• A Introduction

  In 1979, then-Professor Antonin G. Scalia contributed to a symposium on “The Quest for Equality”. Professor Scalia’s article was entitled “The Disease as Cure: ‘In order to get beyond racism, we must first take account of race.’”¹x Washington University Law Quarterly, Vol. 35, 1979, p. 147. The symposium papers had been delivered by a diverse group of constitutional law scholars in nine separate programmes during the 1978-1979 academic year. Id., at p. 1-3. It was an ironic title, built around a quotation from Justice Blackmun’s separate opinion in Regents of the University of California v. Bakke,²x Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 407 (1978) (Blackmun, J., concurring in part and dissenting in part). with which Professor Scalia obviously and deeply disagreed. In a voice that would soon become familiar to the readers of the United States Reports, Professor Scalia began his commentary with the following words: “[a]s you know, every panel needs an anti-hero, and I fill that role on this one. I have grave doubts about the wisdom of where we are going in affirmative action, and in equal protection generally. I frankly find this area an embarrassment to teach.”³x A. Scalia, ‘The Disease as Cure: “In Order to Get Beyond Racism, We Must First Take Account of Race”’, Washington University Law Review, Vol. 1979, No. 1, p. 147. The problem, according to Professor Scalia, was that the Supreme Court’s decisions were “tied together by threads of social preference and predisposition”, rather than by “threads of logic and analysis”.⁴x Id. He concluded the opening paragraph with a flourish: “Frankly, I don’t have it in me to play the game of distinguishing and reconciling the cases in this utterly confused field.”⁵x Id.
  Of course, Justice Scalia’s observations were not those of a dispassionate observer. Even in 1979 his commitments were clear. His objections were due as much to the trend and substance of the jurisprudence as to the doctrinal incoherence that he attributed to it. Nonetheless, while one might be inclined to dismiss Justice Scalia’s observations as so much hyperbole, there is some justice in them. Justice Scalia was writing at a time of particular flux and considerable confusion in the area of equal protection law.⁶x In terms of state governmental action, the ultimate source of equal protection and due process protection is Section 1 of the Fourteenth Amendment to the United States Constitution, which was adopted after the Civil War and provides, in relevant part, that “[n]o State shall … deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws.” U.S. Const., Amend. XIV, § 1 (1868). For federal governmental action, the ultimate source of protection is the Due Process Clause of the Fifth Amendment, which was adopted in 1791 as part of the Bill of Rights and provides in relevant part that “[n]o person shall … be deprived of life, liberty, or property, without due process of law”. U.S. Const., Amend. V (1791). The Supreme Court has made clear that the Due Process Clause of the Fifth Amendment should be interpreted to encompass a guarantee of equal protection as well as due process. See, e.g., Bolling v. Sharpe, 347 U.S. 497 (1954) (requiring public school desegregation in the District of Columbia under the equal protection component of the Due Process Clause of the Fifth Amendment). See also K. L. Karst, ‘The Fifth Amendment’s Guarantee of Equal Protection’, North Carolina Law Review, Vol. 55, 1977, p. 541-562 (discussing the equal protection component of the Due Process Clause.). In 1973, for example, a majority of the Court in Frontiero v. Richardson ⁷x 411 U.S. 677 (1973). had struck down a gender-based classification pertaining to spousal benefits for members of the armed forces, but the Court could not agree on the appropriate test for determining the validity of that classification. At the time, the received wisdom was that race-based legislative classifications should be ‘strictly’ scrutinized, while other classifications should be subject to a less rigorous ‘rational basis’ test.⁸x See Craig v. Boren, 426 U.S. 190, 220-21 (1976) (Rehnquist, J., dissenting) (“I would think we have had enough difficulty with the two standards of review which our cases have recognized – the norm of ‘rational basis,’ and the ‘compelling state interest’ [or strict scrutiny] required where a ‘suspect classification’ is involved – so as to counsel weightily against the insertion of still another ‘standard’ between those two.”). But see San Antonio v. Rodriguez, 411 U.S. 1, 98 (1973) (Marshall, J., dissenting) (“The Court apparently seeks to establish today that equal protection cases fall into one of two neat categories which dictate the appropriate standard of review -- strict scrutiny or mere rationality. But this Court’s decisions … defy such easy categorization.”) In Frontiero, four Justices thought that strict scrutiny should be applied to gender-based classifications, but that position did not command a majority of the Court.⁹x The Frontiero plurality consisted of Justices Brennan, Douglas, White and Marshall. See Frontiero, 411 U.S. at p. 688. Justice Brennan subsequently wrote the majority opinion in Craig. Only three years later, in Craig v. Boren,¹⁰x 429 U.S. 190 (1976). Subsequently, in Trimble v. Gordon, 430 U.S. 762 (1977), the Court struck down a section of the Illinois Probate Code that discriminated against children born outside of a legally recognized marriage. Writing for the Court, Justice Powell stated that “[i]n a case like this, the Equal Protection Clause requires more than the mere incantation of a proper state purpose”. Id., at p. 769. Justice Rehnquist vigorously dissented in an opinion joined by three other Justices. Id., at p. 776. Katie Eyer has recently argued that the majorities in Craig and Trimble did not think of themselves as articulating a new, intermediate standard of review, but understood the cases as simply applying a more robust form of rational basis review. “[C]ases that were at the time understood by the Court itself as applying minimum tier standards have been reimagined today as outside the minimum tier canon – as cases in which the Court was acting at, but not actually, applying rational basis review.” See K.R. Eyer, ‘Constitutional Crossroads and the Canon of Rational Basis Review’, University of California, Davis Law Review, Vol. 48, 2014, p. 535. According to Professor Eyer, “the vision of minimum tier review that has come to dominate canonical accounts – a form of review so deferential as to be meaningless – has been made possible only by the exclusion from the canon of cases in which a more robust form of review was applied. In fact, when viewed over the broad sweep of history – including, but not limited to the Court’s early sex, illegitimacy, and sexual orientation cases – there is a deep history on the Court of taking groups and rights seriously, even outside of the context of formally heighteed review.” Id., at p. 536. the Court again rejected the application of strict scrutiny to classifications based on gender but seemingly adopted a third, intermediate level of review for evaluating the constitutionality of gender-based classifications. Using that standard, the Court struck down the gender-based classification at issue in Craig. Also in 1976, the Court decided Washington v. Davis,¹¹x 426 U.S. 229 (1976). The Court reinforced that approach in Village of Arlington Heights v. Metropolitan Hous. Dev. Corp., 429 U.S. 252 (1977), and Pers. Admin. of Mass. v. Feeney, 442 U.S. 256 (1979). which held that a party challenging the constitutionality of state action that was neutral on its face but discriminatory in its effect could not prevail without proving discriminatory intent. Finally, in 1978, in Regents of the University of California v. Bakke,¹²x 438 U.S. 265 (1978). a deeply divided Court was unable to agree on the proper standard for evaluating the constitutionality of race-conscious university admissions plans. Of the nine Justices, only Justice Powell concluded both that the Constitution did not categorically preclude colleges and universities from taking an applicant’s race into account in admissions decisions and that the particular admissions programme involved in the case was unconstitutional.¹³x Justice Powell concluded that the Constitution and laws did not categorically prohibit race-conscious admissions decisions, because student diversity could be a legitimate academic concern of colleges and universities, but that the UC Davis medical school’s specific admissions programme nonetheless failed to meet constitutional standards. Id., at p. 314-315, 319-320. According to Justice Powell, the UC Davis plan, which was simply a ‘set-aside’ or quota system, did not allow for the individualized assessment of applicants and therefore failed to satisfy strict scrutiny, which, also according to Justice Powell, was equally applicable to both malevolent and ‘benign’ considerations of race. While Justice Powell wrote the lead opinion in Bakke, no other Justice joined in both of his central conclusions. Four Justices – Chief Justice Burger and Justices Stewart, Rehnquist, and Stevens – did not reach the constitutional question but thought (id., at p. 408-421) that a relevant federal statute (Title VI of the 1964 Civil Rights Act) prohibited any consideration of race in the admissions process. Four other Justices – Justices Brennan, White, Marshall, and Blackmun – thought that the programme did not violate Title VI or the Constitution, but those four Justices did so based on their view (not shared by Justice Powell) that so-called ‘benign’ discrimination should not be subject to strict scrutiny. Id., at p. 324-408. In all, six opinions were filed in the case. Justice Powell recognized that this dizzying array of opinions did nothing to make the Court’s holding transparent. For that reason, he delivered an oral summary of the opinions from the bench, but he did not include that summary in his published opinion. See J. Goldstein, The Intelligible Constitution, Oxford, Oxford University Press, 1992, p. 97-104. (Describing Justice Powell’s oral summary and explanation.)
  In one sense, then, Justice Scalia’s observation about the unsatisfactory state of equal protection law was not wide of the mark in 1979,¹⁴x Of course, then Professor Scalia was not alone in expressing frustration with the state of equal protection law. In 1972, Gerald Gunther detected a “mounting discontent [among the Justices] with the rigid two-tier formulations of the Warren Court’s equal protection doctrine”. G. Gunther, ‘The Supreme Court, 1971 Term – Foreword: In Search of Evolving Doctrine on a Changing Court: A Model For A Newer Equal Protection’, Harvard Law Review, Vol. 86, 1972, p. 12 & 17. In Dandridge v. Williams, 397 U.S. 471, 520-521 (1970) (Marshall, J., dissenting), and San Antonio v. Rodriguez, 411 U.S. 1, 98-99 (1973) (Marshall, J., dissenting), Justice Marshall had expressed disagreement with the idea that the Court’s jurisprudence could (or should) be understood in terms of precisely defined levels of scrutiny. In Rodriguez, he wrote: “[t]he Court apparently seeks to establish today that equal protection cases fall into one of two neat categories which dictate the appropriate standard of review. But this Court’s decisions … defy such easy categorization. A principled reading of what this Court has done reveals … a spectrum of standards … [and] variations in the degree of care with which the Court will scrutinize particular classifications.” Id., at p. 98-99. In Craig v. Boren, 429 U.S. 190 (1976), Justice Rehnquist objected to the articulation of what he took to be an intermediate standard of review on the very different ground that “[t]he Court’s conclusion that a law which treats males less favorably than females ‘must serve important governmental objectives and must be substantially related to achievement of those objectives’ apparently comes out of thin air. The Equal Protection Clause contains no such language, and none of our previous cases adopt that standard.” Id., at p. 220 (Rehnquist, J., dissenting). Justice Rehnquist’s objection is somewhat peculiar, given the absence of explicit support in the text of the Equal Protection Clause for either the rational basis test or the strict scrutiny test. Chief Justice Rehnquist took a more nuanced view of a similar issue in Dickerson v. United States, 530 U.S. 428, 442 (2000), where he declined to overrule Miranda v. Arizona, 384 U.S. 436 (1966), and also found unconstitutional a congressional statute that purported to do so. H. Jefferson Powell has offered an insightful commentary on the disagreement between Justice Scalia and Chief Justice Rehnquist in that case: Justice Scalia “insisted that the Court lacked the authority to invalidate the act of Congress, there being no violation of the Constitution. For the Court, Chief Justice Rehnquist conceded that the warnings required by Miranda are not ‘required by the Constitution, in the sense that nothing else will suffice to satisfy constitutional requirements’; the warnings were a strategic device created and imposed by the Court because the Justices believed existing practice ran an unacceptably high risk of permitting unconstitutional criminal convictions.” H.J. Powell, ‘Reasoning About the Irrational: The Roberts Court and the Future of Constitutional Law’, Washington Law Review, Vol. 86, 2011, p. 233. and the situation has not improved greatly in the intervening period. Indeed, this is an area fraught with complexity. Among other things, the law has become even more complex in some respects because of the Court’s new emphasis on cabining Congress’s enforcement power under Section 5 of the Fourteenth Amendment.¹⁵x See U.S. Const. Amend. XIV, § 5 (“The Congress shall have the power to enforce, by appropriate legislation, this article.”). In Katzenbach v. Morgan, 384 U.S. 641, 650 (1966), the Court held that, under Section 5, Congress had “the same broad powers expressed by the Necessary and Proper Clause” of Art. I, Section 8, Clause 18 of the Constitution, as construed by the Court in McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 421 (1819). In City of Boerne v. Flores, 521 U.S. 507, 520 (1997), however, the Court announced that legislation enacted pursuant to Section 5 would be reviewed under a more demanding ‘congruence and proportionality’ test. Moreover, although the current law has often been presented as relatively straightforward – being centred on a three-tiered analytical framework – the reality might better be described, as two leading commentators put it, as one in which the Court uses “at least three standards of review … in equal protection decisions”.¹⁶x See R.D. Rotunda & J.E. Nowak, Treatise on Constitutional Law: Substance and Procedure, 4th ed. Vol. 3, West, 2012 p.306 (emphasis added). Moreover, the Supreme Court’s decisions have frequently been opaque, with the Court sometimes failing to explain the outcome in terms of the three canonical standards of review, and sometimes even failing to clarify whether a particular decision rests on the Equal Protection Clause or the Due Process Clause.¹⁷x The same analytical framework is used in substantive due process cases, but, while the canonical version of equal protection analysis consists of three tiers, the substantive due process rubric omits the intermediate tier. See, e.g., E. Chemerinsky, Constitutional Law: Principles and Policies, Philadelphia, Wolters Kluwer, 2015, p. 564-566, 570-572. The Court’s occasional lack of clarity with respect to the applicable standard of review is demonstrated by its decision in New York City Transit Auth. v. Beazer, 440 U.S. 568 (1979), in which the Court upheld an agency’s anti-drug use employment policy without indicating the relevant level of scrutiny. Several more recent cases are also instructive. In United States v. Windsor, 133 S. Ct. 2675, 2695 (2013), e.g., the Court held, in an opinion by Justice Kennedy, that a federal statute defining marriage as the union of one man and one woman unconstitutionally deprived same-sex couples of the liberty protected by the Fifth Amendment, but the Court did not specify the level of scrutiny being used to reach that conclusion. In Lawrence v. Texas, 539 U.S. 558 (2003), Justice Kennedy, writing for the majority, found that the Texas sodomy statute violated substantive due process, but did not specifically identify the level of scrutiny utilized in making that determination. At the end of the opinion, however, Justice Kennedy did say that the statute “furthers no legitimate state interest which can justify its intrusion into the personal and private life of the individual.” Id., at p. 578. Justice O’Connor concurred in the judgment, but would have held the Texas statute unconstitutional under the Equal Protection Clause, finding that it failed the appropriate, “more searching form of rational basis review” because “moral disapproval is [not] a legitimate state interest to justify by itself a statute that bans homosexual sodomy, but not heterosexual sodomy”. Id., at p. 580, 582. In dissent, Justice Scalia pointed to the absence of any discussion of strict scrutiny in Justice Kennedy’s opinion – and to Justice Kennedy’s statement that the law was not supported by any ‘legitimate state interest’ – to argue that the proper test was rational basis review, which he thought that the Texas statute clearly met. Id., at p. 594, 599. Justice Scalia also rejected Justice O’Connor’s equal protection argument and her suggestion that a “more searching form of rational basis review” was appropriate. Id., at p. 599-602. In Obergefell v. Hodges, 135 S. Ct. 2584 (2015), the Court, in an opinion by Justice Kennedy, held that state laws prohibiting same-sex marriage violated equal protection and due process, but did not explain the analytical basis upon which that holding rested. Justice Kennedy attempted to compensate for that lack of analysis by writing that “[t]he Due Process Clause and the Equal Protection Clause are connected in a profound way, though they set forth independent principles. Rights implicit in liberty and rights secured by equal protection may rest on different precepts and are not always coextensive, yet in some instances each may be instructive as to the meaning and reach of the other. In any particular case one Clause may be thought to capture the essence of the right in a more accurate and more comprehensive way, even as the two Clauses may converge in the identification and definition of the right.” Id., at p. 2603. One of four dissenting Justices, the Chief Justice argued that “[t]he majority’s decision is an act of will, not legal judgment.” Id., at p. 2612. In addition, he noted that “[a]bsent from [the equal protection] portion [of the majority opinion] … is anything resembling our usual framework for deciding equal protection cases.” Id., at p. 2623. In addition, the Court may sometimes describe its analytic method in terms of one standard of review, while apparently applying another.¹⁸x See, e.g., City of Cleburne v. Cleburne Living Center, 473 U.S. 432 (1985) (purporting to apply rational basis review, but seemingly applying a stricter test); Reed v. Reed, 404 U.S. 71 (1971) (same). Some observers would make a similar point with respect to the Supreme Court’s decision in Fisher v. University of Texas, 136 S. Ct. 2198 (2016), where the Court purported to apply strict scrutiny but arguably departed from the classic version of that standard. More fundamentally, the underlying purpose of equal protection analysis points to something more complex, and the Court’s jurisprudence is consistent with that reality.
  The central meaning of equal protection – that like cases should be treated alike, and those that are different should be treated differently – is both a characteristic and an aspiration of the rule of law. It also seems to presuppose a common understanding as to the criteria by which sameness and difference are to be evaluated. But we know that this common understanding frequently breaks down, as issues of similarity or dissimilarity show themselves to be matters of degree and judgment, and that reality has important ramifications when the search for similarity is not simply an abstract exercise but an essential part of “say[ing] what the law is”.¹⁹x Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177 (1803) (Marshall, C.J.). If law is both will and reason, this is a place where those competing forces clearly intersect, and they do not intersect in a vacuum. At least in the US context, they intersect in a complex environment deeply affected by principles of separation of powers and federalism and a fraught and contested history. Because of that, the courts may view their role in enforcing the Equal Protection Clause as having to decide not whether a classification or distinction is correct or justified in any absolute sense, but whether it is justified in light of what might be called the margin of appreciation that is properly due to the determinations of another set of actors – those belonging to a different branch or level of government.²⁰x This view has long been identified with James Bradley Thayer. See J.B. Thayer, ‘The Origin and Scope of the American Doctrine of American Constitutional Law’, Harvard Law Review, Vol. 7, 1893, p. 129-156. Thayer was a friend and professional colleague of Justice Holmes, a teacher of Justice Brandeis, and a strong influence on Justice Frankfurter. See G.E. White, Justice Oliver Wendell Holmes: Law and the Inner Self, Oxford, Oxford University Press, 1993, p. 378-380; P.W. Kahn, Legitimacy and History, New Haven, Yale University Press, 1992, p. 84. See also Powell, 2011, p. 222 (Noting “the role of constitutional doctrine, judicial standards of scrutiny or modes of analysis that the Court creates in order to implement constitutional norms without claiming that the standards or modes of review are themselves identical to those norms”.). On this view, the judge is not asked to decide whether he or she would find certain differences or similarities relevant and material, if the decision were his or hers to make de novo.²¹x Thus, rational basis review, which is the default standard, is generally thought to be strongly deferential. See, e.g., Fitzgerald v. Racing Assoc. of Central Iowa, 539 U.S. 103, 108 (2003) (“Neither could the Iowa Supreme Court deny that the 1994 legislation, seen as a whole, can rationally be understood to do what that court says it seeks to do, namely, advance the racetracks’ economic interests. Its grant to the racetracks of authority to operate slot machines should help the racetracks economically to some degree – even if its simultaneous imposition of a tax on slot machine adjusted revenues means that the law provides less help than respondents might like. At least a rational legislator might so believe.”); Nordlinger v. Hahn, 505 U.S. 1, 11-12 (1997) (“[T]he Equal Protection Clause is satisfied so long as there is a plausible policy reason for the classification, the legislative facts on which the classification is apparently based rationally may have been considered to be true by the governmental decision maker, and the relationship of the classification to its goal is not so attenuated as to render the classification arbitrary or irrational.”). Instead, the judge’s task is to determine and apply the proper standard of review (and, thus, the appropriate margin of appreciation to be given) to the work of other governmental actors who have sought to determine the relevance and materiality of specific similarities and dissimilarities in various circumstances.²²x On this view, courts must accommodate two competing values: the deference due to the political branches, on the one hand, and faithfulness to the courts’ obligation to interpret and enforce the Constitution, on the other. Achieving that balance may be difficult in practice. See, e.g., J. Tussman & J. tenBroek, ‘The Equal Protection of the Laws’, California Law Review, Vol. 37, 1949, p. 366. Indeed, the Court has sometimes applied the rational basis test in a way that suggests that it believes that the primary determination as to rationality is for the Court rather than the legislature. In F.S. Royster Guano Co. v. Virginia, 453 U.S. 412, 415-417 (1920), e.g., the Court recited the test and the conclusion that there was no conceivable basis for the distinction at issue. Justice Brandeis dissented, joined by Justice Holmes, noting that they could conceive of a justification for the statutory distinction, which they therefore found ‘not illusory’. Id., at p. 418.
  Just as the law relating to equal protection has become more complicated in the past forty years, so too has the law with respect to substantive due process, which shares a common analytic framework.²³x See, e.g., Chemerinsky, 2015, p. 570. Overall, some of the additional complexity in this area undoubtedly stems from the Court’s specific applications of the three-tiered approach; while formally repeating the tripartite formula, the Court’s results sometimes strongly suggest that all Gaul is actually divided into more than three parts.²⁴x See supra notes 9, 14, and 16 and accompanying text. But an equally, if not more, significant part of the complexity is attributable to the Court’s outright departures from that analytic framework in the abortion and voting rights contexts, that is, the Court’s retreat from the strict scrutiny analysis that it once applied to those areas, and its development of the ‘undue burden’ standard as something akin to an alternative intermediate scrutiny test in the reproductive rights context.²⁵x See Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833, 874 (1992) (O’Connor, Kennedy, & Souter, JJ.) (plurality opinion). (“Only where state regulation imposes an undue burden on a woman’s ability to make this decision does the power of the State reach into the heart of the liberty protected by the Due Process Clause.”) That retreat is no doubt the result of many factors, but prominent among them is a widespread contemporary acceptance of the view that courts must occupy a more modest role in the protection of individual rights than was previously thought to be the case.²⁶x Perhaps partially in response to what was deemed by some to be an excessive judicial concern with individual and minority rights during the period of the Warren and Burger Courts, some recent Justices have taken a much broader view of majoritarian rule and a more circumscribed view of the Court’s role in protecting individual and minority rights. See, e.g., B.A. Murphy, Scalia: A Court of One, New York, Simon & Schuster, 2014, p. 234 (Quoting Justice Scalia on the limited constitutional protection allegedly afforded to minority rights); G. Will, ‘Where Scalia Was Wrong’, National Review, 1 February 2017, http://www.nationalreview.com/article/444488/antonin-scalias-natural-rights-error-neil-gorsuch-supreme-court-nomination (Arguing that Justice Scalia misunderstood the importance of minority rights in the American system of constitutional democracy). Long before he joined the Court, Justice Rehnquist was a stern critic of Warren Court decisions such as Brown v. Bd. of Educ., 347 U.S. 483 (1954), and he may well have given inaccurate testimony during his confirmation hearings about a pre-decisional memorandum that he drafted for Justice Jackson, for whom he served as a law clerk while the case was pending. The memorandum argued that Plessy v. Ferguson, 163 U.S. 537 (1886), which held that racial segregation did not violate the equal protection clause, was correctly decided. At his confirmation hearings, Justice Rehnquist insisted that the memorandum reflected Justice Jackson’s views rather than his own. See B. Snyder & J.Q. Barrett, ‘Rehnquist’s Missing Letter: A Former Law Clerk’s 1955 Thoughts on Justice Jackson and Brown’, Boston College Law Review, Vol. 53, p. 632. Among other things, the memorandum asserted that, although in theory “a majority may not deprive a minority of its constitutional right … in the long run it is the majority who will decide what the rights of the minority are.” D.M. O’Brien, Justice Robert H. Jackson’s Unpublished Opinion in Brown v. Board, University of Kansas Press, 2017, p. 73 (Quoting Memorandum). Moreover, the Court has recently begun to discuss in explicit terms the need for maintaining public approval of its work, which may be seen to set some parameters with respect to the Court’s vindication of individual and minority rights. In Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833, 868 (1992), e.g., the plurality emphasized the Court’s need for maintaining popular support and belief in its legitimacy: “It is true that diminished legitimacy may be restored, but only slowly. Unlike the political branches, a Court thus weakened could not seek to regain its position with a new mandate from the voters, and even if the Court could somehow go to the polls, the loss of its principled character could not be retrieved by the casting of so many votes. Like the character of an individual, the legitimacy of the Court must be earned over time. So, indeed, must be the character of a Nation of people who aspire to live according to the rule of law. Their belief in themselves as such a people is not readily separable from their understanding of the Court invested with the authority to decide their constitutional cases and speak before all others for their constitutional ideals. If the Court’s legitimacy should be undermined, then, so would the country be in its very ability to see itself through its constitutional ideals. The Court’s concern with legitimacy is not for the sake of the Court, but for the sake of the Nation to which it is responsible.” See B. Friedman, The Will of the People: How Public Opinion Has Influenced the Supreme Court and Shaped the Meaning of the Constitution, Farrar, Strauss, and Giroux, 2009, p. 328-330 (discussing plurality’s observations concerning legitimacy in Casey).
  The correct answer to many questions is now thought to be simple: let the democratic process work it out. While that means that the courts should generally defer to the choices made by the political branches of government, our understanding of such an imperative is necessarily circumscribed by a recognition that constitutional democracy entails something more than simple majority rule, even when the power of the majority is tempered by the constitutional devices of federalism and separation of powers; constitutional democracy contemplates a system in which majority power is further tempered by the rule of law and provisions for the protection of minorities.²⁷x See Letter of James Madison to Thomas Jefferson, October 17, 1788, in J. Madison, Writings, J. Rakove (Ed.), New York City, The Library of America, 1999, p. 421 (“In our Governments, the real power lies in the majority of the Community, and the invasion of private rights is chiefly to be apprehended, not from acts of Government, contrary to the sense of its constituents, but from acts in which the Government is the mere instrument of the major number of the constituents.”); W.F. Murphy, Constitutional Democracy: Creating and Maintaining a Just Political Order, Baltimore, The Johns Hopkins University Press, 2007, p. 10 (“Although the people’s freely chosen representatives should govern, those officials must respect certain substantive limitations on their authority”).The United States Constitution leaves the states with “wide leeway when experimenting with the appropriate allocation of legislative power”, Hunter v. Pittsburgh, 207 U.S. 161, 178 (1907), and some state constitutions provide for direct democracy in the form of referenda and initiatives, see, e.g., Cal. Const. art. 2, §§ 8 and 9, but the United States Constitution notably does not. Peter Schrag has provided a useful account of California’s experience with direct democracy. See P. Schrag, Paradise Lost: California’s Experience, America’s Future, New York City, The New Press, 1998 (discussing California’s experience with direct democracy). Most important, the principle of deference to the political branches is less weighty when governmental majorities use their lawmaking authority to distort the political process itself in a way that disenfranchises their opponents and perpetuates their hold on power. In other words, deference makes sense but only “so long as the people have their say in the public forum and at the ballot box”.²⁸x See G. Gunther, ‘The Supreme Court, 1971 Term – Foreword: In Search of Evolving Doctrine on a Changing Court: A Model For A Newer Equal Protection’, Harvard Law Review, Vol. 86, 1972, p. 44. See also B. Sullivan, ‘FOIA and the First Amendment: Representative Democracy and the People’s Elusive “Right to Know”’, Maryland Law Review, Vol. 72, 2012, p. 1-84 (emphasizing importance of access to government information in constitutional democracy). Indeed, nothing is more basic to constitutional democracy than the citizen’s right to speak in the public forum and to cast a meaningful vote for candidates of his or her choice. As the Court said more than 130 years ago in Yick Wo v. Hopkins,²⁹x 118 U.S. 356, 370 (1886). “the political franchise of voting” is fundamental because it is “preservative of all rights”. Those rights must therefore be protected as zealously as possible in any case, but particularly so when the threat to meaningful political participation comes not in the form of lawless actions by private individuals, but as legislation properly enacted by a majority through punctilious compliance with the forms of law. It is far from clear that current Fourteenth Amendment jurisprudence, complicated though it is, adequately protects the right to meaningful political participation through the electoral process. That is the central concern of this essay.
  This article has five parts. First, it discusses the canonical tiered approach to levels of scrutiny and its limitations as an analytical framework. Second, it considers the Court’s rejection of strict scrutiny in the reproductive rights area and the Court’s adoption of the ‘undue burden’ test in its place. Third, the article discusses an analogous movement away from strict scrutiny in the area of voting rights. With extreme political polarization³⁰x See, e.g., Pew Research Center, Political Polarization in the American Public: How Increasing Ideological Uniformity and Partisan Antipathy Affects Politics, Compromise and Everyday Life (June 2014), http://assets.pewresearch.org/wp-content/uploads/sites/5/2014/06/6-12-2014-Political-Polarization-Release.pdf. and gerrymandered legislatures having become the norm in the US,³¹x See, e.g., A.J. McGann et al, Gerrymandering in America: The House of Representatives, the Supreme Court, and the Future of Popular Sovereignty, Cambridge, Cambridge University Press, 2016, p. 2. (“We now have a remarkable situation. Drawing districts with different population sizes is prohibited by the Constitution. However, achieving the same partisan advantage by cleverly manipulating the shape of the districts apparently is permitted.”) legislative majorities are able to enact voting regulations specifically aimed at disadvantaging, in the exercise of the franchise, members of groups they think are unlikely to support them. For a number of reasons, the Court has been unwilling or unable to find a satisfactory solution to this problem, which strikes at the heart of constitutional democracy. Fourth, the article reviews a thoughtful proposal for strengthening the ‘undue burden’ test in the context of reproductive choice³²x See E. Freeman, ‘Giving Casey Its Bite Back: The Role of Rational Basis Review in Undue Burden Analysis’, Harvard Civil Rights-Civil Liberties Law Review, Vol. 48, 2013, p. 279-323. and considers whether a similar methodology might be used to enhance the constitutional protection of voting rights. Finally, a brief summary and conclusion are presented.

• B Two Tiers, Three Tiers and the Theoretical Background

  The beginning student of US constitutional law soon learns that the key to unlocking many (but not all) of the secrets of contemporary equal protection and substantive due process law is to be found in the three-tiered standard of review model that Justice White described in 1985 in City of Cleburne v. Cleburne Living Center, Inc.³³x 473 U.S. 432 (1985). In Cleburne, the sponsors of a proposed group home for developmentally disabled persons challenged the city’s requirement that they seek a special use permit that was not generally required. They argued that mental disability should be considered a ‘quasi-suspect classification’, and therefore subject to a standard of review more exacting than ‘rational basis’. While the Court rejected that argument, it held that the special use permit requirement did not even satisfy the rational basis standard because it appeared to rest only on an irrational prejudice against developmentally disabled persons. Id., at p. 446, 450. Ironically, Cleburne illustrates the danger of putting too much faith in the canonical three-tiered approach, because, while Justice White accurately summarized the three-tiered approach, the result in the case seemingly cannot be explained in those terms. Instead, the mode of analysis used in the Court’s decision appears to be more demanding than the canonical form of rational basis review that the Court purportedly applied. :

  The Equal Protection Clause … commands that no State shall “deny to any person within its jurisdiction the equal protection of the laws,” which is essentially a direction that all persons similarly situated should be treated alike. … Section 5 of the Amendment empowers Congress to enforce this mandate, but absent controlling congressional direction, the courts have themselves devised standards for determining the validity of state legislation or other official action that is challenged [on this ground]. The general rule is that legislation is presumed to be valid and will be sustained if the classification … is rationally related to a legitimate state interest. … When social or economic legislation is at issue, the Equal Protection Clause allows the States wide latitude, … and the Constitution presumes that even improvident decisions will eventually be rectified by the democratic processes.
  The general rule gives way, however, when a statute classifies by race, alienage, or national origin. These factors are so seldom relevant to the achievement of any legitimate state interest that laws grounded in such considerations are deemed to reflect prejudice and antipathy – a view that those in the burdened class are not as worthy or deserving as others. For these reasons and because such discrimination is unlikely to be soon rectified by legislative means, these laws are subjected to strict scrutiny and will be sustained only if they are suitably tailored to serve a compelling state interest. … Similar oversight by the courts is due when state laws impinge on personal rights protected by the Constitution.
  Legislative classifications based on gender also call for a heightened standard of review. That factor generally provides no sensible ground for differential treatment. “[W]hat differentiates sex from such nonsuspect statuses as intelligence or physical disability … is that the sex characteristic frequently bears no relation to ability to perform or contribute to society.” … [S]tatutes distributing benefits and burdens between the sexes in different ways very likely reflect outmoded notions of the relative capabilities of men and women. A gender classification fails unless it is substantially related to a sufficiently important governmental interest. … [O]fficial discriminations resting on [illegitimacy] are also subject to somewhat heightened review. Those restrictions “will survive equal protection scrutiny to the extent they are substantially related to a legitimate state interest.” …
  [On the other hand,] … where individuals in the group affected by a law have distinguishing characteristics relevant to interests the State has the authority to implement, the courts have been very reluctant, as they should be in our federal system and with our respect for the separation of powers, to closely scrutinize legislative choices as to whether, how, and to what extent those interests should be pursued. In such cases, the Equal Protection Clause requires only a rational means to serve a legitimate end.³⁴x Id., at p. 439-442. The Court explained that age had not been deemed to be a suspect or quasi-suspect classification because the aged have not experienced a “‘history of purposeful unequal treatment’ or been subjected to unique disabilities on the basis of stereotyped characteristics not truly indicative of their abilities”. Id., at p. 441.

