Copyright (c) 2002 Virginia Law Review Association
Virginia Law Review
88 Va. L. Rev. 951LENGTH:
ARTICLE: EQUAL PROTECTION INCORPORATIONNAME:
Michael C. Dorf*BIO:
* Professor of Law, Columbia University School of Law. For helpful comments, discussions, questions, and suggestions, I am grateful to the participants in workshops at Columbia Law School, the University of Miami School of Law, and the Rutgers-Camden School of Law, as well as Albert Alschuler, Vincent Blasi, Richard Briffault, Sherry Colb, John Hart Ely, Martin Flaherty, William Forbath, Barry Friedman, Pat Gudridge, Samuel Issacharoff, Doron Kalir, John Manning, Henry Monaghan, Gerald Neuman, Robert Post, Seana Shiffrin, Reva Siegel, Allan Stein, Peter Strauss, Susan Sturm, William Treanor, Jeremy Waldron, Robert Williams, and Kenji Yoshino. Tal Golomb provided excellent research assistance. SUMMARY:
... IN order to preserve a broad field of play for legislative and administrative action, courts subject most forms of official discrimination to minimal scrutiny under the Equal Protection Clause. ... Does this point apply to the voting rights amendments generally? As a matter of historical fact, the (re)interpretation I propose works more readily for race and sex than for age because while the Fifteenth and Nineteenth Amendments
are best understood as stepping stones on the path from second-class citizenship to full equality for African-Americans and women, the Twenty-Sixth Amendment was neither a part nor the culmination of a grand struggle to end age discrimination. ... Whatever we might conclude about age discrimination on the whole, we have a rough and ready response to the objection that equal protection incorporation pays insufficient attention to constitutional text: on the contrary, just like due process incorporation, it uses constitutional text as a source of value to begin, though not necessarily to end, a process of holistic construction. ... A remaining question is whether adopting equal protection incorporation means that no categories beyond those expressly singled out by the constitutional text are presumptively invalid. ... Unlike the voting rights amendments, the Equal Protection Clause is not expressly written in the form of an antidiscrimination principle. ... TEXT:
IN order to preserve a broad field of play for legislative and administrative action, courts subject most forms of official discrimination to minimal scrutiny under the Equal Protection Clause. 1
They reserve exacting scrutiny for laws and policies that employ a small number of so-called suspect and semi-suspect classifications, such as race and sex. Why courts apply heightened scrutiny to laws and policies employing these classifications, rather than others such as age, disability, and sexual orientation, is not entirely clear. Although the Supreme Court has articulated criteria for identifying suspect and semi-suspect classifications, none of these criteria, standing alone, is satisfactory, and the Court has not found any principled means of combining them.
To fill the justificatory gap, this Article will advance a judicial reading of the Equal Protection Clause that I call equal protection incorporation. The basic idea is borrowed from the Court's due process jurisprudence. Just as the enumerated provisions of the Bill of Rights served as a useful guide to the Supreme Court in determining the scope of liberty protected by the Due Process Clause, so, the core of the argument goes, the forms of discrimination specifically barred by the Constitution's text - such as the prohibitions on race and sex discrimination in voting set forth in the Fifteenth and Nineteenth Amendments
- should guide interpretation of the Equal Protection Clause. 2
Provisions like the Fifteenth and Nineteenth Amendments
provide a textual basis for distinguishing [*953] between presumptively valid and presumptively invalid forms of discrimination.
Equal protection incorporation can be readily understood in three ways: as carrying out the judge's traditional task of making sense of a document as a whole, 3
as a type of structural inference, 4
or, to use a term coined by Professor Amar, as a form of "intratextualism." 5
I recognize that these are somewhat distinct interpretive methodologies. For example, Amar's intratextualism relies on a fine-grained parsing of constitutional text, juxtaposing the way in which the Constitution uses the same phrase in various contexts, 6
while some versions of Professor Black's method of structural inference use the repetition of various concepts to infer broad values, which then inform concrete doctrinal decisionmaking. 7
Professor Ely's inference that representative government is the central constitutional principle is an example of the latter approach. 8
Notwithstanding such differences, this Article employs these methods more or less interchangeably, drawing fine-grained inferences where they seem appropriate and broader ones in other circumstances. I do not regard the shift in level of focus as arbitrary, however. Sometimes it is reasonable to treat fine gradations of language as connoting important distinctions, while other times there will be reasons to focus on broader themes. I regard these methods as compatible as long as there is a good argument for whatever level of generality is chosen. In any event, the holistic methods employed in this Article all share the premise that the meaning of the constitutional text is not exhausted by whatever concepts an isolated phrase connotes to the reader; further guidance can often be gleaned from the balance of the constitutional text. 9
[*954] Part I of this Article will argue that existing accounts of equal protection leave the decision whether to treat a classification as suspect - and most other decisions as well - to almost completely unguided normative judgment. To constrain that judgment, Part II will define equal protection incorporation by analogy to due process incorporation. Part III will then defend the general idea of using other provisions of the Constitution to give content to the Equal Protection Clause. The method is decidedly nonoriginalist. For example, I do not claim that most of the framers and ratifiers of the Nineteenth Amendment
believed that the Amendment - either directly or by its implications for interpretation of the Fourteenth Amendment - required exacting judicial scrutiny of sex classifications outside the context of voting. 10
However, any acceptable account of the Equal Protection Clause will be nonoriginalist, and the incorporationist account I defend has the great virtue of connecting doctrine to constitutional text.
The principal doctrinal payoff of the Article comes in Part IV, which will ask what forms of discrimination are singled out as most invidious under an incorporationist account of equal protection. It will argue that the First, Fourteenth, and Nineteenth Amendments
clearly mark discrimination based on religion, race, color, previous condition of servitude, and sex as presumptively unconstitutional, while drawing the same inference much more tentatively regarding age discrimination and the Twenty-Sixth Amendment.
