The Yale Law Journal

June 2014

Separate Spheres

abstract. This essay is about the mixed legacy, or incomplete achievement, of the landmark legal changes of the Second Reconstruction. This mixed legacy is one of the central themes of The Civil Rights Revolution, the third volume of Bruce Ackerman’s We the People series. The book provides a sweeping account of constitutional change in the 1960s and early 1970s, focusing on both the remarkable legislative accomplishments of that period and the limitations and disappointments that accompanied them. Ackerman argues that these limitations were baked in: The landmark statutes of the Second Reconstruction failed to attend to, or combat, forms of discrimination and disadvantage that travel across different social contexts, and thus cannot provide a platform for addressing such problems today. This essay offers a different perspective on the legislative achievements of the Second Reconstruction. “Interspherical impacts”—the cumulative effects of discrimination and disadvantage across multiple spheres of civil society—are a pressing social problem, and one the law today often fails to rectify, or even to recognize. But these limitations were not an inherent part of the constitutional change that occurred during the civil rights era. Indeed, this essay argues that concern about interspherical impacts motivated many of the key statutes and legal decisions of the period, and that these statutes and decisions provide a foundation for developing twenty-first century legal understandings that are responsive to forms of discrimination and disadvantage that migrate across different spheres.

author. Assistant Professor, University of Texas School of Law. I am grateful to Joey Fishkin, Sandy Levinson, and Reva Siegel for helpful comments and conversation about this essay. My greatest thanks go to Bruce Ackerman, for writing the book that inspired this symposium, and for teaching me so much about constitutional law.


Attorney General Eric Holder made news recently by calling for an end to the practice of felony disenfranchisement1—a practice that currently curtails the voting rights of 5.8 million Americans, a number greater than the individual populations of most states.2 Holder argued that any abrogation of a fundamental right on such a wide scale should concern us, but in this case, there is additional cause for concern, because the burdens imposed by felony disenfranchisement are not distributed evenly across the population. Well over a third of individuals who have lost the right to vote as a result of criminal conviction are African American.3 Holder argued that this disparate racial impact was not a coincidence. He noted that “[a]fter Reconstruction, many Southern states enacted disenfranchisement schemes to specifically target African Americans and diminish the electoral strength of newly-freed populations.”4 These schemes largely accomplished their goal: In the late nineteenth century, and for much of the twentieth century, vast numbers of African Americans “had their rights rescinded, their dignity diminished, and the full measure of their citizenship revoked for the rest of their lives.”5 Holder argued that felony disenfranchisement laws continue to have this racial effect today, and are thus “too significant to ignore” and “too unjust to tolerate.”6 “[W]e’ve outlawed legal discrimination [and] ended ‘separate but equal,’”7 but now, he contended, we need to confront the ways in which discrimination and disadvantage in one context may spill over into other contexts and perpetuate patterns of racial subordination generally associated with earlier eras in the nation’s history.8

The Attorney General’s commentary on the relationship between voting rights and the criminal justice system is part of a much broader, ongoing conversation about the cumulative effects of discrimination and disadvantage across different social and legal contexts. In recent years, commentators have focused in particular on the way in which felony convictions affect one’s life chances in contexts far removed from criminal law. As Holder noted, loss of the right to vote is only one of a constellation of “burdensome collateral consequences . . . [imposed on] formerly incarcerated individuals.”9 Having a criminal conviction makes it more difficult, and sometimes impossible, to access all sorts of opportunities, across contexts such as employment, housing, education, and social welfare.10 But one need not be convicted of a felony to experience this phenomenon. Discrimination and disadvantage in one context frequently spill over and create obstacles to opportunity in other contexts. Public elementary and secondary schools in this country are no longer formally segregated on the basis of race, but decades of redlining and employment discrimination may nonetheless render neighborhoods with well-funded, high-performing schools inaccessible to racial minorities.11 The law has for some time barred race discrimination in employment, but such protections may do little to help people who never hear of job opportunities or develop the right connections because they are unable to afford to live in the residential areas and attend the schools where employers and their children reside.12 President Lyndon Johnson made this point fifty years ago, on the eve of the passage of the Voting Rights Act, when he asserted that the forms of discrimination and disadvantage experienced by racial minorities in this country cannot “be understood as isolated infirmities. They are a seamless web. They cause each other. They result from each other. They reinforce each other.”13