  One might assume from Justice White’s account that this three-tiered mode of analysis has always provided the polestar for equal protection enforcement – and that it has always been followed religiously – but that is not the case. Intermediate scrutiny did not become part of the Court’s formal analytic framework until at least 1976, when the Court decided Craig v. Boren,³⁵x 429 U.S. 190 (1976). Earlier, in Reed v. Reed, 404 U.S. 71, 76 (1971), the Court had struck down a gender-based classification, purportedly because it failed to satisfy the rational basis test. Also during the 1970s, a plurality of the Court flirted with the idea that gender-based classifications should be evaluated under strict scrutiny, but that view did not command a majority of the Court. See Frontiero v. Richardson, 411 U.S. 677, 688 (1973) (Brennan, J.) (plurality opinion) (“[C]lassifications based on sex, like classifications based on race, alienage, or national origin, are inherently suspect, and must therefore be subject to strict judicial scrutiny.”) As previously noted, some scholars would attach a later date to the emergence of intermediate scrutiny on the ground that the majority in Craig was actually utilizing a more demanding form of rational basis review, rather than articulating a new, intermediate level of scrutiny. See K.R. Eyer, ‘Constitutional Crossroads and the Canon of Rational Basis Review’, University of California, Davis Law Review, Vol. 48, 2014, p. 535. In any event, some Justices clearly did not believe that there was a two-tiered system prior to Craig. In San Antonio v. Rodriguez, 411 U.S. 1, 98-99 (1973) (Marshall, J., dissenting), e.g., Justice Marshall insisted that a ‘principled reading’ of the Court’s jurisprudence revealed a ‘spectrum of standards’ and ‘variations in the degree of care with which the Court will scrutinize various classifications’. In any event, the Court extended the sweep of intermediate scrutiny review to include classifications based on illegitimacy in Clark v. Jeter, 486 U.S. 456, 461-465 (1988). and, at least as a formal matter, the term “[s]trict scrutiny did not appear in racial discrimination equal protection cases until 1978”.³⁶x See S.A. Siegel, ‘The Origin of the Compelling State Interest Test and Strict Scrutiny’, American Journal of Legal History, Vol. 48, 2006, p. 402 (“In that year, Justice Powell, who was not speaking for the Court, employed strict scrutiny in casting the deciding vote in Board of Regents of the University of California v. Bakke [438 U.S. 265 (1978)].”). But see R.H. Fallon, ‘Strict Judicial Scrutiny’, University of California Law Review, Vol. 54, 2007, p. 1276 (noting that, in Bolling v. Sharpe, 347 U.S. 497 (1954) (the companion case to Brown v. Board of Education, 347 U.S. 483 (1954), which involved school desegregation in the District of Columbia), the Court “said that ‘[c]lassifications based solely upon race must be scrutinized with particular care, since they are contrary to our traditions’ and thus … are ‘constitutionally suspect’”.). In Korematsu v. United States, 323 U.S. 214, 216 (1944), of course, Justice Black stated that “courts must subject [legal restrictions which curtail the civil rights of a single racial group] to the most rigid scrutiny”, and Justice Douglas used the term ‘strict scrutiny’ in Skinner v. Oklahoma, 316 U.S. 353, 541 (1942), but that case did not involve a racial classification. Nor, as previously noted, would it be fair to say that the framework has been followed religiously.³⁷x See supra notes 17, 18, and 35 and accompanying text. In any event, as the foregoing excerpt from Cleburne suggests, the Court’s general approach is informed by an understanding that the central meaning of equal protection is that “all persons similarly circumstanced should be treated alike”,³⁸x Cleburne, 437 U.S. at 439. See also F.S. Royster Guano Co. v. Virginia, 253 U.S. 412, 415 (1920). and, presumably, that what counts for ‘likeness’ and ‘non-likeness’ – or, to put it more precisely, relevant and material similarity and difference – should be amenable to normative justification.³⁹x See Plyler v. Doe, 457 U.S. 202, 217 (1982) (“The initial discretion to determine what is ‘different’ and what is ‘the same’ resides in the legislatures of the States. A legislature must have substantial latitude to establish classifications that roughly approximate the nature of the problem perceived, that accommodate competing concerns both public and private, and that account for limitations on the practical ability of the State to remedy every ill.”). See also I. Berlin, ‘Equality’, in H. Hardy (Ed.), Concepts and Categories: Philosophical Essays by Isaiah Berlin, New York City, Viking Press, 1979, p. 98 (“The goodness of the reasons will depend upon the degree of value or importance attached to the purposes or motives adduced in justifying the exceptions, and these will vary as the moral convictions – the general outlooks – of different individuals or societies vary.”). But cf. Aristotle, The Politics of Aristotle, E. Barker (Ed.), Oxford, Oxford University Press, 1958, p. 131. (“[T]here is … no good reason for basing a claim to the exercise of authority on any and every kind of superiority. Some may be swift and others slow; but this is no reason why the one should have more [political rights], and the other less. It is in athletic contests that the superiority of the swift receives its reward.”) In that sense, the Court’s formulation hearkens back to Aristotle⁴⁰x See, e.g., Aristotle, Ethica Nicomachea, Bk.5, §1131a (W.D. Ross, trans. 1925) (“And the same equality will exist between the persons and between the things concerned; for as the latter – the things concerned – are related, so are the former; if they are not equal, they will not have what is equal, but this is the origin of quarrels and complaints – when either equals have and are awarded unequal shares, or unequals equal shares.”). See also P. Westen, ‘The Empty Idea of Equality’, Harvard Law Review, Vol. 95, 1982, p. 543 (summarizing Aristotle: “[e]quality in morals means this: things that are alike should be treated alike, while things that are unalike should be treated unalike in proportion to their unalikeness.”) and an interpretive tradition influenced by Aristotle.⁴¹x See, e.g., Tussman & tenBroek, 1949, p. 341-381 (“[L]aws may classify. And ‘the very idea of classification is that of inequality.’ … [T]he Court has neither abandoned the demand for equality nor denied the legislative right to classify. … It has resolved the[se] contradictory demands … by a doctrine of reasonable classification. … The Constitution … does require … that those who are similarly situated be similarly treated. The measure of the reasonableness of a classification is the degree of its success in treating similarly those similarly situated. The difficulties concealed in this proposition will [need to be] analyzed.”)
  But the recognition that similar cases must be treated similarly, and that the differences between cases must be closely evaluated for relevance and materiality, is only half the story. Once one recognizes that the fact of similarity and dissimilarity (and the relevance and materiality of such similarities and dissimilarities) is a matter of judgment, it becomes important to decide whose judgment should matter. In other words, should the judgment that ultimately prevails be that of the political branches of government, which have acted on the basis of their perception and evaluation of the facts, in light of their own policy values and choices, and consistent, presumably, with their duty to uphold the Constitution?⁴²x See United States Constitution, Art. VI cl. 3 (“The Senators and Representatives before mentioned, and the Members of the several State Legislatures, and all executive and judicial Officers, both of the United States and of the several States, shall be bound by Oath or Affirmation, to support this Constitution.”). Interestingly, some scholars have recently argued that the rational relationship test, by deferring to the people’s representatives, does too little to protect the people themselves. See, e.g., R.E. Barnett, ‘Why Popular Sovereignty Requires the Due Process of Law to Challenge ‘Irrational or Arbitrary’ Statutes’, Georgetown Journal of Law and Public Policy, Vol. 14, 2016, p. 368. (“For the people cannot be presumed to have ‘entrusted to the government’ the power to irrationally or arbitrarily restrict their liberties.”) Or should the controlling judgment be that of the judges, whose fact-finding abilities are limited (insofar as legislative facts and the real stuff of public policy are concerned), and whose individual policy values and choices are institutionally irrelevant⁴³x That is not to say, of course, that judges function like machines in the area of legal interpretation; their views are influenced in one way or another, to one degree or another, by education and experience. See B. Sullivan, ‘The Power of Imagination: Diversity and the Education of Lawyers and Judges’, University of California, Davis Law Review, Vol. 51, 2018, p. 1109. Moreover, as Justice Breyer has noted, the opinions of individual Justices “have emphasized different constitutional themes, objectives, or approaches over time.” See S. Breyer, Active Liberty: Interpreting our Democratic Constitution, New York City, Alfred A. Knopf, 2005, p. 9. For his own part, Justice Breyer would give special emphasis to the concept of constitutional liberty, which he takes to mean “not only freedom from government coercion, but also the freedom to participate in the government itself”. Id., at p. 3. In Justice Breyer’s view, the current Court has not paid sufficient attention to the second aspect, namely the individual citizen’s right to participate in government. Id., at p. 10-11. – but who necessarily have the last word on constitutionality in the US system?⁴⁴x See, e.g., Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177-178 (1803) (C.J. Marshall). (“It is emphatically the province and duty of the Judicial Department to say what the law is. Those who apply the rule to particular cases must, of necessity, expound and interpret the rule. … This is of the very essence of judicial duty.”). See also M.J. Klarman, How ‘Great Were the “Great” Marshall Court Opinions?’, Virginia Law Review, Vol. 87, 2001, p. 1113-117 (showing that judicial review was well established before Marbury).
  As we have seen,⁴⁵x See supra notes 21-24 and accompanying text. given the nature of the US constitutional system, the answer must be that there is – and can be – no simple or categorical answer to that question. It cannot be the case, for example, that the courts should always feel free to substitute their judgment for that of the political branches.⁴⁶x The essential constitutional concern with the question ‘Who decides?’ necessarily precludes the universal application of a de novo proportionality test. See, e.g., N. Gertner, ‘On Competence, Legitimacy, and Proportionality’, University of Pennsylvania Law Review, Vol. 160, 2012, p. 1587. (“Proportionality analysis is simply not within the competence of the American judiciary. Worse yet it is not even within their legitimate role.”) Such a test would greatly simplify the work of the courts, but it would not satisfy the threshold concern as to legitimacy. That is not to say, of course, that proportionality is irrelevant or unimportant. See, e.g., E.T. Sullivan & R.S. Frase, Proportionality Principles in American Law: Controlling Excessive Government Actions, Oxford, Oxford University Press, 2009, p. 6 (noting increasing reliance on proportionality review in US law). But neither can it be the case that the courts should always defer to the political branches. Legislation is normally entitled to a presumption of constitutionality under US law, as Justice White noted in Cleburne, but that presumption sometimes can – and should – count for more or less. Sometimes the courts will look more closely at the rationale for official action and sometimes less so. Sometimes the courts will even supply a rationale where the political branches have either failed to do so or have offered a rationale that simply lacks the power to persuade. In the most general terms, the point of the three-tiered approach is to provide some degree of regularity and predictability as to those circumstances in which the courts should scrutinize more or less closely the work of the political branches.
  Given the substantial degree of doctrinal fluidity that beginning law students encounter in the introductory constitutional law course (in contrast, perhaps, to some other introductory courses, where the relevant doctrine appears more stable), students are normally delighted to encounter this easily grasped and seemingly straightforward three-tiered approach. The students’ delight may be short-lived, of course, as they not only attempt to apply the approach to concrete cases, but also discover that the test itself may not be as simple, as straightforward or as universally applicable as they were led to believe. In addition, some students will wonder about the origins and legitimacy of the test. Where does it say in the text of the Constitution, for example, that possible violations of equal protection and due process are to be identified and analysed in this way?⁴⁷x See, e.g., Craig v. Boren, 429 U.S. 190, 211-212 (1976) (Stevens, J., concurring) (“There is only one Equal Protection Clause. It requires every State to govern impartially. It does not direct the courts to apply one standard of review in some cases and a different standard in other cases.”) How does one derive this tiered approach from the seemingly categorical (if also opaque) guarantees set forth in the relevant clauses?⁴⁸x See, e.g., National Mutual Ins. Co. v. Tidewater Transfer Co., Inc., 337 U.S. 582, 626 (1949) (Frankfurter, J., dissenting) (“Great concepts like ‘Commerce … among the several States,’ ‘due process of law,’ ‘liberty,’ ‘property’ were purposely left to gather meaning from experience. For they relate to the whole domain of social and economic fact, and the statesmen who founded this Nation knew too well that only a stagnant society remains unchanged.”) In one sense, Justice White answered those questions in the quoted excerpt from Cleburne: Section 5 of the Fourteenth Amendment authorizes Congress to enact legislation to enforce the substantive provisions of the Amendment, but those substantive provisions are also self-executing and judicially enforceable.⁴⁹x See, e.g., W. E. Nelson, The Fourteenth Amendment: From Political Principle to Judicial Doctrine, Cambridge, Harvard University Press, 1988, p. 55 (“One noteworthy feature of this proposal, from which the Fourteenth Amendment was ultimately derived, was its apparent adoption of the suggestion of Representative Giles W. Hotchkiss … that any new constitutional provision be framed as a self-executing guarantee of rights, and not merely as a grant of power to Congress to legislate for the protection of rights. … [Hotchkiss] wanted to be certain that rights would be enforced by the judiciary even if Congress fell under Democratic control.”) Many would argue that certain recent decisions reflect the opposite problem, that is, that Congress may have become more solicitous than the Court with respect to the values embodied in the Civil War Amendments. See, e.g., Shelby County v. Holder, 133 S. Ct. 2612 (2013) (striking down the central part of the Voting Rights Act of 1965, which had been re-enacted in 2006 by votes of 98-0 in the Senate and 390-33 in the House of Representatives and signed into law by President George W. Bush). In other words, the Court is responsible for interpreting Section 1, thereby executing its duty “to say what the law is”,⁵⁰x Marbury v. Madison, 5 U.S. (1 Cranch) at p. 177. but Congress and the courts otherwise share responsibility for enforcing the substantive provisions of the Fourteenth Amendment. Both are authorized to formulate appropriate rules and techniques for doing so, and conflicts may therefore arise.⁵¹x See, e.g., R.B. Siegel & R.C. Post, ‘Legislative Constitutionalism and Section Five Power: Policentric Interpretation of the Family and Medical Leave Act’, Yale Law Journal, Vol. 112, 2003, p. 1945 (“Because Section 5 of the Fourteenth Amendment vests in Congress ‘power to enforce, by appropriate legislation, the provisions of this article,’ the great rights contained in Section 1 … are enforced by both Congress and the Court. How to conceive of the relationship between the legislative power established in Section 5 and the judicial power authorized by Section 1 is one of the deep puzzles of American constitutional law.”) See also W.D. Araiza, Enforcing the Equal Protection Clause: Congressional Power, Judicial Doctrine, and Constitutional Law, New York, New York University Press, 2015, p. 17 (The “core insight [of City of Boerne v. Flores, 521 U.S. 507 (1997)] – that enforcement legislation must exhibit some relationship to Court-stated Fourteenth Amendment law – appears here to stay. A court’s scrutiny of that relationship may well be deferential. … Similarly, the Court may have to adjust its understanding of what that underlying Fourteenth Amendment law actually says, and thus what constitutes the target for congruence and proportionality review.”); J.T. Noonan, Narrowing the Nation’s Power: The Supreme Court Sides with the States, Berkeley, University of California Press, 2002, p. 6 (The congruence and proportionality test “means that the federal judiciary, from the Supreme Court itself down to the federal district court in Guam, may, and indeed must, treat Congress the way courts treat an administrative agency, whose work will be set aside on appeal if the court finds the record made by the agency not substantial enough to justify the agency’s rulings”.) In another sense, of course, the answer to the inquiry about origins is more complicated. One could begin by tracing the modern doctrine from Justice Stone’s famous footnote four in United States v. Carolene Products Co.,⁵²x 304 U.S. 144 (1938). In Carolene Products, the Court applied an extremely deferential standard of review to an early statute pertaining to the sale in interstate commerce of ‘filled milk’, that is, milk that was produced through the extraction of its natural cream content and the substitution of another kind of fat or oil for the natural component. In footnote 4, Justice Stone indicated that a more muscular form of constitutional review might sometimes be warranted: “[t]here may be narrower scope for operation of the presumption of constitutionality when legislation appears on its face to be within a specific prohibition of the Constitution, such as those of the first ten Amendments. … It is unnecessary to consider now whether legislation which restricts those political processes which can ordinarily be expected to bring about repeal of undesirable legislation, is to be subjected to more exacting judicial scrutiny. … Nor need we enquire whether similar considerations enter into the review of statutes directed at particular religious … or national, … or racial minorities [or] whether prejudice against discrete and insular minorities may be a special condition … [calling] for a … more searching judicial inquiry.” Id., at p. 152, n. 4. and the Court’s subsequent use of the term ‘strict scrutiny’ in Skinner v. Oklahoma,⁵³x 316 U.S. 353 (1942). In his opinion for the Court, Justice Douglas used the term ‘strict scrutiny’, Id., at p. 541, but recognized that the classification (which allowed for the sterilization of thrice-convicted chicken thieves, but not for embezzlers, regardless of the degree of recidivism) was not supported by any rational basis. Id., at p. 538-539. Chief Justice Stone and Justice Jackson concurred in the result but thought that the case should have been decided under the due process clause, whereas Justice Douglas based his opinion on the equal protection clause. See id., at p. 543-547. but the route from those sources to the Court’s present articulation of the three-tiered approach is far from straightforward.⁵⁴x See, e.g., Fallon, 2007, p. 1267-1337; Siegel, 2006, p. 355-407. And the issue is further complicated, of course, by a divergence of views concerning the appropriate constitutional relationships between the courts and the political branches, on the one hand, and the national and state governments, on the other hand.
  A better question might be why the three-tiered approach takes the precise form that it does. As previously noted, the beginning constitutional law student also learns that the ultimate purpose of this framework is, as Justice White signalled in the cited passage from Cleburne, to articulate a standard approach to judicial review for constitutionality of legislation and other forms of government action that is responsive to a variety of sometimes competing concerns: the need to protect constitutional rights; the desirability of holding government accountable to the rule of law; the need to give effect to the respective functions and roles of the judiciary and the political branches in the constitutional system;⁵⁵x See, e.g., Thayer, 1893, p. 144 (1893) (The Court “can only disregard the Act when those who have the right to make laws have not merely made a mistake, but have made a very clear one – so clear that it is not open to rational question. … This rule recognizes that, having regard to the great, complex, unfolding exigencies of government, much which will seem unconstitutional to one man, or body of men, may reasonably not seem so to another; that the constitution often admits of different interpretations; that there is often a range of choice or judgment; that in such cases the constitution does not impose upon the legislature any one specific opinion, but leaves open this range of choice; and that whatever choice is rational is constitutional”.) Compare Lochner v. New York, 198 U.S. 45, 76 (1905) (J. Holmes, dissenting). (“I think that the word liberty in the Fourteenth Amendment is perverted when it is held to prevent the natural outcome of a dominant opinion, unless it can be said that a rational and fair man necessarily would admit that the statute proposed would infringe fundamental principles as they have been understood by the traditions of our people and our law. … A reasonable man might think [the statute at issue] a proper measure on the score of health. Men whom I certainly could not pronounce unreasonable would uphold it as a first installment of a general regulation of the hours of work. Whether in the latter aspect it would be open to the charge of inequality I think it unnecessary to discuss.”) and the need to avoid unnecessary friction between the state and national components of a federal system.⁵⁶x The Court explained the deference due to state legislation in Skinner v. Oklahoma, 316 U.S. 535 (1942): “[i]t was stated in Buck v. Bell [274 U.S. 200, 208 (1927)] that the claim that state legislation violates the equal protection clause … is ‘the usual last resort of constitutional arguments.’ … [T]he States … need not provide ‘abstract symmetry’. … They may mark and set apart the classes and types of problems according to the needs and as dictated or suggested by experience. … ‘We must remember that the machinery of government would not work if it were not allowed a little play in its joints.’ … [And] the equal protection clause does not prevent the legislature from recognizing ‘degrees of evil’ … ‘[T]he law does all that is needed when it does all that it can, indicates a policy, applies it to all within the lines, and seeks to bring within the lines all similarly situated so far and so fast as its means allow.’” Id., at p. 539-540. See E. Chemerinsky, The Case Against the Supreme Court, City of Westminster, Penguin Publishing, 2014, p. 1-5 (discussing the facts of Buck v. Bell); J.M. Wisdom, ‘The Frictionmaking, Exacerbating Political Role of Federal Courts’, Southwestern Law Journal, Vol. 21, 1967, p. 411-428 (discussing role of federal courts in protecting civil rights plaintiffs against unconstitutional actions by state officials). The problem presented by legislation and regulation is particularly acute because they inevitably classify, whether explicitly or not, and the kind of line drawing involved in the crafting of legislation and regulations involves questions of judgment and policy generally thought to be matters for the political branches. At the same time, of course, those classifications touch on constitutional rights and structures.
  Students also learn that a court’s threshold determination as to which of the three tests should be applied in the circumstances of a particular case will often determine the outcome. A plaintiff might well prevail, for example, if the court decides that the government’s action is subject to strict scrutiny, while the same plaintiff would face certain defeat on the same record if the rational basis standard were to be applied. Such is the power of the tests and the stark differences among them. Thus, rational basis review, the most permissive level of scrutiny, has often been disparaged as ‘a rubber stamp’⁵⁷x See, e.g., Freeman, 2013, p. 282. But see K. R. Eyer, ‘Constitutional Crossroads and the Canon of Rational Basis Review’, University of California, Davis Law Review, Vol. 48, 2014, p. 535-36 (arguing that the current view of rational basis review as ‘toothless’ is in part due to academic amnesia and the omission from the canon of earlier cases in which the courts employed a more muscular form of rationality review). for government action, while strict scrutiny, the most exacting level of review, has famously been characterized as “strict in theory and fatal in fact”.⁵⁸x See Gunther, 1972, p. 8. But see Adarand Constructors v. Pena, 515 U.S. 200, 237 (1995) (“[W]e wish to dispel the notion that strict scrutiny is ‘strict in theory, but fatal in fact.’”).
  One commentator has explained that “[c]ourts consider rational basis review the default standard. To uphold state action under rational basis, a court must only determine that the challenged legislation is reasonably related to a legitimate state interest. … Typically, courts uphold legislation if any conceivable circumstance exists to justify it, and concoct statutory rationales if the state’s proffered interest does not pass constitutional muster. Rational basis applies to equal protection claims that do not implicate gender, suspect classifications or fundamental rights; it also applies in the due process context where no fundamental rights are implicated.”⁵⁹x See Freeman, 2013, p. 282-283. It bears emphasis that courts are not limited under the rational basis test to evaluating the reasons the legislature gave for enacting the legislation. Far from rewarding the thoroughness or thoughtfulness of the legislature, the rational basis standard rewards the creativity of litigators for the state who are called upon to generate some plausible post hoc justification for the government’s action. Moreover, if even the state’s litigators cannot ‘concoct’ a plausible post hoc justification, the courts may concoct one for themselves. In short, “[r]ational basis review places the burden of persuasion on the party challenging a law, who must disprove ‘every conceivable basis which might support it’.”⁶⁰x Heller v. Doe, 509 U.S. 312, 320 (1993) (emphasis added).
  Scholars and courts often group together intermediate and strict scrutiny under the heading of ‘heightened scrutiny’,⁶¹x See, e.g., K. Yoshino, ‘The New Equal Protection’, Harvard Law Review, Vol. 124, 2011, p. 756 (The Court’s framework of tiered scrutiny “distinguishes between classifications that draw ‘heightened scrutiny’ and classifications that draw ‘rational basis review.’”). As Professor Yoshino notes, the Court has not added to the list of characteristics worthy of heightened review since 1977. Id., at p. 756. “The claim that the canon has closed on heightened scrutiny classifications must be tempered by acknowledging the Court’s use of a more aggressive form of rational basis review,” which academic commentators have referred to as ‘rational basis with bite’. Id., at p. 759. but they do not operate in the same way. As Justice White noted in Cleburne, the intermediate scrutiny standard is somewhat more demanding, with respect to both establishing the degree of importance of the state interest thought to be furthered by the challenged classification and the showing of a strong connection between the classification and the end to be achieved. The Supreme Court has relied on intermediate scrutiny in cases involving classifications based on gender and illegitimacy,⁶²x See, e.g., Craig v. Boren, 429 U.S. 190, 197 (1976) (“[C]lassifications by gender must serve important governmental objectives and must be substantially related to those objectives.”). holding that such classifications will fail “unless [they are] substantially related to a sufficiently important governmental interest”.⁶³x City of Cleburne v. Cleburne Living Center, Inc., 473 US 432, 441 (1985). Commentators have identified three tests for determining whether a classification merits intermediate scrutiny review: “[f]irst, is the classifying trait, like race, an immutable personal characteristic – an accident of birth beyond a person’s control or responsibility – rendering it presumptively unjust for the government to use the trait as a basis for allocating rewards or penalties? Second, is the trait, like race, broadly irrelevant to legitimate generalization, rendering discrimination on this basis not only unfair, but also indefensible in a wide range of governmental settings? And third, is the disadvantaged group, like African-Americans and other racial minorities, a group that lacks political power and therefore warrants special judicial solicitude, that is, special protection from the ordinary operation of the political process?” See D.O. Conkle, ‘Evolving Values, Animus, and Same-Sex Marriage’, Indiana Law Journal, Vol. 89, 2014, p. 34. But the Court has been reluctant to extend this more searching standard of review to additional kinds of classifications.⁶⁴x For example, some commentators have argued that sexual orientation should be subject to intermediate scrutiny as well. See, e.g., S.L. Sobel, ‘When Windsor Isn’t Enough: Why the Court Must Clarify Equal Protection Analysis for Sexual Orientation Classifications’, Cornell Journal of Law and Public Policy, Vol. 24, 2015, p. 493-531; K. LaCour, ‘License to Discriminate: How a Washington Florist is Making the Case for Applying Intermediate Scrutiny to Sexual Orientation’, Seattle University Law Review, Vol. 38, 2014, p. 122-124. Some lower courts have also applied an intermediate standard of review in sexual orientation cases, but the Supreme Court has chosen to invalidate certain classifications based on sexual orientation, without specifying the appropriate level of scrutiny. See, e.g., Windsor v. United States, 699 F.3d 169, 181 (2d Cir. 2012) (heightened scrutiny required), aff’d on other grounds, 133 S. Ct. 2675 (2013) (discrimination based on sexual orientation held unconstitutional, without specifying the appropriate level of scrutiny); Lawrence v. Texas, 539 U.S. 558 (2003) (same). See also Romer v. Evans, 517 U.S. 620 (1996) (invalidating state constitutional provision that withdrew previously granted protection against discrimination based on sexual orientation because it was motivated by a bare desire to harm a politically unpopular group).
  Strict scrutiny is the most demanding of the three levels of review. It has been applied to certain so-called ‘fundamental rights’, such as the right to vote⁶⁵x See, e.g., Harper v. Va. State Bd. of Elections, 383 U.S. 663, 667, 670 (1966) (“Long ago, in Yick Wo v. Hopkins, … the Court referred to ‘the political franchise of voting’ as a ‘fundamental political right, because preservative of all rights.’ … We have long been mindful that, where fundamental rights and liberties are asserted under the Equal Protection Clause, classifications which might invade or restrain them must be closely scrutinized and carefully confined.”). and to ‘suspect classifications’, such as race, national origin and religion, which are thought to warrant a higher degree of judicial interrogation.⁶⁶x See, e.g., Adarand Construction, Inc. v. Pena, 515 U.S. 200, 227 (1995) (holding that all racial classifications, whether benign or not, are subject to strict scrutiny); Clark v. Jeter, 486 U.S. 456, 461 (1988) (indicating that classifications based on illegitimacy are similarly subject to strict scrutiny). When a classification warrants strict scrutiny, the government bears a particularly heavy burden. As the Court said recently in setting aside certain race-conscious pupil assignment plans in Parents Involved in Seattle Schools v. Seattle School District No. 1,⁶⁷x 551 U.S. 701 (2007).

  It is well established that when the government distributes burdens or benefits on the basis of individual racial classifications, that action is reviewed under strict scrutiny. … As the Court recently reaffirmed, “racial classifications are simply too pernicious to permit any but the most exact connection between justification and classification.” … In order to satisfy this searching standard of review, the [defendants] must demonstrate that the use of individual racial classifications in the assignment plans here under review is “narrowly tailored” to achieve a “compelling” government interest.⁶⁸x Id., at p. 720. In Parents Involved, the Court struck down certain school assignment plans that sought to achieve racial balance in the public schools. The Court applied strict scrutiny and ultimately found the plans to be unconstitutional, because they took race into account. In a controversial plurality opinion, Chief Justice Roberts suggested that this result was dictated by the Court’s decision in Brown v. Board of Education, 347 U.S. 483, 494 (1954), which he took to hold that any consideration of race in school assignments was subject to strict scrutiny and ordinarily unconstitutional. Parents Involved, 551 U.S. at p. 746. Others have thought Brown to be concerned with the problem of racial classification in aid of prejudice and discrimination or subordination, rather than with the mere existence of racial classification. See e.g. R.B. Siegel, ‘Equality Talk: Antisubordination and Anticlassification Values in Constitutional Struggles over Brown’, Harvard Law Review, Vol. 117, 2004, p. 1470-1547. In Palmore v. Sidoti, 466 U.S. 429, 432 (1984), Chief Justice Burger discussed the rationale for applying strict scrutiny in terms more consistent with Professor Siegel’s anti-subordination theory: “[c]lassifying persons according to their race is more likely to reflect racial prejudice than legitimate public concerns.” In Palmore, the Court set aside a state court judgment that awarded custody to a child’s father simply because the child’s mother had married an African-American after the failure of the relationship that produced the child. The state courts had reasoned that being part of a mixed-race household was not ‘in the best interests’ of the child. Id., at p. 433. The Supreme Court recognized that the child might experience prejudice because of his family situation but concluded: “[t]he question … is whether the reality of private biases and the possible injury they might inflict are permissible considerations for removal of an infant child from the custody of its natural mother. We have little difficulty concluding that they are not.” Id., at p. 433.