[*955] Part IV will conclude by sketching implications of equal protection incorporation for forms of discrimination not expressly enumerated in the text, offering two variants of the basic approach. The first, which will appeal to textualists in the tradition of Justice Hugo Black, would reserve heightened scrutiny for distinctions drawn on the basis of those classifications expressly enumerated in the constitutional text. The second, which I favor, will appeal to those who follow in the tradition of the second Justice Harlan's dissent in Poe v. Ullman. 11
It uses the constitutional text as a constraint on completely open-ended adjudication, but permits some interpolation, extrapolation, and reasoning by analogy from specific text. Given current debates, the principal practical difference between the two approaches would concern the proper treatment of discrimination on the basis of sexual orientation and disability.
Parts V and VI will ask whether equal protection incorporation has implications beyond identifying presumptively invalid classifications. These Parts will identify two primary implications. Part V will argue that under the incorporationist model, the Equal Protection Clause is best read as an antidiscrimination principle rather than an antisubordination principle. The Fifteenth, Nineteenth, and Twenty-Sixth Amendments are written as antidiscrimination norms. Although one could argue that this distinguishes these provisions from the Equal Protection Clause, the general argument for incorporation in Part III will aim to show that the various equality provisions of the Constitution should be read as shedding light on one another. Similarly, Part V will argue that the antidiscrimination wording of the Fifteenth, Nineteenth, and Twenty-Sixth Amendments also informs the meaning of the Equal Protection Clause. These provisions demonstrate a fairly strong commitment to understanding equality in antidiscrimination terms, although no severe doctrinal consequences necessarily follow from that commitment. For example, it is possible to use the antidiscrimination principle to justify both Washington v. Davis, 12
which held that a [*956] disparate racial impact does not, by itself, trigger strict scrutiny, 13
and City of Richmond v. J. A. Croson Co., 14
which held that racial classifications which benefit minorities must be subject to the same exacting scrutiny as those that burden minorities. 15
Alternatively, it is possible to think that either one or both of these cases was wrongly decided because the Supreme Court misunderstood the antidiscrimination principle. A properly understood antidiscrimination principle, on this view, need not render irrelevant the question whether challenged state action disadvantages a subordinated group. Hence, Part V will conclude that equal protection incorporation does not point decisively towards any one answer to the question of how the Equal Protection Clause bears on disparate impact and affirmative action.
Finally, Part VI will argue that equal protection incorporation is consistent with robust congressional power to define rights under the Equal Protection Clause more broadly than the Court's own doctrines define them. The driving force behind the Rehnquist Court's narrow construction of Section Five of the Fourteenth Amendment is its worry that without tethering the congressional power to the Court's own jurisprudence, the Fourteenth Amendment could become a de facto plenary power. Part VI will argue that equal protection incorporation can address this concern without adopting the narrow view of the Section Five power found in the Court's recent decisions.
I. Interpreting the Equal Protection Clause
The Equal Protection Clause of the Fourteenth Amendment grew out of an American tradition honoring the general principle of equality that goes at least as far back as President Andrew Jackson, 16
if not to the Declaration of Independence, 17
or even the [*957] Mayflower Compact. 18
In an earlier age it might have been possible to argue that the Clause requires only that the law, whatever its content, be "equally binding upon every member of the community," 19
but the time when the meaning of "equal protection" could be separated from the meaning of "equality" has long passed, if it ever existed. Today, any convincing account of the Equal Protection Clause must also be, in substantial part, an account of the general concept of equality.
At the conceptual level, however, equality is either entirely empty 20
or so hotly contested that it can be invoked with (equal?) aplomb by those on either side of our most divisive national questions: Does affirmative action remedy or create inequality? Is a right to abortion necessary for equality of the sexes or anathema to the equal right to life of the unborn? Is the proscription of some, but not all, categories of discrimination in civil rights statutes the paradigmatic instance of legally required equality or the conferral of special rights? Answers to such questions do not come from the abstract concept of equality but, at best, from more particularized conceptions of equality and other norms.
To what sources should a contemporary reader of the Constitution turn in choosing an appropriate conception of equality? After constitutional text, courts and scholars typically list history as the next most authoritative source of guidance, but here too the conventional [*958] approach provides scant help. 21
There is general agreement that the central, original purpose of the Equal Protection Clause, indeed of the entire Fourteenth Amendment, was to protect African-Americans against the Black Codes (whether directly or through congressional legislation). 22
Today, however, virtually no one thinks the meaning of the Equal Protection Clause can be restricted to its original purpose, narrowly defined. The Clause is majestically inclusive in its language, not confined to burdens on African-Americans, inequalities based on race, or even unequal treatment among citizens.
Moreover, there is broad consensus that, whatever its merits in other contexts, a jurisprudence based on a narrowly defined original understanding of the Equal Protection Clause is morally unacceptable because it would license such odious institutions as racially segregated schools, antimiscegenation laws, and the most egregious forms of discrimination against women. This is not just a problem for liberals. Whatever they may think in their heart of hearts, conservatives who are generally sympathetic to originalism cannot openly say that Brown v. Board of Education 23
was wrongly decided. Accordingly, they concoct implausible accounts of the Reconstruction Era understanding of segregation. 24
Moreover, [*959] those conservatives who wish to preserve not only the decisions invalidating segregation, but also those invalidating most forms of affirmative action, must blind themselves still further by ignoring Reconstruction Era institutions such as the Freedman's Bureau, which practiced something like affirmative action. 25
Strikingly, not a single Supreme Court opinion invalidating affirmative action confronts this evidence that such programs are consistent with the original understanding of the Fourteenth Amendment. Thus, neither liberals nor conservatives are willing to be guided by the nineteenth-century understanding of the Equal Protection Clause in its narrowest form. Furthermore, as soon as one moves to a somewhat higher level of generality - but not one that is so general as to provide no real guidance, like "equal protection means treating people who are similarly situated in the same way" - the disagreement over the appropriate conception of equality reemerges.