As the Attorney General’s recent comments indicate, antidiscrimination law, as it is currently interpreted, does not always recognize these interconnections. Laws barring race discrimination offer little protection to those who are prevented from obtaining public housing, food stamps, or student loans on the basis of past criminal convictions14—even when such policies have a wildly disproportionate impact on racial minorities. Modern equal protection doctrine bars most overt racial classifications, but does little to protect racial minorities from regulations such as felony disenfranchisement laws, which do not formally classify on the basis of race—even in instances where the differential effects of those regulations on particular racial groups are quite pronounced.15 Thus it seems right to celebrate the half-century of progress inaugurated by the landmark civil rights statutes of the 1960s—the Civil Rights Act, the Voting Rights Act, and the Fair Housing Act—while also recognizing, as the Attorney General recently did,16 that the antidiscrimination project to which this nation committed itself fifty years ago is still unfolding. Those landmark statutes significantly reshaped the way race is regulated in this country, but they did not solve the problem of accumulated disadvantage, and they have not yet enabled all individuals to “share fully and equally, in American society”17 or “to be treated in every part of our national life as a person equal in dignity and promise to all others.”18

The mixed legacy—or, rather, incomplete achievement—of the landmark legal changes of the 1960s is a central theme of Bruce Ackerman’s new book, We the People: The Civil Rights Revolution.19 The book meditates on the remarkable achievements of the civil rights era, focusing both on what they did for the nation and on what they failed to do. Ackerman argues that these failures were baked in: the landmark statutes of the Second Reconstruction, for all that they accomplished, failed to attend to cross-contextual discrimination and disadvantage, and thus do not provide a platform for addressing such problems today. This essay takes up this argument. Ackerman is surely right that “interspherical impacts”—the cumulative effects of disadvantage and discrimination across multiple spheres of civil society—are a pressing social problem, and one that law today often fails to rectify, or even to recognize. But this essay argues that these failures, or limitations, are not an inherent part of the constitutional change that occurred in the civil rights era. Indeed, this essay argues that concern about interspherical impacts motivated some of the key statutes and legal decisions of the 1960s and early 1970s, and that these statutes and decisions provide a foundation for developing a set of legal understandings that is responsive to forms of discrimination and disadvantage that travel across different spheres.

I. ackerman’s argument

In order to appreciate Ackerman’s claims regarding the nature of the constitutional change that occurred in the 1960s, it is useful to situate The Civil Rights Revolution in the context of his broader project. This book is the latest installment of We the People, a magisterial project spanning multiple decades and volumes (three and counting20) that seeks, among other things, to explain how constitutional change happens in the United States. The jumping-off point for the project is the observation that formal Article V amendments have, at best, played a minor role in the Constitution’s considerable evolution over the past two centuries.21 Indeed, We the People contends that some of the most significant constitutional changes in American history—those associated with the Civil War and the New Deal—have occurred outside the formal amendment process.22 Ackerman argues that these changes have, instead, been the product of “constitutional moments,” stretches of time in which Americans have come together and self-consciously altered the nation’s founding document without recourse to Article V amendment procedures.23 The third installment of We the People argues that the civil rights revolution, which occurred in the two decades after Brown v. Board of Education,24 was one of those moments.

One of the central aims of this volume is the constitutional canonization of the landmark statutes and legal decisions associated with the civil rights revolution. Ackerman describes the Second Reconstruction as “one of the greatest acts of popular sovereignty in American history.”25 He views the constitutional changes associated with this latter-day Reconstruction as no less significant than those associated with the First Reconstruction—and as no less constitutional simply because they were codified primarily in the form of landmark statutes and court decisions rather than Article V amendments. Indeed, The Civil Rights Revolution seeks to shift the center of gravity in constitutional law away from the Founding and First Reconstruction and toward the Second Reconstruction.26 Ackerman reminds us that it was in the 1960s, and not the 1860s, that Americans rejected a constitutional framework in which restaurants could refuse to serve black customers, railroads and bus companies could systemically humiliate black riders, and states could bar those who lacked sufficient funds to pay a poll tax from exercising their right to vote. It was in the 1960s that all three branches of government and a mobilized citizenry came together “to affirm their support for a series of landmark statutes that broke the back of Jim Crow in this country.”27

Throughout the book, Ackerman calls on contemporary lawyers and judges, and all those who make constitutional arguments, to stop looking back, always, to the eighteenth and nineteenth centuries, and to look instead to the civil rights era when attempting to discern the meaning of the modern-day Constitution. Making this shift is critical not simply because the civil rights revolution moved us “far beyond the constitutional principles”28 of those earlier eras, but also because “political determination and legal insight will be required if America is to sustain its commitment to the Second Reconstruction over the coming decades.”29 The Second Reconstruction was, in Ackerman’s view, a great constitutional achievement, and one that requires protection from a contemporary Supreme Court that sometimes seems intent on overturning its guarantees.30