  As with intermediate scrutiny, the Court has declined in recent years to extend strict scrutiny review to additional rights or classifications.
  But there are complications. Once students have mastered the rudiments of the canonical approach, they will be asked to look a bit more carefully at the Court’s application of the three levels of scrutiny to see whether the Court’s jurisprudence is really consistent with the three-tiered typology that the Court often treats as if it were exhaustive.
  The Court’s decision in Cleburne is a suitable starting point. In that case, the Court expressly rejected the applicability of intermediate scrutiny to classifications affecting the rights of mentally disabled persons, holding that such classifications should simply be reviewed under the deferential rational basis standard. But the standard of review actually applied by the Cleburne Court seems far removed from the exceptionally deferential, textbook version of rational basis review. The Court’s mode of analysis in Cleburne is indeed more probing than rational basis and has come to exemplify a standard of review commonly known as ‘rational basis with bite’.⁶⁹x The phenomenon seems to have been identified for the first time by Gerald Gunther in 1972. See Gunther, 1972, p. 1-306. Professor Gunther noted that the Court sometimes “found bite in the equal protection clause after explicitly voicing the traditionally toothless minimal scrutiny standard”. Id., at p. 17-18. See also R. Holoszyc-Pimental, ‘Reconciling Rational Basis Review: When Does Rational Basis Bite?’, New York University Law Review, Vol. 90, 2015, p. 2071-2117 (tracing development of jurisprudence); J.B. Smith, ‘The Flaws of Rational Basis with Bite: Why the Supreme Court Should Acknowledge Its Application of Heightened Scrutiny to Classifications Based on Sexual Orientation’, Fordham Law Review, Vol. 73, 2005, p. 2769-2814 (advocating that Court should acknowledge its use of a more searching version of rational basis review in cases involving discrimination based on sexual orientation); R.C. Farrell, ‘Successful Rational Basis Claims in the Supreme Court from the 1971 Term Through Romer v. Evans’, Indiana Law Review, Vol. 32, 1999, p. 370 (tracing development of jurisprudence); G. L. Pettinga, ‘Rational Basis with Bite: Intermediate Scrutiny by Any Other Name’, Indiana Law Journal, Vol. 62, 1987, p. 779-803 (tracing development of jurisprudence).
  Some uncertainty may also exist concerning the contours of the intermediate scrutiny standard of review. In her opinion for the Court in Virginia v. United States,⁷⁰x 518 U.S. 515 (1996). for example, Justice Ginsburg summarized “the Court’s current directions for [evaluating] cases of official classification based on gender”, by stating that

  Focusing on the differential treatment or denial of opportunity for which relief is sought, the reviewing court must determine whether the proffered justification is “exceedingly persuasive.” The burden of justification is demanding and it rests entirely on the State. … The State must show “at least that the [challenged] classification serves ‘important governmental objectives and that the discriminatory means employed’ are ‘substantially related to the achievement of those objectives.’” … The justification must be genuine, not hypothesized or invented post hoc in response to litigation. And it must not rely on overbroad generalizations about the different talents, capacities, or preferences of males and females.⁷¹x Id., at p. 532-533.

  As Justice Ginsburg notes in the foregoing passage, the Court has sometimes articulated the test relevant to sex or gender discrimination as one that places on the State the burden of proffering an “exceedingly persuasive” justification for the classification,⁷²x See, e.g., J.E.B. v. Alabama, 511 U.S. 127, 136-137 (1994); Mississippi University for Women v. Hogan, 458 U.S. 718, 724 (1982): Kirchberg v. Feenstra, 450 U.S. p. 460-461 (1981); Personnel Admin. of Mass. v. Feeney, 442 U.S. 256, 273 (1979). but the Court has otherwise described the state’s burden in such cases in the more traditional terms associated with intermediate scrutiny, namely, the obligation to demonstrate that a discriminatory classification is ‘substantially related’ to the achievement of ‘important governmental objectives’. Justice Ginsburg does not distinguish between the two tests and seems to treat them as substantially the same. In his separate concurrence, however, Chief Justice Rehnquist took issue with Justice Ginsburg’s reliance on the “exceedingly persuasive justification” formulation. The Chief Justice wrote, “[i]t is unfortunate that the Court thereby introduces an element of uncertainty respecting the appropriate test.”⁷³x United States v. Virginia, 518 U.S. 559 (Rehnquist, C.J., concurring in judgment). He continued: “[w]hile terms like ‘important governmental objective’ and ‘substantially related’ are hardly models of precision, they have more content and specificity than does the phrase ‘exceedingly persuasive justification.’ … To avoid introducing potential confusion, I would have adhered more closely to our traditional … standard that a gender-based classification ‘must bear a close and substantial relationship to important governmental objectives.’”⁷⁴x Id., at p. 559. It is unclear, of course, whether Justice Ginsburg was attempting to state a more demanding formulation of the state’s burden – as the Chief Justice seems to have assumed – or was simply stating the test as she thought the Court had developed.
  Finally, the Court has seriously split in recent years with respect to the proper application of strict scrutiny. In Fisher v. University of Texas,⁷⁵x 136 S. Ct. 2198 (2016). which upheld the university’s affirmative action programme against the claim that it constituted impermissible race-based discrimination, the dissenting Justices did not simply disagree about the outcome, but viewed it as profoundly incompatible with any competent application of the strict scrutiny standard. Justice Thomas, for example, observed that “[t]he Court’s decision … is irreconcilable with strict scrutiny, rests on pernicious assumptions about race, and departs from many of our precedents.”⁷⁶x Id., at p. 2215 (Thomas, J., dissenting). Similarly, Justice Alito observed that

  UT’s race-conscious admissions program cannot satisfy strict scrutiny. UT says that the program furthers its interest in the educational benefits of diversity, but it has failed to define that interest with any clarity or to demonstrate that its program is narrowly tailored to achieve that or any other particular interest. By accepting UT’s rationales as sufficient to meet its burden, the majority licenses UT’s perverse assumptions about different groups of minority students – the precise assumptions strict scrutiny is supposed to stamp out.⁷⁷x Id., at p. 2220 (Alito, J., dissenting).

  To underscore the point, Justice Alito suggested that the majority’s application of strict scrutiny was unfaithful to one of its most basic aspects, that is, the principle that the burden of proof rests with the state: “[t]ellingly, the Court frames its analysis as if petitioner bears the burden of proof. … But it is not the petitioner’s burden to show that the consideration of race is unconstitutional. To the extent the record is inadequate, the responsibility lies with UT.”⁷⁸x Id., at p. 2238. Justice Alito continued: “[f]or ‘[w]hen a court subjects governmental action to strict scrutiny, it cannot construe ambiguities in favor of the State,’ … particularly where, as here, the summary judgment posture obligates the Court to view the facts in the light most favorable to petitioner”.⁷⁹x Id., Justice Alito continued, noting that, “[e]ven though UT has never provided any coherent explanation for its asserted need to discriminate on the basis of race, and even though UT’s position relies on a series of unsupported and noxious racial assumptions, the majority concludes that UT has met its heavy burden. This conclusion is remarkable – and remarkably wrong.” Id., at p. 2243.
  Whether Justices Alito and Thomas are correct in perceiving a weakening of the strict scrutiny standard remains to be seen. What seems clear, however, is that the Court may be divided with respect to the level of specificity that is necessary to satisfy the strict scrutiny standard, at least in some circumstances, such as cases of “benign” racial “discrimination” involving access to higher education. If that is the case, the Court certainly has not explained it in those terms and is unlikely to do so in light of prior jurisprudence. On the other hand, the prior jurisprudence would suggest that the holding in Fisher is a fragile one and might amount to little more than “a restricted railway ticket, good for this day and train only”,⁸⁰x See Smith v. Allwright, 321 U.S. 649, 669 (1944) (Roberts, J., dissenting). to use Justice Owen Roberts’s memorable phrase.
  On closer inspection, therefore, the three-tiered approach to review appears less tidy and straightforward than it did at first blush. But the ‘undue burden’ standard – and the displacement of strict scrutiny – complicates matters even more.

• C From Strict Scrutiny to Undue Burden: The Right to Choose

  In 1973, the Supreme Court decided Roe v. Wade,⁸¹x 410 U.S. 113 (1973). in which the Court held that a woman’s right to choose whether to continue a pregnancy to term was a fundamental right protected by the Constitution.⁸²x Id., at p. 153-155. Justice Blackmun summarized the grounds on which the constitutionality of the Texas statutes was challenged: “[t]he principal thrust of appellant’s attack … is that they improperly invade a right, said to be possessed by the pregnant woman, to choose to terminate her pregnancy. Appellant would discover this right in the concept of personal “liberty” embodied in the Fourteenth Amendment’s Due Process Clause; or in personal, marital, familial, and sexual privacy said to be protected by the Bill of Rights or its penumbras, see Griswold v. Connecticut, 381 U.S. 479 (1965); Eisenstadt v. Baird, 405 U.S. 438 (1972); Id., at p.460 (White, J., concurring in result); or among those rights reserved to the people by the Ninth Amendment, Griswold v. Connecticut, 381 U.S. at p. 486 (Goldberg, J., concurring).” Roe, 410 U.S. at p. 129. The argument, based on the existence of a liberty interest embodied in the Fourteenth Amendment, properly finds its source in Justice Harlan’s dissenting opinion in Poe v. Ullman, 367 U.S. 497, 543 (1961) (Harlan, J., dissenting). As Justice Souter later observed in Washington v. Glucksberg, 521 U.S. 702, 752, 756 n.2 (1997) (Souter, J., concurring in judgment), the Supreme Court’s modern substantive due process jurisprudence is uniquely indebted to Justice Harlan’s dissenting opinion in Poe. As with other fundamental rights, however, the Court recognized that the right to choose an abortion was not absolute and that the scope of the right was subject to adjustment in light of other important governmental interests. For example, “[t]he State has a legitimate interest in seeing to it that abortion, like any other medical procedure, is performed under circumstances that insure maximum safety for the patient. This interest obviously extends at least to the performing physician and his staff, to the facilities involved, to the availability of after-care, and to adequate provision for any complication or emergency that might arise.”⁸³x Roe, 410 U.S. at p. 150. But the Court further recognized that the state also has another legitimate interest: “as long as … potential life is involved, the State may assert interests beyond the protection of the pregnant woman alone.”⁸⁴x Id., Although the Court recognized the state’s interest in protecting ‘prenatal life’, the Court rejected the view that a fetus was a person within the meaning of the Fourteenth Amendment. Id., at p. 157-158. That remains the case today. Nonetheless, because it understood reproductive choice to be a fundamental right, the Roe Court held that strict scrutiny was the appropriate standard for reviewing any state-imposed limitations. Such limitations “may [therefore] be justified only by a compelling state interest and … must be narrowly drawn to express only the legitimate state interests at stake”.⁸⁵x Id., at p. 155. The Court then articulated its now-famous trimester-based approach, whereby it divided pregnancy into three trimesters and stated that the balance between the interests of the woman and the state should be calibrated differently in each of the three trimesters.⁸⁶x Id., at p. 162-166. The Court summarized its approach: “(a) For the stage prior to approximately the end of the first trimester, the abortion decision and its effectuation must be left to the medical judgment of the pregnant woman’s attending physician. (b) For the stage subsequent to approximately the end of the first trimester, the State, in promoting its interest in the health of the mother, may, if it chooses, regulate the abortion procedure in ways that are reasonably related to maternal health. (c) For the stage subsequent to viability, the State in promoting its interest in the potentiality of human life may, if it chooses, regulate, and even proscribe, abortion except where it is necessary, in appropriate medical judgment, for the preservation of the life or health of the mother.” Id., at p. 164-165. The Court reasoned:

  With respect to the State’s important and legitimate interest in the health of the mother, the “compelling” point, in the light of present medical knowledge, is at approximately the end of the first trimester. This is so because of the now-established medical fact … that, until the end of the first trimester mortality in abortion may be less than mortality in normal childbirth. It follows that, from and after this point, a State may regulate the abortion procedure to the extent that the regulation reasonably relates to the preservation and protection of maternal health. Examples of permissible state regulation in this area are requirements as to the qualifications of the person who is to perform the abortion; as to the licensure of that person; as to the facility in which the procedure is to be performed, that is, whether it must be a hospital or may be a clinic or some other place of less-than-hospital status; as to the licensing of the facility; and the like.
  This means, on the other hand, that, for the period of pregnancy prior to this “compelling” point, the attending physician, in consultation with his patient, is free to determine, without regulation by the State, that, in his medical judgment, the patient’s pregnancy should be terminated…. [T]he judgment may be effectuated by an abortion free of interference by the State.
  With respect to the State’s important and legitimate interest in potential life, the “compelling” point is at viability. This is so because the fetus then presumably has the capability of meaningful life outside the mother’s womb. State regulation protective of fetal life after viability thus has both logical and biological justifications. If the State is interested in protecting fetal life after viability, it may go so far as to proscribe abortion during that period, except when it is necessary to preserve the life or health of the mother.⁸⁷x Id., at p. 163-164.

  Although only two Justices dissented from the Court’s decision in Roe,⁸⁸x Justices Rehnquist and White both dissented, believing that the issues in Roe should be left to the legislative process. Id., at p. 221 (White, J., dissenting); id., at p. 223 (Rehnquist, J., dissenting). several filed separate concurring opinions.⁸⁹x Id., at p. 208 (Burger, C.J., concurring); id., at p. 217 (Douglas, J., concurring); id., at p. 167 (Stewart, J., concurring). As time went by, the decision became a lightning rod and was probably as controversial, both within the legal community and among the general public, as any Supreme Court decision since Brown v. Board of Education ⁹⁰x 347 U.S. 483 (1954). See, e.g., M. Ziegler, ‘Beyond Backlash: Legal History, Polarization, and Roe v. Wade’, Washington and Lee Law Review, Vol. 71, 2014, p. 969-1021. or Engel v. Vitale.⁹¹x 370 U.S. 421 (1962). In Engel, the Court held that New York school officials violated the Establishment Clause by requiring students to recite a government-authored prayer at the beginning of the school day, which had been a long-standing tradition, in one form or another, throughout the United States.
  The Court revisited the subject almost ten years later in City of Akron v. Akron Center for Reproductive Health.⁹²x 462 U.S. 416 (1982). By then, Justice Stevens had taken Justice Douglas’s seat on the Court, and Justice O’Connor had replaced Justice Stewart. In an opinion by Justice Powell, the Court invalidated certain provisions of the Akron ordinance, but specifically reaffirmed Roe and its trimester scheme.⁹³x Id., at p. 420. In a challenge to portions of Akron’s ordinances regulating the conduct of abortions, the Court “affirm[ed] the judgment of the Court of Appeals invalidating those sections … that deal with parental consent, informed consent, a 24-hour waiting period, and the disposal of fetal remains [and reversed that] portion of the judgment [that] sustain[ed] Akron’s requirement that all second trimester abortions be performed in a hospital”. Id., at p. 452. According to Justice Powell, the Akron ordinances were inconsistent with the Court’s understanding in Roe that “[f]rom approximately the end of the first trimester of pregnancy, the State ‘may regulate the abortion procedure to the extent that the regulation reasonably relates to the preservation and protection of maternal health’” and that, even in the second trimester, the regulation must be consistent with “accepted medical practice” and “legitimately related to the objective the State seeks to accomplish”. Id., at p. 430-431. The case is significant, however, because of Justice O’Connor’s dissent, in which she argued that the Court’s test was unworkable and should be replaced with an ‘undue burden’ test. In an opinion joined by Justices Rehnquist and White, Justice O’Connor wrote: “Our recent cases indicate that a regulation imposed on ‘a lawful abortion is not unconstitutional unless it unduly burdens the right to seek an abortion.’ … In my view, this ‘unduly burdensome’ standard should be applied to the challenged regulations throughout the entire pregnancy without reference to the particular ‘stage’ of pregnancy involved. If the particular regulation does not ‘unduly burde[n]’ the fundamental right, … then our evaluation of that regulation is limited to our determination that the regulation rationally relates to a legitimate state purpose.”⁹⁴x Id., at p. 453. She also argued that the test was inconsistent with the Court’s more general fundamental rights jurisprudence. See id., at p. 452-453.
  Three years later, when the Court invalidated several portions of a Pennsylvania abortion law in Thornburgh v. American College of Obstetricians and Gynecologists,⁹⁵x 476 U.S. 747 (1986). In Thornburgh, the Court, speaking through Justice Blackmun, specifically reaffirmed Roe and invalidated several provisions of a Pennsylvania abortion law. Id., at p. 759 (“The States are not free, under the guise of protecting maternal health or potential life, to intimidate women into continuing pregnancies. Appellants claim that the statutory provisions before us today further legitimate compelling interests of the Commonwealth. Close analysis of those provisions, however, shows that they wholly subordinate constitutional privacy interests and concerns with maternal health in an effort to deter a woman from making a decision that, with her physician, is hers to make.”) Justice Stevens filed an important concurrence, in which he responded to several points made by Justice White’s dissent. Chief Justice Burger, Justice Rehnquist and Justice O’Connor also filed dissenting opinions. Justice Rehnquist joined in both Justice White’s and Justice O’Connor’s dissents. Justice O’Connor once more dissented and again invoked the notion of ‘undue burden’, which she now defined as an “absolute obstacle or severe limitation” on the right:

  The State has compelling interests in ensuring maternal health and in protecting potential human life, and these interests exist “throughout pregnancy.” … Under this Court’s fundamental-rights jurisprudence, judicial scrutiny of state regulation of abortion should be limited to whether the state law bears a rational relationship to legitimate purposes such as the advancement of these compelling interests, with heightened scrutiny reserved for instances in which the State has imposed an “undue burden” on the abortion decision. … An undue burden will generally be found “in situations involving absolute obstacles or severe limitations on the abortion decision,” not wherever a state regulation “may inhibit’ abortions to some degree.” … And if a state law does interfere with the abortion decision to an extent that is unduly burdensome, so that it becomes “necessary to apply an exacting standard of review,” … the possibility remains that the statute will withstand the stricter scrutiny.⁹⁶x Id., at p. 828. Justice O’Connor continued: “[t]hese principles for evaluating state regulation of abortion were not newly minted in my dissenting opinion in Akron. Apart from Roe’s outmoded trimester framework, the ‘unduly burdensome’ standard had been articulated and applied with fair consistency by this Court in cases such as Harris v. McRae, 448 U.S. 297, 314 (1980), Maher v. Roe, 432 U.S. 464, 473 (1977), Beal v. Doe, 432 U.S. 438, 446 (1977), and Bellotti v. Baird, 428 U.S. 132, 147 (1976). In Akron and Ashcroft, the Court, in my view, distorted and misapplied this standard, see Akron, 462 U.S. at p. 452-453 (O’Connor, J., dissenting), but made no clean break with precedent, and indeed ‘follow[ed] this approach’ in assessing some of the regulations before it in those cases. Id., at p. 463 (O’Connor, J., dissenting).” Thornburgh, 476 U.S. at p. 828-829.

  When the Court took up Webster v. Reproductive Health Services ⁹⁷x 492 U.S. 490 (1989). three years later, Justices Scalia and Kennedy had joined the Court, and Justice Rehnquist had become Chief Justice. In Webster, the state specifically asked the Court to overrule Roe, and many thought that would happen. The Court declined to do so, however, because a majority of the Justices did not believe that the case presented an appropriate occasion for reconsidering Roe.⁹⁸x Chief Justice Rehnquist, in a plurality opinion joined by Justices White and Kennedy, strongly criticized the holding in Roe: “We have not refrained from reconsideration of a prior construction of the Constitution that has proved ‘unsound in principle and unworkable in practice.’ … [T]he rigid Roe [trimester] framework is hardly consistent with the notion of a Constitution cast in general terms, as ours is, and usually speaking in general principles, as ours does. The key elements of the Roe framework – trimesters and viability – are not found in the text of the Constitution, or in any place else one would expect to find a constitutional principle. Since the bounds of the inquiry are essentially indeterminate, the result has been a web of legal rules that have become increasingly intricate, resembling a code of regulations rather than a body of constitutional doctrine. … [T]he trimester framework has left this Court to serve as the country’s ‘ex officio medical board.’” Id., at p. 518-19. He also faulted the line drawn in Roe between the state’s pre-viability and viability interests. Id., at p. 519. But the testing requirement at issue was “reasonably designed to ensure that abortions are not performed where the fetus is viable – an end which all concede is legitimate – and … sufficient to sustain its constitutionality”. Id., at p. 520. Writing separately, Justice Scalia also thought that Roe should be explicitly overruled. Id., at p. 532. Justice Blackmun, Justice Brennan and Justice Marshall dissented. Justice Blackmun saw the writing on the wall: “[f]or today, at least, the law of abortion stands undisturbed. … But the signs are … very ominous, and a chill wind blows.” Id., at p. 557. Justice O’Connor observed in a critical concurring opinion that there was no need to accept the “invitation to reexamine the constitutional validity of Roe” because the challenged “viability testing requirements [did not] conflict with any of the Court’s past decisions concerning state regulation of abortion”.⁹⁹x Id., at p. 525. But she also reconfirmed the vulnerability of Roe, saying, “there will be time enough to examine Roe [when the issue is properly presented]. And to do so carefully”.¹⁰⁰x Id., at p. 526. Significantly, Justice O’Connor once again invoked the ‘undue burden’ test:

  I dissented from the Court’s opinion in Akron because it was my view that, even apart from Roe’s trimester framework, … the Akron majority had distorted and misapplied its own standard for evaluating state regulation of abortion which the Court had applied with fair consistency in the past: that, previability, “a regulation imposed on a lawful abortion is not unconstitutional unless it unduly burdens the right to seek an abortion.”
  It is clear to me that requiring the performance of examinations and tests useful to determining whether a fetus is viable, when viability is possible, and when it would not be medically imprudent to do so, does not impose an undue burden on a woman’s abortion decision.¹⁰¹x Id., at p. 529-530.

  The most significant post-Roe ruling came down in 1992, when the Court decided Planned Parenthood of Southeastern Pennsylvania v. Casey.¹⁰²x 505 U.S. 833 (1992). In its petition for Supreme Court review, Planned Parenthood framed the question presented as “Whether the Supreme Court overruled Roe v. Wade, holding that a woman’s right to choose abortion is a fundamental right protected by the United States Constitution?” Planned Parenthood v. Casey, 505 U.S. 833 (1992), Petition for A Writ of Certiorari at i (No.91-744). As Jeffrey Toobin has pointed out, the question was extremely provocative, essentially suggesting that the Court might have decided that Roe was no longer binding precedent, without being forthright about it. See J. Toobin, The Nine: Inside the Secret World of the United States Supreme Court, New York, Anchor, 2007, p. 49. The United States took the position that Roe should be overruled. See Planned Parenthood v. Casey, 505 U.S. 833 (1992), Brief for the United States as Amicus Curiae (No. 91-744). In a highly unusual move, three Justices – Justices O’Connor, Kennedy, and Souter – signed a joint, plurality opinion.¹⁰³x Id., at p. 843. Justices Blackmun and Stevens each filed opinions (see id., at p. 911 Stevens, J., concurring in part and dissenting in part; id., at p. 922 Blackmun, J., concurring in part, concurring in the judgment in part, and dissenting in part), as did Chief Justice Rehnquist (speaking for himself and Justices White, Scalia and Thomas) and Justice Scalia (speaking for himself, Chief Justice Rehnquist and Justices White and Thomas). See id., at p. 944 (Rehnquist, C.J., concurring in the judgment in part and dissenting in part); id., at p. 979 (Scalia, J., concurring in the judgment in part and dissenting in part). They emphasized the critical importance of precedent in constitutional law, the link between stability in the law and public confidence in the Court, and the narrow circumstances in which precedents might properly be set aside. The plurality found that those circumstances were not present here.¹⁰⁴x Id., at p. 854-69. Among other things, the plurality found that scientific advances had made the Roe Court’s trimester scheme obsolete and that the analytic framework should be centred on viability. Id., at p. 860, 870. The trimester scheme also was seen to “undervalue […] the potential life within the woman”. Id., at p. 875. In any event, the plurality “reject[ed] the trimester framework, which we do not consider to be part of the essential holding of Roe”. Id., at p. 873. Although the plurality emphasized the need to follow Roe, they also sought to distil its ‘essential holding’ from its ‘non-essential’ aspects, and to give effect only to the former.¹⁰⁵x Id., at p. 846. The plurality wrote that “[a] decision to overrule Roe’s essential holding under the existing circumstances would address error, if error there was, at the cost of both profound and unnecessary damage to the Court’s legitimacy, and to the Nation’s commitment to the rule of law. It is therefore imperative to adhere to the essence of Roe’s original decision, and we do so today.” Id., at p. 869. Chief Justice Rehnquist mocked the plurality’s “newly minted variation on stare decisis”, id., at p. 944 (Rehnquist, C.J., concurring in the judgment and dissenting in part) and catalogued all of the plurality’s disagreements with Roe. Id., at p. 953-954. Most significantly, “Roe decided that a woman had a fundamental right to an abortion. The joint opinion rejects that view. Roe decided that abortion regulations were to be subjected to ‘strict scrutiny’ and would be justified only in the light of ‘compelling state interests.’ The joint opinion rejects that view.” Id., at p. 954. The plurality described Roe’s ‘essential holding’ as follows:

  Roe’s essential holding, the holding we reaffirm, has three parts. First is a recognition of the right of the woman to choose to have an abortion before viability and to obtain it without undue interference from the State. Before viability, the State’s interests are not strong enough to support a prohibition of abortion or the imposition of a substantial obstacle to the woman’s effective right to elect the procedure. Second is a confirmation of the State’s power to restrict abortions after fetal viability, if the law contains exceptions for pregnancies which endanger the woman’s life or health. And third is the principle that the State has legitimate interests from the outset of the pregnancy in protecting the health of the woman and the life of the fetus that may become a child. These principles do not contradict one another; and we adhere to each.¹⁰⁶x Id., at p. 846.

  Summing up, the plurality stated its understanding of the central meaning of Roe: “it is a constitutional right of the woman to have some freedom to terminate her pregnancy.”¹⁰⁷x Id., at p. 869. In a somewhat strange turn of phrase, the plurality then noted that “the basic decision in Roe was based on a constitutional analysis which we cannot now repudiate”. Id. (emphasis added). In the same vein, the plurality observed that “[t]he woman’s right to terminate her pregnancy before viability is the most central principle of Roe v. Wade. It is a rule of law and component of liberty that we cannot renounce.” Id., at p. 871 (emphasis added). Chief Justice Rehnquist construed these expressions, among others, as evidence that the plurality lacked enthusiasm for defending the merits of Roe. See id., at p. 954 (the plurality “cannot bring itself to say that Roe was correct as an original matter”). But that point may understate the significance of the plurality’s insistence on what it took to be the ‘most central principle’ of the case. The plurality added: “[t]he woman’s liberty is not so unlimited, however, that from the outset the State cannot show its concern for the life of the unborn, and at a later point in fetal development the State’s interest in life has sufficient force so that the right of the woman to terminate the pregnancy can be restricted.”¹⁰⁸x Id., at p. 871. The plurality further noted that,

  That portion of the decision in Roe [emphasizing the state’s ‘important and legitimate interest in potential life’] has been given too little acknowledgment and implementation by the Court in its subsequent cases. Those cases decided that any regulation touching upon the abortion decision must survive strict scrutiny, to be sustained only if drawn in narrow terms to further a compelling state interest. … Not all of the cases decided under that formulation can be reconciled with the holding in Roe itself that the State has legitimate interests in the health of the woman and in protecting the potential life within her. In resolving this tension, we choose to rely upon Roe, as against the later cases.¹⁰⁹x Id.

  More specifically, the plurality observed that the state may regulate (but not prohibit) abortion before viability, but that it can prohibit abortions once viability has been reached. Significantly, the plurality compared the law relating to abortion with that concerning the right to vote:

  As our jurisprudence relating to all liberties save perhaps abortion has recognized, not every law which makes a right more difficult to exercise is, ipso facto, an infringement of that right. An example clarifies the point. We have held that not every ballot access limitation amounts to an infringement on the right to vote. Rather, the States are granted substantial flexibility in establishing the framework within which voters choose the candidates for whom they wish to vote.¹¹⁰x Id., at p. 873-874.