Perhaps, then, it is a mistake for judges to seek a coherent overarching conception of equality. Courts might instead simply decide equal protection cases on an ad hoc basis. After all, there may well be no account of equal protection that is acceptable to all of the Justices of the Supreme Court and explains all, or even most, of the Court's jurisprudence. In this view, incompletely theorized agreement, as a modus vivendi, is all that one can hope for. 26
Such minimalism is, at best, a descriptive account of the output of a multi-member institution like a court (or a legislature) that must reach compromises among persons with different convictions. It is hardly a prescription for how an individual judge, legislator, or citizen should think about the meaning of equality, or anything [*960] else. Furthermore, with few exceptions, even minimalists are not nihilists. 27
Minimalists, along with virtually everyone else, accept that purposeful discrimination against members of a traditionally subordinate group on the basis of race denies equal protection. This raises the question which, if any, of the italicized terms are essential to such a judgment, but this question cannot be answered except by reference to a substantive conception of equality.
A somewhat different sort of minimalist might say that the judicially enforceable interpretation of the Equal Protection Clause should not extend beyond the core of consensus. The competing conceptions of equality would battle it out in the political domain. On such a view, racially neutral laws that have a disparate impact would be valid absent an illicit subjective purpose, but so too would most forms of affirmative action because there is no political consensus about how the general equality norm bears on either of these issues.
Whatever the attraction of such a conventionalist jurisprudence in other contexts, 28
it creates substantial difficulty where equality is concerned. The contemporary consensus that purposeful discrimination against members of a traditionally subordinate group on the basis of race denies equal protection is of relatively recent vintage. Chief Justice Earl Warren's appealing rhetoric in Brown notwithstanding, 29
different treatment on the basis of race is not inherently unequal; it is, after all, possible to conceive of a hypothetical society in which segregation connotes no subordinate status for one group or the other. 30
Only by rejecting separate-but-equal in favor of some conception of equality (but exactly what we are not quite [*961] sure) could the Brown Court overrule Plessy v. Ferguson. 31
Brown then becomes a deep puzzle because the now widely accepted principle that purposeful discrimination against members of a traditionally subordinate group on the basis of race denies equal protection was not a matter of consensus when Brown was decided. The conventionalist-minimalist therefore must say that Brown was wrongly decided, even if it has become right by virtue of subsequent acceptance. One can think, as I do, that public acceptance plays a substantial role in determining the path of constitutional and other judge-made law, but making the correctness of the defining decision of a half-century turn on retrospective acceptance is at least somewhat problematic.
In any event, conventionalism is itself a controversial conception of the Equal Protection Clause (or perhaps of constitutional interpretation more generally) for which a justification may be demanded. The argument that conventionalism respects democracy better than some other interpretive approach is just that - an argument. It takes no great imagination to envision conceptions of democracy in which even (indeed especially) widely accepted discriminatory decisions provide the occasion for judicial interference with majoritarian decisionmaking.
Judges, legislators, and others trying to give effect to the Equal Protection Clause would therefore appear to be left with nowhere to turn except their own subjective sense that some conception of equality is the best understanding of the broader concept. Even if the judge (or other interpreter) takes "best" to refer not only to his own subjective sense of justice, but also to the entire fabric of American constitutional law, the diversity of views regarding what equality entails leaves the normative choice of a conception of equality substantially unguided. 32
[*962] Given their insulation from politics, the seemingly intractable ambiguity of the Equal Protection Clause is a special problem for courts. Supreme Court doctrine responds in two ways. First, the Court defines equal protection in formal rather than substantive terms, adopting a narrow antidiscrimination principle 33
instead of a broader antisubordination principle. 34
Second, and more immediately relevant to my purposes here, the Court understands the Equal Protection Clause to be primarily targeted at discrimination against individuals based on a small number of forbidden grounds. To avoid undue judicial interference with legislative and administrative action, courts require only minimal rationality of most state action challenged under the Equal Protection Clause. The principal exceptions to low-level scrutiny are inequalities bearing upon the exercise of fundamental rights, a subject I address only briefly and in the margin, 35
and laws or policies that employ (either expressly [*963] or surreptitiously) a so-called suspect or semi-suspect classification. 36
To be sure, the Justices occasionally invoke equal protection [*964] to invalidate government action that does not target persons on the basis of one of the forbidden grounds. 37
Such rare cases are probably best understood as the exceptions that prove the rule, as evidenced by the intellectual gymnastics in which commentators have engaged in order to render them consistent with the main line of doctrine. These cases are described as exemplars of "covert use of heightened scrutiny," 38
"rational basis with teeth," 39
or a freestanding principle barring government from acting on "animus" to individuals or members of a group, 40
regardless of whether the animus attaches to traits that are independently deemed suspect.
Accordingly, when a litigant asks a court to invalidate some government action on equal protection grounds, the crucial question is whether the government has utilized what I shall call a presumptively invalid classification. How do the courts determine which classifications are presumptively invalid? The embarrassing fact of the matter is that they simply make it up. Granted, the Supreme Court has identified criteria for suspectness. Standing alone, however, [*965] none of these criteria is satisfactory, nor has the Court articulated any principled means of combining them.
The leading formulation comes from the United States v. Carolene Products Co. 41
footnote, where the Court opined that "discrete and insular minorities" constitute what came to be known as suspect classes. 42
In the classic process theory account of Carolene Products, heightened judicial scrutiny is justified by the fact that prejudice prevents a discrete and insular minority from coalition building in the manner described in The Federalist No. 10. 43
Political science, however, teaches that some discrete and insular groups should be better able to achieve their political aims than groups, like women, that are diffused throughout the general population. 44
Moreover, women, as is commonly observed, are neither insular nor a minority. 45
Courts also sometimes ask whether discrimination is based on an "immutable characteristic," 46
but immutability is hardly a necessary condition for suspectness. For example, religion is mutable, but constitutional doctrine, whether under the Equal Protection Clause or the First Amendment, nonetheless appropriately subjects religious discrimination to strict scrutiny. Indeed, even if medical technology made it possible to change one's skin color or sex through a safe, inexpensive, and painless procedure, that would [*966] hardly justify race-or sex-based discrimination against those people who opted not to undergo the procedure. 47
Finally, in deciding which classifications are presumptively invalid, courts sometimes inquire into whether there is a history of discrimination on the basis of a trait (such as race). Yet this criterion is also problematic, or at best indeterminate, because it raises again the question of the appropriate level of generality. For example, under current doctrine, the history of discrimination against African-Americans leads the Court to conclude that race is a suspect classification, and that conclusion in turn is used to justify strict scrutiny of laws that disadvantage white men, who have historically fared well.
