen as part of a much broader history: that of American women’s struggle for group recognition in the eyes of the law. Dukes was another installment in a long battle over sex class — one we thought we’d won forty years ago.
In 1848, Elizabeth Cady Stanton held these truths to be self-evident: that men and women were created equal; that women, like men, were endowed with a trio of inalienable rights (to life, liberty, and the pursuit of happiness); and that if a government abridged or denied those rights, it was one’s duty to cast off that faulty government and help establish one that wouldn’t. At the time, women were, as Stanton observed, “civilly dead.” They could not vote. If married, they could not represent themselves in court. They could not serve on juries or hold office. Their property, including wages, belonged to their husbands when they married. They could not practice law, let alone attend any school that would teach them how. And so at the Seneca Falls Convention of 1848, the first major women’s rights convention in history, Stanton argued that the long train of abuses and usurpations against women had gone on long enough. Delusions of liberty were liberty’s enemy, as were those who presented themselves as liberty’s guardian. “Now do not think, gentlemen,” Stanton declared in an 1860 address to the New York State Legislature, “[that] we wish you to do a great many troublesome things for us, that you need spend a whole session in fixing up a code of laws to satisfy a class of most unreasonable women. We, as a class, are tired of one kind of protection, that which leaves us everything to do, to dare, and to suffer, and strips us of all means for its accomplishment. We would not tax man to take care of us.” Women didn’t need special treatment, Stanton argued; only full and equal citizenship.
The Equal Rights Amendment, introduced to Congress in 1923, would have fulfilled Stanton’s demand for women’s equality under the law. It stated simply that “equality of rights under the law shall not be denied or abridged by the United States or by any state on account of sex.” Had it passed, it would have forced lawmakers to strike from the books any laws that denied women rights on the basis of sex, providing a blanket solution to sex discrimination. Without the amendment — which was finally approved by Congress in 1972 but fell short of ratification in 1982 — women were left with the remedial, complicated, and confusing “code” for ladies that Stanton despised: the mosaic of tailored laws that recognized women as a class, but an inferior one — a class that needed to be compensated for its innate weakness and attendant disadvantages.
Though by no means the first of their kind, the “protective labor laws” that proliferated during the first two decades of the 20th century exemplified how the old ideals of sex-specific legislation were ushered into the new era of suffragism. These laws came on the heels of the late-19th-century Lochner era, a time when courts were declaring laws restricting night work, maximum hours, and minimum wage for both male and female workers unconstitutional under the Due Process Clause of the Fourteenth Amendment, which protected freedom of contract. These Due Process claims dominated in the courts until advocates for reform began to focus their efforts on women — hoping that sympathy for the ladies could create an “opening wedge” for legislation that protected all workers, male and female alike. Their moment came with Muller v. Oregon in 1908, when a state law limiting the number of hours per day women could work in mechanical facilities, factories, and laundries came before the Supreme Court.
Appealing to what a later Supreme Court would call the “romantic paternalism” of the era, the lawyers defending the restriction assembled a 113-page sociological brief — the first instance of the use of sociological evidence in a Supreme Court case (and, perhaps, a cautionary tale about the pitfalls of sociological jurisprudence) — citing “facts of common knowledge” from doctors, academics, and factory inspectors. According to the brief, overwork “is more disastrous to the health of women than of men, and entails upon them more lasting injury.” This “deterioration” was ostensibly “handed down to succeeding generations,” meaning that threats to women’s bodies were also threats to the future population. Given this “fact” — and conflating motherhood with womanhood — the Court deemed it within the police power of the state to regulate women’s working conditions, even though doing so violated women’s freedom of contract. “As healthy mothers are essential to vigorous offspring,” Justice Brewer wrote in the Muller opinion, “the physical well-being of woman becomes an object of public interest and care in order to preserve the strength and vigor of the race.”
The subsequent sex-specific laws that grew out of Muller — which “protected” women by limiting how much weight they could lift on the job, how many hours they could work, and so on — were for years seen as noble and just, a way of leveling the playing field for the weaker sex. But protective laws obstructed women’s rights as much as they secured them. Restrictions for women workers were heartily supported by male-dominated unions, which sought to diminish the competition posed by unskilled female labor. For decades to follow, the influence of these laws rippled outward to affect all spheres of women’s lives, putting them at a disadvantage in fields as far-flung as family and criminal law. Protective legislation moreover inscribed female difference into constitutional law, after Muller — taking women several steps further from the legal equality that many American feminists had been pursuing since Seneca Falls. These women had wanted recognition as a class, but as one equal to men. What they got instead, in the guise of a “leg up,” was a ghetto.