  Analogizing the right to choose an abortion to the right to vote, the plurality concluded that, contrary to Roe, the woman’s right to choose was not absolute at any stage of her pregnancy. The plurality noted that, “[b]efore viability, Roe and subsequent cases treat all governmental attempts to influence a woman’s decision on behalf of the potential life within her as unwarranted. This treatment is, in our judgment, incompatible with the recognition that there is a substantial state interest in potential life throughout pregnancy”.¹¹¹x Id., at p. 876. The plurality further noted that “[t]he very notion that the State has a substantial interest in potential life leads to the conclusion that not all regulations must be deemed unwarranted. Not all burdens on the right to decide whether to terminate a pregnancy will be undue”.¹¹²x Id. Thus, the plurality concluded, “the undue burden standard is the appropriate means of reconciling the State’s interest with the woman’s constitutionally protected liberty”.¹¹³x Id.
  As we have seen, Justice O’Connor had made reference to the ‘undue burden’ standard in separate opinions in earlier cases, but the phrase would now be defined in a somewhat different way. In Thornburgh, Justice O’Connor had said that “[a]n undue burden will generally be found ‘in situations involving absolute obstacles or severe limitations on the abortion decision,’ not wherever a state regulation ‘may inhibit’ abortions to some degree".¹¹⁴x Thornburgh v. American College of Obstetricians & Gynecologists, 476 U.S. at p. 828 (O’Connor, J., dissenting). In addition, the plurality took up Justice O’Connor’s suggestion in City of Akron that “this ‘unduly burdensome’ standard should be applied to the challenged regulations throughout the entire pregnancy without reference to the particular ‘stage’ of pregnancy involved.” City of Akron v. Akron Center for Reproductive Health, 462 U.S. at p. 453 (O’Connor, J., dissenting). In Casey, by contrast, the plurality stated that “[a] finding of an undue burden is a shorthand for the conclusion that a state regulation has the purpose or effect of placing a substantial obstacle in the path of a woman seeking an abortion of a nonviable fetus”.¹¹⁵x Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. at p. 877. The plurality also specifically held that “measures designed to advance [the State’s] interest [in ensuring that the woman’s choice is properly informed] will not be invalidated as long as their purpose is to persuade the woman to choose childbirth over abortion”, and the measures do not unduly burden “her right of choice”.¹¹⁶x Id., at p. 878. The plurality also recognized the state’s interest in promulgating appropriate medical regulations but stated that “unnecessary” regulations that have the “purpose or effect of presenting a substantial obstacle to a woman seeking an abortion [would constitute] an undue burden on the right”.¹¹⁷x Id. Moreover, while “a State may not prohibit any woman from making the ultimate decision to terminate her pregnancy before viability”, the state may, subsequent to viability, “regulate, and even proscribe, abortion except where it is necessary … for the preservation of the life or health of the mother’”.¹¹⁸x Id., at p. 878-879 (quoting Roe v. Wade, 410 U.S. at 164-165). Finally, the plurality explained that, in determining whether an obstacle is an undue burden, the “proper focus … is the group for whom the law is a restriction, not the group for whom the law is irrelevant”, because the validity of legislation is “measured … by its impact on those whose conduct it affects”.¹¹⁹x Id., at p. 894.
  Justice Stevens, who concurred and dissented in part, observed that “[c]ontrary to the suggestion of the joint opinion, … it is not a ‘contradiction’ to recognize that the State may have a legitimate interest in potential human life and, at the same time, to conclude that that interest does not justify the regulation of abortion before viability (although other interests, such as maternal health, may). The fact that the State’s interest is legitimate does not tell us when, if ever, that interest outweighs the pregnant woman’s interest in personal liberty. It is appropriate, therefore, to consider more carefully the nature of the interests at stake”.¹²⁰x Id., at p. 914. Justice Blackmun (the author of Roe) also concurred and dissented in part. He fundamentally disagreed with the standard of review adopted by the plurality: “[t]oday, no less than yesterday, the Constitution and decisions of this Court require that a State’s abortion restrictions be subjected to the strictest of judicial scrutiny.”¹²¹x Id., at p. 925. In addition, according to Justice Blackmun, the “application of [the trimester] analytical framework is no less warranted than when it was approved by seven Members of this Court in Roe. Strict scrutiny of state limitations on reproductive choice still offers the most secure protection of the woman’s right to make her own reproductive decisions, free from state coercion.” Id., at p. 930.
  Chief Justice Rehnquist and Justice Scalia each wrote an opinion concurring in part and dissenting in part; both also joined in the other’s opinion, and Justices White and Thomas also joined both opinions.¹²²x See id., at p. 944 (Rehnquist, C.J., concurring in part and dissenting in part); id., at p. 979 (Scalia, J., concurring in part and dissenting in part). The Chief Justice thought that Roe was wrongly decided,¹²³x Id., at p. 944. but that the case also was distinguishable because it involved a prohibition of abortion, whereas Casey involved only its regulation.¹²⁴x Id., at p. 945. Rejecting the plurality’s ‘undue burden’ standard as “an unjustified constitutional compromise, one which leaves the Court in a position to closely scrutinize all types of abortion regulations despite the fact that it lacks the power to do so under the Constitution”,¹²⁵x Id. Chief Justice Rehnquist was particularly critical of the plurality’s ‘undue burden’ standard: “Roe v. Wade adopted a ‘fundamental right’ standard under which state regulations could survive only if they met the requirement of ‘strict scrutiny.’ While we disagree with that standard, it at least had a recognized basis in constitutional law at the time Roe was decided. The same cannot be said for the ‘undue burden’ standard, which is created largely out of whole cloth by the authors of the joint opinion. It is a standard which even today does not command the support of a majority of this Court. And it will not, we believe, result in the sort of ‘simple limitation,’ easily applied, which the joint opinion anticipates. … In sum, it is a standard which is not built to last.” Id., at p. 964-965. the Chief Justice thought that “the correct analysis is that … [a] woman’s interest in having an abortion is a form of liberty protected by the Due Process Clause, but States may regulate abortion in ways rationally related to a legitimate state interest”.¹²⁶x Id., at p. 966, citing Williamson v. Lee Optical Co. of Oklahoma, Inc., 348 U.S. 483, 491 (1955). In other words, abortion regulations should be measured according to the most deferential possible standard of review. For his part, Justice Scalia emphasized that the regulation of abortion was a matter for resolution by the political process.¹²⁷x Id., at p. 1002. (“[B]y foreclosing all democratic outlet for the deep passions this issue arouses, by banishing the issue from the political forum that gives all participants, even the losers, the satisfaction of a fair hearing and an honest fight, by continuing the imposition of a rigid national rule instead of allowing for regional differences, the Court merely prolongs and intensifies the anguish.”)
  Finally, in Stenberg v. Carhart,¹²⁸x 530 U.S. 914 (2000). a majority of the Court explicitly adopted the ‘undue burden’ test, which it applied to strike down a state statute prohibiting a controversial procedure sometimes called ‘partial birth abortion’. Speaking for the majority, Justice Breyer wrote:

  Millions of Americans believe that life begins at conception and consequently that an abortion is akin to causing the death of an innocent child; they recoil at the thought of a law that would permit it. Other millions fear that a law that forbids abortion would condemn many American women to lives that lack dignity, depriving them of equal liberty and leading those with least resources to undergo illegal abortions with the attendant risks of death and suffering. Taking account of these virtually irreconcilable points of view, aware that constitutional law must govern a society whose different members sincerely hold directly opposing views, and considering the matter in light of the Constitution’s guarantees of fundamental individual liberty, this Court … has determined and then redetermined that the Constitution offers basic protection to the woman’s right to choose. …
  Three established principles determine the issue before us. … First, before ‘‘viability the woman has a right to choose to terminate her pregnancy.’’
  Second, ‘‘a law designed to further the State’s interest in fetal life which imposes an undue burden on the woman’s decision before fetal viability’’ is unconstitutional. An ‘‘undue burden is shorthand for the conclusion that a state regulation has the purpose or effect of placing a substantial obstacle in the path of a woman seeking an abortion of a nonviable fetus.’’
  Third, ‘‘‘subsequent to viability, the State in promoting its interest in the potentiality of human life may, if it chooses, regulate, and even proscribe, abortion except where it is necessary, in appropriate medical judgment, for the preservation of the life or health of the mother.’”¹²⁹x Id., at p. 920-21. Justices Stevens, O’Connor and Ginsburg filed concurring opinions, while Chief Justice Rehnquist and Justices Kennedy, Scalia and Thomas filed dissenting opinions.

  In subsequent cases, the Court has continued to apply the undue burden test in cases pertaining to the constitutionality of regulations relating to reproductive choice.¹³⁰x See Whole Woman’s Health v. Hellerstedt, 136 S.Ct. 2292, 2309-2310 (2016) (holding that certain regulatory provisions constituted an undue burden because they did not afford “medical benefits sufficient to justify the burdens [that they imposed] upon access”); Gonzales v. Carhart, 550 U.S. 124, 167-168 (2007) (distinguishing the statute invalidated in Stenberg and holding that those challenging the facial validity of a federal statute prohibiting so-called ‘partial birth abortions’ had failed to demonstrate that the statute “would be unconstitutional in a large fraction of relevant cases”). Some commentators have criticized the test on various grounds, including the difficulty of its application.¹³¹x See, e.g., Freeman, 2013, p. 279 (noting that Casey “has engendered confusion rather than clarity” and that “the correct method of implementing [its] test remains murky”). In this regard, Freeman notes that the “courts have applied Casey inconsistently and unfaithfully, creating a tangled body of abortion precedent and rendering the undue burden standard insufficient to protect women’s reproductive autonomy”. Id., p. 279. For example, Erwin Chemerinsky thinks that the test is inconsistent with the four-part analysis that the Court typically uses in cases involving individual liberties: “[f]irst, is there a fundamental right? Second, is the right infringed? Third, is the government’s action justified by a sufficient purpose? And fourth, are the means sufficiently related to the end sought?”¹³²x See Chemerinsky, 2015, p. 828. According to Dean Chemerinsky, the undue burden test collapses the last three of these questions into one, which does not make the test more manageable or transparent:

  Obviously ‘undue burden’ pertains to whether there is an infringement of the right, but … Casey also uses it to analyze whether the law is justified. No level of scrutiny is articulated by the joint opinion: there is no statement that the goal of the law must be compelling or important or that the means have to be necessary or substantially related to the end. Undue burden is thus confusing to apply because it melds together three distinct issues.¹³³x Id., at p. 863.

  Dean Chemerinsky also suggests that the test has an internal tension in that a law will be deemed to place an undue burden on a woman’s choice if its ‘purpose or effect’ is to place ‘a substantial obstacle’ in the path of a woman seeking a pre-viability termination of her pregnancy, but measures to assure that the woman’s choice is informed will be upheld “as long as their purpose is to persuade the woman to choose childbirth over abortion” and they do not unduly burden her right.¹³⁴x Id., at p. 864. Dean Chemerinsky further argues: “[e]very law adopted to limit abortion is for the purpose of discouraging abortions and encouraging childbirths. How is it to be decided which of these laws is invalid as an undue burden and which is permissible? The joint opinion simply says that the regulation “‘must not be an undue burden on the right.’ But this, of course, is circular; it offers no guidance as to which laws are an undue burden and which are not”.¹³⁵x Id. Finally, Dean Chemerinsky questions how many people would have to be adversely affected before a statute would be determined to be unconstitutional.¹³⁶x Id.

• D Down from Strict Scrutiny: Regulating the Right to Vote

  In some ways, the modern history of the right to vote parallels the history of the right to choose whether to terminate a pregnancy. Neither the Constitution of 1787 nor the Bill of Rights specifically protects the right to vote. On the contrary, the constitutional text leaves to the separate states the matter of qualifications for voting, even in federal elections: “the Electors [in federal elections] in each State shall have the Qualifications requisite for Electors of the most numerous Branch of the State Legislature.”¹³⁷x U.S. Const. Art. I, § 2, cl. 1. In addition, the Constitution provides that the president and vice-president shall be chosen by an electoral college, rather than by the voters, and it leaves to the states the determination as to how the members of the electoral college should be selected. Id., at Art. II, § 1, Amend. XII, XX. And the Constitution originally provided that members of the Senate were to be chosen by the members of the state legislatures. Id., at Art. I, § 2, cl. 1. The Seventeenth Amendment provided for the direct election of Senators in 1919. Id., at Amend XVII. Although that remains the case, the Constitution has been amended several times to prohibit the states from relying on certain criteria to deny persons the right to vote. Thus, in 1870, the people adopted the Fifteenth Amendment, which prohibited the states from withholding the right to vote “on account of race, color, or previous condition of servitude”.¹³⁸x Id., at Amend. XV. In 1920, the people adopted the Nineteenth Amendment, which prohibited the states from withholding the right to vote “on account of sex”.¹³⁹x Id., at Amend. XIX. In 1964, the people adopted the Twenty-fourth Amendment, which prohibited the states from withholding the right to vote “by reason of failure to pay any poll tax or other tax”.¹⁴⁰x Id., at Amend. XXIV. And in 1971, the people adopted the Twenty-sixth Amendment, which guaranteed that “[t]he right of citizens of the United States, who are eighteen years of age or older, to vote shall not be denied or abridged by the United States or by any State on account of age.”¹⁴¹x Id., at Amend. XXVI. In addition, the Constitution originally provided that the members of the federal House of Representatives would be directly elected by the people, while members of the Senate would be chosen by the state legislatures. See U.S. Const., Art. I, § 3, cl. 1. In 1913, however, the Seventh Amendment was adopted to provide for the direct election of senators. Id., at Amend. XVII. In addition, Congress enacted the Voting Rights Act of 1965, which currently provides, among other things, that “No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color, or in contravention of [certain other] guarantees.”¹⁴²x Voting Rights Act of 1965, Section 2, currently codified at 52 U.S.C. § 10301 et seq. In Shelby County v. Holder, 570 U.S. 2 (2013), the Court found that the pre-clearance coverage formula, a key section of the Voting Rights Act that required certain ‘covered’ jurisdictions to secure prior approval for changes in voter qualifications and other matters relating to the franchise, was unconstitutional because the coverage formula was based on stale data, so that it was no longer responsive to current needs and therefore an impermissible burden on the constitutional principles of federalism and equal sovereignty of the states. Many previously covered jurisdictions have recently adopted measures to make it more difficult to vote. See Brennan Center for Justice, ‘Election 2016: Restrictive Voting Laws By The Numbers’, 28 September 2016, available at: https://www.brennancenter.org/analysis/election-2016-restrictive-voting-laws-numbers (“Starting after the 2010 election, legislators in nearly half the states passed a wave of laws making it harder to vote. These new restrictions ranged from cuts to early voting to burdens on voter registration to strict voter ID requirements. While courts stepped in before the 2012 election to block many of these laws, the Supreme Court’s 2013 decision in Shelby County gutting the most powerful protections of the Voting Rights Act made it even easier for states to put in place restrictive voting laws.”) On the other hand, Atiba Ellis points out that “politicians, typically of a conservative persuasion, have echoed the voter fraud argument since the November 2000 election and resulting Bush v. Gore, 531 U.S. 98 (2000) debacle.” A.R. Ellis, ‘The Meme of Voter Fraud’, Catholic University Law Review, Vol. 63, 2014, p. 881-882.
  These constitutional and statutory changes have given rise, as one commentator has said, to “a triumphant narrative about voting and citizenship that Americans embrace”.¹⁴³x Id., at p. 898. In other words, Americans take pride in a narrative that emphasizes the progressive legal expansion of the franchise over the course of American history. But the historical truth is that legal expansions of the franchise invariably have been followed by the invention of new barriers to its exercise.¹⁴⁴x Id., at p. 897 (describing devices such as “poll taxes, literacy tests, and similar exclusionary tools” used to target newly enfranchised minority voters, and compensations made by law to exempt favored voters who would otherwise be affected by the tools). Professor Ellis also points out that “the meme of voter fraud represents the latest round of America’s evolution from an exclusion-based republic to an inclusive republic supporting full participation of all citizens.” Id., at p. 893. “Various arguments and beliefs advocating the exclusion of ‘unworthy’ voters have existed over time.”¹⁴⁵x Id., at p. 883. Moreover, those arguments and beliefs have regularly been used by those in power to justify the exclusion from the franchise, either legally or practically, of those thought to be their political adversaries. Over the years, efforts by those in power to exclude from the franchise those who are thought unlikely to support those in power have taken many forms: literacy tests, poll taxes, the exclusion of persons previously convicted of crimes, regulations relating to voter rolls, ballot access, and the conduct of elections, and, more recently, voter identification laws and political gerrymandering.¹⁴⁶x It is obviously beyond the scope of this article to deal comprehensively with all of the particular constitutional and legal issues raised by these practices. See Rotunda & Nowak, 2009, p. 219-349.
  As long ago as 1886, the Supreme Court noted that “the political franchise of voting” is rightly “regarded as a fundamental political right, because preservative of all rights”.¹⁴⁷x Yick Wo v. Hopkins, 118 U.S. 356, 370 (1886). More recently, in 1964, the Court in Wesberry v. Sanders ¹⁴⁸x 376 U.S. 1 (1964). observed that “[n]o right is more precious in a free country than that of having a voice in the election of those who make the laws under which, as good citizens, we must live. Other rights, even the most basic, are illusory if the right to vote is undermined.”¹⁴⁹x Id., at p. 17. In 1966, in Harper v. Virginia State Board of Elections,¹⁵⁰x 383 U.S. 663 (1966). the Court held that the imposition of a $1.50 poll tax on eligible voters was unconstitutional. The Court held “that a State violates the Equal Protection Clause … whenever it makes the affluence of the voter or payment of any fee an electoral standard”.¹⁵¹x Id., at p. 666. The Court overruled its 1937 decision in Breedlove v. Suttles, 302 U.S. 377 (1937), which upheld the constitutionality of provisions that conditioned voting on the payment of a poll tax. The Court also distinguished its earlier decision in Lassiter v. Northampton County Bd. of Elec., 360 U.S. 45 (1959), in which the Court upheld a North Carolina literacy test. The Court stated that “the Lassiter case does not govern the result here, because, unlike a poll tax, the ‘ability to read and write * * * has some relation to standards designed to promote intelligent use of the ballot.’” Id., at 51. The Court further observed that “[w]e have long been mindful that where fundamental rights and liberties are asserted … , classifications which might invade or restrain them must be closely scrutinized and carefully confined. … Those principles apply here. … [W]ealth or fee-paying has … no relation to voting qualifications; the right to vote is too precious, too fundamental to be so burdened or conditioned.”¹⁵²x Harper v. Va. State Bd. of Elections, at p. 670.
  In 1969, in Kramer v. Union Free School District,¹⁵³x 395 U.S. 621 (1969). the Court applied strict scrutiny to invalidate a state law that restricted voting in school board elections to those who held real property or had custody of children enrolled in the schools. Also in 1969, in Cipriano v. City of Houma,¹⁵⁴x 395 U.S. 701 (1969). the Court applied strict scrutiny to strike down a Louisiana law that conditioned the right to vote with respect to bond issues on the ownership of property. In 1970, when the Court held in Evans v. Cornman ¹⁵⁵x 398 U.S. 419 (1970). that residents of a federal enclave could not be prevented from voting in state elections, the Court noted that the right to vote was uniquely precious inasmuch as it is “protective of all fundamental rights and privileges”.¹⁵⁶x Id., at p. 422. And, in 1972, the Court applied strict scrutiny to strike down certain Tennessee residency requirements in Dunn v. Blumstein.¹⁵⁷x 405 U.S. 320 (1972). But see Sayler Land Co. v. Tulare Lake Basin Water Storage District, 410 U.S. 719 (1973) (upholding limitation on right to vote in water district elections to property owners and permitting votes to be apportioned according to assessed valuation of land within the district). In an amicus curiae brief in Crawford v. Marion County Election Bd., Nos. 07-21 and 07-25, Dean Chemerinsky attempted to draw a distinction between the foregoing cases and the Court’s later jurisprudence (which is summarized below), based on whether the deprivation of the right to vote was direct or indirect. If it were direct (as in the forgoing cases), strict scrutiny would apply. If not, the balancing test of Burdick v. Takushi, 504 U.S. 428 (1992) would control. See Crawford v. Marion County Election Bd., Nos. 07-21 and 07-25 (U.S.), Brief of Professor Erwin Chemerinsky as Amicus Curiae in Support of Neither Party (filed 13 November 2007). The Court did not credit that distinction, which might have provided one answer to the problem, while leaving a potentially large universe of possibly serious infringements outside the purview of strict scrutiny review. Indeed, the distinction seems to provide the basis for redressing simple-minded violations of voting rights while countenancing those that are more ingenious.
  As with the liberty interest in reproductive choice that the Court identified as fundamental in Roe, it appeared that the Court would henceforth treat the right to vote as a fundamental right subject to strict scrutiny review under the First and Fourteenth Amendments.¹⁵⁸x Infringements on the right to vote may be conceptualized in either equal protection or First Amendment terms, but the same analysis applies. See Rotunda & Nowak, 2009, p. 222. That would make sense for two reasons. First, unlike other fundamental rights, the right to vote exists only within a legal framework. As Atiba Ellis has pointed out, “[u]nlike other fundamental rights, the right to vote actually requires governmental participation in order to effectively and meaningfully manifest the right. Therefore the right-bearer depends upon the government for actualization of the right”.¹⁵⁹x Ellis, 2014, p. 913-914. Second, as Professor Ellis also points out, “politicians have an incentive to define the electorate to whom they wish to be accountable”.¹⁶⁰x Id., at p. 894. Indeed, “[p]oliticians and policymakers throughout American political history manipulated the rules of entry to the franchise in order to control voter turnout.” Id., at p. 893-94. It is for that reason that the kind of deference to the political process that Justice Frankfurter advocated in Colegrove v. Green, 328 U.S. 549, 552 (1949), seems inadequate. See Rotunda & Nowak, 2009, p. 310-312 (describing evolution of law with respect to justiciability beginning with Colegrove). Given the importance of the right to meaningful participation in the electoral process, it would make sense that regulations and restrictions on the effective exercise of that right should require a justification more substantial “than the mere incantation of a proper state purpose”.¹⁶¹x See Trimble v. Gordon, 430 U.S. 762, 769 (1977) (Powell, J.). As with the liberty interest in reproductive choice, however, the Court soon indicated that the right to vote would not invariably be given the most muscular form of constitutional protection. Thus, as two leading constitutional scholars have noted, the Court has been reluctant to give strict scrutiny its customary meaning in this context: “in this context ‘strict scrutiny’ means only that judges must independently review the voting regulation or restriction. If [the] restriction is in fact related to important or overriding state interests, the Court will sustain that regulation or restriction.”¹⁶²x See Rotunda & Nowak, 2009, p. 221. In other words, “strict scrutiny analysis in this area may only require the state to demonstrate that its regulation is narrowly tailored to promote an interest that is significant enough to outweigh any incidental restriction on the right to vote or the right of political association”.¹⁶³x Id., at p. 222. “Laws that totally prohibit a class of persons from voting in a general election or laws that are designed to restrict the voting power of a particular class of persons in a general election are unlikely to survive such a standard. Laws that regulate the electoral system to promote substantial state interests in the conduct of efficient and honest elections need to be examined on a case-by-case basis.” Id. In both areas, of course, the right of the individual is not absolute, but is seen to stand in tension with a legitimate state interest. In the one case, the state was said to have a legitimate interest not only in the woman’s health, but also in the promotion of childbirth and the protection of potential life. In the other case, the state was said to have a legitimate interest in a fair and efficient electoral system. Indeed, the very efficacy of the right to vote depended on it. But, unlike the situation with reproductive choice, there was no consideration akin to the viability of the fetus to help structure the inquiry into the proper accommodation of the individual and governmental interests.
  In Anderson v. Celebrezze,¹⁶⁴x 460 U.S. 780 (1983). a third-party candidate for president challenged a March filing deadline that Ohio law imposed on independent candidates who wished to stand for election in the November general election. In a 5-4 decision, the Court found that the early filing deadline was unconstitutional. Speaking through Justice Stevens, the Court noted that “‘[i]n approaching candidate restrictions, it is essential to examine in a realistic light the extent and nature of their impact on voters.’”¹⁶⁵x Id., at p. 786. According to Justice Stevens, “the right to vote is ‘heavily burdened’ if that vote may be cast only for major-party candidates at a time when other parties or other candidates are ‘clamoring for a place on the ballot.’”¹⁶⁶x Id., at p. 787. Justice Stevens further observed:

  Although these rights of voters are fundamental, not all restrictions imposed by the States on candidates’ eligibility for the ballot impose constitutionally suspect burdens on voters’ rights to associate or to choose among candidates. We have recognized that, “as a practical matter, there must be a substantial regulation of elections if they are to be fair and honest and if some sort of order, rather than chaos, is to accompany the democratic processes.” … Each provision [of sometimes complex election codes] inevitably affects – at least to some degree – the individual’s right to vote and his right to associate with others for political ends. Nevertheless, the State’s important regulatory interests are generally sufficient to justify reasonable, nondiscriminatory restrictions.¹⁶⁷x Id., at p. 788.

  To determine whether an election regulation satisfies constitutional requirements, the Court said, a court must first consider the character and magnitude of the asserted injury to constitutional rights. The court must then identify and evaluate the precise interests put forward as justifications for the burdens imposed by the rule. Finally, the court “must not only determine the legitimacy and strength of each of those interests, it must also consider the extent to which those interests make it necessary to burden the plaintiff’s rights”.¹⁶⁸x Id., at p. 789. According to the Court, “[t]he results of this evaluation will not be automatic; as we have recognized, there is ‘no substitute for the hard judgments that must be made.’”¹⁶⁹x Id., at p. 789-790.
  According to the Court, “the March filing deadline places a particular burden on an identifiable segment of Ohio’s independent-minded voters.”¹⁷⁰x Id., at p. 792. Moreover, “[a] burden that falls unequally on new or small political parties or on independent candidates … discriminates against those candidates – and of particular importance – against those voters whose political preferences lie outside the existing political parties.”¹⁷¹x Id., at p. 793-794. The Court also noted that the Ohio law not only burdened the rights of independent voters and candidates, but also “place[d] a significant state-imposed restriction on a nationwide electoral process”.¹⁷²x Id., at p. 795. The state proffered three justifications for the early filing date, but the Court found them unpersuasive, holding that “[u]nder any realistic appraisal, ‘the extent and nature’ of the burdens Ohio has placed on the voters’ freedom of choice and freedom of association, in an election of nationwide importance, unquestionably outweigh the State’s minimal interest in imposing a March deadline”.¹⁷³x Id., at p. 806.
  In dissent, Justice Rehnquist thought that the appropriate rule was that “so long as the Ohio ballot access laws are rational and allow nonparty candidates reasonable access to the general election ballot, this Court should not interfere with Ohio’s exercise of its Art. II, § 1, cl. 2, power.”¹⁷⁴x Id., at p. 808. Justice Rehnquist further argued that the Court had never before determined in this kind of case that the states must “meet some kind of ‘narrowly tailored’ standard,” but that the courts’ role was simply “to ensure that the State ‘in no way freezes the status quo, but implicitly recognizes the potential fluidity of American political life’”.¹⁷⁵x Id., at p. 817. According to Justice Rehnquist, “[i]f it does not freeze the status quo, then the State’s laws will be upheld if they are ‘tied to a particularized legitimate purpose, and [are] in no sense invidious or arbitrary.’”¹⁷⁶x Id.
  In 1993, the Court decided Burdick v. Takushi,¹⁷⁷x 504 U.S. 428 (1992). which, by a 6-3 vote, upheld a Hawaii statute that prohibited write-in votes. Justice White, one of the dissenters in Anderson, wrote for the majority, noting that the party challenging the statute “proceeds from the erroneous assumption that a law that imposes any burden upon the right to vote must be subject to strict scrutiny. Our cases do not so hold”.¹⁷⁸x Id., at p. 432. The Court interpreted the test set forth in Anderson as a two-part test, whereby the rigour of the inquiry depended on the extent to which the challenged regulation burdens constitutional rights:

  [W]hen those rights are subjected to “severe” restrictions, the regulation must be “narrowly drawn to advance a state interest of compelling importance.” … But when a state election law provision imposes only ‘reasonable, nondiscriminatory restrictions” upon the … rights of voters, “the State’s important regulatory interests are generally sufficient to justify” the restrictions.¹⁷⁹x Id., at p. 434.

  Justice White conceded that “the Hawaii election laws, like all election regulations have an impact on the right to vote,” but he concluded that “it can hardly be said that [these laws] limit access to the ballot by party or independent candidates or unreasonably interfere with the right of voters to associate and have candidates of their choice placed on the ballot. Indeed, petitioner understandably does not challenge the manner in which the State regulates access to the ballot”.¹⁸⁰x Id., at p. 434-435. While Justice White emphasized that Hawaii’s overall system provided adequate ballot access, the plaintiff had challenged “the write-in prohibition [on the ground that it] deprives him of the opportunity to cast a meaningful ballot”.¹⁸¹x Id., at p. 437. “At bottom,” according to Justice White, the plaintiff “claims that he is entitled to cast and … [have counted] ‘a protest vote’ for Donald Duck, … and that any impediment to this asserted ‘right’ is unconstitutional.”¹⁸²x Id., at p. 938. But “a prohibition on write-in voting will be presumptively valid, since any burden on the right to vote for the candidate of one’s choice will be light and normally will be counterbalanced by the very state interests supporting the ballot access scheme.”¹⁸³x Id., at p. 441.
  In his dissent, Justice Kennedy agreed that the majority had properly stated the relevant balancing test, but he thought that the proper application of that test led to the conclusion that “the write-in ban deprives some voters of any substantial voice in selecting candidates for the entire range of offices at issue”.¹⁸⁴x Id., at p. 446. According to Justice Kennedy, the record in the case showed that the Hawaii law placed a ‘significant burden’ on the rights of voters to vote for whomever they wished and therefore prevented voters who preferred to vote for persons not listed on the ballot “from participating in Hawaii elections in a meaningful manner”.¹⁸⁵x Id., at p. 442-443. Justice Kennedy continued:

  For those who are affected by write-in bans, the infringement on their right to vote for the candidate of their choice is total. The fact that write-in candidates are longshots more often than not makes no difference; the right to vote for one’s preferred candidate exists regardless of the likelihood that the candidate will be successful.¹⁸⁶x Id., at p. 447.

  Justice Kennedy then discussed the state’s justifications, which he found insubstantial compared with the ‘significant burden’ that the ban places on these voters.¹⁸⁷x Id., at p. 448.
  In 2007, in Crawford v. Marion County Election Board,¹⁸⁸x 472 F.3d 949 (2007). the United States Court of Appeals for the Seventh Circuit addressed the constitutionality of an Indiana law that required persons wishing to vote in person at polling places to present a special, government-issued photo identification (‘ID’) card.¹⁸⁹x Id., at p. 950. The statute did not place the same restriction on persons who were eligible to cast an absentee ballot or voted in a nursing home. Id. In addition, voters could cast provisional ballots and return within 10 days with appropriate documentation. Id. To secure a state-issued ID card, it is necessary to have a certified birth certificate, and “it’s not particularly easy for a poor, elderly person who lives in South Bend, but was born in Arkansas, to get a certified copy of his birth certificate.” Id., at p. 955 (Evans, J. dissenting). The majority speculated that “[t]he benefits of voting to the individual voter are elusive (a vote in a political election rarely has any instrumental value …), and even very slight costs in time or bother or out-of-pocket expense deter people from voting, or at least from voting in elections they’re not much interested in. So some people who have not bothered to obtain a photo ID will not bother to so just to be allowed to vote, and a few who have a photo ID but forget to bring it to the polling place will say what the hell and not vote, rather than go home and get the ID and return to the polling place.” Id., at p. 951. Voters had previously been required to verify their identities by signing the poll book, which would be checked against signatures on file. Several plaintiffs challenged the law “as an undue burden on the right to vote”.¹⁹⁰x Id., at p. 950. In an opinion by Judge Richard Posner, a distinguished jurist and legal scholar, a divided panel held that “[a] strict standard would be especially inappropriate in a case such as this, in which the right to vote is on both sides of the ledger”.¹⁹¹x Id., at p. 952. The Seventh Circuit further observed that:

  The Indiana law is not like a poll tax, where on one side is the right to vote and on the other side the state’s interest of defraying the cost of elections or in limiting the franchise to people who really care about voting or in excluding poor people or in discouraging people who are black. The purpose of the Indiana statute is to reduce voting fraud, and voting fraud impairs the right of legitimate voters to vote by diluting their votes – dilution being recognized to be an impairment of the right to vote. … On one side of the balance in this case is the effect of requiring a photo ID in inducing eligible voters to disfranchise themselves. That effect, so far as the record shows, is slight. …
  On the other side of the balance is voting fraud, specifically the form of voting fraud in which a person shows up at polls claiming to be someone else. … Without requiring a photo ID, there is little if any chance of preventing this kind of fraud because busy poll workers are unlikely to scrutinize signatures and argue with people who deny having forged someone else’s signature.¹⁹²x Id., at p. 952-953.

  The district court had found that approximately 43,000 Indiana residents, or slightly less than 1% of its voting age population, had no qualifying ID.¹⁹³x See Indiana Democratic Party v. Rokita, 458 F.Supp.2d 775, 782-84 (S.D. Ind. 2006). The record showed that “as far as anyone knows, no one in Indiana, and not many people elsewhere, are known to have been prosecuted for impersonating a registered voter,” but the Seventh Circuit panel found the explanation for that fact in either “the endemic underenforcement of minor criminal laws” or “the extreme difficulty of apprehending a voter impersonator.” The panel apparently discounted the possibility that the Indiana law was either a solution in search of a problem or an effort to discriminate against poor and minority voters. The panel also explained the absence of any published reports of voter fraud as “reflect[ing] nothing more than the vagaries of journalists’ and other investigators’ choice of scandals to investigate”.¹⁹⁴x Crawford, 472 F.3d at p. 953. Judge Posner has subsequently confessed that his resolution of the case was incorrect. See Richard A. Posner, Reflections on Judging, Cambridge, Harvard University Press 2013, p. 851.
  In a spirited dissent, Judge Evans wrote: “[t]he Indiana voter photo ID law is a not-too-thinly-veiled attempt to discourage election-day turnout by certain folks believed to skew Democratic. We should subject this law to strict scrutiny – or at least, in the wake of Burdick … something akin to ‘strict scrutiny light’ – and strike it down as an undue burden on the right to vote.”¹⁹⁵x Crawford, 472 F.3d at p. 955. Judge Evans observed that there was little or no evidence of the type of polling-place fraud that photo ID laws seek to stop, but that “this law will make it more difficult for some eligible voters – I have no idea how many, but 4 percent is a number that has been bandied about – to vote … [a]nd this group is mostly comprised of people who are poor, elderly, minorities, disabled, or some combination thereof”.¹⁹⁶x Id. He continued: “Burdick adopts a flexible standard, and as I read it, strict scrutiny may still be appropriate in cases where the burden, as it is here, is great and the state’s justification for it, again as it is here, is hollow.”¹⁹⁷x Id., at p. 956.
  The court of appeals denied rehearing en banc by a vote of 7 to 4. In an opinion for the four dissenting judges, Judge Wood wrote that:

  [T]he panel assumes that Burdick also means that strict scrutiny is no longer appropriate in any election case. As Judge Evans makes clear, however, Burdick holds no such thing. To the contrary, Burdick simply established a threshold inquiry that a court must perform before it decides what level of scrutiny is required for the particular case before it. … [W]hen there is a serious risk that an election law has been passed with the intent of imposing an additional significant burden on the right to vote of a specific group of voters, the court must apply strict scrutiny. …
  The state’s justification for the new voting requirement is voter fraud – specifically, the problem of fraud on the part of people who show up in person at the polling place. Yet the record shows that the existence of this problem is a disputed question of fact. It is also a crucial question for the inquiry that Burdick demands, because if the burden on voting is great and the benefit for the asserted state interest is small as an empirical matter, the law cannot stand. …
  Burdick requires an inquiry into the “precise interests put forward by the State as justifications for the burden imposed,” but in this case, the “facts” asserted by the state in support of its voter fraud justification were taken as true without any examination to see if they reflected reality.¹⁹⁸x Crawford v. Marion County Election Board, 484 F.3d 436, 437-439 (7th Cir. 2007). Judge Wood pointed out that, contrary to Judge Posner’s understanding, “as a matter of law, the Supreme Court’s voting cases do not support a rule that depends in part for support on the idea that no one vote matters. Voting is a complex act that both helps to decide elections and involves individual citizens in the group act of self-governance.” Id., at p. 438.