Alternatively, if one says that the history of discrimination must be specific to a particular group, other nettlesome questions of definition arise. For example, should recent black immigrants from Africa, the West Indies, and elsewhere count as members of the traditionally subordinated group of African-Americans whose ancestors were enslaved in the United States? Should it matter whether the ancestors of such immigrants themselves were enslaved somewhere else? Does the history of discrimination against the Chinese and Japanese in the American West and elsewhere render suspect contemporary discrimination against Southeast Asians? If not, what about ethnically Chinese Southeast Asians? As the population becomes ever more heterogeneous, such examples multiply. Thus, it is not especially surprising that even those Justices who have been sympathetic to race-conscious measures that aim to assist members of traditionally subordinated groups have advocated intermediate scrutiny of such measures rather than mere rationality review. 48
At the end of the day, we are left with a messy hodgepodge. If a litigant argues that discrimination against some group - such as gays, the elderly, or the disabled - should trigger heightened scrutiny, [*967] courts will ask whether the group constitutes a discrete and insular minority, whether it lacks political power proportionate to its numbers, whether the trait that defines membership in the group is immutable, or whether the group has suffered a history of discrimination. Or, more in keeping with the symmetry of current doctrine, if a litigant frames the argument in terms of discrimination on the basis of some trait - such as sexual orientation, age, or disability - courts will ask similar questions about the general trait.
In addition, courts are likely to search for analogies. Race is suspect; sex is viewed as pretty much, but not exactly like race; therefore, sex is quasi-suspect. Age, however, is too far from race or sex, and everyone gets old anyway, so the Justices (who are not young but are themselves rarely the victims of age discrimination) conclude that age is not suspect. None of the criteria has anything remotely like an on/off character, and thus the whole process is highly subjective.
Yet the identification of suspect and semi-suspect classifications need not proceed entirely subjectively. This Article argues that the constitutional text itself can play a substantial role in identifying presumptively invalid grounds for distributing governmental benefits and burdens.
Three provisions - the prohibition of religious tests in Article VI and the Free Exercise and Establishment Clauses of the First Amendment - mark discrimination on the basis of religion as suspect. In addition, the Fifteenth Amendment prohibits denial of the right to vote on the basis of "race, color, or previous condition of servitude." The Nineteenth Amendment
does the same with respect to "sex," and the Twenty-Sixth with respect to "age." 49
Taken together, these provisions evince powerful textual support for the proposition that discrimination on the basis of religion, race, color, [*968] previous condition of servitude, sex, and age are presumptively invalid. 50
II. From Due Process Incorporation to Equal Protection Incorporation
Incorporation of the Bill of Rights is one of the few great success stories of modern constitutional law. Judges and constitutional scholars almost universally agree that, whatever else it does or does not do, the Fourteenth Amendment makes most of the provisions of the Bill of Rights applicable to the states and their subdivisions. 51
It is difficult to imagine anyone saying today what Justice Oliver Wendell Holmes, Jr. said in a 1918 letter to Judge Learned Hand: that "free speech stands no differently than freedom from vaccination." 52
In response to such a claim, contemporary readers of the Constitution would likely make arguments about the virtues of free speech, no doubt echoing the views expressed by Justice Holmes himself in his dissent in Abrams v. United States 53
less than five months after his letter to Hand. 54
Our contemporaries would not, however, rely on normative arguments alone. They would also point to the text of the First Amendment, saying, in effect, that the [*969] constitutional text places free speech on a different footing from freedom from vaccination.
As a matter of principle, there is much that is wrong with the argument that the enumeration of a right in one of the first eight amendments requires its application against the states. In light of the oxymoronic character of the doctrine of substantive due process, 55
the Fourteenth Amendment's Due Process Clause is a very awkward vehicle for incorporating substantive rights like freedom of speech or religion. Among other problems, treating the Fourteenth Amendment's Due Process Clause as shorthand for most of the provisions of the Bill of Rights renders those provisions redundant with respect to the federal government, which is constrained by the Fifth Amendment's identically worded Due Process Clause.
This textual objection could perhaps be met by shifting the task of incorporation to the Fourteenth Amendment's Privileges or Immunities Clause, 56
but other objections remain. For one thing, although there is strong evidence that the proponents of the Fourteenth Amendment expected it to incorporate the Bill of Rights, 57
[*970] controversy remains over how to interpret that evidence. 58
Moreover, if enumeration in the Bill of Rights is the key to applying a right against the states, what justifies the Court in failing to incorporate some provisions - such as the Fifth Amendment right to indictment by a grand jury and the Seventh Amendment right to a civil jury trial? 59
On the one hand, selective incorporation of the sort actually practiced by the Supreme Court undermined incorporation's ability to constrain judicial discretion, which, after all, was what attracted incorporation's chief proponent, Justice Hugo Black, to the theory in the first place. On the other hand, had the Court adopted Black's "the whole Bill of Rights, and nothing but the Bill of Rights" version of incorporation, it would have run headlong into the Ninth Amendment. If it means anything, the Ninth Amendment means that the fact that a right is mentioned in express terms in the Bill of Rights does not by itself constitute a sufficient reason to grant that right greater recognition than one not mentioned in the Bill of Rights.