Despite Ackerman’s deep appreciation for and dedication to the constitutional achievements of the 1960s, he asserts quite pointedly, and repeatedly, that “[t]he struggle to preserve the civil rights legacy . . . should not blind us to its very serious limitations.”31 The most serious of these limitations, on Ackerman’s account, is the fact that the landmark statutes did not take account of interspherical impacts. Ackerman views this failure as an intrinsic and defining feature of the civil rights revolution. He argues that one of the hallmarks of the Second Reconstruction is that it “divided the world into different spheres of life: public accommodations, education, employment, housing, voting,” rather than taking a “one-size-fits-all” approach to these issues.32 Indeed, one of the central claims of The Civil Rights Revolution is that the landmark statutes of the 1960s imposed “different regimes on different spheres,” and that, in comparison with the First Reconstruction, the Second Reconstruction insisted on a “far more contextual understanding of the constitutional meaning of equality in different spheres of social and political life.”33

In some ways, Ackerman regards this sphere-by-sphere approach as a signal achievement of the Second Reconstruction. It enabled government actors to develop legal principles and technologies that were attentive to the different challenges and underlying realities present in each of the different spheres instead of approaching every problem with the same hammer.34 The book celebrates the kind of contextual sophistication officials in the civil rights era employed when devising strategies for tackling the different challenges presented by different manifestations of racial inequality.35

Ackerman argues, however, that there was a significant drawback to this sphere-by-sphere approach, which is that the civil rights revolution “did not take account of the cumulative impact of inequalities across spheres.”36 Congress, courts, and administrative agencies in the 1960s and early 1970s worked to guarantee equality within each sphere, but, on Ackerman’s account, they were not attentive to the ways in which discrimination and deprivation in one sphere can leave people at a distinct disadvantage in others. For instance, Ackerman suggests that although Title VII of the Civil Rights Act of 1964 barred employers from discriminating against minority job applicants with the same qualifications as white applicants, these protections did little to help minorities who never had a chance to develop equal qualifications in the first place, due to residential segregation, impoverished schools, or other forms of inequality outside the context of employment.37 Thus, Ackerman argues that, in part because the civil rights revolution carved the world into separate spheres and failed to account for interspherical impacts, it left vast numbers of Americans behind, particularly those who were both non-white and poor. He asserts that the most privileged racial minorities benefitted the most from the Second Reconstruction, but that this landmark change in our constitutional order offered much less to people who faced deprivations across multiple spheres—people whose circumstances rendered them less able to take advantage of the new protections.38

The Civil Rights Revolution ends with a rousing call for a new constitutional moment, or a Third Reconstruction, that would do this work—that would recognize the connections between the spheres—in a way the Second Reconstruction did not.39 Ackerman argues that it is past time to extend the promise of equality to people who were left behind fifty years ago, when Americans last came together to rethink our most fundamental constitutional commitments.

It is undoubtedly true that not everyone has shared equally in the great social and legal changes wrought by the Second Reconstruction. One could find support for this proposition in any number of subsequent court decisions; in the doctrinal structures that currently guide antidiscrimination law; or in contexts in which law simply offers little or no protection—particularly along the lines of class. Consider one small example from my own institution, the University of Texas. Twenty years ago, when Texas’s epic battle over affirmative action began, it became apparent that despite federal civil rights laws, and an affirmative action program, vast numbers of Texans still lacked access to the state’s flagship university. The undergraduate student body at UT hailed overwhelmingly from wealthier-than-average families;40 some whole counties had never sent a student to college in Austin.41 The latest research on college admissions in this country suggests that parental earnings, place of residence, and socioeconomic status continue significantly to impair students’ access to selective colleges.42 This research is part of a vast body of data demonstrating that interspherical impacts, or the cumulative effects of disadvantage across spheres, continue to generate inequality—despite the tremendous legislative accomplishments of the President Texas sent to the White House fifty years ago.43

One of the great contributions of The Civil Rights Revolution, particularly given rising levels of income inequality over the past few decades, is its emphasis on the importance of reasoning across spheres in order to achieve genuine equality. The book makes a persuasive case that addressing interspherical impacts should be the centerpiece of the civil rights project in the twenty-first century.

This essay raises some questions, however, about the book’s assertion that the Second Reconstruction was fundamentally insensitive to interspherical impacts, and that a new constitutional moment is required in order to do the work left undone in the 1960s. For a start: Is it truly the case that the limitations The Civil Rights Revolution identifies in our legal tradition regarding interspherical reasoning are inherent in the meaning of the Second Reconstruction? Or, instead, might these limitations be a product—or, only one side—of an ongoing debate about the meaning of the major constitutional changes that occurred fifty years ago? The idea that the changes associated with this critical era in constitutional lawmaking were confined within hermetically sealed spheres is only one way of understanding the constitutional meaning-making that occurred in the era of Lyndon Johnson and Martin Luther King, Jr. This essay argues that it is possible to identify within the Second Reconstruction itself—particularly as its meaning has been elaborated over time—pathways and precedents for thinking more seriously about interspherical impacts than the law currently does.