  On further review, the Supreme Court affirmed the Seventh Circuit panel decision.¹⁹⁹x See Crawford v. Marion County Election Board, 533 U.S. 181 (2008). Justice Stevens wrote the lead opinion for the majority, but only the Chief Justice and Justice Kennedy joined in his reasoning. Justice Stevens acknowledged that “even rational restrictions on the right to vote are invidious if they are unrelated to voter qualifications,” but that “‘even-handed restrictions that protect the integrity and reliability of the electoral process itself’ are not invidious and satisfy the standard set forth in Harper”.²⁰⁰x Id., at p. 189-190. Justice Stevens added: “[h]owever slight that burden may appear, as Harper demonstrates, it must be justified by relevant and legitimate state interests ‘sufficiently weighty to justify the limitation.’”²⁰¹x Id., at p. 191. In Justice Stevens’s view, “a court must identify and evaluate the interests put forward … as justifications for the burdens imposed by [the state’s] rule, and then make ‘the hard judgment’ that our adversary system demands.”²⁰²x Id., at p. 190. There was, of course, no record evidence to show that voter impersonation fraud was a problem in Indiana or anywhere else, as Justice Stevens expressly conceded.²⁰³x As Justice Stevens put it, “[t]he only kind of voter fraud [the law] addresses is in-person voter impersonation at polling places. The record contains no evidence of any such fraud actually occurring in Indiana at any time in its history.” Id., at p. 194. Nor was there any evidence to show that the Indiana law would provide an effective means for dealing with that phantom problem or improve the situation in any way. Justice Stevens nonetheless discounted the significance of both points: “[w]hile the most effective method of preventing election fraud may well be debatable, the propriety of doing so is not.” Id., at p. 296. Nonetheless, Justice Stevens thought that the record evidence failed to establish the facial invalidity of the Indiana voter ID law: “[w]hen we consider only the statute’s broad application to all Indiana voters we conclude that it ‘imposes only a limited burden on voters’ rights,’” and “[t]he ‘precise interests’ [advanced by Indiana] are … sufficient to defeat petitioners’ challenge.”²⁰⁴x Id., at p. 202-203.
  Justice Scalia, together with Justices Thomas and Alito, concurred in the judgment, but their reasoning departed significantly from that of Justice Stevens. Justice Scalia wrote:

  The lead opinion assumes petitioners’ premise that the voter-identification law “may have imposed a special burden on” some voters, … but holds that petitioners have not assembled evidence to show that the special burden is severe enough to warrant strict scrutiny. … That is true enough, but for the sake of clarity and finality (as well as adherence to precedent), I prefer to decide these cases on the grounds that petitioners’ premise is irrelevant and that the burden at issue is minimal and justified.²⁰⁵x Id., at p. 204.

  Justice Souter dissented in an opinion joined by Justice Ginsburg. Justice Souter thought that cases involving administrative restrictions on voting necessarily raise “two competing interests,” one being “the fundamental right to vote,” which requires that the judiciary “train a skeptical eye on any qualification of that right.”²⁰⁶x Id., at p. 210. “As against [that] unfettered right,” Justice Souter continued, lies the constitutional imperative that “‘there must be a substantial regulation of elections if they are to be fair and honest and if some sort of order, rather than chaos, is to accompany the democratic processes.’”²⁰⁷x Id. Thus, Justice Souter thought that, as the Court held in Burdick, “‘[h]owever slight [the] burden may appear, … it must be justified by relevant and state interests sufficiently weighty to justify the limitation.’”²⁰⁸x Id., at p. 211. Applying this test, Justice Souter would have held that the Indiana law “threaten[ed] to impose nontrivial burdens on the voting right of tens of thousands of the State’s citizens … and a significant percentage of those individuals are likely to be deterred from voting.”²⁰⁹x Id., at p. 209. According to Justice Souter, the Indiana statute therefore failed to satisfy the test set out in Burdick:

  [A] state may not burden the right to vote merely by invoking abstract interests, be they legitimate … or even compelling, but must make a particular, factual showing that threats to its interests outweigh the particular impediments it has imposed. The State has made no such justification here, and as to some aspects of its law, it has hardly even tried.²¹⁰x Id.

  Justice Breyer also dissented. He would have “balance[d] the voting-related interests that the statute affects asking ‘whether the statute burdens any one such interest in a manner out of proportion to the statute’s salutary effects upon the others (perhaps, but not necessarily, because of a clearly superior, less restrictive alternative).’”²¹¹x Id., at p. 237. Pursuant to that standard, Justice Breyer would have held that “the statute is unconstitutional because it imposes a disproportionate burden upon those eligible voters who lack a driver’s license or other statutorily valid form of photo ID”.²¹²x Id.
  Justice Scalia’s concurring opinion well illustrates the difficulties with the majority’s understanding and application of the appropriate constitutional test. First, Justice Scalia observes that, since strict scrutiny applies only if the burden placed on voters is severe, “the first step is to decide whether a challenged law severely burdens the right to vote.”²¹³x Id., at p. 205. In this sense, Justice Scalia builds on Justice Stevens’s holding that strict scrutiny review is not required in this type of case, where the restrictions placed on the right to vote are not “unrelated to voter qualifications”. Id., at p. 189. Similarly, the court of appeals had held that “the law should [not] be held by the same strict standard applicable to a poll tax because the burden on voters was offset by the benefit of reducing the risk of fraud.” Id., at p. 188. In other words, the first step is not to determine whether there is any problem to be solved (a significant omission here, given the absence of evidence to show that voter impersonation fraud was now or ever had been a problem in Indiana or elsewhere),²¹⁴x Id., at p. 194. (“The only kind of voter fraud that [the Indiana law] addresses is in-person voter impersonation at polling places. The record contains no evidence of any such fraud occurring in Indiana at any time in its history.”) nor whether the law actually serves any legitimate purpose. Instead, according to Justice Scalia, the first step is simply to assess the severity of the burden that the law imposes. In other words, a limitation that the Court deems not to be “severe” will pass constitutional muster even if the problem to be solved is imaginary, and there is no evidence to suggest that the limitation will accomplish any good whatsoever. That seems a seriously inadequate means of protecting a “fundamental right,” particularly the right to vote, which is, as the Court said in Yick Wo, “preservative of all rights”.²¹⁵x See Yick Wo v. Hopkins, 118 U.S. at p. 370. Second, burdens are not severe, according to Justice Scalia, unless they “go beyond the merely inconvenient;” they must be “‘so burdensome’ as to be ‘virtually impossible’ to satisfy.”²¹⁶x Crawford, 533 U.S. at p. 205. That is a similar view, of course, to that taken by Justice O’Connor when she invoked the undue burden test in the reproductive choice context in Thornburgh. In that case, she used the expression to mean an “absolute obstacle[…] or severe limitation[…]” on the right.²¹⁷x See Thornburgh, 476 U.S. at p. 828. The Casey plurality adopted a less demanding version of the test. See Casey, 505 U.S. at p. 876. That was not, of course, the version of the ‘undue burden’ test that a plurality of the Court adopted in Casey or that a majority of the Court subsequently adopted in Stenberg.
  In addition, Justice Scalia thinks that the severity of a burden is to be measured in terms of its “reasonably foreseeable effect on voters generally”,²¹⁸x Crawford, 505 U.S. at p. 206 (emphasis in original). not on any particular, identifiable demographic group or subgroup, such as the elderly, the poor, or those born in another state.²¹⁹x The trial judge “found that petitioners had ‘not introduced evidence of a single, Indiana resident who will be unable to vote as a result of [the law] or who will have his or her right to vote unduly burdened by its requirements’”. Id., at p. 187. The trial court refused to credit the testimony of an expert witness, who testified that the law could affect up to 989,000 registered voters who lacked a government-issued ID, but the trial court nonetheless estimated that only about 43,000 (or less than 1%) of Indiana residents would be affected. Id., at p. 187-188. That, according to Justice Scalia, is because “our precedents refute the view that individual impacts are relevant to determining the severity of the burden it imposes.”²²⁰x Id., at p. 205. It seems clear, however, that not even the Virginia poll tax could have been struck down if a majority of the Court in Harper v. Virginia Board of Elections ²²¹x 383 U.S. 663 (1966). had followed the approach outlined by Justice Scalia in Crawford. Moreover, the Court has held, in the reproductive rights context that “the proper focus of constitutional inquiry is the group for whom the law is a restriction, not the group for whom the law is irrelevant”.²²²x See Casey, 505 U.S. at p. 894. Finally, Justice Scalia argues that “[t]he Fourteenth Amendment does not regard neutral laws as invidious ones, even when their burdens purportedly fall disproportionately on a protected class. A fortiori, it does not do so when, as here, the classes complaining of disparate impact are not even protected.”²²³x Id., at p. 207 (emphasis omitted). Thus, according to Justice Scalia, “weighing the burden of a nondiscriminatory voting law upon each voter and concomitantly requiring exceptions for vulnerable voters would effectively turn back decades of equal-protection jurisprudence.”²²⁴x Id. But the fact that the right to vote has long been deemed to be a fundamental right rebuts that point, as does the fact that the point is likewise inconsistent with the jurisprudence relating to reproductive choice. Most fundamentally, perhaps, Justice Scalia expressed the view that the regulation of voting should be left to state officials, whose “judgment must prevail unless [the law] imposes a severe and unjustified overall burden upon the right to vote, or is intended to disadvantage a particular class”.²²⁵x Id., at p. 208. That again sounds like the version of the undue burden test that Justice O’Connor proposed in Thornburgh – which the Court has not adopted in the area of reproductive choice, let alone in the voting rights area.
  Needless to say, the problem was not solved in Crawford, and it continues to manifest itself wherever one party has control of the machinery of government and chooses to use that machinery to enact legislation that limits the rights of others to participate fully and effectively in the electoral process. Whether that legislation conditions voting on the presentation of identification documents that are not readily available to all or allows for partisan gerrymandering, the purpose is the same: preventing meaningful participation in the political process by those thought not to be supporters of those who make the rules. Given the number of instances in which state governments have chosen to enact such legislation in recent years, it is clear that the problem will not go away until the Court imposes a more realistic and rigorous test for evaluating such legislation. To date, the Court has been unwilling to do so, as shown by its decision in Crawford. Moreover, the Court has affirmatively demonstrated hostility to the notion of judicial protection of the right to vote, as indicated by its decision in Shelby County v. Holder,²²⁶x 570 U.S. 2 (2013). which struck down a key component of the Voting Rights Act of 1965. But the problem remains, and it cries out for an effective judicial solution. In this Term alone, the Court will face two partisan gerrymandering cases, one involving a Republican gerrymander in Wisconsin, the other a Democratic gerrymander in Maryland.²²⁷x In Whitford v. Gill, 218 F.Supp.3d 837 (W.D. Wisc. 2016), a three-judge district court struck down a Republican redistricting plan in Wisconsin. In Benisek v. Lamone, 266 F.Supp.3d 799 (E.D. Md. 2017), a three-judge district court declined to strike down a Democratic redistricting plan in Maryland. See Gill v. Whitford, No. 16-1161, jurisdictional statement filed 24 March 2017; Benisek v. Lamore, No. 17-333, jurisdictional statement filed 1 September 2017.

• E The Forest and the Trees: Protecting the Fundamental Right to Vote

  The people have repeatedly recognized the importance of the right to vote by enacting amendments to the Constitution to prohibit the withholding of the right to vote from various groups.²²⁸x In 1870, as previously noted, the people of the United States acknowledged the importance of the right to vote by adopting the Fifteenth Amendment to the Constitution, which guarantees that “[t]he right of citizens of the United States to vote shall not be denied or abridged by the United States or by any State on account of race, color, or previous condition of servitude.” U.S. Const., Amend. XV. Similar amendments have since been adopted to prohibit exclusions from the franchise based on gender, the failure to pay “any poll tax or other tax,” or on account of age if the putative voter is “eighteen years of age or older.” Id., Amend. XIX (1920) (gender), Amend. XXIV (1964) (poll or other tax), Amend. XXVI (1971) (age). In addition, Congress enacted the Voting Rights Act in 1965. See Voting Rights Act of 1965, 79 Stat. 437, codified, as amended, 52 U.S.C. §101001, et seq. But see Shelby County v. Holder, 570 U.S. 2 (2013) (holding that the coverage formula of Section 4(b) is unconstitutional because it is based on data over 40 years old, making it no longer responsive to current needs and therefore an impermissible burden on the constitutional principles of federalism and the “equal sovereignty of the states”). In addition, in 1913, the people amended the Constitution to provide that United States Senators would henceforth be elected by the people of the several states, rather than by the state legislatures. In addition, in 1913, the people amended the Constitution to provide that United States Senators would henceforth be elected by the people of the several states, rather than by the state legislatures. U.S. Const., Amend. XVII. The Supreme Court has also recognized the central importance of the franchise in a constitutional democracy. In 1886, the Supreme Court observed that “the political franchise of voting” is rightly “regarded as a fundamental political right because preservative of all rights”.²²⁹x Yick Wo v. Hopkins, 118 U.S. 356, 370 (1886). As we have also seen, the modern Supreme Court has continued to refer to the right to vote as a “fundamental right”, and it initially held that restrictions on the right to vote should be subject to strict scrutiny, in its accepted sense. The backdrop for such judicial pronouncements, of course, was the persistent efforts, both ingenious and simple-minded, whereby those in control of the electoral machinery had exploited that control to make meaningful participation in the political process more difficult or impossible for members of groups thought for one reason or another to be politically antagonistic to those in control.²³⁰x See, e.g., Louisiana v. United States, 380 U.S. 145 (1965) (invalidating Louisiana statute that authorized the registrar of voters to determine whether a voter’s ‘understanding’ of the federal or state constitution was sufficient to permit him or her to vote); Gomillion v. Lightfoot, 364 U.S. 339 (1960) (invalidating state statute that created a 28-sided city boundary by which nearly all African-American voters would be excluded without excluding any whites); Guinn v. United States, 238 U.S. 347 (1915) (invalidating the so-called ‘grandfather clause’, an Oklahoma constitutional amendment that provided that “no person who was, on January 1, 1866, or at any time prior thereto, entitled to vote under any form of government, or who at that time resided in some foreign nation, and no lineal descendant of such person, shall be denied the right to register and vote because of his inability to do so read and write sections of such constitution.”). In other words, notwithstanding the clear trend towards greater inclusiveness in the formal legal definition of the electorate, and the progressive dismantling of de jure barriers to voting, those opposed to the enlargement of the franchise (or simply hostile to one identifiable group or another) have repeatedly found new ways of stifling the electoral voices of those whose votes they fear. They are able to do so because of the simple fact that “[u]nlike other fundamental rights, the right to vote actually requires governmental participation in order to effectively and meaningfully manifest the right”.²³¹x Ellis, 2014, p. 913-914. In other words, elections necessarily require electoral regulations and machinery, and state officials have been given broad discretion in designing and implementing that machinery. Recognizing that the state theoretically acts on behalf of all voters when it regulates voting to protect the regularity and integrity of the electoral process, the Court has held that the state may justifiably impose ‘reasonable, nondiscriminatory restrictions’²³²x See Anderson v. Celebreeze, 460 U.S. 780, 788 (1983). on the electoral process. But the inquiry mandated by that principle turns out to be considerably more difficult – and the protection it affords to the right of a citizen to cast a meaningful vote less certain and sure – than its simple words would suggest. The gulf between promise and reality is simply too great, at least if one takes seriously the centrality of the right to vote.²³³x See generally J.H. Ely, Democracy and Distrust: A Theory of Judicial Review, Cambridge, Harvard University Press, 1980.
  As Crawford demonstrates, the difficulty rests in ensuring the adequate protection of the fundamental right to vote while also allowing the state the regulatory power it needs to conduct elections on a neutral and even-handed basis. As Crawford also shows, the Court has thus far failed to formulate a test that does not in practice encourage state officials to abuse that power. Whether one adopts Justice Stevens’s version or that of Justice Scalia, the Crawford test provides scant protection for the right to vote and can be easily manipulated by those who control the electoral machinery. Among other things, neither Justice Stevens nor Justice Scalia would even require the state to show at the threshold that there is a problem to be solved. Whatever the Court might say, the reality after Crawford seems to be that the existence of such a problem can simply be assumed. One can always justify additional measures to perfect the voting process, notwithstanding the fact that those additional measures deprive some people of the right to vote, if one is not called upon to show that the absence of perfection is actually problematic. In that case, one can always justify new regulations based on the possibility, which is necessarily perpetual, of ‘improving the election machinery’. The same is true if, and contrary to the common understanding of the words used, the existence of an ‘undue burden’ will be seen to depend only on the absolute size of the burden created, without regard to any possible balancing of the burden against benefits allegedly to be achieved. In those circumstances, no medicine could possibly be too strong. Moreover, the Crawford concept of burden seemingly relates to an effect on the population at large rather than on the effect on those on whom the burden actually falls. Given those features, the illusory nature of the protection offered by the Crawford test is clear. Indeed, the test used in Crawford seems as undemanding as the rational basis standard used with respect to ordinary commercial activities. Indeed, in Williamson v. Lee Optical of Oklahoma, Inc.,²³⁴x 348 U.S. 483 (1955). the canonical rational basis case in which the Court famously upheld an under-inclusive state statute pertaining to the regulation of eye care professionals on the ground that the state was entitled to pursue regulation ‘one step at a time’,²³⁵x Id., at p. 489. the Court emphasized that “It is enough that there is an evil at hand for correction, and that it might be thought that the particular legislative measure was a rational way to correct it.”²³⁶x Id., at p. 488. See also id., at p. 489. (“The problem of legislative classification is a perennial one, admitting of no doctrinaire definition. Evils in the same field may be of different dimensions and proportions, requiring different remedies. Or so the legislature may think.”) Although the Court gave excessive deference to the legislature in that case,²³⁷x Id., at p. 488. (“We emphasize again what Chief Justice Waite said in Munn v. State of Illinois …, ‘For protection against abuses by legislatures, the people must resort to the polls, not to the courts.’") it did seem to suggest the necessity for showing the existence of ‘an evil at hand for correction’, that is, a genuine problem to be solved, and some rational relationship between that problem and the means chosen to correct it.
  A wealth of judicial statements suggests that the solution to one problem or another should rest with the political process.²³⁸x See, e.g., Obergefell v. Hodges, 135 S.Ct. 2584, 2627 (2015) (J. Scalia, dissenting) (“Until the courts put a stop to it, public debate over same-sex marriage displayed American democracy at its best. Individuals on both sides of the issue passionately, but respectfully, attempted to persuade their fellow citizens to accept their views. Americans considered the arguments and put the question to a vote. The electorates of 11 States, either directly or through their representatives, chose to expand the traditional definition of marriage. Many more decided not to. Win or lose, advocates for both sides continued pressing their cases, secure in the knowledge that an electoral loss can be negated by a later electoral win. That is exactly how our system of government is supposed to work.”); Planned Parenthood of Southeastern Pennsylvania, Inc. v. Casey, 505 U.S. 833, 1002 (1992) (“[B]y foreclosing all democratic outlet for the deep passions this [reproductive choice] issue arouses, by banishing the issue from the political forum that gives all participants, even the losers, the satisfaction of a fair hearing and an honest fight, by continuing the imposition of a rigid national rule instead of allowing for regional differences, the Court merely prolongs and intensifies the anguish.”). In most cases, that is surely an appropriate response. In a representative democracy, we necessarily look in the first instance to the political process to synthesize, prioritize and resolve our public problems. That is in the nature of our government, and it is an approach that works much of the time. We act at our peril when we seek to short-circuit its customary processes. But fundamental rights and suspect classifications present a special case, as Justice Stone recognized in Carolene Products.²³⁹x See United States v. Carolene Products, 304 U.S. 144, 153 n. 4 (1938). And that is especially true when the very problem to be solved is the political process – its integrity, its inclusiveness and its fundamental fairness. Moreover, that is where we stand with respect to the current state of the right to vote. The current reality is that those who control the electoral machinery often use that control for their own purposes, self-interest, and perpetuation in office, and for discriminatory or partisan ends. They are truly ‘judges in their own cases’,²⁴⁰x See John Locke, Of Civil Government: Second Treatise, Chicago, Regnery/Gateway, 1955, § 13, p. 11. and the courts, having first diluted the meaning of ‘strict scrutiny’, and then having rejected the application of even that weak version to all but a fraction of voting rights cases,²⁴¹x See Rotunda & Nowak, 2012 p. 222. appear powerless to ensure fairness in this centrally important and foundational area of civic life. The Court must either rediscover the importance of strict scrutiny in voting cases, which seems highly unlikely, or it must devise a new approach for affording greater protection to the fundamental right to vote.
  Some guidance in that regard may be found in a recent article by Emma Freeman,²⁴²x See Freeman, 2013, p. 279-323. who takes issue with what she sees as the Court’s (possibly inadvertent)²⁴³x Id., at p. 280. diminution of the constitutional protection afforded to a woman’s right to reproductive choice and suggests a refinement to the Court’s approach, whereby she hopes to give an additional degree of “bite” to this constitutional protection.²⁴⁴x Id., at p. 281. Although the two areas are obviously dissimilar in many respects, they do share some important commonalities and may be susceptible to analogous methodological treatment. In contemplating approaches that might provide more muscular protection for the right to vote, it is therefore appropriate to consider Freeman’s approach, which aims “to imbue the [undue burden] test with as much rigor as it can tolerate”.²⁴⁵x Id.
  As Freeman notes, the problem that concerned the Court in Casey was the need to reconcile two competing interests: the woman’s right to terminate her pregnancy, on the one hand, and the state’s interests in protecting both the woman’s health and potential life, on the other hand.²⁴⁶x Id., at p. 321. In the Casey Court’s view, Roe had given sufficient weight to the woman’s interest but not to those of the state. The Casey plurality therefore intended to correct that error by adopting a standard of constitutional review that gave appropriate weight to both interests.²⁴⁷x Id. While the Casey plurality chose not to adopt a balancing test, there is no doubt but that it continued to take seriously the woman’s right to choose.²⁴⁸x Id., at p. 321-322. Freeman writes: “[t]hough it is difficult to speculate about the Court’s reluctance to adopt a proportionality-based test, that hesitation may be symptomatic of the judiciary’s conception of its own perceived boundaries.” Id., at p. 322. Indeed, Freeman believes that the Casey plurality intended to provide a high degree of protection to the woman’s right to terminate her pregnancy,²⁴⁹x For example, Freeman argues that “[i]t is evident that the plurality believed state regulations on abortion must further a genuinely legitimate state interest. Though they did not incorporate this inquiry into their articulation of the undue burden test, the plurality opinion repeatedly assessed the state’s interest to ensure its validity. The state had no legitimate interest, e.g., in “giv[ing] to a man the kind of dominion over his wife that parents exercise over their children.’” Id., at p. 294, quoting Casey, 505 U.S. at p. 898 (plurality opinion). but inadvertently created a standard that failed to meet that objective.²⁵⁰x Freeman suggests that the Casey plurality intended for the ‘undue burden’ test to constitute an intermediate level of scrutiny, “lying somewhere between the deferential rational basis and the punishing strict scrutiny”. Id., at p. 298. According to Freeman, the plurality did not intend “to retreat wholly from Roe’s protection of a woman’s independence and discretion”, but “sought to construct a less strict but still vigorous standard capable of defending the abortion right”. Id., Finally, “[i]t is precisely because undue burden is a form of intermediate review that Justice Blackmun expressed his preference for Roe’s strict scrutiny and Chief Justice Rehnquist for Webster’s rational basis.” Id. at p. 299-300. In Freeman’s view, a significant part of the problem rests in the fact that the Casey approach contains no nexus requirement. “Omitting nexus analysis denies Casey its rightful bite because regulations without unduly burdensome purposes or effects may still fail to further reasonably a legitimate state interest.”²⁵¹x Id., at p. 316. Under Casey, the courts must look to the purpose and effect of the regulation but not to the relationship between them, Freeman writes:

  Even rational basis review, the most forgiving standard of constitutional scrutiny, nominally requires courts to establish as adequate the connection, or “nexus,” between the state’s legislative ends and its legislative means. Though purportedly as stringent as intermediate scrutiny, undue burden lacks such a nexus inquiry: under Casey, courts must analyze a statute’s purpose and its effects, but need not assess the relationship between the two. … The Court’s imprecise discussion of that test has led the appellate courts to apply the test in ways that poorly safeguard women’s reproductive choice.²⁵²x Id., at p. 279-280. Freeman points out that traditional rational basis review and heightened rational basis review both contain a nexus requirement, but that courts implementing the former “merely invoke, but do not in fact apply, nexus analysis”, whereas courts implementing the latter “actually examine the relationship between state means and ends”. Id., at p. 285. Freeman also observes that “[b]ecause the Court has been reluctant to acknowledge overtly the existence of rational basis with bite, much less identify the factors that trigger such enhanced review, the standard’s boundaries remain blurry.” Id., at p. 287. In any event, heightened rationality review differs in three ways from traditional rational basis review: “the ‘bite’ renders the courts less deferential to the legislature, less tolerant of over- and under-inclusive classifications, and less open to state experimentation.” Id., at p. 285. Finally, Freeman notes that because of the Court’s extensive use of the language of rational basis review, “some scholars remain ‘unclear whether Casey’s undue burden standard subjects abortion regulations to intermediate scrutiny, or merely to rational basis review.’” Id., at p. 293. Indeed, as Freeman also observes, even Justice Scalia expressed confusion, noting that the plurality’s “description of the undue burden standard in terms more commonly associated with the rational-basis test will come as a surprise even to those who have followed closely our wanderings in this forsaken wilderness”. Id., quoting Casey, 505 U.S. at 986 n.4 (Scalia, J., concurring in part and dissenting in part).

  To restore the Casey standard to the degree of muscularity that Freeman believes that the plurality intended, she suggests that the courts should first apply what she calls a ‘rational basis with bite’ standard as a threshold requirement before moving on to an application of the ‘purpose and effect’ test. Freeman takes City of Cleburne v. Cleburne Living Center ²⁵³x 457 U.S. 202 (1982). In neither case, of course, did the Court indicate that it was applying any test other than the standard rational basis test. and Plyler v. Doe ²⁵⁴x 473 U.S. 432 (1985). to exemplify what she means by ‘rational basis with bite’. In City of Cleburne, according to Freeman, Justice Brennan departed from standard rational basis analysis by concluding that the statute at issue “could not be considered rational unless it furthered a ‘substantial’ state goal”.²⁵⁵x Freeman, 2013, p. 286. Likewise, in Plyler, according to Freeman, Justice White “analyze[d] each of the city’s purported rationales in great detail and ultimately concluded that the statute appeared to ‘rest on an irrational prejudice against the mentally retarded’” so that “the city lacked ‘any rational basis for believing’ that a group home for retarded persons would ‘pose any special threat to the city’s legitimate interests’”.²⁵⁶x Id., at p. 286-287.
  The next step in the analysis depends on whether the regulation satisfies this heightened rational basis review. Freeman continues:

  Rational basis with bite [or heightened rationality review] includes a searching nexus analysis that enables courts to invalidate challenged legislation. If the legislation survives heightened rationality review, the court should then assess the permissibility of its purpose and the severity of its effects. Should the legislation fail heightened rationality review, however, the court should invalidate the statute without proceeding to the purpose and effects test. Derived in equal measure from Casey’s text and Roe’s promise, the method aims to balance loyalty to precedent with advocacy for the abortion right.²⁵⁷x Id., at p. 280.

  Importantly, Freeman observes that “only rational basis with bite can truly assess – rather than simply presume – the legitimacy of a state’s interest.”²⁵⁸x Id., at p. 294, quoting Casey, 505 U.S. at p. 898 (plurality opinion). That is so because heightened rationality or rational basis with bite review (in contrast to ordinary rational basis review) does not permit the state to prevail in a challenge to its regulation by simply invoking a ‘conceivable’ justification that played no role whatever in the state’s determination but was merely the product of resourceful and imaginative litigation lawyers who were called upon to justify the measure after the fact. If one can generalize from Plyler and City of Cleburne, what seems to be significant to the Court in heightened rationality review is not the articulation of reasons that are merely ‘conceivable’, but the identification of the government’s real reasons for taking the action it took, as well as the demonstration of a connection between those reasons and a ‘substantial’ goal of the state. In addition, the Court seems to expect that the state will come forward with those reasons. Under heightened rationality review, the party challenging the regulation does not have the obligation to prove the negative; that is, it is not required to “disprove ‘every conceivable basis which might support it’”.²⁵⁹x Heller v. Doe, 509 U.S. 312, 320 (1993). Thus, it is necessary for the Court to assess – rather than just presume – the reality and legitimacy of the state’s interest, and it is necessary to determine whether the means selected to further a real and legitimate state interest are a reasonable means of furthering that interest. Freeman further writes:

  Inherent in these statements is a substantial analysis of the relationship between the state’s interest and its legislation. Although traditional rational basis review technically contains such analysis, only rational basis with bite actually applies it: courts implementing rationality review purport to assess the connection between the state’s means and its ends, but they do not in fact do so. Nexus analysis only truly comes to fruition through rational basis with bite.²⁶⁰x Freeman, 2013, p. 294-295.