These are strong arguments in principle, but they are ultimately beside the point. In a constitutional order that is still haunted by Lochner v. New York 60
(or for some by Roe v. Wade 61
), judges feel obliged to look to constitutional text as a source of values that are substantially, if not entirely, external to their own consciences. Constitutional text matters, not only because it constrains judicial [*971] discretion, but also because, as a matter of contingent historical fact in a nation with a terse, difficult-to-amend Constitution, those values that find their way into the constitutional text are likely to be the ones most fundamentally embraced by the American people. At the same time, it is unrealistic and unwise to expect courts to follow the textual implications of the Bill of Rights slavishly, at least where, as in the case of incorporation, the literal text to be interpreted is the very general Fourteenth Amendment, rather than the Bill of Rights itself. Thus we arrive at the current understanding of the Fourteenth Amendment's Due Process Clause (or, if you prefer, its Privileges or Immunities Clause or the Amendment as a whole): Enumeration in the Bill of Rights creates a rebuttable presumption that a right applies against the states, and lack of enumeration creates a rebuttable presumption against recognizing a proposed right against state or federal action. This "selective incorporation plus" approach, for all of its flaws, has been widely accepted by courts and commentators. 62
In contrast with the Due Process Clause of the Fourteenth Amendment, interpretation of the Equal Protection Clause has proceeded in large part independently of the rest of the constitutional text. That is surprising. The text and history of the Equal Protection Clause are no more definite than that of the Due Process (or Privileges and Immunities) Clause. Just as the Supreme Court in the 1960s turned to the Bill of Rights to inform its understanding of the Fourteenth Amendment's Due Process Clause, so can it turn to other provisions of the Constitution - mostly those governing discrimination with respect to the franchise - to inform its understanding of the Equal Protection Clause.
The next Part addresses what I consider to be the most substantial objections to equal protection incorporation. Before coming to these, however, I need to consider a threshold objection that, upon inspection, proves to be less substantial than it at first appears. The objection goes like this: It is at least plausible to think that the [*972] framers and ratifiers of the Fourteenth Amendment meant the Privileges or Immunities Clause, the Due Process Clause, or the Fourteenth Amendment as a whole as a kind of shorthand for the first eight provisions of the Bill of Rights, but it is inconceivable, because anachronistic, to suppose that those same framers and ratifiers meant the Equal Protection Clause as a shorthand for amendments that would not be proposed for decades.
This objection is sound only if one is a strict originalist. If, however, one is prepared to say - as I would venture most judges and constitutional scholars actually think - that due process incorporation would be sensible regardless of whether the framers and ratifiers of the Fourteenth Amendment meant the Due Process Clause as a shorthand for the first eight provisions of the Bill of Rights, then one must give a different account of due process incorporation. The most plausible account, in my view, is that interpreting the Fourteenth Amendment's Due Process Clause by reference to the first eight provisions of the Bill of Rights provides judges with some textual guidance in an enterprise that would otherwise be almost entirely unbounded. That same decidedly nonoriginalist account can be given for the Equal Protection Clause.
As will be apparent in the succeeding Parts of this Article, while the case for equal protection incorporation is not originalist, history does play a role in the arguments marshaled in defense of equal protection incorporation. One may find those arguments persuasive or not. Unless, however, one is prepared to rest support for due process incorporation on the original understanding of the Fourteenth Amendment, it should not count as a fatal objection to equal protection incorporation that it uses later-enacted text to inform earlier-enacted text. 63
With the justificatory work to be taken up in the next Part, here, in a nutshell, is the incorporationist account of the Equal Protection Clause: Close judicial scrutiny of all government classifications is inconsistent with the presumption of constitutionality that characterizes post-Lochner Era judicial review. In deciding where that [*973] presumption does not apply, as in deciding what liberties are fundamental, courts should be guided by the balance of the constitutional text. That text singles out a handful of illicit criteria, which correspond fairly closely to extant constitutional doctrine. The principal exception is age discrimination, and there, we shall see, the constitutional text itself manifests some ambivalence. Thus, although justified in the case law in a largely ad hoc manner, the broad outlines of equal protection doctrine - the handful of suspect and semi-suspect classifications that trigger heightened judicial scrutiny set against a background principle of rational basis scrutiny - can be better explained by a method that draws its inspiration from a careful reading of the constitutional text as a whole.
III. In Defense of Equal Protection Incorporation
The core of the affirmative argument for equal protection incorporation is almost trivially straightforward: text matters. Precisely why the constitutional text matters, however, is a surprisingly difficult question. No doubt part of the answer involves the fact that the Constitution's text was ratified by the People. 64
But this cannot be the whole story, given the dead-hand problem: Why should ratification by long-dead generations, under voting rules that are grossly unfair by contemporary standards, bind the People today? Answering this question is a central task of any full-blown justification for constitutionalism.
That task is largely beyond the scope of this Article. For my purposes, it is sufficient to acknowledge the relevance of constitutional text as simply a brute fact of our legal order: In positivist terms, when the People (of the here and now) accept the Constitution as legitimate, a substantial piece of what they accept is its text. To be sure, the text is not the only thing that the People accept; they also accept some version of authoritative construction by the [*974] Supreme Court, 65
but only to a point. It is doubtful that the view that "the Constitution is what the judges say it is" 66
has ever been cheerily endorsed by a substantial portion of the population. Accordingly, judges cannot regularly depart too far from constitutional text. To put the point affirmatively, the legitimacy of judicial decisionmaking is enhanced to the extent that it is rooted in, even if not dictated by, authoritative text.
This brings us immediately to the chief objection to equal protection incorporation's use of the voting rights amendments. Incorporation, the objection charges, plays fast and loose with the text. In particular, it overlooks the obvious fact that the voting rights amendments apply only to voting. Accordingly, the categories singled out by the Fifteenth, Nineteenth, and Twenty-Sixth Amendments have no special salience for evaluating government conduct outside the area of voting.
This objection might draw further force from history. When the Fourteenth Amendment was adopted, and even well into the twentieth century, the civil rights it protected were generally understood to be distinct from political rights, like voting. 67
The civil/political distinction was reflected in Section Two of the Fourteenth Amendment itself, which permitted disenfranchisement of adult male African-Americans - the principal intended beneficiaries of Section One of the Fourteenth Amendment - so long as the disenfranchising states paid the price of reduced representation in Congress. The fact that the Fifteenth and Nineteenth Amendments
[*975] were needed at all, the objection continues, shows that voting was considered distinct from the general command of equal protection, so that the wording of these Amendments (and the Twenty-Sixth) cannot be used to gloss the meaning of the Equal Protection Clause.