II. the development of interspherical reasoning

To begin to develop an account of this alternative understanding of the constitutional change associated with the Second Reconstruction, let me begin where The Civil Rights Revolution leaves off: with the women’s movement. This was a movement that knew a little something about separate spheres. For well over a century, the chief obstacle to equality in the context of sex was an ideology that divided the world into two ostensibly equal and complimentary spheres: a public sphere of work and politics, which was reserved for men, and a private sphere of home and family, considered women’s domain. When the women’s movement began to garner national attention in the mid-1960s, it took aim at this ideology, arguing that in order to achieve true equality between the sexes, the law needed to move beyond the concept of separate spheres and begin to address the interspherical impacts that rendered women second-class citizens across a wide range of social and legal contexts.

In 1966, a number of people committed to this project founded the National Organization for Women (NOW).44 NOW, which was formed in direct response to the government’s failure to enforce Title VII’s prohibition of sex discrimination, argued that equality for women, in the context of the workplace and throughout society, would require changes in the regulation of the family. NOW’s founding Statement of Purpose called for the creation of “new social institutions which will enable women to enjoy true equality of opportunity and responsibility in society, without conflict with their responsibilities as mothers and homemakers.”45 Among the innovations NOW deemed essential to sex equality were “a nationwide network” of child-care centers, “which will make it unnecessary for women to retire completely from society until their children are grown, and national programs to provide retraining for women who have chosen to care for their own children full-time.”46 NOW’s key claim was that gender equity in spheres such as education and employment depended on structural changes in the sphere of the family, because without such changes, women would continue to lack practical access to opportunities widely available to men.

In the summer of 1970, the movement put these arguments to the nation in a massive Women’s Strike for Equality, which took place in Washington, D.C., New York City, and dozens of other cities throughout the country.47 The Strike, which attracted tens of thousands of participants, was intended to illustrate that fundamental changes in the social organization of the family were a prerequisite to sex equality, and that such changes required the implementation of laws and policies designed to alleviate the pressure on women (and men) to conform to traditional sex roles. To this end, the strikers made three core demands: “(1) free abortion on demand, (2) free 24-hour childcare centers, and (3) equal opportunity in jobs and education.”48 Reproductive rights and childcare were essential, the movement argued, because equal opportunity in school and at work would remain elusive as long as women were expected to subordinate all other activities to the care of home and family.49 To dramatize the interspherical nature of their claims, the movement scheduled the Strike to coincide with the fiftieth anniversary of the Nineteenth Amendment.50 To transform the promise of full citizenship and constitutional equality for women into a reality, the movement argued, it was necessary to attend to the ways in which the regulation of the family hampered women’s ability to seize the rights and opportunities granted to them by law.

In the late 1960s and early 1970s, the movement brought these interspherical arguments to Congress and the courts. On the legislative side, feminists in Congress—Bella Abzug in particular—argued that equality in the public sphere, especially in education and employment, would remain illusory as long as the regulation of the family remained unchanged. Feminist legislators also drew connections between various contexts outside the home, arguing, for instance, that educational reforms were necessary in order for women to participate equally in work and politics. These efforts resulted in an unprecedented burst of sex equality lawmaking by the Ninety-Second Congress that ran to nearly a dozen statutes, some of which responded directly to arguments about the ways in which discrimination or deprivation in one sphere affected women’s status across a range of other spheres.51 Proponents of Title IX52 argued that equal opportunity for women in education and athletics was necessary to enable them to take advantage of opportunities in the contexts of business and politics after graduation. The Comprehensive Child Development Act53—which would have provided free or subsidized daycare for American children if President Nixon hadn’t vetoed it—and a bill instituting a tax deduction for child-care expenses,54 which the President did sign, responded directly to the argument that real equal opportunity for women in spheres such as education and employment would remain out of reach unless the legislature altered the status quo with regard to the provision of care within the family.