With the women’s movement in the 1970s, much of this changed. Three landmark cases — Reed v. Reed, Frontiero v. Richardson, and Craig v. Boren2— prompted the Court to begin scrutinizing sex-specific laws, and led them to undo much of the protective legislation that had been put in place in previous decades. Frontiero v. Richardson challenged a federal law that automatically granted male members of the armed forces “dependency” benefits for their wives, but demanded that servicewomen prove their husbands were dependents. Deeming the law unconstitutional under the Due Process Clause of the Fifth Amendment, Justice Brennan’s ruling for the majority formally recognized, for the first time ever, the law’s complicity in upholding sex discrimination:
There can be no doubt that our Nation has had a long and unfortunate history of sex discrimination. Traditionally, such discrimination was rationalized by an attitude of “romantic paternalism” which, in practical effect, put women, not on a pedestal, but in a cage . . . It is true, of course, that the position of women in America has improved markedly in recent decades. Nevertheless, it can hardly be doubted that, in part because of the high visibility of the sex characteristic, women still face pervasive, although at times more subtle, discrimination in our educational institutions, in the job market, and, perhaps most conspicuously, in the political arena.
Justice Brennan upheld Frontiero’s equal protection challenge, and took the case a step further by insisting that the Court put the law in question to “strict scrutiny” — the most rigorous test of judicial review. In order to pass strict scrutiny, a law must prove that it serves a “compelling” (meaning necessary or crucial) state interest; that it is as narrowly tailored as possible to serve that interest; and that the law is the least restrictive means for achieving that interest. Failing those three tests, the law is deemed unconstitutional. In Frontiero, Brennan determined that the motive behind the discriminatory policy — ostensibly saving on administrative costs by assuming that all wives of male members of the armed forces were dependents, rather than simply asking — was insufficiently compelling. After all, the government had no proof that this assumption was true, or that the policy saved them any money. (For all they knew, the government might be losing money on the policy.) Stating what should have been obvious, Brennan said, “Any statutory scheme which draws a sharp line between the sexes, solely for the purpose of achieving administrative convenience” was in violation of equal protection.
Brennan’s ruling in Frontiero marked the first time that strict scrutiny had been applied to a case of sex discrimination — and thus the first time women were considered a “suspect class” alongside racial and religious minorities, non-natives, and the poor. “Suspect classification” was a concept that had come out of a footnote to US v. Carolene Products in 1938 — a significant case about a surprisingly boring law, regarding what kind of milk could be shipped in interstate commerce — and drew a distinction between the presumed constitutionality of legislation dealing with everyday economic activities and that which pertained to personal liberties. Footnote 4 — the most famous footnote in US constitutional law — suggested that any laws suspected to restrict the rights of individuals, especially those of “discrete and insular minorities,” ought to be “subjected to more exacting judicial scrutiny.” After Carolene Products, the Court adopted two levels of such scrutiny: strict scrutiny, and the significantly less rigorous “rational basis review,” which merely suggested that a challenged law be “rationally related to a legitimate government interest.” If a law seemed to single out a “suspect class,” it would automatically be submitted to strict scrutiny — and in most cases be stricken down. If not, it would face the much more forgiving rational basis test.
Although Brennan used strict scrutiny to evaluate the law Frontiero challenged, three of his (male) fellow justices — though ultimately concurring that the law was unconstitutional — could not be persuaded that strict scrutiny was necessary, or appropriate, for deciding this particular case. In order to be deemed “suspect,” following the logic of Footnote 4 — and the subsequent cases that built Court doctrine upon it — a class must have experienced a history of discrimination, share visible and “immutable” characteristics that bear little or no influence on their ability to contribute meaningfully to society, and constitute a small and “politically powerless” minority. As a group, women met nearly all these criteria. But as far as numbers went, they were not, and are not, a minority.
Brennan’s elegant defense could not elide the fact that the requirements of suspect classification posed a curious paradox for women. The phrasing of Footnote 4 implied that a class of people large enough to constitute a population majority — as women did — could not feasibly be “oppressed,” or oppressed enough to require extra attention at the level of judicial review. The criterion was logical, but it couldn’t accommodate the illogical reality of patriarchy, whose influence on every aspect of daily life is so big as to be essentially invisible. It seemed as if the law could not comprehend the scope of the inequality it had confronted. When dealing with sex discrimination, women seeking legal redress would have to resort to the existing structures built to address racial and religious discrimination — structures that didn’t always match up.
That women were treated like a minority — or, at the least, not treated as the equals of men — was perfectly obvious in the 1970s, at least to justices like Brennan, whose ruling four years later in Craig v. Boren helped to refine the relevant legal standards with further concept of “intermediate scrutiny” for the benefit of a “quasi-suspect class” like women. More difficult to satisfy than rational basis review, intermediate scrutiny demanded that a law demonstrating bias serve “important governmental objectives,” and that “the discriminatory means employed be substantially related to the achievement of those objectives” — with the imputed caveat that extra care should be taken by the Court to determine whether the objective hadn’t itself been influenced by bias. As a means of plugging up the embedded minority-majority contradiction in Footnote 4, the “quasi-suspect class” was more inclusive, and intermediate scrutiny was decent and flexible. Justices after Brennan, including Ruth Bader Ginsburg, managed to apply versions of immediate scrutiny to sex discrimination cases that in many ways resembled strict scrutiny in their rigor.