The objection might be a good one if we were interested in unearthing the subjective intent of the drafters and ratifiers of the voting rights amendments. For example, many or even most of the framers and ratifiers of the Nineteenth Amendment
may have believed that the Amendment had no impact on sex classifications outside the context of voting. As a matter of historical fact, this conjecture is open to doubt. 68
Nonetheless, even if we concede that the Fifteenth and Nineteenth Amendments
were not originally understood to extend beyond the franchise - whether directly or by their implications for the meaning of the Fourteenth Amendment - that hardly forecloses reading them in such a manner today. If historical accuracy connotes fidelity to the narrowly defined subjective intent of the framers and ratifiers of particular provisions, then any morally acceptable account of the Equal Protection Clause will be deeply ahistorical given nineteenth-(and early twentieth-) century views about race and sex.
Equal protection incorporation therefore strives for an account of fidelity through history, 69
an account that updates prior understandings by integrating new constitutional text, changed circumstances, and major doctrinal shifts. In contrast, by emphasizing distinctions among classes of citizens and rights, the objection under consideration assumes that it is still appropriate to imagine that the American people can be divided into "First-Class Citizens" and "members of the larger society." 70
This assumption is not only problematic on moral grounds; it has not stood the test of history. The civil/political distinction was already under pressure within a decade after the adoption of the Fifteenth Amendment. In Strauder v. West Virginia, 71
the Court used the Equal Protection [*976] Clause to invalidate a murder conviction on the ground that African-Americans were ineligible for jury service under state law; yet, traditionally, the right to serve on a jury was seen as a political right on par with the right to vote. 72
Had the Court hewed closely to the civil/political distinction, it would have rejected Strauder's claim. Yet under the influence of the Fifteenth Amendment, the Strauder Court expressly understood equal protection as applicable to political rights like jury service. 73
To be sure, in some quarters the distinction between civil and political rights persisted well into the twentieth century. Certainly by the twenty-first century, however, the notion that there can be distinct classes of citizens has become anathema to public attitudes and constitutional law. The modern "assumption that citizenship means the right to vote was most clearly underscored in the way the Court in Reynolds v. Sims nonchalantly equated citizenship with suffrage in the phrase "[a] citizen, a qualified voter.'" 74
Whatever one might think about particular Supreme Court decisions construing the Equal Protection Clause in the context of voting - such as Bush v. Gore 75
or Shaw v. Reno 76
- the old notion that the Fourteenth Amendment has nothing to do with the franchise is certainly dead. Thus, any effort to derive a currently operative meaning of equal protection from the text of the Constitution as a [*977] whole cannot rest on the assumption that the Fourteenth Amendment and the voting rights amendments address different subjects.
Or can it? A determined textualist might note that while the Equal Protection Clause protects persons, the voting rights amendments protect citizens, suggesting again that the voting rights amendments have a narrower scope than the Equal Protection Clause. Moreover, this objection continues, the reference to citizenship reminds us that the distinction between civil and political rights has not entirely vanished. Although there can be no classes of citizens, states can constitutionally deny aliens political rights such as the right to vote, run for political office, or serve on juries. 77
Thus, the objection concludes, the voting rights amendments really are about nothing other than political rights.
This version of the objection is stronger but hardly insuperable. Let us grant that a denial of the right to vote to aliens would not violate any provision of the Constitution, and would be, by virtue of the wording of the voting rights amendments, essentially immunized from constitutional challenge. 78
Suppose, however, that a state were to extend the franchise to male but not female aliens. Is it clear that the Nineteenth Amendment's
reference to citizenship means that such a gender-based disenfranchisement of aliens would receive the blessing of the Equal Protection Clause? And if the Equal Protection Clause would condemn this action - as I believe the courts would rightly hold - is the Nineteenth Amendment
irrelevant to that judgment?
More broadly, the textualist objection we are now considering ultimately misconceives equal protection incorporation as an effort to discern the meaning of the voting rights amendments rather than the way in which they may be used to guide understanding of the [*978] Equal Protection Clause. Consider a comparison with due process incorporation. The First Amendment's express reference to Congress renders implausible an interpretation of the First Amendment that makes it applicable to the states of its own force. Yet that is no obstacle to using the values identified in the First Amendment as a means of guiding the Court in its effort to identify rights that are rooted sufficiently deeply in American traditions to rank as fundamental, and therefore applicable to the states via the Fourteenth Amendment's Due Process Clause. Likewise, the fact that the voting rights amendments of their own force only apply to citizens and voting should not count as an obstacle to their use as a means of guiding the Court in its effort to identify the forms of discrimination or subordination that are so incompatible with notions of equal status as to trigger heightened scrutiny under the Equal Protection Clause.
Finally, in response to the objection based on the textual reference to citizenship in the voting rights amendments, it is worth noting that the distinction underwriting the objection actually supports the use to which I would put these amendments. The objection asserts that there are some grounds upon which political, but not civil, rights can be restricted. It follows a fortiori that if some classification - such as race or sex - cannot be a basis for denying political rights, neither can it be a basis for denying civil rights, which can be denied on fewer, not greater numbers of grounds.
These arguments respond to the objection that the very terms of the voting rights amendments preclude their use in elaborating a general equality provision, but there remains a different kind of objection. This other objection is rooted in a process theory of the sort propounded by Professor Ely (although Professor Ely himself disagreed with the objection for reasons of the sort I provide below). 79
Pursuant to such a theory, one might think that heightened judicial scrutiny is least needed precisely where the Constitution - through the voting rights amendments - preserves a group's ability to make use of ordinary politics.
This objection, however, does not take into account the fact that under current doctrine, a blanket denial of the franchise on almost [*979] any basis - even one that is not a trigger of heightened scrutiny outside the voting area - will result in heightened scrutiny under the voting-as-fundamental-right prong of equal protection doctrine. Suppose that a state were to deny the franchise to all persons who lacked college degrees or to persons who did not own their own homes. Even though education and property qualifications for voting have in the past been defended, is there any doubt that these restrictions would be subject to, and would fail, heightened scrutiny?
Or imagine a world that is exactly like our own except that the Constitution does not contain the Fifteenth and Nineteenth Amendments.