Meanwhile, in court, one of Ruth Bader Ginsburg’s central aims as a litigator for the ACLU in the 1970s was to persuade the Justices that laws regulating the family—laws that were intended to benefit women by assuming their dependency on a male breadwinner—were unconstitutional because they reinforced women’s second-class status in a plethora of other spheres.55 Ginsburg noted in the briefs she submitted to the Court that, as of the early 1970s, the law consistently enforced traditional sex roles at the intersection of work and family. It awarded husbands the exclusive right to control family assets and to determine the family’s domicile; expected wives to adopt their husbands’ names upon marriage; and permitted girls to marry at a younger age than boys, thereby according the latter “more time to prepare for bigger, better and more useful pursuits.”56 Criminal law, too, reinforced traditional sex and family roles through penalties for promiscuity and prostitution that targeted women and girls; and “tax law present[ed] a significant disincentive to the woman who contemplate[d] combining a career with marriage and a family.”57 In a series of cases challenging the constitutionality of laws restricting certain government benefits to women, Ginsburg asserted that the state was not helping women, but rather, “fortif[ying] the assumption, harmful to women, that labor for pay and attendant benefits is primarily the prerogative of men.”58 Drawing on the recent achievements of the Second Reconstruction in the context of race,59 Ginsburg argued that the Constitution guaranteed women, too, full and equal citizenship, and that the only way to secure constitutional equality was to attend to the interspherical impacts of policies that enforced traditional sex roles.

Ginsburg won nearly all of the cases she litigated in the 1970s, and this concern about interspherical impacts appeared, at least in embryonic form, in some of the Burger Court’s sex discrimination decisions.60 It has surfaced, in a more robust form, in some of the Court’s more recent decisions. In 1996, in United States v. Virginia,61 the Court invalidated VMI’s all-male admissions policy not simply, or even primarily, because it classified on the basis of sex (indeed, the Court suggested that some sex classifications in the context of education might be perfectly acceptable62), but because the policy perpetuated historic inequalities across a wide range of spheres. The Court noted that a VMI education was a gateway to opportunity in business, politics, and the military, and that denying women a chance to compete for admission prevented them from accessing “the powerful political and economic ties of the VMI alumni network”63 and made it harder for them to attract the attention “of business owners, corporations, VMI graduates and non-graduate employers . . . interested in hiring VMI graduates.”64

The Court engaged in this interspherical kind of reasoning again a few years later in Nevada Department of Human Resources v. Hibbs,65 which upheld the Family and Medical Leave Act (FMLA) as a proper means of enforcing the Equal Protection Clause. The FMLA grants affirmative benefits, in the form of twelve weeks of family leave, to employees of both sexes.66 The Court had never held that an employer’s decision not to provide employees with family leave violated constitutional sex discrimination law. But it upheld the FMLA as a valid exercise of Congress’s power to “enforce” the Fourteenth Amendment because it concluded that such benefits were essential to providing women—especially those who are mothers—with actual, substantive equality in the workplace.67 In other words, the Court recognized in Hibbs that it was necessary to take account of background conditions in the sphere of the family in order to ensure genuine equality at work.

By citing these instances of interspherical reasoning by both Congress and the Court, I do not mean to suggest that the American legal system is consistently attentive to the cumulative impacts of sex inequalities across spheres. It is not. Courts and legislatures today frequently overlook the interspherical impacts of laws limiting reproductive rights, for example.68 My argument is simply that attention to and concern about interspherical impacts is one major theme in sex equality legislation and constitutional sex discrimination law. It is, at the very least, a plausible interpretation of what equal protection requires—and it is an interpretation that currently plays a role in both legislative and judicial reasoning.

III. interspherical reasoning in the civil rights era

The question arises, then, whether this kind of reasoning about equality was a new departure when the women’s movement began championing it in the 1970s, or whether the seeds of this way of thinking were planted during the Second Reconstruction. The remainder of this essay argues for the latter claim: that the women’s movement’s concern about interspherical impacts—a concern at least partially echoed by all three branches of government, and evident in sex discrimination law today—builds on precedent from the civil rights revolution. The women’s movement in the 1960s and ’70s adapted and developed the argument that genuine equality requires taking into account interspherical impacts by emphasizing how regulation of the family affects men and women’s opportunities in other spheres. But it did not invent that argument out of whole cloth. Concern about the cumulative effects of inequalities across spheres was very much a part of the thinking that gave rise to the landmark civil rights statutes of the Second Reconstruction.

Consider the Voting Rights Act of 1965.69 Proponents of the Voting Rights Act (VRA) argued that safeguarding the right to vote should be a top priority because voting was “preservative of all rights.”70 This idea is inherently interspherical; it accords voting a special place in the hierarchy of rights precisely because the right to vote is crucial to the achievement of equality in a range of other spheres. The Court drew on this interspherical conception of the right to vote one year after the passage of the VRA in Katzenbach v. Morgan,71 which upheld the section of the Act providing that no person who has successfully completed sixth grade in an American school in which the predominant language is other than English shall be disqualified from voting on the basis of a literacy test. The Court in Morgan held that this provision was a proper exercise of Congress’s power to enforce the Fourteenth Amendment because it “may be viewed as a measure to secure for the Puerto Rican community residing in New York nondiscriminatory treatment by government,” not only in the sphere of voting, but also in “the provision or administration of governmental services, such as public schools, public housing and law enforcement.”72 The idea was that “enhanced political power will be helpful in gaining nondiscriminatory treatment in public services for the entire Puerto Rican community”73—that is, fair treatment in the sphere of voting would help to secure fair treatment in other spheres.