  Under Freeman’s two-part scheme, the court’s inquiry will terminate if the regulation does not meet the heightened rational basis standard.²⁶¹x Id., at p. 301. In the event that the regulation does meet that standard, however, the court will then consider the constitutionality of the regulation under the purpose and effects test. In the abortion context, the purpose prong requires an evaluation of “the state’s reason for enacting the challenged statute, determining whether it sought to make abortions more difficult to procure”.²⁶²x Id., at p. 295. Freeman continues: “[t]he effects prong, on the other hand, looks not to the state’s rationale but to the statute’s concrete consequences. How might a twenty-four-hour waiting provision, for instance, practically affect a woman seeking to obtain an abortion? In short, the purpose and effect prongs are the two routes through which a regulation may prove unduly burdensome.”²⁶³x Id., at p. 296. And the effect of the regulation will be measured on the basis of its effect on those who are affected by it, not on those for whom it does not matter.²⁶⁴x See, e.g., Whole Woman’s Health v. Hellerstedt, 136 S. Ct. 2292, 2320 (2016). Freeman also emphasizes the importance of engaging in the two inquiries in the order in which she has discussed them: “[t]he undue burden standard will not retain its intended rigor unless rational basis analysis is properly situated before, rather than within, the two prongs of that test. … Whereas the purpose and effects test examines the state’s means and its ends separately, asking whether each is sufficient in itself, the nexus inquiry of rational basis with bite examines the connection between those means and those ends, assessing the adequacy of the relationship between them.”²⁶⁵x Freeman, 2013, p. 302. Freeman finds support for her interpretation of the undue burden test in Justice Stevens’s partial concurrence in Casey, in which he remarked that, notwithstanding the opacity of the plurality’s opinion, “[t]he future may also demonstrate that a standard that analyzes both the severity of a regulatory burden and the legitimacy of its justification will provide a fully adequate framework for the review of abortion legislation even if the contours of the standard are not authoritatively articulated in any single opinion.” Id., at p. 322, quoting Casey, 505 at 920 n.6 (J. Stevens, concurring in part and dissenting in part). According to Freeman, Justice Stevens recognized that the undue burden standard “tests the weight of the burden, the legitimacy of the state’s regulatory purpose, and the sufficiency of the relationship between them. A regulation that fails any of the above components is an unconstitutionally undue burden on the right to abortion”. Id.
  Clearly, the two-part analysis set forth in Freeman’s article would provide additional protection for the constitutional right originally recognized in Roe. In Freeman’s view, it would also provide the kind of muscular protection for that right that the Casey plurality sought, but failed to achieve, when it created the ‘undue burden’ test. But what lessons, if any, does Freeman’s revamping of the Casey test hold for the similar problem facing the realm of voting rights?
  One impetus for Freeman’s work on the law relating to reproductive choice was her recognition that the Court had expended a great deal of time and effort in reaching its present point with respect to the relevant jurisprudence. A majority of the Court had soon come to the conclusion that Roe was too lopsided in its protection of the woman’s interest and gave insufficient attention to the state’s legitimate interest. For that reason, the Court embarked on a long intellectual journey, seeking what it deemed to be a better balance between the competing rights and interests within the constraints imposed by stare decisis. Although Freeman thought that the proper solution to the problem was to return to Roe’s strict scrutiny test, the Court was unlikely to go back. Indeed, given the long history of the Court’s jurisprudence in this area, it was extremely unlikely that the Court would suddenly return to its previous view that strict scrutiny provides the proper standard of review in the context of regulations relating to reproductive choice. The Court had long recognized that reproductive choice issues require the courts to be attentive to multiple interests, which was thought to make strict scrutiny inappropriate. Thus, although Freeman thought that strict scrutiny was the appropriate standard, she set about constructing an alternative mode of analysis that would ensure more muscular protection for the woman’s interest, while also giving appropriate attention to the state’s interest. In this way, she hoped to garner the support of a majority of the Court. Given the relative theoretical underdevelopment of the undue burden test, Freeman chose to focus her attention on how that test could be more adequately articulated in a way that gave greater protection to the woman’s interest in reproductive choice.
  A somewhat similar, but not identical, situation exists with respect to the protection of voting rights. To start with, both areas are characterized by some need to balance competing interests. In the reproductive choice area, the Court recognizes that the state has a strong interest both in the promotion of childbirth and in the protection of maternal health – interests that may well conflict with the woman’s interest in reproductive choice. Similarly, the Court recognizes that the right to vote would be illusory if there were no regulation of voting. Elections – and the right to vote – would be a farce if, for example, anyone could cast a vote without having to identify herself in some way, or, alternatively, could cast as many votes as she could fit into her schedule. In both cases, there are competing concerns, but the Court treats them quite differently in practice, even today. Even under existing jurisprudence, a law regulating reproductive rights will not be upheld merely because the state says that there might be a problem and a more onerous regulation might ameliorate the problem if it does exist.²⁶⁶x See, e.g., Whole Woman’s Health v. Hellerstedt, 136 S. Ct. 2292, 2309-2310 (2016) (holding that certain regulatory provisions constituted an undue burden because they did not afford “medical benefits sufficient to justify the burdens [that they imposed] upon access”). Thus, in Whole Woman’s Health, for example, the Court found no basis for doubting the district court’s conclusion that there was “no significant health-related problem that the law helped to cure”.²⁶⁷x Id., at p. 2311. If a problem does not exist, the need to solve it cannot provide the basis for a rational regulation. Yet that is the only standard that makes sense in attempting to describe the Court’s decision in Crawford. There the Court accepted a merely theoretical concern as a valid justification, while a similar, also purely theoretical concern was not deemed sufficient in Whole Woman’s Health.
  As Crawford shows, the Court’s voting rights jurisprudence has proved too anaemic in practice to provide the kind of protection that the right to vote deserves and demands. The Court clearly recognizes the central importance of the right to vote, and the Court certainly must realize the ease with which that right can be nullified by those who control the machinery of the electoral process – but the Court seems incapable of affording adequate protection to it. There is a serious need for a new approach – one that can actually be administered by the lower courts, with a view to guaranteeing the integrity of the electoral process while also ensuring that individuals are not unfairly excluded from the franchise, whether by means that are ‘sophisticated’ or by those that are ‘simple-minded’.²⁶⁸x See, e.g., Lane v. Wilson, 307 U.S. 268, 275 (1939) (Frankfurter, J.) (“The [Fifteenth] Amendment nullifies sophisticated as well as simple-minded modes of discrimination.”)
  There are several possibilities. First, the Court could return to the strict scrutiny standard that the Court adopted in Harper. Of course, the Court has moved a long way away from that standard and is unlikely to return to it. As was the case with Roe, the Court obviously believes that strict scrutiny pays insufficient attention to the state’s competing interest, which can be seen, in the case of voting, as the state’s interest in an honest and efficient voting system. For that reason, the Court presumably believes that its current approach is preferable, notwithstanding its failure to give adequate protection to the right to vote.²⁶⁹x As previously noted (see supra note 157), Dean Chemerinsky offered the Court another alternative, based on a distinction between direct and indirect violations of the right to vote, but the Court rejected that alternative as well. But a new approach is clearly warranted. Thus, another possibility is that the Court could adopt something along the lines of Freeman’s approach. This would allow the Court to place this area of the law within the class of cases to which undue burden analysis is applied, while also providing a greater level of protection for the right to vote.²⁷⁰x See G. Metzger, ‘Unburdening the Undue Burden Standard: Orienting Casey in Constitutional Jurisprudence’, Columbia Law Review, Vol. 94, 1994, p. 2038-2040 (discussing burden analysis in dormant commerce clause and First Amendment jurisprudence). The two-step analysis that Freeman would apply in the reproductive choice context would apply equally well to voting rights regulations. First, the regulations would be subject to the type of heightened rationality review that Freeman draws from Cleburne and Plyler. At that stage, the state could not rest on any “conceivable” purpose to support its legislation or regulation. The state would have the burden of coming forward with what really was on the minds of those who crafted the regulation. Where the regulation was meant to be remedial, the state would not be entitled to rely on a presumption that a problem existed. The state would be required to identify the problem that it sought to solve and to show that the problem was a real one, not something that might conceivably be a problem for someone somewhere in some theoretical universe, but an actual problem that existed in the state that enacted the legislation or promulgated the regulation. In this way, the reality and legitimacy of the state’s interest would be made a matter of proof rather than presumption. The court would also be able to address, in a similarly grounded way, the appropriateness of the means chosen to further the state’s real and legitimate interest.
  If the state could not satisfy this heightened rational basis test, that would be the end of the matter. If, on the other hand, the state succeeded in meeting that test, the court would be required to proceed further. At the second stage, the court would consider the constitutionality of the regulation under the purpose and effects test that Freeman has described or pursuant to a more general balancing test. In either event, the court would be allowed to delve more deeply into the problem, particularly with respect to the costs and benefits of the state regulation, while at the same time demonstrating the degree of respect due to the political processes of the state. At this point, however, the state would not be able to justify a regulation that yields only the smallest of benefits, while imposing a heavy burden on an identifiable group for whom the regulation matters. As the Court has made clear in the reproductive rights context, the inquiry must focus on those who are burdened by the regulation, not on those for whom it makes no difference at all. In addition, the court would be entitled at this second stage to inquire further into the purpose of the regulation. In this way, appropriate attention could be given to the state’s legitimate interest in the integrity of the electoral process, but the mere incantation of a theoretical interest would not suffice to disenfranchise those whose political beliefs and interests are thought not to be congruent with those of the officials who have the power to make the rules.

• F Conclusion

  As we have seen, the process by which meaning is given in individual cases to the requirements of substantive due process and equal protection is not tidy. The three-tiered approach may be more honoured in the breach, and it tells only part of the story in any event, while perhaps only the outlines of the ‘undue burden’ test have thus far been disclosed by the jurisprudence. But, far from agreeing with Justice Scalia, we can conclude that this area of the law is not ‘an embarrassment’. And, unlike Justice Rehnquist, we can take no comfort in basing our rejection of innovation on the ground that the words of a new test are not contained in the text of the Constitution – when that was also true of the words of the old test too.‍²⁷¹x See Craig v. Boren, 429 U.S. 190, 220 (1976) (Rehnquist, J., dissenting).
  The mere fact that equal protection law requires the use of a number of tests and therefore lacks doctrinal tidiness is not necessarily an evil in want of a cure. Aristotle seems to have understood this point when he acknowledged that different subjects admit different degrees of certainty and therefore warrant different modes of inquiry.²⁷²x See Aristotle, Nicomachean Ethics, Martin Ostwald (Trans.), Indianapolis, Boobs-Merrill, 1962, p. 18-19. Aristotle also would have recognized that approaches must change as the life they are meant to govern also change. Last year’s influenza vaccine will do no good in dealing with this year’s strain of the disease. The seemingly uncertain or changing nature of the doctrine reflects the complexity of the circumstances to which the doctrine must be applied as well as the multiplicity of values that must be considered in formulating that doctrine. This seeming uncertainty or change also reflects the fact that circumstances may require that some matters of judgment be committed for structural reasons to one kind of decision maker rather than another. A multiplicity of tests may cause confusion, and an unwarranted multiplicity of tests should not be countenanced for that reason. On the other hand, a multiplicity of tests may be required to accommodate reality. In that case, evil does not lie in a multiplicity of tests, but only in the lack of explanation and transparency as to their need and use.
  That is not to say that current doctrine necessarily is coherent or in good form, and the apparent density of this area of the law reflects that fact as well. Such complex doctrines are always in a state of evolution and always warrant improvement to better reflect an appropriate balance of values. That is particularly true at the present moment with respect to the law relating to voting rights. In Wesberry v. Sanders,²⁷³x 376 U.S. 1, 17 (1964). the Court correctly observed that “[n]o right is more precious in a free country than that of having a voice in the election of those who make the laws under which, as good citizens, we must live. Other rights, even the most basic, are illusory if the right to vote is undermined.” Representative democracy itself depends on affording the highest possible degree of protection to that right, which is ‘preservative of all rights’.²⁷⁴x Yick Wo v. Hopkins, 118 U.S. 356, 370 (1886). But, notwithstanding these inspiring words, the Court clearly has afforded an excessive degree of deference to state officials, allowing citizens to be disenfranchised on the flimsiest of grounds. As we have seen, the right to vote cannot adequately be protected when the state places the responsibility for making the rules in the hands of those whose own interests are at stake, and the courts effectively absolve the state from any obligation to show that it acted in response to a real problem or otherwise explain its actions. The law in that area clearly needs an overhaul.

Noten

• 1 Washington University Law Quarterly, Vol. 35, 1979, p. 147. The symposium papers had been delivered by a diverse group of constitutional law scholars in nine separate programmes during the 1978-1979 academic year. Id., at p. 1-3.

• 2 Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 407 (1978) (Blackmun, J., concurring in part and dissenting in part).

• 3 A. Scalia, ‘The Disease as Cure: “In Order to Get Beyond Racism, We Must First Take Account of Race”’, Washington University Law Review, Vol. 1979, No. 1, p. 147.

• 4 Id.

• 5 Id.

• 6 In terms of state governmental action, the ultimate source of equal protection and due process protection is Section 1 of the Fourteenth Amendment to the United States Constitution, which was adopted after the Civil War and provides, in relevant part, that “[n]o State shall … deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws.” U.S. Const., Amend. XIV, § 1 (1868). For federal governmental action, the ultimate source of protection is the Due Process Clause of the Fifth Amendment, which was adopted in 1791 as part of the Bill of Rights and provides in relevant part that “[n]o person shall … be deprived of life, liberty, or property, without due process of law”. U.S. Const., Amend. V (1791). The Supreme Court has made clear that the Due Process Clause of the Fifth Amendment should be interpreted to encompass a guarantee of equal protection as well as due process. See, e.g., Bolling v. Sharpe, 347 U.S. 497 (1954) (requiring public school desegregation in the District of Columbia under the equal protection component of the Due Process Clause of the Fifth Amendment). See also K. L. Karst, ‘The Fifth Amendment’s Guarantee of Equal Protection’, North Carolina Law Review, Vol. 55, 1977, p. 541-562 (discussing the equal protection component of the Due Process Clause.).

• 7 411 U.S. 677 (1973).

• 8 See Craig v. Boren, 426 U.S. 190, 220-21 (1976) (Rehnquist, J., dissenting) (“I would think we have had enough difficulty with the two standards of review which our cases have recognized – the norm of ‘rational basis,’ and the ‘compelling state interest’ [or strict scrutiny] required where a ‘suspect classification’ is involved – so as to counsel weightily against the insertion of still another ‘standard’ between those two.”). But see San Antonio v. Rodriguez, 411 U.S. 1, 98 (1973) (Marshall, J., dissenting) (“The Court apparently seeks to establish today that equal protection cases fall into one of two neat categories which dictate the appropriate standard of review -- strict scrutiny or mere rationality. But this Court’s decisions … defy such easy categorization.”)

• 9 The Frontiero plurality consisted of Justices Brennan, Douglas, White and Marshall. See Frontiero, 411 U.S. at p. 688. Justice Brennan subsequently wrote the majority opinion in Craig.

• 10 429 U.S. 190 (1976). Subsequently, in Trimble v. Gordon, 430 U.S. 762 (1977), the Court struck down a section of the Illinois Probate Code that discriminated against children born outside of a legally recognized marriage. Writing for the Court, Justice Powell stated that “[i]n a case like this, the Equal Protection Clause requires more than the mere incantation of a proper state purpose”. Id., at p. 769. Justice Rehnquist vigorously dissented in an opinion joined by three other Justices. Id., at p. 776. Katie Eyer has recently argued that the majorities in Craig and Trimble did not think of themselves as articulating a new, intermediate standard of review, but understood the cases as simply applying a more robust form of rational basis review. “[C]ases that were at the time understood by the Court itself as applying minimum tier standards have been reimagined today as outside the minimum tier canon – as cases in which the Court was acting at, but not actually, applying rational basis review.” See K.R. Eyer, ‘Constitutional Crossroads and the Canon of Rational Basis Review’, University of California, Davis Law Review, Vol. 48, 2014, p. 535. According to Professor Eyer, “the vision of minimum tier review that has come to dominate canonical accounts – a form of review so deferential as to be meaningless – has been made possible only by the exclusion from the canon of cases in which a more robust form of review was applied. In fact, when viewed over the broad sweep of history – including, but not limited to the Court’s early sex, illegitimacy, and sexual orientation cases – there is a deep history on the Court of taking groups and rights seriously, even outside of the context of formally heighteed review.” Id., at p. 536.

• 11 426 U.S. 229 (1976). The Court reinforced that approach in Village of Arlington Heights v. Metropolitan Hous. Dev. Corp., 429 U.S. 252 (1977), and Pers. Admin. of Mass. v. Feeney, 442 U.S. 256 (1979).

• 12 438 U.S. 265 (1978).

• 13 Justice Powell concluded that the Constitution and laws did not categorically prohibit race-conscious admissions decisions, because student diversity could be a legitimate academic concern of colleges and universities, but that the UC Davis medical school’s specific admissions programme nonetheless failed to meet constitutional standards. Id., at p. 314-315, 319-320. According to Justice Powell, the UC Davis plan, which was simply a ‘set-aside’ or quota system, did not allow for the individualized assessment of applicants and therefore failed to satisfy strict scrutiny, which, also according to Justice Powell, was equally applicable to both malevolent and ‘benign’ considerations of race. While Justice Powell wrote the lead opinion in Bakke, no other Justice joined in both of his central conclusions. Four Justices – Chief Justice Burger and Justices Stewart, Rehnquist, and Stevens – did not reach the constitutional question but thought (id., at p. 408-421) that a relevant federal statute (Title VI of the 1964 Civil Rights Act) prohibited any consideration of race in the admissions process. Four other Justices – Justices Brennan, White, Marshall, and Blackmun – thought that the programme did not violate Title VI or the Constitution, but those four Justices did so based on their view (not shared by Justice Powell) that so-called ‘benign’ discrimination should not be subject to strict scrutiny. Id., at p. 324-408. In all, six opinions were filed in the case. Justice Powell recognized that this dizzying array of opinions did nothing to make the Court’s holding transparent. For that reason, he delivered an oral summary of the opinions from the bench, but he did not include that summary in his published opinion. See J. Goldstein, The Intelligible Constitution, Oxford, Oxford University Press, 1992, p. 97-104. (Describing Justice Powell’s oral summary and explanation.)

• 14 Of course, then Professor Scalia was not alone in expressing frustration with the state of equal protection law. In 1972, Gerald Gunther detected a “mounting discontent [among the Justices] with the rigid two-tier formulations of the Warren Court’s equal protection doctrine”. G. Gunther, ‘The Supreme Court, 1971 Term – Foreword: In Search of Evolving Doctrine on a Changing Court: A Model For A Newer Equal Protection’, Harvard Law Review, Vol. 86, 1972, p. 12 & 17. In Dandridge v. Williams, 397 U.S. 471, 520-521 (1970) (Marshall, J., dissenting), and San Antonio v. Rodriguez, 411 U.S. 1, 98-99 (1973) (Marshall, J., dissenting), Justice Marshall had expressed disagreement with the idea that the Court’s jurisprudence could (or should) be understood in terms of precisely defined levels of scrutiny. In Rodriguez, he wrote: “[t]he Court apparently seeks to establish today that equal protection cases fall into one of two neat categories which dictate the appropriate standard of review. But this Court’s decisions … defy such easy categorization. A principled reading of what this Court has done reveals … a spectrum of standards … [and] variations in the degree of care with which the Court will scrutinize particular classifications.” Id., at p. 98-99. In Craig v. Boren, 429 U.S. 190 (1976), Justice Rehnquist objected to the articulation of what he took to be an intermediate standard of review on the very different ground that “[t]he Court’s conclusion that a law which treats males less favorably than females ‘must serve important governmental objectives and must be substantially related to achievement of those objectives’ apparently comes out of thin air. The Equal Protection Clause contains no such language, and none of our previous cases adopt that standard.” Id., at p. 220 (Rehnquist, J., dissenting). Justice Rehnquist’s objection is somewhat peculiar, given the absence of explicit support in the text of the Equal Protection Clause for either the rational basis test or the strict scrutiny test. Chief Justice Rehnquist took a more nuanced view of a similar issue in Dickerson v. United States, 530 U.S. 428, 442 (2000), where he declined to overrule Miranda v. Arizona, 384 U.S. 436 (1966), and also found unconstitutional a congressional statute that purported to do so. H. Jefferson Powell has offered an insightful commentary on the disagreement between Justice Scalia and Chief Justice Rehnquist in that case: Justice Scalia “insisted that the Court lacked the authority to invalidate the act of Congress, there being no violation of the Constitution. For the Court, Chief Justice Rehnquist conceded that the warnings required by Miranda are not ‘required by the Constitution, in the sense that nothing else will suffice to satisfy constitutional requirements’; the warnings were a strategic device created and imposed by the Court because the Justices believed existing practice ran an unacceptably high risk of permitting unconstitutional criminal convictions.” H.J. Powell, ‘Reasoning About the Irrational: The Roberts Court and the Future of Constitutional Law’, Washington Law Review, Vol. 86, 2011, p. 233.

• 15 See U.S. Const. Amend. XIV, § 5 (“The Congress shall have the power to enforce, by appropriate legislation, this article.”). In Katzenbach v. Morgan, 384 U.S. 641, 650 (1966), the Court held that, under Section 5, Congress had “the same broad powers expressed by the Necessary and Proper Clause” of Art. I, Section 8, Clause 18 of the Constitution, as construed by the Court in McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 421 (1819). In City of Boerne v. Flores, 521 U.S. 507, 520 (1997), however, the Court announced that legislation enacted pursuant to Section 5 would be reviewed under a more demanding ‘congruence and proportionality’ test.

• 16 See R.D. Rotunda & J.E. Nowak, Treatise on Constitutional Law: Substance and Procedure, 4th ed. Vol. 3, West, 2012 p.306 (emphasis added).

• 17 The same analytical framework is used in substantive due process cases, but, while the canonical version of equal protection analysis consists of three tiers, the substantive due process rubric omits the intermediate tier. See, e.g., E. Chemerinsky, Constitutional Law: Principles and Policies, Philadelphia, Wolters Kluwer, 2015, p. 564-566, 570-572. The Court’s occasional lack of clarity with respect to the applicable standard of review is demonstrated by its decision in New York City Transit Auth. v. Beazer, 440 U.S. 568 (1979), in which the Court upheld an agency’s anti-drug use employment policy without indicating the relevant level of scrutiny. Several more recent cases are also instructive. In United States v. Windsor, 133 S. Ct. 2675, 2695 (2013), e.g., the Court held, in an opinion by Justice Kennedy, that a federal statute defining marriage as the union of one man and one woman unconstitutionally deprived same-sex couples of the liberty protected by the Fifth Amendment, but the Court did not specify the level of scrutiny being used to reach that conclusion. In Lawrence v. Texas, 539 U.S. 558 (2003), Justice Kennedy, writing for the majority, found that the Texas sodomy statute violated substantive due process, but did not specifically identify the level of scrutiny utilized in making that determination. At the end of the opinion, however, Justice Kennedy did say that the statute “furthers no legitimate state interest which can justify its intrusion into the personal and private life of the individual.” Id., at p. 578. Justice O’Connor concurred in the judgment, but would have held the Texas statute unconstitutional under the Equal Protection Clause, finding that it failed the appropriate, “more searching form of rational basis review” because “moral disapproval is [not] a legitimate state interest to justify by itself a statute that bans homosexual sodomy, but not heterosexual sodomy”. Id., at p. 580, 582. In dissent, Justice Scalia pointed to the absence of any discussion of strict scrutiny in Justice Kennedy’s opinion – and to Justice Kennedy’s statement that the law was not supported by any ‘legitimate state interest’ – to argue that the proper test was rational basis review, which he thought that the Texas statute clearly met. Id., at p. 594, 599. Justice Scalia also rejected Justice O’Connor’s equal protection argument and her suggestion that a “more searching form of rational basis review” was appropriate. Id., at p. 599-602. In Obergefell v. Hodges, 135 S. Ct. 2584 (2015), the Court, in an opinion by Justice Kennedy, held that state laws prohibiting same-sex marriage violated equal protection and due process, but did not explain the analytical basis upon which that holding rested. Justice Kennedy attempted to compensate for that lack of analysis by writing that “[t]he Due Process Clause and the Equal Protection Clause are connected in a profound way, though they set forth independent principles. Rights implicit in liberty and rights secured by equal protection may rest on different precepts and are not always coextensive, yet in some instances each may be instructive as to the meaning and reach of the other. In any particular case one Clause may be thought to capture the essence of the right in a more accurate and more comprehensive way, even as the two Clauses may converge in the identification and definition of the right.” Id., at p. 2603. One of four dissenting Justices, the Chief Justice argued that “[t]he majority’s decision is an act of will, not legal judgment.” Id., at p. 2612. In addition, he noted that “[a]bsent from [the equal protection] portion [of the majority opinion] … is anything resembling our usual framework for deciding equal protection cases.” Id., at p. 2623.

• 18 See, e.g., City of Cleburne v. Cleburne Living Center, 473 U.S. 432 (1985) (purporting to apply rational basis review, but seemingly applying a stricter test); Reed v. Reed, 404 U.S. 71 (1971) (same). Some observers would make a similar point with respect to the Supreme Court’s decision in Fisher v. University of Texas, 136 S. Ct. 2198 (2016), where the Court purported to apply strict scrutiny but arguably departed from the classic version of that standard.

• 19 Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177 (1803) (Marshall, C.J.).

• 20 This view has long been identified with James Bradley Thayer. See J.B. Thayer, ‘The Origin and Scope of the American Doctrine of American Constitutional Law’, Harvard Law Review, Vol. 7, 1893, p. 129-156. Thayer was a friend and professional colleague of Justice Holmes, a teacher of Justice Brandeis, and a strong influence on Justice Frankfurter. See G.E. White, Justice Oliver Wendell Holmes: Law and the Inner Self, Oxford, Oxford University Press, 1993, p. 378-380; P.W. Kahn, Legitimacy and History, New Haven, Yale University Press, 1992, p. 84. See also Powell, 2011, p. 222 (Noting “the role of constitutional doctrine, judicial standards of scrutiny or modes of analysis that the Court creates in order to implement constitutional norms without claiming that the standards or modes of review are themselves identical to those norms”.).

• 21 Thus, rational basis review, which is the default standard, is generally thought to be strongly deferential. See, e.g., Fitzgerald v. Racing Assoc. of Central Iowa, 539 U.S. 103, 108 (2003) (“Neither could the Iowa Supreme Court deny that the 1994 legislation, seen as a whole, can rationally be understood to do what that court says it seeks to do, namely, advance the racetracks’ economic interests. Its grant to the racetracks of authority to operate slot machines should help the racetracks economically to some degree – even if its simultaneous imposition of a tax on slot machine adjusted revenues means that the law provides less help than respondents might like. At least a rational legislator might so believe.”); Nordlinger v. Hahn, 505 U.S. 1, 11-12 (1997) (“[T]he Equal Protection Clause is satisfied so long as there is a plausible policy reason for the classification, the legislative facts on which the classification is apparently based rationally may have been considered to be true by the governmental decision maker, and the relationship of the classification to its goal is not so attenuated as to render the classification arbitrary or irrational.”).

• 22 On this view, courts must accommodate two competing values: the deference due to the political branches, on the one hand, and faithfulness to the courts’ obligation to interpret and enforce the Constitution, on the other. Achieving that balance may be difficult in practice. See, e.g., J. Tussman & J. tenBroek, ‘The Equal Protection of the Laws’, California Law Review, Vol. 37, 1949, p. 366. Indeed, the Court has sometimes applied the rational basis test in a way that suggests that it believes that the primary determination as to rationality is for the Court rather than the legislature. In F.S. Royster Guano Co. v. Virginia, 453 U.S. 412, 415-417 (1920), e.g., the Court recited the test and the conclusion that there was no conceivable basis for the distinction at issue. Justice Brandeis dissented, joined by Justice Holmes, noting that they could conceive of a justification for the statutory distinction, which they therefore found ‘not illusory’. Id., at p. 418.

• 23 See, e.g., Chemerinsky, 2015, p. 570.

• 24 See supra notes 9, 14, and 16 and accompanying text.

• 25 See Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833, 874 (1992) (O’Connor, Kennedy, & Souter, JJ.) (plurality opinion). (“Only where state regulation imposes an undue burden on a woman’s ability to make this decision does the power of the State reach into the heart of the liberty protected by the Due Process Clause.”)

• 26 Perhaps partially in response to what was deemed by some to be an excessive judicial concern with individual and minority rights during the period of the Warren and Burger Courts, some recent Justices have taken a much broader view of majoritarian rule and a more circumscribed view of the Court’s role in protecting individual and minority rights. See, e.g., B.A. Murphy, Scalia: A Court of One, New York, Simon & Schuster, 2014, p. 234 (Quoting Justice Scalia on the limited constitutional protection allegedly afforded to minority rights); G. Will, ‘Where Scalia Was Wrong’, National Review, 1 February 2017, http://www.nationalreview.com/article/444488/antonin-scalias-natural-rights-error-neil-gorsuch-supreme-court-nomination (Arguing that Justice Scalia misunderstood the importance of minority rights in the American system of constitutional democracy). Long before he joined the Court, Justice Rehnquist was a stern critic of Warren Court decisions such as Brown v. Bd. of Educ., 347 U.S. 483 (1954), and he may well have given inaccurate testimony during his confirmation hearings about a pre-decisional memorandum that he drafted for Justice Jackson, for whom he served as a law clerk while the case was pending. The memorandum argued that Plessy v. Ferguson, 163 U.S. 537 (1886), which held that racial segregation did not violate the equal protection clause, was correctly decided. At his confirmation hearings, Justice Rehnquist insisted that the memorandum reflected Justice Jackson’s views rather than his own. See B. Snyder & J.Q. Barrett, ‘Rehnquist’s Missing Letter: A Former Law Clerk’s 1955 Thoughts on Justice Jackson and Brown’, Boston College Law Review, Vol. 53, p. 632. Among other things, the memorandum asserted that, although in theory “a majority may not deprive a minority of its constitutional right … in the long run it is the majority who will decide what the rights of the minority are.” D.M. O’Brien, Justice Robert H. Jackson’s Unpublished Opinion in Brown v. Board, University of Kansas Press, 2017, p. 73 (Quoting Memorandum). Moreover, the Court has recently begun to discuss in explicit terms the need for maintaining public approval of its work, which may be seen to set some parameters with respect to the Court’s vindication of individual and minority rights. In Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833, 868 (1992), e.g., the plurality emphasized the Court’s need for maintaining popular support and belief in its legitimacy: “It is true that diminished legitimacy may be restored, but only slowly. Unlike the political branches, a Court thus weakened could not seek to regain its position with a new mandate from the voters, and even if the Court could somehow go to the polls, the loss of its principled character could not be retrieved by the casting of so many votes. Like the character of an individual, the legitimacy of the Court must be earned over time. So, indeed, must be the character of a Nation of people who aspire to live according to the rule of law. Their belief in themselves as such a people is not readily separable from their understanding of the Court invested with the authority to decide their constitutional cases and speak before all others for their constitutional ideals. If the Court’s legitimacy should be undermined, then, so would the country be in its very ability to see itself through its constitutional ideals. The Court’s concern with legitimacy is not for the sake of the Court, but for the sake of the Nation to which it is responsible.” See B. Friedman, The Will of the People: How Public Opinion Has Influenced the Supreme Court and Shaped the Meaning of the Constitution, Farrar, Strauss, and Giroux, 2009, p. 328-330 (discussing plurality’s observations concerning legitimacy in Casey).

• 27 See Letter of James Madison to Thomas Jefferson, October 17, 1788, in J. Madison, Writings, J. Rakove (Ed.), New York City, The Library of America, 1999, p. 421 (“In our Governments, the real power lies in the majority of the Community, and the invasion of private rights is chiefly to be apprehended, not from acts of Government, contrary to the sense of its constituents, but from acts in which the Government is the mere instrument of the major number of the constituents.”); W.F. Murphy, Constitutional Democracy: Creating and Maintaining a Just Political Order, Baltimore, The Johns Hopkins University Press, 2007, p. 10 (“Although the people’s freely chosen representatives should govern, those officials must respect certain substantive limitations on their authority”).The United States Constitution leaves the states with “wide leeway when experimenting with the appropriate allocation of legislative power”, Hunter v. Pittsburgh, 207 U.S. 161, 178 (1907), and some state constitutions provide for direct democracy in the form of referenda and initiatives, see, e.g., Cal. Const. art. 2, §§ 8 and 9, but the United States Constitution notably does not. Peter Schrag has provided a useful account of California’s experience with direct democracy. See P. Schrag, Paradise Lost: California’s Experience, America’s Future, New York City, The New Press, 1998 (discussing California’s experience with direct democracy).

• 28 See G. Gunther, ‘The Supreme Court, 1971 Term – Foreword: In Search of Evolving Doctrine on a Changing Court: A Model For A Newer Equal Protection’, Harvard Law Review, Vol. 86, 1972, p. 44. See also B. Sullivan, ‘FOIA and the First Amendment: Representative Democracy and the People’s Elusive “Right to Know”’, Maryland Law Review, Vol. 72, 2012, p. 1-84 (emphasizing importance of access to government information in constitutional democracy).

• 29 118 U.S. 356, 370 (1886).

• 30 See, e.g., Pew Research Center, Political Polarization in the American Public: How Increasing Ideological Uniformity and Partisan Antipathy Affects Politics, Compromise and Everyday Life (June 2014), http://assets.pewresearch.org/wp-content/uploads/sites/5/2014/06/6-12-2014-Political-Polarization-Release.pdf.