In such a world, the voting-as-fundamental-right prong of equal protection doctrine would surely invalidate denials of the franchise based on race and sex. 80
Accordingly, today the voting rights amendments may have little or no force in their own domain. Thus, their primary influence is, or should be, their effect on interpretation of the Equal Protection Clause.
There is nothing inconsistent with process theory in this conclusion. The fact that a characteristic was historically a basis for denial of the franchise - thereby necessitating a constitutional suffrage amendment in an earlier era - shows that people are likely to be vulnerable to discrimination on that basis even after the suffrage amendment. Among other things, this explains why the Carolene Products footnote includes separate categories for non-self-correcting defects in the electoral system and laws that burden discrete and insular minorities. Securing the franchise to a group or prohibiting voting discrimination on the basis of some characteristic is not by itself sufficient to guarantee equal treatment of that group or non-discrimination on the basis of the characteristic.
The foregoing responses not only address the objections thus far considered; they also point the way to a broader affirmative argument. That argument begins with the proposition that the Constitution's commitments to equality in voting are tied up with and implement its commitment to equality in general. In my reconstructed [*980] narrative, the historical struggles that gave rise to the Fifteenth and Nineteenth Amendments
are seen as struggles for equality, not just for technical changes in the rules of voting. Accordingly, the Fourteenth Amendment's Equal Protection Clause should be read in the light of, rather than in juxtaposition to, the voting rights amendments.
Notwithstanding the distinction between civil and political rights, this reinterpretation is almost self-evident with respect to race. Proposed by the Reconstruction Congress closely on the heels of the Fourteenth Amendment, it is almost impossible not to understand the Fifteenth Amendment as furthering a general vision of racial equality.
The linkage may be more obscure in the case of sex, but only because much of the historical context in which the Nineteenth Amendment
was enacted has been largely forgotten. Yet, when we do even a little bit of digging, we discover that the Nineteenth Amendment
was also understood as part of the project of fleshing out the Fourteenth. As Professor Siegel explains:
The Nineteenth Amendment
grew out of struggles over the Fourteenth Amendment and was a long-resisted, fully deliberated, collective commitment to include women as equal members of the constitutional community ... . The Nineteenth Amendment
may "only" concern voting; but that is hardly responsive to the historicist's objection. For nineteenth-century Americans, voting was the central question of women's citizenship - "the woman question." Nineteenth-century Americans knew what woman suffrage signified, even if its full significance to them is no longer legible to us today. 81
Professor Siegel's most powerful evidence demonstrates that many of the opponents of the Nineteenth Amendment
believed that it would intrude into the family and disentrench patriarchy. 82
Exactly what the Nineteenth Amendment
meant to the people who adopted it, however, remains a tricky question. Opponents of a measure often exaggerate its impact in an effort to defeat it, only to turn around and urge its narrow construction after enactment. For this reason, looking for the meaning of a legal text in the understandings [*981] of its opponents is every bit as uncertain an enterprise as constructing the collective understanding of its proponents. Suppose that the Equal Rights Amendment ("ERA") had been ratified. Would it necessarily have required a gender-neutral military draft and integrated public restrooms simply because opponents raised such possibilities to defeat the ERA? 83
My point is not that the original understanding of the Nineteenth Amendment
or any other constitutional provision is thoroughly unknowable. History can and should be studied as part of the process of constitutional interpretation because history teaches object lessons, and those object lessons take on special importance when a historical struggle yields textual change. 84
Moreover, when we look to history understood in this way, we find that it supports, or at least does not undermine, the enterprise of extrapolating a general gender-equality norm from the Nineteenth Amendment.
Does this point apply to the voting rights amendments generally? 85
As a matter of historical fact, the (re)interpretation I propose works more readily for race and sex than for age because while the Fifteenth and Nineteenth Amendments
are best understood as stepping stones on the path from second-class citizenship [*982] to full equality for African-Americans and women, the Twenty-Sixth Amendment was neither a part nor the culmination of a grand struggle to end age discrimination. It was enacted at a time when the Supreme Court was already in the business of using the Equal Protection Clause to invalidate those forms of discrimination in the distribution of the franchise that reflected more general offenses to equal citizenship.
It is thus quite possible that the framers and ratifiers of the Twenty-Sixth Amendment saw the provision as remedying the peculiar injustice of denying a political voice to eighteen-to twenty-year-olds who were being drafted to fight and die for their country. These framers and ratifiers may not have thought that they were acting on a deeper principle of equality. 86
Nonetheless, since the Twenty-Sixth Amendment was enacted just four years after Congress passed the Age Discrimination in Employment Act of 1967, 87
undoubtedly some of the former's supporters saw its concerns as connected to a broader prohibition on age discrimination.
Moreover, even if the framers and ratifiers of the Twenty-Sixth Amendment did not believe that their actions reflected a general principle prohibiting age discrimination, that fact should hardly be dispositive on the question of its interpretation now or in the future. Eighty-four years elapsed between the adoption of the [*983] Fifteenth Amendment and Brown v. Board of Education. 88
Fifty-three years elapsed between the adoption of the Nineteenth Amendment
and Frontiero v. Richardson. 89
It has only been thirty-one years since the adoption of the Twenty-Sixth Amendment. If the historical pattern holds, it may take more time for the impulse underlying that Amendment to gain full recognition in equal protection doctrine - even if equal protection incorporation already provides a textual basis for the shift.
This last point merits emphasis. Constitutional text, after all, is the key to equal protection incorporation (and much constitutional interpretation generally). As a matter of text, the Twenty-Sixth Amendment is very similar to the Fifteenth and Nineteenth Amendments.
By its literal terms, the Twenty-Sixth Amendment prohibits denying the franchise to the old as well as to the young, thereby stating a typically symmetrical antidiscrimination principle. 90
It is thus a plausible basis from which to infer that age discrimination should be considered as presumptively denying equal protection, 91
although, as we shall see, other constitutional provisions make the conclusion that age is a suspect classification somewhat problematic.
Whatever we might conclude about age discrimination on the whole, we have a rough and ready response to the objection that [*984] equal protection incorporation pays insufficient attention to constitutional text: on the contrary, just like due process incorporation, it uses constitutional text as a source of value to begin, though not necessarily to end, a process of holistic construction.