Interspherical concerns about voting ran in the other direction as well. The Eighty-Ninth Congress was concerned not only that deprivations of the right to vote would have deleterious consequences for people outside the context of voting, but also that the cumulative effects of disadvantage in other contexts would burden the right to vote in constitutionally salient ways. Congressional debate focused in particular on the effect of educational inequities on voter registration. Proponents of the VRA’s restriction of the use of literacy tests as a qualification for voting argued that such tests “would, inevitably, work unfairly against Negroes—Negroes who have for decades been systematically denied educational opportunity equal to that available to the white population.”74 Advocates noted that such tests worked a cruel irony, as they seemed to “reward” white leadership in districts that had long deprived black communities of anything approximating equal access to education.75 These advocates argued that we cannot “have the white Mississippi legislature operating an inferior and illegally segregated public school system for Negro children, and then turning around and saying to these same wronged citizens, ‘You are not educated enough to decide who shall govern you.’”76

Opponents of the VRA did not respond to these arguments only, or even primarily, by claiming that what happened in the context of education had no bearing on the context of voting. Indeed, VRA opponents seemed tacitly to accept that concerns about interspherical impacts would be legitimate—if there were any to worry about in this case. Rather than attempting to isolate voting from other spheres, opponents argued that the law’s supporters had their facts wrong: black schools in the South were equal to white schools. In fact, opponents were sometimes so insistent on the equal quality and stature of black schools in the South that they seemed to want to relitigate Brown.77 This prompted proponents of the VRA to introduce into the Congressional Record hundreds of pages of reports documenting current and historical inequities between white and black schools in southern states.78 These reports recorded, in painstaking detail, discrepancies in teacher pay, teacher training, expenditure per pupil, curricular breadth, length of school year, and many other differences between black and white schools in the former Confederacy. Ultimately, the argument that racial disparities in educational opportunity persisted, and that they necessitated new protections in the context of voting, won the day. The VRA passed, with restrictions on literacy tests—reflecting congressional concern about the cumulative effects of inequalities across spheres—intact.

When opponents of the VRA challenged the constitutionality of the Act’s literacy-test restrictions, the Court echoed Congress’s argument that attention to interspherical impacts was necessary in order to ensure genuine equality of opportunity in the context of voting. Indeed, the Court’s decision in Gaston County v. United States79 is a model of interspherical reasoning. The problem with imposing a literacy test in Gaston County, North Carolina, the Court held, was that the County had for decades relegated African Americans to underfunded and inferior schools,80 which left them, on the whole, less able than whites to pass a literacy test.81 The Court noted that the Eighty-Ninth “Congress was fully cognizant of the potential effect of unequal educational opportunities upon the exercise of the franchise,” and that, in fact, “[t]his causal relationship was . . . one of the principal arguments made in support of the [VRA’s] test-suspension provisions.”82 Like Congress, the Court declared itself unwilling to allow the unequal educational opportunities afforded the County’s black residents to “become the excuse for continuing”83 to deprive them of the right to vote. “‘Impartial’ administration of the literacy test requirement today,” the Court held, “would serve only to perpetuate these inequities in a different form.”84

In The Civil Rights Revolution,Ackerman singles out Gaston County as the only instance in which the landmark decisions and statutes of the Second Reconstruction departed from “a sphere-by-sphere approach to racial injustice”85 and recognized the ways in which inequality in one sphere may negatively affect equal opportunity in another. But Gaston County was far from sui generis in this regard. Interspherical reasoning played a considerably more central role in the Second Reconstruction than The Civil Rights Revolution suggests.