• 31 See, e.g., A.J. McGann et al, Gerrymandering in America: The House of Representatives, the Supreme Court, and the Future of Popular Sovereignty, Cambridge, Cambridge University Press, 2016, p. 2. (“We now have a remarkable situation. Drawing districts with different population sizes is prohibited by the Constitution. However, achieving the same partisan advantage by cleverly manipulating the shape of the districts apparently is permitted.”)

• 32 See E. Freeman, ‘Giving Casey Its Bite Back: The Role of Rational Basis Review in Undue Burden Analysis’, Harvard Civil Rights-Civil Liberties Law Review, Vol. 48, 2013, p. 279-323.

• 33 473 U.S. 432 (1985). In Cleburne, the sponsors of a proposed group home for developmentally disabled persons challenged the city’s requirement that they seek a special use permit that was not generally required. They argued that mental disability should be considered a ‘quasi-suspect classification’, and therefore subject to a standard of review more exacting than ‘rational basis’. While the Court rejected that argument, it held that the special use permit requirement did not even satisfy the rational basis standard because it appeared to rest only on an irrational prejudice against developmentally disabled persons. Id., at p. 446, 450. Ironically, Cleburne illustrates the danger of putting too much faith in the canonical three-tiered approach, because, while Justice White accurately summarized the three-tiered approach, the result in the case seemingly cannot be explained in those terms. Instead, the mode of analysis used in the Court’s decision appears to be more demanding than the canonical form of rational basis review that the Court purportedly applied.

• 34 Id., at p. 439-442. The Court explained that age had not been deemed to be a suspect or quasi-suspect classification because the aged have not experienced a “‘history of purposeful unequal treatment’ or been subjected to unique disabilities on the basis of stereotyped characteristics not truly indicative of their abilities”. Id., at p. 441.

• 35 429 U.S. 190 (1976). Earlier, in Reed v. Reed, 404 U.S. 71, 76 (1971), the Court had struck down a gender-based classification, purportedly because it failed to satisfy the rational basis test. Also during the 1970s, a plurality of the Court flirted with the idea that gender-based classifications should be evaluated under strict scrutiny, but that view did not command a majority of the Court. See Frontiero v. Richardson, 411 U.S. 677, 688 (1973) (Brennan, J.) (plurality opinion) (“[C]lassifications based on sex, like classifications based on race, alienage, or national origin, are inherently suspect, and must therefore be subject to strict judicial scrutiny.”) As previously noted, some scholars would attach a later date to the emergence of intermediate scrutiny on the ground that the majority in Craig was actually utilizing a more demanding form of rational basis review, rather than articulating a new, intermediate level of scrutiny. See K.R. Eyer, ‘Constitutional Crossroads and the Canon of Rational Basis Review’, University of California, Davis Law Review, Vol. 48, 2014, p. 535. In any event, some Justices clearly did not believe that there was a two-tiered system prior to Craig. In San Antonio v. Rodriguez, 411 U.S. 1, 98-99 (1973) (Marshall, J., dissenting), e.g., Justice Marshall insisted that a ‘principled reading’ of the Court’s jurisprudence revealed a ‘spectrum of standards’ and ‘variations in the degree of care with which the Court will scrutinize various classifications’. In any event, the Court extended the sweep of intermediate scrutiny review to include classifications based on illegitimacy in Clark v. Jeter, 486 U.S. 456, 461-465 (1988).

• 36 See S.A. Siegel, ‘The Origin of the Compelling State Interest Test and Strict Scrutiny’, American Journal of Legal History, Vol. 48, 2006, p. 402 (“In that year, Justice Powell, who was not speaking for the Court, employed strict scrutiny in casting the deciding vote in Board of Regents of the University of California v. Bakke [438 U.S. 265 (1978)].”). But see R.H. Fallon, ‘Strict Judicial Scrutiny’, University of California Law Review, Vol. 54, 2007, p. 1276 (noting that, in Bolling v. Sharpe, 347 U.S. 497 (1954) (the companion case to Brown v. Board of Education, 347 U.S. 483 (1954), which involved school desegregation in the District of Columbia), the Court “said that ‘[c]lassifications based solely upon race must be scrutinized with particular care, since they are contrary to our traditions’ and thus … are ‘constitutionally suspect’”.). In Korematsu v. United States, 323 U.S. 214, 216 (1944), of course, Justice Black stated that “courts must subject [legal restrictions which curtail the civil rights of a single racial group] to the most rigid scrutiny”, and Justice Douglas used the term ‘strict scrutiny’ in Skinner v. Oklahoma, 316 U.S. 353, 541 (1942), but that case did not involve a racial classification.

• 37 See supra notes 17, 18, and 35 and accompanying text.

• 38 Cleburne, 437 U.S. at 439. See also F.S. Royster Guano Co. v. Virginia, 253 U.S. 412, 415 (1920).

• 39 See Plyler v. Doe, 457 U.S. 202, 217 (1982) (“The initial discretion to determine what is ‘different’ and what is ‘the same’ resides in the legislatures of the States. A legislature must have substantial latitude to establish classifications that roughly approximate the nature of the problem perceived, that accommodate competing concerns both public and private, and that account for limitations on the practical ability of the State to remedy every ill.”). See also I. Berlin, ‘Equality’, in H. Hardy (Ed.), Concepts and Categories: Philosophical Essays by Isaiah Berlin, New York City, Viking Press, 1979, p. 98 (“The goodness of the reasons will depend upon the degree of value or importance attached to the purposes or motives adduced in justifying the exceptions, and these will vary as the moral convictions – the general outlooks – of different individuals or societies vary.”). But cf. Aristotle, The Politics of Aristotle, E. Barker (Ed.), Oxford, Oxford University Press, 1958, p. 131. (“[T]here is … no good reason for basing a claim to the exercise of authority on any and every kind of superiority. Some may be swift and others slow; but this is no reason why the one should have more [political rights], and the other less. It is in athletic contests that the superiority of the swift receives its reward.”)

• 40 See, e.g., Aristotle, Ethica Nicomachea, Bk.5, §1131a (W.D. Ross, trans. 1925) (“And the same equality will exist between the persons and between the things concerned; for as the latter – the things concerned – are related, so are the former; if they are not equal, they will not have what is equal, but this is the origin of quarrels and complaints – when either equals have and are awarded unequal shares, or unequals equal shares.”). See also P. Westen, ‘The Empty Idea of Equality’, Harvard Law Review, Vol. 95, 1982, p. 543 (summarizing Aristotle: “[e]quality in morals means this: things that are alike should be treated alike, while things that are unalike should be treated unalike in proportion to their unalikeness.”)

• 41 See, e.g., Tussman & tenBroek, 1949, p. 341-381 (“[L]aws may classify. And ‘the very idea of classification is that of inequality.’ … [T]he Court has neither abandoned the demand for equality nor denied the legislative right to classify. … It has resolved the[se] contradictory demands … by a doctrine of reasonable classification. … The Constitution … does require … that those who are similarly situated be similarly treated. The measure of the reasonableness of a classification is the degree of its success in treating similarly those similarly situated. The difficulties concealed in this proposition will [need to be] analyzed.”)

• 42 See United States Constitution, Art. VI cl. 3 (“The Senators and Representatives before mentioned, and the Members of the several State Legislatures, and all executive and judicial Officers, both of the United States and of the several States, shall be bound by Oath or Affirmation, to support this Constitution.”). Interestingly, some scholars have recently argued that the rational relationship test, by deferring to the people’s representatives, does too little to protect the people themselves. See, e.g., R.E. Barnett, ‘Why Popular Sovereignty Requires the Due Process of Law to Challenge ‘Irrational or Arbitrary’ Statutes’, Georgetown Journal of Law and Public Policy, Vol. 14, 2016, p. 368. (“For the people cannot be presumed to have ‘entrusted to the government’ the power to irrationally or arbitrarily restrict their liberties.”)

• 43 That is not to say, of course, that judges function like machines in the area of legal interpretation; their views are influenced in one way or another, to one degree or another, by education and experience. See B. Sullivan, ‘The Power of Imagination: Diversity and the Education of Lawyers and Judges’, University of California, Davis Law Review, Vol. 51, 2018, p. 1109. Moreover, as Justice Breyer has noted, the opinions of individual Justices “have emphasized different constitutional themes, objectives, or approaches over time.” See S. Breyer, Active Liberty: Interpreting our Democratic Constitution, New York City, Alfred A. Knopf, 2005, p. 9. For his own part, Justice Breyer would give special emphasis to the concept of constitutional liberty, which he takes to mean “not only freedom from government coercion, but also the freedom to participate in the government itself”. Id., at p. 3. In Justice Breyer’s view, the current Court has not paid sufficient attention to the second aspect, namely the individual citizen’s right to participate in government. Id., at p. 10-11.

• 44 See, e.g., Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177-178 (1803) (C.J. Marshall). (“It is emphatically the province and duty of the Judicial Department to say what the law is. Those who apply the rule to particular cases must, of necessity, expound and interpret the rule. … This is of the very essence of judicial duty.”). See also M.J. Klarman, How ‘Great Were the “Great” Marshall Court Opinions?’, Virginia Law Review, Vol. 87, 2001, p. 1113-117 (showing that judicial review was well established before Marbury).

• 45 See supra notes 21-24 and accompanying text.

• 46 The essential constitutional concern with the question ‘Who decides?’ necessarily precludes the universal application of a de novo proportionality test. See, e.g., N. Gertner, ‘On Competence, Legitimacy, and Proportionality’, University of Pennsylvania Law Review, Vol. 160, 2012, p. 1587. (“Proportionality analysis is simply not within the competence of the American judiciary. Worse yet it is not even within their legitimate role.”) Such a test would greatly simplify the work of the courts, but it would not satisfy the threshold concern as to legitimacy. That is not to say, of course, that proportionality is irrelevant or unimportant. See, e.g., E.T. Sullivan & R.S. Frase, Proportionality Principles in American Law: Controlling Excessive Government Actions, Oxford, Oxford University Press, 2009, p. 6 (noting increasing reliance on proportionality review in US law).

• 47 See, e.g., Craig v. Boren, 429 U.S. 190, 211-212 (1976) (Stevens, J., concurring) (“There is only one Equal Protection Clause. It requires every State to govern impartially. It does not direct the courts to apply one standard of review in some cases and a different standard in other cases.”)

• 48 See, e.g., National Mutual Ins. Co. v. Tidewater Transfer Co., Inc., 337 U.S. 582, 626 (1949) (Frankfurter, J., dissenting) (“Great concepts like ‘Commerce … among the several States,’ ‘due process of law,’ ‘liberty,’ ‘property’ were purposely left to gather meaning from experience. For they relate to the whole domain of social and economic fact, and the statesmen who founded this Nation knew too well that only a stagnant society remains unchanged.”)

• 49 See, e.g., W. E. Nelson, The Fourteenth Amendment: From Political Principle to Judicial Doctrine, Cambridge, Harvard University Press, 1988, p. 55 (“One noteworthy feature of this proposal, from which the Fourteenth Amendment was ultimately derived, was its apparent adoption of the suggestion of Representative Giles W. Hotchkiss … that any new constitutional provision be framed as a self-executing guarantee of rights, and not merely as a grant of power to Congress to legislate for the protection of rights. … [Hotchkiss] wanted to be certain that rights would be enforced by the judiciary even if Congress fell under Democratic control.”) Many would argue that certain recent decisions reflect the opposite problem, that is, that Congress may have become more solicitous than the Court with respect to the values embodied in the Civil War Amendments. See, e.g., Shelby County v. Holder, 133 S. Ct. 2612 (2013) (striking down the central part of the Voting Rights Act of 1965, which had been re-enacted in 2006 by votes of 98-0 in the Senate and 390-33 in the House of Representatives and signed into law by President George W. Bush).

• 50 Marbury v. Madison, 5 U.S. (1 Cranch) at p. 177.

• 51 See, e.g., R.B. Siegel & R.C. Post, ‘Legislative Constitutionalism and Section Five Power: Policentric Interpretation of the Family and Medical Leave Act’, Yale Law Journal, Vol. 112, 2003, p. 1945 (“Because Section 5 of the Fourteenth Amendment vests in Congress ‘power to enforce, by appropriate legislation, the provisions of this article,’ the great rights contained in Section 1 … are enforced by both Congress and the Court. How to conceive of the relationship between the legislative power established in Section 5 and the judicial power authorized by Section 1 is one of the deep puzzles of American constitutional law.”) See also W.D. Araiza, Enforcing the Equal Protection Clause: Congressional Power, Judicial Doctrine, and Constitutional Law, New York, New York University Press, 2015, p. 17 (The “core insight [of City of Boerne v. Flores, 521 U.S. 507 (1997)] – that enforcement legislation must exhibit some relationship to Court-stated Fourteenth Amendment law – appears here to stay. A court’s scrutiny of that relationship may well be deferential. … Similarly, the Court may have to adjust its understanding of what that underlying Fourteenth Amendment law actually says, and thus what constitutes the target for congruence and proportionality review.”); J.T. Noonan, Narrowing the Nation’s Power: The Supreme Court Sides with the States, Berkeley, University of California Press, 2002, p. 6 (The congruence and proportionality test “means that the federal judiciary, from the Supreme Court itself down to the federal district court in Guam, may, and indeed must, treat Congress the way courts treat an administrative agency, whose work will be set aside on appeal if the court finds the record made by the agency not substantial enough to justify the agency’s rulings”.)

• 52 304 U.S. 144 (1938). In Carolene Products, the Court applied an extremely deferential standard of review to an early statute pertaining to the sale in interstate commerce of ‘filled milk’, that is, milk that was produced through the extraction of its natural cream content and the substitution of another kind of fat or oil for the natural component. In footnote 4, Justice Stone indicated that a more muscular form of constitutional review might sometimes be warranted: “[t]here may be narrower scope for operation of the presumption of constitutionality when legislation appears on its face to be within a specific prohibition of the Constitution, such as those of the first ten Amendments. … It is unnecessary to consider now whether legislation which restricts those political processes which can ordinarily be expected to bring about repeal of undesirable legislation, is to be subjected to more exacting judicial scrutiny. … Nor need we enquire whether similar considerations enter into the review of statutes directed at particular religious … or national, … or racial minorities [or] whether prejudice against discrete and insular minorities may be a special condition … [calling] for a … more searching judicial inquiry.” Id., at p. 152, n. 4.

• 53 316 U.S. 353 (1942). In his opinion for the Court, Justice Douglas used the term ‘strict scrutiny’, Id., at p. 541, but recognized that the classification (which allowed for the sterilization of thrice-convicted chicken thieves, but not for embezzlers, regardless of the degree of recidivism) was not supported by any rational basis. Id., at p. 538-539. Chief Justice Stone and Justice Jackson concurred in the result but thought that the case should have been decided under the due process clause, whereas Justice Douglas based his opinion on the equal protection clause. See id., at p. 543-547.

• 54 See, e.g., Fallon, 2007, p. 1267-1337; Siegel, 2006, p. 355-407.

• 55 See, e.g., Thayer, 1893, p. 144 (1893) (The Court “can only disregard the Act when those who have the right to make laws have not merely made a mistake, but have made a very clear one – so clear that it is not open to rational question. … This rule recognizes that, having regard to the great, complex, unfolding exigencies of government, much which will seem unconstitutional to one man, or body of men, may reasonably not seem so to another; that the constitution often admits of different interpretations; that there is often a range of choice or judgment; that in such cases the constitution does not impose upon the legislature any one specific opinion, but leaves open this range of choice; and that whatever choice is rational is constitutional”.) Compare Lochner v. New York, 198 U.S. 45, 76 (1905) (J. Holmes, dissenting). (“I think that the word liberty in the Fourteenth Amendment is perverted when it is held to prevent the natural outcome of a dominant opinion, unless it can be said that a rational and fair man necessarily would admit that the statute proposed would infringe fundamental principles as they have been understood by the traditions of our people and our law. … A reasonable man might think [the statute at issue] a proper measure on the score of health. Men whom I certainly could not pronounce unreasonable would uphold it as a first installment of a general regulation of the hours of work. Whether in the latter aspect it would be open to the charge of inequality I think it unnecessary to discuss.”)

• 56 The Court explained the deference due to state legislation in Skinner v. Oklahoma, 316 U.S. 535 (1942): “[i]t was stated in Buck v. Bell [274 U.S. 200, 208 (1927)] that the claim that state legislation violates the equal protection clause … is ‘the usual last resort of constitutional arguments.’ … [T]he States … need not provide ‘abstract symmetry’. … They may mark and set apart the classes and types of problems according to the needs and as dictated or suggested by experience. … ‘We must remember that the machinery of government would not work if it were not allowed a little play in its joints.’ … [And] the equal protection clause does not prevent the legislature from recognizing ‘degrees of evil’ … ‘[T]he law does all that is needed when it does all that it can, indicates a policy, applies it to all within the lines, and seeks to bring within the lines all similarly situated so far and so fast as its means allow.’” Id., at p. 539-540. See E. Chemerinsky, The Case Against the Supreme Court, City of Westminster, Penguin Publishing, 2014, p. 1-5 (discussing the facts of Buck v. Bell); J.M. Wisdom, ‘The Frictionmaking, Exacerbating Political Role of Federal Courts’, Southwestern Law Journal, Vol. 21, 1967, p. 411-428 (discussing role of federal courts in protecting civil rights plaintiffs against unconstitutional actions by state officials).

• 57 See, e.g., Freeman, 2013, p. 282. But see K. R. Eyer, ‘Constitutional Crossroads and the Canon of Rational Basis Review’, University of California, Davis Law Review, Vol. 48, 2014, p. 535-36 (arguing that the current view of rational basis review as ‘toothless’ is in part due to academic amnesia and the omission from the canon of earlier cases in which the courts employed a more muscular form of rationality review).

• 58 See Gunther, 1972, p. 8. But see Adarand Constructors v. Pena, 515 U.S. 200, 237 (1995) (“[W]e wish to dispel the notion that strict scrutiny is ‘strict in theory, but fatal in fact.’”).

• 59 See Freeman, 2013, p. 282-283.

• 60 Heller v. Doe, 509 U.S. 312, 320 (1993) (emphasis added).

• 61 See, e.g., K. Yoshino, ‘The New Equal Protection’, Harvard Law Review, Vol. 124, 2011, p. 756 (The Court’s framework of tiered scrutiny “distinguishes between classifications that draw ‘heightened scrutiny’ and classifications that draw ‘rational basis review.’”). As Professor Yoshino notes, the Court has not added to the list of characteristics worthy of heightened review since 1977. Id., at p. 756. “The claim that the canon has closed on heightened scrutiny classifications must be tempered by acknowledging the Court’s use of a more aggressive form of rational basis review,” which academic commentators have referred to as ‘rational basis with bite’. Id., at p. 759.

• 62 See, e.g., Craig v. Boren, 429 U.S. 190, 197 (1976) (“[C]lassifications by gender must serve important governmental objectives and must be substantially related to those objectives.”).

• 63 City of Cleburne v. Cleburne Living Center, Inc., 473 US 432, 441 (1985). Commentators have identified three tests for determining whether a classification merits intermediate scrutiny review: “[f]irst, is the classifying trait, like race, an immutable personal characteristic – an accident of birth beyond a person’s control or responsibility – rendering it presumptively unjust for the government to use the trait as a basis for allocating rewards or penalties? Second, is the trait, like race, broadly irrelevant to legitimate generalization, rendering discrimination on this basis not only unfair, but also indefensible in a wide range of governmental settings? And third, is the disadvantaged group, like African-Americans and other racial minorities, a group that lacks political power and therefore warrants special judicial solicitude, that is, special protection from the ordinary operation of the political process?” See D.O. Conkle, ‘Evolving Values, Animus, and Same-Sex Marriage’, Indiana Law Journal, Vol. 89, 2014, p. 34.

• 64 For example, some commentators have argued that sexual orientation should be subject to intermediate scrutiny as well. See, e.g., S.L. Sobel, ‘When Windsor Isn’t Enough: Why the Court Must Clarify Equal Protection Analysis for Sexual Orientation Classifications’, Cornell Journal of Law and Public Policy, Vol. 24, 2015, p. 493-531; K. LaCour, ‘License to Discriminate: How a Washington Florist is Making the Case for Applying Intermediate Scrutiny to Sexual Orientation’, Seattle University Law Review, Vol. 38, 2014, p. 122-124. Some lower courts have also applied an intermediate standard of review in sexual orientation cases, but the Supreme Court has chosen to invalidate certain classifications based on sexual orientation, without specifying the appropriate level of scrutiny. See, e.g., Windsor v. United States, 699 F.3d 169, 181 (2d Cir. 2012) (heightened scrutiny required), aff’d on other grounds, 133 S. Ct. 2675 (2013) (discrimination based on sexual orientation held unconstitutional, without specifying the appropriate level of scrutiny); Lawrence v. Texas, 539 U.S. 558 (2003) (same). See also Romer v. Evans, 517 U.S. 620 (1996) (invalidating state constitutional provision that withdrew previously granted protection against discrimination based on sexual orientation because it was motivated by a bare desire to harm a politically unpopular group).

• 65 See, e.g., Harper v. Va. State Bd. of Elections, 383 U.S. 663, 667, 670 (1966) (“Long ago, in Yick Wo v. Hopkins, … the Court referred to ‘the political franchise of voting’ as a ‘fundamental political right, because preservative of all rights.’ … We have long been mindful that, where fundamental rights and liberties are asserted under the Equal Protection Clause, classifications which might invade or restrain them must be closely scrutinized and carefully confined.”).

• 66 See, e.g., Adarand Construction, Inc. v. Pena, 515 U.S. 200, 227 (1995) (holding that all racial classifications, whether benign or not, are subject to strict scrutiny); Clark v. Jeter, 486 U.S. 456, 461 (1988) (indicating that classifications based on illegitimacy are similarly subject to strict scrutiny).

• 67 551 U.S. 701 (2007).

• 68 Id., at p. 720. In Parents Involved, the Court struck down certain school assignment plans that sought to achieve racial balance in the public schools. The Court applied strict scrutiny and ultimately found the plans to be unconstitutional, because they took race into account. In a controversial plurality opinion, Chief Justice Roberts suggested that this result was dictated by the Court’s decision in Brown v. Board of Education, 347 U.S. 483, 494 (1954), which he took to hold that any consideration of race in school assignments was subject to strict scrutiny and ordinarily unconstitutional. Parents Involved, 551 U.S. at p. 746. Others have thought Brown to be concerned with the problem of racial classification in aid of prejudice and discrimination or subordination, rather than with the mere existence of racial classification. See e.g. R.B. Siegel, ‘Equality Talk: Antisubordination and Anticlassification Values in Constitutional Struggles over Brown’, Harvard Law Review, Vol. 117, 2004, p. 1470-1547. In Palmore v. Sidoti, 466 U.S. 429, 432 (1984), Chief Justice Burger discussed the rationale for applying strict scrutiny in terms more consistent with Professor Siegel’s anti-subordination theory: “[c]lassifying persons according to their race is more likely to reflect racial prejudice than legitimate public concerns.” In Palmore, the Court set aside a state court judgment that awarded custody to a child’s father simply because the child’s mother had married an African-American after the failure of the relationship that produced the child. The state courts had reasoned that being part of a mixed-race household was not ‘in the best interests’ of the child. Id., at p. 433. The Supreme Court recognized that the child might experience prejudice because of his family situation but concluded: “[t]he question … is whether the reality of private biases and the possible injury they might inflict are permissible considerations for removal of an infant child from the custody of its natural mother. We have little difficulty concluding that they are not.” Id., at p. 433.

• 69 The phenomenon seems to have been identified for the first time by Gerald Gunther in 1972. See Gunther, 1972, p. 1-306. Professor Gunther noted that the Court sometimes “found bite in the equal protection clause after explicitly voicing the traditionally toothless minimal scrutiny standard”. Id., at p. 17-18. See also R. Holoszyc-Pimental, ‘Reconciling Rational Basis Review: When Does Rational Basis Bite?’, New York University Law Review, Vol. 90, 2015, p. 2071-2117 (tracing development of jurisprudence); J.B. Smith, ‘The Flaws of Rational Basis with Bite: Why the Supreme Court Should Acknowledge Its Application of Heightened Scrutiny to Classifications Based on Sexual Orientation’, Fordham Law Review, Vol. 73, 2005, p. 2769-2814 (advocating that Court should acknowledge its use of a more searching version of rational basis review in cases involving discrimination based on sexual orientation); R.C. Farrell, ‘Successful Rational Basis Claims in the Supreme Court from the 1971 Term Through Romer v. Evans’, Indiana Law Review, Vol. 32, 1999, p. 370 (tracing development of jurisprudence); G. L. Pettinga, ‘Rational Basis with Bite: Intermediate Scrutiny by Any Other Name’, Indiana Law Journal, Vol. 62, 1987, p. 779-803 (tracing development of jurisprudence).

• 70 518 U.S. 515 (1996).

• 71 Id., at p. 532-533.

• 72 See, e.g., J.E.B. v. Alabama, 511 U.S. 127, 136-137 (1994); Mississippi University for Women v. Hogan, 458 U.S. 718, 724 (1982): Kirchberg v. Feenstra, 450 U.S. p. 460-461 (1981); Personnel Admin. of Mass. v. Feeney, 442 U.S. 256, 273 (1979).

• 73 United States v. Virginia, 518 U.S. 559 (Rehnquist, C.J., concurring in judgment).

• 74 Id., at p. 559.

• 75 136 S. Ct. 2198 (2016).

• 76 Id., at p. 2215 (Thomas, J., dissenting).

• 77 Id., at p. 2220 (Alito, J., dissenting).

• 78 Id., at p. 2238.

• 79 Id., Justice Alito continued, noting that, “[e]ven though UT has never provided any coherent explanation for its asserted need to discriminate on the basis of race, and even though UT’s position relies on a series of unsupported and noxious racial assumptions, the majority concludes that UT has met its heavy burden. This conclusion is remarkable – and remarkably wrong.” Id., at p. 2243.

• 80 See Smith v. Allwright, 321 U.S. 649, 669 (1944) (Roberts, J., dissenting).

• 81 410 U.S. 113 (1973).

• 82 Id., at p. 153-155. Justice Blackmun summarized the grounds on which the constitutionality of the Texas statutes was challenged: “[t]he principal thrust of appellant’s attack … is that they improperly invade a right, said to be possessed by the pregnant woman, to choose to terminate her pregnancy. Appellant would discover this right in the concept of personal “liberty” embodied in the Fourteenth Amendment’s Due Process Clause; or in personal, marital, familial, and sexual privacy said to be protected by the Bill of Rights or its penumbras, see Griswold v. Connecticut, 381 U.S. 479 (1965); Eisenstadt v. Baird, 405 U.S. 438 (1972); Id., at p.460 (White, J., concurring in result); or among those rights reserved to the people by the Ninth Amendment, Griswold v. Connecticut, 381 U.S. at p. 486 (Goldberg, J., concurring).” Roe, 410 U.S. at p. 129. The argument, based on the existence of a liberty interest embodied in the Fourteenth Amendment, properly finds its source in Justice Harlan’s dissenting opinion in Poe v. Ullman, 367 U.S. 497, 543 (1961) (Harlan, J., dissenting). As Justice Souter later observed in Washington v. Glucksberg, 521 U.S. 702, 752, 756 n.2 (1997) (Souter, J., concurring in judgment), the Supreme Court’s modern substantive due process jurisprudence is uniquely indebted to Justice Harlan’s dissenting opinion in Poe.

• 83 Roe, 410 U.S. at p. 150.

• 84 Id., Although the Court recognized the state’s interest in protecting ‘prenatal life’, the Court rejected the view that a fetus was a person within the meaning of the Fourteenth Amendment. Id., at p. 157-158. That remains the case today.

• 85 Id., at p. 155.

• 86 Id., at p. 162-166. The Court summarized its approach: “(a) For the stage prior to approximately the end of the first trimester, the abortion decision and its effectuation must be left to the medical judgment of the pregnant woman’s attending physician. (b) For the stage subsequent to approximately the end of the first trimester, the State, in promoting its interest in the health of the mother, may, if it chooses, regulate the abortion procedure in ways that are reasonably related to maternal health. (c) For the stage subsequent to viability, the State in promoting its interest in the potentiality of human life may, if it chooses, regulate, and even proscribe, abortion except where it is necessary, in appropriate medical judgment, for the preservation of the life or health of the mother.” Id., at p. 164-165.

• 87 Id., at p. 163-164.

• 88 Justices Rehnquist and White both dissented, believing that the issues in Roe should be left to the legislative process. Id., at p. 221 (White, J., dissenting); id., at p. 223 (Rehnquist, J., dissenting).

• 89 Id., at p. 208 (Burger, C.J., concurring); id., at p. 217 (Douglas, J., concurring); id., at p. 167 (Stewart, J., concurring).

• 90 347 U.S. 483 (1954). See, e.g., M. Ziegler, ‘Beyond Backlash: Legal History, Polarization, and Roe v. Wade’, Washington and Lee Law Review, Vol. 71, 2014, p. 969-1021.

• 91 370 U.S. 421 (1962). In Engel, the Court held that New York school officials violated the Establishment Clause by requiring students to recite a government-authored prayer at the beginning of the school day, which had been a long-standing tradition, in one form or another, throughout the United States.

• 92 462 U.S. 416 (1982).

• 93 Id., at p. 420. In a challenge to portions of Akron’s ordinances regulating the conduct of abortions, the Court “affirm[ed] the judgment of the Court of Appeals invalidating those sections … that deal with parental consent, informed consent, a 24-hour waiting period, and the disposal of fetal remains [and reversed that] portion of the judgment [that] sustain[ed] Akron’s requirement that all second trimester abortions be performed in a hospital”. Id., at p. 452. According to Justice Powell, the Akron ordinances were inconsistent with the Court’s understanding in Roe that “[f]rom approximately the end of the first trimester of pregnancy, the State ‘may regulate the abortion procedure to the extent that the regulation reasonably relates to the preservation and protection of maternal health’” and that, even in the second trimester, the regulation must be consistent with “accepted medical practice” and “legitimately related to the objective the State seeks to accomplish”. Id., at p. 430-431.

• 94 Id., at p. 453. She also argued that the test was inconsistent with the Court’s more general fundamental rights jurisprudence. See id., at p. 452-453.

• 95 476 U.S. 747 (1986). In Thornburgh, the Court, speaking through Justice Blackmun, specifically reaffirmed Roe and invalidated several provisions of a Pennsylvania abortion law. Id., at p. 759 (“The States are not free, under the guise of protecting maternal health or potential life, to intimidate women into continuing pregnancies. Appellants claim that the statutory provisions before us today further legitimate compelling interests of the Commonwealth. Close analysis of those provisions, however, shows that they wholly subordinate constitutional privacy interests and concerns with maternal health in an effort to deter a woman from making a decision that, with her physician, is hers to make.”) Justice Stevens filed an important concurrence, in which he responded to several points made by Justice White’s dissent. Chief Justice Burger, Justice Rehnquist and Justice O’Connor also filed dissenting opinions. Justice Rehnquist joined in both Justice White’s and Justice O’Connor’s dissents.

• 96 Id., at p. 828. Justice O’Connor continued: “[t]hese principles for evaluating state regulation of abortion were not newly minted in my dissenting opinion in Akron. Apart from Roe’s outmoded trimester framework, the ‘unduly burdensome’ standard had been articulated and applied with fair consistency by this Court in cases such as Harris v. McRae, 448 U.S. 297, 314 (1980), Maher v. Roe, 432 U.S. 464, 473 (1977), Beal v. Doe, 432 U.S. 438, 446 (1977), and Bellotti v. Baird, 428 U.S. 132, 147 (1976). In Akron and Ashcroft, the Court, in my view, distorted and misapplied this standard, see Akron, 462 U.S. at p. 452-453 (O’Connor, J., dissenting), but made no clean break with precedent, and indeed ‘follow[ed] this approach’ in assessing some of the regulations before it in those cases. Id., at p. 463 (O’Connor, J., dissenting).” Thornburgh, 476 U.S. at p. 828-829.