But if the Court had to jerry-build all these tangled accommodations, why not just apply strict scrutiny? In part, the justices’ resistance in Frontiero was due to what they presumed would be the impending passage of the ERA, which promised to grant women suspect classification automatically. With the ERA gaining momentum (at the time), the dissenting justices felt no pressing need to settle the question of women’s classification status in Frontiero; Congress and the Constitution would take care of it. Yet the answer they did eventually come up with in Craig v. Boren — and only after it was clear the ERA was losing momentum, having been ratified by thirty states in 1972, then eight, then three, then one, then in 1976, none at all — was more like an open question, as their insistence on “intermediate” scrutiny seemed to reveal a deeper reluctance to validate women as the legal class the Court had, in other contexts, already told them they were.3 While in theory women could be recognized as a class by the law, and governed as a class, it just so happened that they earned that distinction only when doing so served the patriarchy of the day.
So what do women want — class recognition, or not? As Muller’s legacy proved, lumping women together under the uniform banner of their sex could have disastrous consequences, perpetuating the very distinctions that motivate inequality in the first place. But recognizing women as a class on the basis of their being women living under patriarchy — and here it seems useful to invoke the term “sex class” to mean exactly this — has also worked powerfully in the service of equality, especially when women themselves have been the ones advocating for that recognition. If protective labor laws ultimately discriminated against women, cordoning them off in such a way that made it easier to oppress them, arguments for suspect classification discriminated for women, recognizing a distinction already codified in society to assist in its undoing. Recognizing sex class, in other words, could make up for its unrecognized but real abuses elsewhere. When time came to draw class lines in the law, who held the chalk made all the difference.
With Dukes, it was women themselves — not male lawmakers or justices who sought to grant them special legislation — who put forward the case for their common experience, and this was significant. Of course, what had spurred them to do so were Wal-Mart’s attempts to fend off that other beacon of nongendered class struggle, the union. As Liza Featherstone reported in her excellent book on the Dukes case, Selling Women Short (2004), Wal-Mart has employed a number of union-busting tactics, both legal and illegal, to discourage organizing — like telling their employees that they’ll be fired for discussing their wages with one another (one of many ways of keeping women oblivious to sexist pay discrepancies), or demanding that managers report suspicious “union activity” to headquarters. (According to one training manual, a flyer announcing a meeting of — this is the example given by an official corporate handbook — the “baby shower committee” should immediately raise a red flag.) If a specific Wal-Mart store seems especially sympathetic to unionization, a “labor team” from headquarters in Bentonville, Arkansas is promptly flown out on a company plane to hold mandatory anti-union meetings with the workers, reminding them to “think for themselves.” While the United Food and Commercial Workers Union has made significant strides in helping to organize Wal-Mart workers, a majority of female employees, intimidated by the company’s scare tactics or simply afraid of losing their jobs, may not have seen unionizing as an option.
In a way, this vehemence was a blessing, for it was only when the Dukes women encountered the sort of class-making allowed by law — first viewed as a last resort — that they were able to tap into a solidarity of sex class previously unknown to them. Many of these women came from low-income, middle-American, conservative Christian backgrounds, and had professedly little or no experience with political life. Before she became the face of Wal-Mart’s female workforce, Betty Dukes thought “sex discrimination” was a bad word, believing it had something to do with sex or rape. For the women involved in Dukes—including those who didn’t fit this general description—the case was a political awakening. It granted them license to think about sexism, about feminism, about labor equality, and about other political issues that affected them personally as nothing before had. Gretchen Adams, the former deli manager who quit Wal-Mart six months after the case was filed, is now an organizer for the UFCW.
The value of this cannot be overstated. In feminism at least, ours is an era supposed to be post–“women-as-a-class.” The perceived crisis of the women’s movement since the late 1970s (or so the story goes) has been the slow erosion of “sisterhood” — caused in part by legitimate concerns over those it excluded, in part by infighting and factionalism demanding that feminists name their camps: sex-pos or anti-porn, gender-constructivist or biological essentialist. Like a magazine survey, the dialogue asks, Which type are you? The notion that women constitute a “totalizing” class is always the enemy — considered too sweeping, ungenerous, naïve, or worse: willfully blind to difference and variety of experience. It’s a fair critique — but this account, too, is a caricature, and can do more damage to the conversation about women’s rights than it does good. In scale, scope, and sheer persistence, Dukes reveals how flat this version of our history really is. Women’s oppression transcends partisan politics and niche identity categories, and women who live in a patriarchal society will continue to share common nuisances in law and fact as long as that patriarchy persists. Dukes is heartening evidence that some women are willing to fight on the large scale, and that sex class is something we can, and should, reclaim.