The kind of interspherical reasoning about the right to vote exemplified by Gaston County resurfaced a year later when Congress amended the VRA to ban the use of literacy tests not just in the jurisdictions covered by the 1965 Act, but nationwide.86 In so doing, Congress repeatedly and forcefully affirmed its view that taking account of racial inequalities in education was essential to the pursuit of genuine equality in the context of voting.87 When this provision too was challenged in Court, the Justices responded by unanimously upholding the nationwide ban. In this case, Oregon v. Mitchell,88 the challenge came from Arizona, which contended that its schools, unlike those in Gaston County, had not been segregated by law and that its literacy test had always been fairly administered89—thus depriving Congress of the power to curtail its use of the test. In rejecting this argument, Justice Black noted “that concern with educational inequality was perhaps uppermost in the minds of the congressmen who sponsored the Act,”90 and that Congress had before it substantial evidence demonstrating “that literacy tests reduce voter participation in a discriminatory manner not only in the South but throughout the Nation.”91 Justice Brennan echoed these observations, noting that “[e]xtensive testimony before both Houses indicated that racial minorities have long received inferior educational opportunities throughout the United States,”92 and that a survey of northern and western states by the United States Commission on Civil Rights revealed that literacy tests had a disparate impact on racial minorities in those states too.93 Thus, by 1970, it seems fair to say that concerns about interspherical impacts—at least those generated by racial inequalities in education—were fairly well-established in the context of voting rights law.94 These concerns became part of subsequent interpretations and refinements of the Voting Rights Act by both Congress and the Court.95

Voting was not the only arena in which Congress expressed concern about interspherical impacts during the Second Reconstruction. Such concerns also played a significant role in the passage of the Fair Housing Act of 1968.96 The Act’s proponents argued not only that discrimination in housing was harmful in and of itself, but also, and perhaps even more importantly, that it had negative collateral consequences for equality in spheres such as education and employment.97 Shortly before Congress took up the question of fair housing, the United States Commission on Civil Rights published a major report on the problem of racial isolation in public schools in which it called for federal housing legislation in order to address the problem of educational inequity.98 Congressional proponents of the Fair Housing Act cited this report,99 and emphasized the powerful causal relationship between discrimination in the housing market and the ongoing segregation of public schools. They argued that “[o]pen housing is absolutely essential to the realistic achievement of . . . desegregated schools and equal opportunity,”100 and that “[m]uch of the statutory civil rights progress of recent years will be little more than theoretical until open housing becomes a reality.”101

These arguments drew on testimony offered during the hearings on the Fair Housing Act about how difficult it was to rectify imbalances in the distribution of educational opportunity in the face of severe residential segregation and race-based barriers to entry into many neighborhoods. Senator Walter Mondale, a co-sponsor, with Senator Edward Brooke, of the Fair Housing Act, pressed this point about education particularly, arguing that “[w]e cannot afford to allow our efforts to provide the best education possible to all Americans to be thwarted by actions of private persons, actions which are antisocial, immoral, and which ultimately amount to contravention of our public policy in favor of equal educational opportunity.”102 Indeed, Mondale went so far as to declare that fair housing is “more than merely housing. It is part of an educational bill of rights for all citizens.”103

Advocates of the Fair Housing Act were equally focused on the effects of housing discrimination on the job prospects of racial minorities. Harvard economist John Kain developed the concept of “spatial mismatch” in this period, which posited that serious limitations on black residential choice, combined with the steady dispersal of jobs from central cities, contributed significantly to low rates of employment and low earnings in African American communities.104 Much of the congressional testimony on fair housing in 1968 documented and built on this idea. The Secretary of Housing and Urban Development testified about the vast growth in job opportunities outside of cities;105 proponents also cited a study by the Department of Labor that identified hugely disproportionate unemployment rates in urban areas and another study by the National Committee Against Discrimination in Housing that showed the continuing movement of jobs from the cities to the suburbs.106 Senator Mondale noted, in this vein, that:

[W]e have talked about and will continue to talk about manpower training programs to remedy hard-core unemployment . . . [but] [u]nless the county [sic] is willing to invest billions of dollars in a cheap, fast, and multi-destination transportation system, ghetto residents will not be able to reach most jobs. A much simpler solution to one facet of the employment problem would be to open housing to everyone.107

Thus, when the Fair Housing Act passed, in 1968, it was understood as a means not only of curbing discrimination in housing, but also as a way to combat ongoing disparities in the context of education and employment.

Concerns about interspherical impacts were no less pronounced in the context of employment, and in the interpretation of Title VII of the Civil Rights Act of 1964,108 than they were in the contexts of voting and housing. The Court’s 1971 decision in Griggs v. Duke Power Co.109 strongly reinforces Gaston County’s message that civil rights law cannot achieve its goals without attending to the cumulative effects of disadvantage across spheres. Griggs, of course, involved an employer who, after Title VII went into effect, began to require job applicants to produce a high school degree and pass two professionally-prepared aptitude tests if they wished to gain access to any of its higher-paying jobs.110 These requirements screened out many more African Americans than whites, with the result that the racial composition of Duke Power Company’s workforce after the institution of Title VII bore a striking resemblance to its racial composition before the law went into effect.