• 97 492 U.S. 490 (1989).

• 98 Chief Justice Rehnquist, in a plurality opinion joined by Justices White and Kennedy, strongly criticized the holding in Roe: “We have not refrained from reconsideration of a prior construction of the Constitution that has proved ‘unsound in principle and unworkable in practice.’ … [T]he rigid Roe [trimester] framework is hardly consistent with the notion of a Constitution cast in general terms, as ours is, and usually speaking in general principles, as ours does. The key elements of the Roe framework – trimesters and viability – are not found in the text of the Constitution, or in any place else one would expect to find a constitutional principle. Since the bounds of the inquiry are essentially indeterminate, the result has been a web of legal rules that have become increasingly intricate, resembling a code of regulations rather than a body of constitutional doctrine. … [T]he trimester framework has left this Court to serve as the country’s ‘ex officio medical board.’” Id., at p. 518-19. He also faulted the line drawn in Roe between the state’s pre-viability and viability interests. Id., at p. 519. But the testing requirement at issue was “reasonably designed to ensure that abortions are not performed where the fetus is viable – an end which all concede is legitimate – and … sufficient to sustain its constitutionality”. Id., at p. 520. Writing separately, Justice Scalia also thought that Roe should be explicitly overruled. Id., at p. 532. Justice Blackmun, Justice Brennan and Justice Marshall dissented. Justice Blackmun saw the writing on the wall: “[f]or today, at least, the law of abortion stands undisturbed. … But the signs are … very ominous, and a chill wind blows.” Id., at p. 557.

• 99 Id., at p. 525.

• 100 Id., at p. 526.

• 101 Id., at p. 529-530.

• 102 505 U.S. 833 (1992). In its petition for Supreme Court review, Planned Parenthood framed the question presented as “Whether the Supreme Court overruled Roe v. Wade, holding that a woman’s right to choose abortion is a fundamental right protected by the United States Constitution?” Planned Parenthood v. Casey, 505 U.S. 833 (1992), Petition for A Writ of Certiorari at i (No.91-744). As Jeffrey Toobin has pointed out, the question was extremely provocative, essentially suggesting that the Court might have decided that Roe was no longer binding precedent, without being forthright about it. See J. Toobin, The Nine: Inside the Secret World of the United States Supreme Court, New York, Anchor, 2007, p. 49. The United States took the position that Roe should be overruled. See Planned Parenthood v. Casey, 505 U.S. 833 (1992), Brief for the United States as Amicus Curiae (No. 91-744).

• 103 Id., at p. 843. Justices Blackmun and Stevens each filed opinions (see id., at p. 911 Stevens, J., concurring in part and dissenting in part; id., at p. 922 Blackmun, J., concurring in part, concurring in the judgment in part, and dissenting in part), as did Chief Justice Rehnquist (speaking for himself and Justices White, Scalia and Thomas) and Justice Scalia (speaking for himself, Chief Justice Rehnquist and Justices White and Thomas). See id., at p. 944 (Rehnquist, C.J., concurring in the judgment in part and dissenting in part); id., at p. 979 (Scalia, J., concurring in the judgment in part and dissenting in part).

• 104 Id., at p. 854-69. Among other things, the plurality found that scientific advances had made the Roe Court’s trimester scheme obsolete and that the analytic framework should be centred on viability. Id., at p. 860, 870. The trimester scheme also was seen to “undervalue […] the potential life within the woman”. Id., at p. 875. In any event, the plurality “reject[ed] the trimester framework, which we do not consider to be part of the essential holding of Roe”. Id., at p. 873.

• 105 Id., at p. 846. The plurality wrote that “[a] decision to overrule Roe’s essential holding under the existing circumstances would address error, if error there was, at the cost of both profound and unnecessary damage to the Court’s legitimacy, and to the Nation’s commitment to the rule of law. It is therefore imperative to adhere to the essence of Roe’s original decision, and we do so today.” Id., at p. 869. Chief Justice Rehnquist mocked the plurality’s “newly minted variation on stare decisis”, id., at p. 944 (Rehnquist, C.J., concurring in the judgment and dissenting in part) and catalogued all of the plurality’s disagreements with Roe. Id., at p. 953-954. Most significantly, “Roe decided that a woman had a fundamental right to an abortion. The joint opinion rejects that view. Roe decided that abortion regulations were to be subjected to ‘strict scrutiny’ and would be justified only in the light of ‘compelling state interests.’ The joint opinion rejects that view.” Id., at p. 954.

• 106 Id., at p. 846.

• 107 Id., at p. 869. In a somewhat strange turn of phrase, the plurality then noted that “the basic decision in Roe was based on a constitutional analysis which we cannot now repudiate”. Id. (emphasis added). In the same vein, the plurality observed that “[t]he woman’s right to terminate her pregnancy before viability is the most central principle of Roe v. Wade. It is a rule of law and component of liberty that we cannot renounce.” Id., at p. 871 (emphasis added). Chief Justice Rehnquist construed these expressions, among others, as evidence that the plurality lacked enthusiasm for defending the merits of Roe. See id., at p. 954 (the plurality “cannot bring itself to say that Roe was correct as an original matter”). But that point may understate the significance of the plurality’s insistence on what it took to be the ‘most central principle’ of the case.

• 108 Id., at p. 871.

• 109 Id.

• 110 Id., at p. 873-874.

• 111 Id., at p. 876.

• 112 Id.

• 113 Id.

• 114 Thornburgh v. American College of Obstetricians & Gynecologists, 476 U.S. at p. 828 (O’Connor, J., dissenting). In addition, the plurality took up Justice O’Connor’s suggestion in City of Akron that “this ‘unduly burdensome’ standard should be applied to the challenged regulations throughout the entire pregnancy without reference to the particular ‘stage’ of pregnancy involved.” City of Akron v. Akron Center for Reproductive Health, 462 U.S. at p. 453 (O’Connor, J., dissenting).

• 115 Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. at p. 877.

• 116 Id., at p. 878.

• 117 Id.

• 118 Id., at p. 878-879 (quoting Roe v. Wade, 410 U.S. at 164-165).

• 119 Id., at p. 894.

• 120 Id., at p. 914.

• 121 Id., at p. 925. In addition, according to Justice Blackmun, the “application of [the trimester] analytical framework is no less warranted than when it was approved by seven Members of this Court in Roe. Strict scrutiny of state limitations on reproductive choice still offers the most secure protection of the woman’s right to make her own reproductive decisions, free from state coercion.” Id., at p. 930.

• 122 See id., at p. 944 (Rehnquist, C.J., concurring in part and dissenting in part); id., at p. 979 (Scalia, J., concurring in part and dissenting in part).

• 123 Id., at p. 944.

• 124 Id., at p. 945.

• 125 Id. Chief Justice Rehnquist was particularly critical of the plurality’s ‘undue burden’ standard: “Roe v. Wade adopted a ‘fundamental right’ standard under which state regulations could survive only if they met the requirement of ‘strict scrutiny.’ While we disagree with that standard, it at least had a recognized basis in constitutional law at the time Roe was decided. The same cannot be said for the ‘undue burden’ standard, which is created largely out of whole cloth by the authors of the joint opinion. It is a standard which even today does not command the support of a majority of this Court. And it will not, we believe, result in the sort of ‘simple limitation,’ easily applied, which the joint opinion anticipates. … In sum, it is a standard which is not built to last.” Id., at p. 964-965.

• 126 Id., at p. 966, citing Williamson v. Lee Optical Co. of Oklahoma, Inc., 348 U.S. 483, 491 (1955).

• 127 Id., at p. 1002. (“[B]y foreclosing all democratic outlet for the deep passions this issue arouses, by banishing the issue from the political forum that gives all participants, even the losers, the satisfaction of a fair hearing and an honest fight, by continuing the imposition of a rigid national rule instead of allowing for regional differences, the Court merely prolongs and intensifies the anguish.”)

• 128 530 U.S. 914 (2000).

• 129 Id., at p. 920-21. Justices Stevens, O’Connor and Ginsburg filed concurring opinions, while Chief Justice Rehnquist and Justices Kennedy, Scalia and Thomas filed dissenting opinions.

• 130 See Whole Woman’s Health v. Hellerstedt, 136 S.Ct. 2292, 2309-2310 (2016) (holding that certain regulatory provisions constituted an undue burden because they did not afford “medical benefits sufficient to justify the burdens [that they imposed] upon access”); Gonzales v. Carhart, 550 U.S. 124, 167-168 (2007) (distinguishing the statute invalidated in Stenberg and holding that those challenging the facial validity of a federal statute prohibiting so-called ‘partial birth abortions’ had failed to demonstrate that the statute “would be unconstitutional in a large fraction of relevant cases”).

• 131 See, e.g., Freeman, 2013, p. 279 (noting that Casey “has engendered confusion rather than clarity” and that “the correct method of implementing [its] test remains murky”). In this regard, Freeman notes that the “courts have applied Casey inconsistently and unfaithfully, creating a tangled body of abortion precedent and rendering the undue burden standard insufficient to protect women’s reproductive autonomy”. Id., p. 279.

• 132 See Chemerinsky, 2015, p. 828.

• 133 Id., at p. 863.

• 134 Id., at p. 864.

• 135 Id.

• 136 Id.

• 137 U.S. Const. Art. I, § 2, cl. 1. In addition, the Constitution provides that the president and vice-president shall be chosen by an electoral college, rather than by the voters, and it leaves to the states the determination as to how the members of the electoral college should be selected. Id., at Art. II, § 1, Amend. XII, XX. And the Constitution originally provided that members of the Senate were to be chosen by the members of the state legislatures. Id., at Art. I, § 2, cl. 1. The Seventeenth Amendment provided for the direct election of Senators in 1919. Id., at Amend XVII.

• 138 Id., at Amend. XV.

• 139 Id., at Amend. XIX.

• 140 Id., at Amend. XXIV.

• 141 Id., at Amend. XXVI. In addition, the Constitution originally provided that the members of the federal House of Representatives would be directly elected by the people, while members of the Senate would be chosen by the state legislatures. See U.S. Const., Art. I, § 3, cl. 1. In 1913, however, the Seventh Amendment was adopted to provide for the direct election of senators. Id., at Amend. XVII.

• 142 Voting Rights Act of 1965, Section 2, currently codified at 52 U.S.C. § 10301 et seq. In Shelby County v. Holder, 570 U.S. 2 (2013), the Court found that the pre-clearance coverage formula, a key section of the Voting Rights Act that required certain ‘covered’ jurisdictions to secure prior approval for changes in voter qualifications and other matters relating to the franchise, was unconstitutional because the coverage formula was based on stale data, so that it was no longer responsive to current needs and therefore an impermissible burden on the constitutional principles of federalism and equal sovereignty of the states. Many previously covered jurisdictions have recently adopted measures to make it more difficult to vote. See Brennan Center for Justice, ‘Election 2016: Restrictive Voting Laws By The Numbers’, 28 September 2016, available at: https://www.brennancenter.org/analysis/election-2016-restrictive-voting-laws-numbers (“Starting after the 2010 election, legislators in nearly half the states passed a wave of laws making it harder to vote. These new restrictions ranged from cuts to early voting to burdens on voter registration to strict voter ID requirements. While courts stepped in before the 2012 election to block many of these laws, the Supreme Court’s 2013 decision in Shelby County gutting the most powerful protections of the Voting Rights Act made it even easier for states to put in place restrictive voting laws.”) On the other hand, Atiba Ellis points out that “politicians, typically of a conservative persuasion, have echoed the voter fraud argument since the November 2000 election and resulting Bush v. Gore, 531 U.S. 98 (2000) debacle.” A.R. Ellis, ‘The Meme of Voter Fraud’, Catholic University Law Review, Vol. 63, 2014, p. 881-882.

• 143 Id., at p. 898.

• 144 Id., at p. 897 (describing devices such as “poll taxes, literacy tests, and similar exclusionary tools” used to target newly enfranchised minority voters, and compensations made by law to exempt favored voters who would otherwise be affected by the tools). Professor Ellis also points out that “the meme of voter fraud represents the latest round of America’s evolution from an exclusion-based republic to an inclusive republic supporting full participation of all citizens.” Id., at p. 893.

• 145 Id., at p. 883.

• 146 It is obviously beyond the scope of this article to deal comprehensively with all of the particular constitutional and legal issues raised by these practices. See Rotunda & Nowak, 2009, p. 219-349.

• 147 Yick Wo v. Hopkins, 118 U.S. 356, 370 (1886).

• 148 376 U.S. 1 (1964).

• 149 Id., at p. 17.

• 150 383 U.S. 663 (1966).

• 151 Id., at p. 666. The Court overruled its 1937 decision in Breedlove v. Suttles, 302 U.S. 377 (1937), which upheld the constitutionality of provisions that conditioned voting on the payment of a poll tax. The Court also distinguished its earlier decision in Lassiter v. Northampton County Bd. of Elec., 360 U.S. 45 (1959), in which the Court upheld a North Carolina literacy test. The Court stated that “the Lassiter case does not govern the result here, because, unlike a poll tax, the ‘ability to read and write * * * has some relation to standards designed to promote intelligent use of the ballot.’” Id., at 51.

• 152 Harper v. Va. State Bd. of Elections, at p. 670.

• 153 395 U.S. 621 (1969).

• 154 395 U.S. 701 (1969).

• 155 398 U.S. 419 (1970).

• 156 Id., at p. 422.

• 157 405 U.S. 320 (1972). But see Sayler Land Co. v. Tulare Lake Basin Water Storage District, 410 U.S. 719 (1973) (upholding limitation on right to vote in water district elections to property owners and permitting votes to be apportioned according to assessed valuation of land within the district). In an amicus curiae brief in Crawford v. Marion County Election Bd., Nos. 07-21 and 07-25, Dean Chemerinsky attempted to draw a distinction between the foregoing cases and the Court’s later jurisprudence (which is summarized below), based on whether the deprivation of the right to vote was direct or indirect. If it were direct (as in the forgoing cases), strict scrutiny would apply. If not, the balancing test of Burdick v. Takushi, 504 U.S. 428 (1992) would control. See Crawford v. Marion County Election Bd., Nos. 07-21 and 07-25 (U.S.), Brief of Professor Erwin Chemerinsky as Amicus Curiae in Support of Neither Party (filed 13 November 2007). The Court did not credit that distinction, which might have provided one answer to the problem, while leaving a potentially large universe of possibly serious infringements outside the purview of strict scrutiny review. Indeed, the distinction seems to provide the basis for redressing simple-minded violations of voting rights while countenancing those that are more ingenious.

• 158 Infringements on the right to vote may be conceptualized in either equal protection or First Amendment terms, but the same analysis applies. See Rotunda & Nowak, 2009, p. 222.

• 159 Ellis, 2014, p. 913-914.

• 160 Id., at p. 894. Indeed, “[p]oliticians and policymakers throughout American political history manipulated the rules of entry to the franchise in order to control voter turnout.” Id., at p. 893-94. It is for that reason that the kind of deference to the political process that Justice Frankfurter advocated in Colegrove v. Green, 328 U.S. 549, 552 (1949), seems inadequate. See Rotunda & Nowak, 2009, p. 310-312 (describing evolution of law with respect to justiciability beginning with Colegrove).

• 161 See Trimble v. Gordon, 430 U.S. 762, 769 (1977) (Powell, J.).

• 162 See Rotunda & Nowak, 2009, p. 221.

• 163 Id., at p. 222. “Laws that totally prohibit a class of persons from voting in a general election or laws that are designed to restrict the voting power of a particular class of persons in a general election are unlikely to survive such a standard. Laws that regulate the electoral system to promote substantial state interests in the conduct of efficient and honest elections need to be examined on a case-by-case basis.” Id.

• 164 460 U.S. 780 (1983).

• 165 Id., at p. 786.

• 166 Id., at p. 787.

• 167 Id., at p. 788.

• 168 Id., at p. 789.

• 169 Id., at p. 789-790.

• 170 Id., at p. 792.

• 171 Id., at p. 793-794.

• 172 Id., at p. 795.

• 173 Id., at p. 806.

• 174 Id., at p. 808.

• 175 Id., at p. 817.

• 176 Id.

• 177 504 U.S. 428 (1992).

• 178 Id., at p. 432.

• 179 Id., at p. 434.

• 180 Id., at p. 434-435.

• 181 Id., at p. 437.

• 182 Id., at p. 938.

• 183 Id., at p. 441.

• 184 Id., at p. 446.

• 185 Id., at p. 442-443.

• 186 Id., at p. 447.

• 187 Id., at p. 448.

• 188 472 F.3d 949 (2007).

• 189 Id., at p. 950. The statute did not place the same restriction on persons who were eligible to cast an absentee ballot or voted in a nursing home. Id. In addition, voters could cast provisional ballots and return within 10 days with appropriate documentation. Id. To secure a state-issued ID card, it is necessary to have a certified birth certificate, and “it’s not particularly easy for a poor, elderly person who lives in South Bend, but was born in Arkansas, to get a certified copy of his birth certificate.” Id., at p. 955 (Evans, J. dissenting). The majority speculated that “[t]he benefits of voting to the individual voter are elusive (a vote in a political election rarely has any instrumental value …), and even very slight costs in time or bother or out-of-pocket expense deter people from voting, or at least from voting in elections they’re not much interested in. So some people who have not bothered to obtain a photo ID will not bother to so just to be allowed to vote, and a few who have a photo ID but forget to bring it to the polling place will say what the hell and not vote, rather than go home and get the ID and return to the polling place.” Id., at p. 951.

• 190 Id., at p. 950.

• 191 Id., at p. 952.

• 192 Id., at p. 952-953.

• 193 See Indiana Democratic Party v. Rokita, 458 F.Supp.2d 775, 782-84 (S.D. Ind. 2006).

• 194 Crawford, 472 F.3d at p. 953. Judge Posner has subsequently confessed that his resolution of the case was incorrect. See Richard A. Posner, Reflections on Judging, Cambridge, Harvard University Press 2013, p. 851.

• 195 Crawford, 472 F.3d at p. 955.

• 196 Id.

• 197 Id., at p. 956.

• 198 Crawford v. Marion County Election Board, 484 F.3d 436, 437-439 (7th Cir. 2007). Judge Wood pointed out that, contrary to Judge Posner’s understanding, “as a matter of law, the Supreme Court’s voting cases do not support a rule that depends in part for support on the idea that no one vote matters. Voting is a complex act that both helps to decide elections and involves individual citizens in the group act of self-governance.” Id., at p. 438.

• 199 See Crawford v. Marion County Election Board, 533 U.S. 181 (2008).

• 200 Id., at p. 189-190.

• 201 Id., at p. 191.

• 202 Id., at p. 190.

• 203 As Justice Stevens put it, “[t]he only kind of voter fraud [the law] addresses is in-person voter impersonation at polling places. The record contains no evidence of any such fraud actually occurring in Indiana at any time in its history.” Id., at p. 194. Nor was there any evidence to show that the Indiana law would provide an effective means for dealing with that phantom problem or improve the situation in any way. Justice Stevens nonetheless discounted the significance of both points: “[w]hile the most effective method of preventing election fraud may well be debatable, the propriety of doing so is not.” Id., at p. 296.

• 204 Id., at p. 202-203.

• 205 Id., at p. 204.

• 206 Id., at p. 210.

• 207 Id.

• 208 Id., at p. 211.

• 209 Id., at p. 209.

• 210 Id.

• 211 Id., at p. 237.

• 212 Id.

• 213 Id., at p. 205. In this sense, Justice Scalia builds on Justice Stevens’s holding that strict scrutiny review is not required in this type of case, where the restrictions placed on the right to vote are not “unrelated to voter qualifications”. Id., at p. 189. Similarly, the court of appeals had held that “the law should [not] be held by the same strict standard applicable to a poll tax because the burden on voters was offset by the benefit of reducing the risk of fraud.” Id., at p. 188.

• 214 Id., at p. 194. (“The only kind of voter fraud that [the Indiana law] addresses is in-person voter impersonation at polling places. The record contains no evidence of any such fraud occurring in Indiana at any time in its history.”)

• 215 See Yick Wo v. Hopkins, 118 U.S. at p. 370.

• 216 Crawford, 533 U.S. at p. 205.

• 217 See Thornburgh, 476 U.S. at p. 828. The Casey plurality adopted a less demanding version of the test. See Casey, 505 U.S. at p. 876.

• 218 Crawford, 505 U.S. at p. 206 (emphasis in original).

• 219 The trial judge “found that petitioners had ‘not introduced evidence of a single, Indiana resident who will be unable to vote as a result of [the law] or who will have his or her right to vote unduly burdened by its requirements’”. Id., at p. 187. The trial court refused to credit the testimony of an expert witness, who testified that the law could affect up to 989,000 registered voters who lacked a government-issued ID, but the trial court nonetheless estimated that only about 43,000 (or less than 1%) of Indiana residents would be affected. Id., at p. 187-188.

• 220 Id., at p. 205.

• 221 383 U.S. 663 (1966).

• 222 See Casey, 505 U.S. at p. 894.

• 223 Id., at p. 207 (emphasis omitted).

• 224 Id.

• 225 Id., at p. 208.

• 226 570 U.S. 2 (2013).

• 227 In Whitford v. Gill, 218 F.Supp.3d 837 (W.D. Wisc. 2016), a three-judge district court struck down a Republican redistricting plan in Wisconsin. In Benisek v. Lamone, 266 F.Supp.3d 799 (E.D. Md. 2017), a three-judge district court declined to strike down a Democratic redistricting plan in Maryland. See Gill v. Whitford, No. 16-1161, jurisdictional statement filed 24 March 2017; Benisek v. Lamore, No. 17-333, jurisdictional statement filed 1 September 2017.

• 228 In 1870, as previously noted, the people of the United States acknowledged the importance of the right to vote by adopting the Fifteenth Amendment to the Constitution, which guarantees that “[t]he right of citizens of the United States to vote shall not be denied or abridged by the United States or by any State on account of race, color, or previous condition of servitude.” U.S. Const., Amend. XV. Similar amendments have since been adopted to prohibit exclusions from the franchise based on gender, the failure to pay “any poll tax or other tax,” or on account of age if the putative voter is “eighteen years of age or older.” Id., Amend. XIX (1920) (gender), Amend. XXIV (1964) (poll or other tax), Amend. XXVI (1971) (age). In addition, Congress enacted the Voting Rights Act in 1965. See Voting Rights Act of 1965, 79 Stat. 437, codified, as amended, 52 U.S.C. §101001, et seq. But see Shelby County v. Holder, 570 U.S. 2 (2013) (holding that the coverage formula of Section 4(b) is unconstitutional because it is based on data over 40 years old, making it no longer responsive to current needs and therefore an impermissible burden on the constitutional principles of federalism and the “equal sovereignty of the states”). In addition, in 1913, the people amended the Constitution to provide that United States Senators would henceforth be elected by the people of the several states, rather than by the state legislatures. In addition, in 1913, the people amended the Constitution to provide that United States Senators would henceforth be elected by the people of the several states, rather than by the state legislatures. U.S. Const., Amend. XVII.

• 229 Yick Wo v. Hopkins, 118 U.S. 356, 370 (1886).

• 230 See, e.g., Louisiana v. United States, 380 U.S. 145 (1965) (invalidating Louisiana statute that authorized the registrar of voters to determine whether a voter’s ‘understanding’ of the federal or state constitution was sufficient to permit him or her to vote); Gomillion v. Lightfoot, 364 U.S. 339 (1960) (invalidating state statute that created a 28-sided city boundary by which nearly all African-American voters would be excluded without excluding any whites); Guinn v. United States, 238 U.S. 347 (1915) (invalidating the so-called ‘grandfather clause’, an Oklahoma constitutional amendment that provided that “no person who was, on January 1, 1866, or at any time prior thereto, entitled to vote under any form of government, or who at that time resided in some foreign nation, and no lineal descendant of such person, shall be denied the right to register and vote because of his inability to do so read and write sections of such constitution.”).

• 231 Ellis, 2014, p. 913-914.

• 232 See Anderson v. Celebreeze, 460 U.S. 780, 788 (1983).

• 233 See generally J.H. Ely, Democracy and Distrust: A Theory of Judicial Review, Cambridge, Harvard University Press, 1980.

• 234 348 U.S. 483 (1955).

• 235 Id., at p. 489.

• 236 Id., at p. 488. See also id., at p. 489. (“The problem of legislative classification is a perennial one, admitting of no doctrinaire definition. Evils in the same field may be of different dimensions and proportions, requiring different remedies. Or so the legislature may think.”)

• 237 Id., at p. 488. (“We emphasize again what Chief Justice Waite said in Munn v. State of Illinois …, ‘For protection against abuses by legislatures, the people must resort to the polls, not to the courts.’")

• 238 See, e.g., Obergefell v. Hodges, 135 S.Ct. 2584, 2627 (2015) (J. Scalia, dissenting) (“Until the courts put a stop to it, public debate over same-sex marriage displayed American democracy at its best. Individuals on both sides of the issue passionately, but respectfully, attempted to persuade their fellow citizens to accept their views. Americans considered the arguments and put the question to a vote. The electorates of 11 States, either directly or through their representatives, chose to expand the traditional definition of marriage. Many more decided not to. Win or lose, advocates for both sides continued pressing their cases, secure in the knowledge that an electoral loss can be negated by a later electoral win. That is exactly how our system of government is supposed to work.”); Planned Parenthood of Southeastern Pennsylvania, Inc. v. Casey, 505 U.S. 833, 1002 (1992) (“[B]y foreclosing all democratic outlet for the deep passions this [reproductive choice] issue arouses, by banishing the issue from the political forum that gives all participants, even the losers, the satisfaction of a fair hearing and an honest fight, by continuing the imposition of a rigid national rule instead of allowing for regional differences, the Court merely prolongs and intensifies the anguish.”).

• 239 See United States v. Carolene Products, 304 U.S. 144, 153 n. 4 (1938).

• 240 See John Locke, Of Civil Government: Second Treatise, Chicago, Regnery/Gateway, 1955, § 13, p. 11.

• 241 See Rotunda & Nowak, 2012 p. 222.

• 242 See Freeman, 2013, p. 279-323.

• 243 Id., at p. 280.

• 244 Id., at p. 281.

• 245 Id.

• 246 Id., at p. 321.

• 247 Id.

• 248 Id., at p. 321-322. Freeman writes: “[t]hough it is difficult to speculate about the Court’s reluctance to adopt a proportionality-based test, that hesitation may be symptomatic of the judiciary’s conception of its own perceived boundaries.” Id., at p. 322.

• 249 For example, Freeman argues that “[i]t is evident that the plurality believed state regulations on abortion must further a genuinely legitimate state interest. Though they did not incorporate this inquiry into their articulation of the undue burden test, the plurality opinion repeatedly assessed the state’s interest to ensure its validity. The state had no legitimate interest, e.g., in “giv[ing] to a man the kind of dominion over his wife that parents exercise over their children.’” Id., at p. 294, quoting Casey, 505 U.S. at p. 898 (plurality opinion).

• 250 Freeman suggests that the Casey plurality intended for the ‘undue burden’ test to constitute an intermediate level of scrutiny, “lying somewhere between the deferential rational basis and the punishing strict scrutiny”. Id., at p. 298. According to Freeman, the plurality did not intend “to retreat wholly from Roe’s protection of a woman’s independence and discretion”, but “sought to construct a less strict but still vigorous standard capable of defending the abortion right”. Id., Finally, “[i]t is precisely because undue burden is a form of intermediate review that Justice Blackmun expressed his preference for Roe’s strict scrutiny and Chief Justice Rehnquist for Webster’s rational basis.” Id. at p. 299-300.

• 251 Id., at p. 316.

• 252 Id., at p. 279-280. Freeman points out that traditional rational basis review and heightened rational basis review both contain a nexus requirement, but that courts implementing the former “merely invoke, but do not in fact apply, nexus analysis”, whereas courts implementing the latter “actually examine the relationship between state means and ends”. Id., at p. 285. Freeman also observes that “[b]ecause the Court has been reluctant to acknowledge overtly the existence of rational basis with bite, much less identify the factors that trigger such enhanced review, the standard’s boundaries remain blurry.” Id., at p. 287. In any event, heightened rationality review differs in three ways from traditional rational basis review: “the ‘bite’ renders the courts less deferential to the legislature, less tolerant of over- and under-inclusive classifications, and less open to state experimentation.” Id., at p. 285. Finally, Freeman notes that because of the Court’s extensive use of the language of rational basis review, “some scholars remain ‘unclear whether Casey’s undue burden standard subjects abortion regulations to intermediate scrutiny, or merely to rational basis review.’” Id., at p. 293. Indeed, as Freeman also observes, even Justice Scalia expressed confusion, noting that the plurality’s “description of the undue burden standard in terms more commonly associated with the rational-basis test will come as a surprise even to those who have followed closely our wanderings in this forsaken wilderness”. Id., quoting Casey, 505 U.S. at 986 n.4 (Scalia, J., concurring in part and dissenting in part).

• 253 457 U.S. 202 (1982). In neither case, of course, did the Court indicate that it was applying any test other than the standard rational basis test.

• 254 473 U.S. 432 (1985).

• 255 Freeman, 2013, p. 286.

• 256 Id., at p. 286-287.

• 257 Id., at p. 280.

• 258 Id., at p. 294, quoting Casey, 505 U.S. at p. 898 (plurality opinion).

• 259 Heller v. Doe, 509 U.S. 312, 320 (1993).

• 260 Freeman, 2013, p. 294-295.

• 261 Id., at p. 301.

• 262 Id., at p. 295.

• 263 Id., at p. 296.

• 264 See, e.g., Whole Woman’s Health v. Hellerstedt, 136 S. Ct. 2292, 2320 (2016).

• 265 Freeman, 2013, p. 302. Freeman finds support for her interpretation of the undue burden test in Justice Stevens’s partial concurrence in Casey, in which he remarked that, notwithstanding the opacity of the plurality’s opinion, “[t]he future may also demonstrate that a standard that analyzes both the severity of a regulatory burden and the legitimacy of its justification will provide a fully adequate framework for the review of abortion legislation even if the contours of the standard are not authoritatively articulated in any single opinion.” Id., at p. 322, quoting Casey, 505 at 920 n.6 (J. Stevens, concurring in part and dissenting in part). According to Freeman, Justice Stevens recognized that the undue burden standard “tests the weight of the burden, the legitimacy of the state’s regulatory purpose, and the sufficiency of the relationship between them. A regulation that fails any of the above components is an unconstitutionally undue burden on the right to abortion”. Id.

• 266 See, e.g., Whole Woman’s Health v. Hellerstedt, 136 S. Ct. 2292, 2309-2310 (2016) (holding that certain regulatory provisions constituted an undue burden because they did not afford “medical benefits sufficient to justify the burdens [that they imposed] upon access”).

• 267 Id., at p. 2311.

• 268 See, e.g., Lane v. Wilson, 307 U.S. 268, 275 (1939) (Frankfurter, J.) (“The [Fifteenth] Amendment nullifies sophisticated as well as simple-minded modes of discrimination.”)

• 269 As previously noted (see supra note 157), Dean Chemerinsky offered the Court another alternative, based on a distinction between direct and indirect violations of the right to vote, but the Court rejected that alternative as well.

• 270 See G. Metzger, ‘Unburdening the Undue Burden Standard: Orienting Casey in Constitutional Jurisprudence’, Columbia Law Review, Vol. 94, 1994, p. 2038-2040 (discussing burden analysis in dormant commerce clause and First Amendment jurisprudence).

• 271 See Craig v. Boren, 429 U.S. 190, 220 (1976) (Rehnquist, J., dissenting).

• 272 See Aristotle, Nicomachean Ethics, Martin Ostwald (Trans.), Indianapolis, Boobs-Merrill, 1962, p. 18-19.

• 273 376 U.S. 1, 17 (1964).

• 274 Yick Wo v. Hopkins, 118 U.S. 356, 370 (1886).