As it had in Gaston County, the Court in Griggs concluded that the racial disparity in individuals’ ability to satisfy these requirements was “directly traceable”111 to disadvantages encountered by African Americans in other spheres. “Because they are Negroes,” the Court reasoned,

petitioners have long received inferior education in segregated schools and this Court expressly recognized these differences in Gaston County . . . There, because of the inferior education received by Negroes in North Carolina, this Court barred the institution of a literacy test for voter registration on the ground that the test would abridge the right to vote indirectly on account of race.112

In Griggs, the Court deployed the same logic—now in the context of employment—to invalidate the diploma and standardized test requirements. It held that Congress’s intention, in passing Title VII, was not to “provide equality of opportunity merely in the sense of the fabled offer of milk to the stork and the fox.”113 Rather, the Court held, when Congress passed the Civil Rights Act, it “required that the posture and condition of the job-seeker be taken into account,”114 for the “Act proscribes not only overt discrimination but also practices that are fair in form, but discriminatory in operation.”115

With these now-famous words and images, the Court in Griggs began to formulate disparate impact doctrine—a doctrine that institutionalizes the law’s concern for the cumulative effects of inequality across spheres. To prevail in a disparate impact suit, a plaintiff need not show explicit or intentional discrimination on the basis of race; he or she need only demonstrate that an employer instituted a policy or screening device that had a disparate impact on a protected group. Often, as in Griggs and Gaston County, that impact will be the product of discrimination or disadvantage experienced by the relevant group in other spheres. It is precisely this kind of accumulation of disadvantage Title VII aims to stop. Disparate impact doctrine is alert to instances in which disadvantage begins to migrate from one sphere to another, and it presses employers to adopt hiring and promotion criteria that do not perpetuate inequalities between workers, but instead, begin to break them down—even when those inequalities have their roots in institutions such as the family and the schools.


By emphasizing the ways in which the landmark statutes of the Second Reconstruction attend to interspherical impacts, my intention is not to suggest that antidiscrimination law today is sensitive to interspherical impacts across the board. Nor do I mean to suggest that the American legal system has made adequate progress in addressing the great and ongoing inequalities Ackerman describes so eloquently in The Civil Rights Revolution. As I noted at the outset, I agree entirely with the book’s suggestion that addressing such inequalities should lie at the center of the civil rights agenda of the twenty-first century. The question is what route, or routes, we might take to address these problems.

One route, the route Ackerman advocates, is a Third Reconstruction—a new constitutional moment in which we amend the Constitution (most likely outside the Article V process) in order to enable the legal system to attend to interspherical impacts. I am quite amenable to the idea of Third Reconstruction, as envisioned by Bruce Ackerman. But is a body of constitutional and statutory law that addresses interspherical impacts really such a departure from the principles of the Second Reconstruction that it necessarily requires a Third?

This essay suggests that alternatively—or, while we wait—there are other routes Americans can take to begin to address these problems. It is true that law today does not do a particularly good job of protecting people against the deleterious effects of interspherical impacts. But the point of this essay is to show that although concern about the cumulative effects of inequalities across spheres may be an underdeveloped aspect of current law, it is not absent from the law. Such concerns motivated a multitude of actors during the Second Reconstruction, ranging from civil rights activists to the legislators who passed the landmark statutes to the President who emphasized that the discrimination and disadvantage experienced by racial minorities were not “isolated infirmities,” but “a seamless web.”116 Even in the absence of a new constitutional moment, the achievements of these Americans half a century ago make it possible to argue that at least some forms of cumulative disadvantage raise constitutional concerns.

We are still today engaged in the process of interpreting the meaning of the Second Reconstruction and the nature of the constitutional change that occurred in that era. The meaning of the Second Reconstruction was not fixed in the 1960s. Its meaning has evolved and will continue to evolve over time, as all sorts of constitutional actors make claims on it. This process began as early as the late 1960s, when the women’s movement started to rely on the constitutional changes of the civil rights revolution to argue that the regulation of the family negatively affected women’s opportunities in other spheres and in so doing deprived them of equal protection. Just recently, the Attorney General made a similar set of arguments about felony disenfranchisement, arguing that the practice violates constitutional principles that are now fifty years old, in part because it allows disadvantage in one sphere to migrate across many others.

In both instances, advocates were drawing on a set of understandings of the Second Reconstruction that foreground the importance of curtailing interspherical impacts. Both argued that realizing the constitutional commitments of those who campaigned for the landmark changes of civil rights era requires addressing cumulative disadvantage across spheres. These are interpretive and normative claims about the meaning of the Second Reconstruction. They are contested, but they are certainly not foreclosed. Through such claims, Americans in the twenty-first century can build on foundations laid in the 1960s to extend the Second Reconstruction’s promise of equality to people for whom equality is still only a promise.