Now we must consider the precisely opposite objection - that incorporation accords too much weight to constitutional text. According to this objection, the particular amendments to the Constitution that have been enacted are almost entirely the product of historical accident. For example, there is a provision expressly authorizing the federal income tax, the Sixteenth Amendment, because (prior to its adoption) the Supreme Court ruled such a tax unconstitutional, 92
but there is no provision specifically authorizing a federal bureaucracy on the scale to which we are accustomed because the Court acquiesced in the creation of the modern administrative state. 93
Or, to choose an example closer to our present subject, the Twenty-Sixth Amendment became part of the Constitution only because the Supreme Court had previously invalidated a federal statutory provision lowering the voting age to eighteen, finding that as applied to non-federal elections the provision was outside of Congress's power. 94
Faced with the daunting possibility of having to run two sets of elections, one for federal offices and one for non-federal offices, the states were eager to ratify the Twenty-Sixth Amendment. 95
Thus, had it not been for a decision of a divided Supreme Court, there would have been no need for the Twenty-Sixth Amendment.
Conversely, the absence of a particular constitutional text may not be very informative. Consider the fact that the Constitution contains no general express prohibition of sex discrimination. There is an obvious explanation for this absence: The social [*985] changes that did not quite produce the Equal Rights Amendment produced a de facto ERA in the Court's equal protection jurisprudence. 96
Indeed, it is possible that the Court's jurisprudence itself played a causal role in the states' failure to ratify the ERA because, at the margin, state legislators who otherwise might have been in favor of ratification could have thought that the Amendment was unnecessary given the Court's willingness to accomplish the same ends via the Equal Protection Clause. 97
More generally, as Professor Strauss, who has recently made a forceful version of this argument, puts it, constitutional amendments are irrelevant. 98
This objection is fair enough if understood as a descriptive claim about how our constitutional order evolves. The common-law method that the Supreme Court has applied to constitutional interpretation permits a large amount of change without formal amendments to the basic text. Supreme Court justices are human beings who inhabit much the same social world as other Americans. When society at large comes to see a federal bureaucracy as necessary (even if only as a necessary evil) or comes to view most forms of discrimination against women as inconsistent with their equal status, it is hardly surprising that judicial doctrine adapts. There is no denying the descriptive claim that much, or even most constitutional change occurs without textual change.
Nonetheless, it hardly follows that, in adapting the Constitution to changed circumstances and values, the Court should treat the extant text as irrelevant. 99
To make that further normative claim is to confuse what Professor H.L.A. Hart called the external and internal perspectives on law. 100
From an external point of view - that of say, a political scientist 101
- one can speculate that the main currents of constitutional [*986] law would have cut roughly the same riverbed regardless of the particular amendments that happened to be enacted. Given the nature of the judicial appointments process, in the long run, at least on matters of great national moment, the federal courts' views about what is constitutional will not often remain more than a standard deviation away from the center of political opinion. 102
Yet these undeniable facts play almost no proper role in constitutional adjudication as practiced from the inside. Imagine a newly appointed Supreme Court justice explaining his decision to cast the decisive vote in a controversial case on the express ground that the President who nominated him and the Senate that confirmed him hoped and expected he would vote that way. Even if not quite an impeachable misdeed, such a decision would reflect a profound misunderstanding of the nature of legal justification. Thoroughgoing legal realism is not a prescription for judging. 103
Someone who thinks, per radical legal realism, that constitutional interpretation is always or nearly always simply a smokescreen for the enactment of the judges' political preferences will have nothing constructive to say about how judges should interpret the Constitution. 104
If, however, legal realism is true only insofar as it debunks the notion that judging can be purely formal and objective - in other words, if the social and political attitudes of judges interact with a discrete constraining force called "law" to produce constitutional decisions - then it is useful to ask what law is and how it constrains preferences. To end our detour into these deep jurisprudential waters, any answer one gives should include a substantial role for enacted text. 105
[*987] Can we construct an affirmative argument for equal protection incorporation based on the foundation of this response to Professor Strauss's claim about the irrelevance of constitutional amendments? We might begin with the observation that constitutional text, including the text of constitutional amendments, matters because it serves as a constraint on judicial discretion, but this is hardly a sufficient justification. We must explain why this particular constraint is an appropriate one.
The answer is readily available: Given the difficulty of the constitutional amendment process, the People will typically only resort to it in addressing questions of value on which there is an unusual degree of consensus. 106
(I put to one side housekeeping measures, such as the Twenty-Fifth Amendment, that do not speak in especially value-laden terms.) Constitutional text matters, in substantial part, because the text reflects a process that inscribes the nation's deepest commitments. Accordingly, the relatively clear textual provisions can serve as a source of values for the interpreter who must make sense of the relatively ambiguous ones.
Thus, we have a rough and ready justification of equal protection incorporation. A judge facing the deep ambiguity of the Equal Protection Clause asks: Given the ubiquity of government classifications and the diversity of circumstances in which people live, how do I know which differences deny equal protection, and how do I avoid answering this question in a completely subjective manner? Equal protection incorporation replies: Begin by looking at the concrete inscriptions found elsewhere in the document - prohibiting religious favoritism and denials of the franchise based on race, color, previous condition of servitude, sex, and age - for they are constitutional values rather than merely your own subjective values, and, given the nature of the amendment process, such constitutional values are likely (though not certain) to reflect the nation's deepest commitments.
IV. Presumptively Invalid Classifications
The main doctrinal utility of equal protection incorporation is that it provides a principled textual basis for deciding which forms of discrimination are presumptively invalid. Moreover, it does so in a way that largely validates the moral intuitions underlying current doctrine, even as it suggests some reforms.
Just what and how many reforms equal protection incorporation suggests depends on which of two variants one adopts. A version that will appeal to formalists and others who place a very high value on textual determinacy would limit heightened judicial scrutiny to the precise categories singled out by the constitutional text. A second version, which might go by the infelicitous name of "equal protection incorporation plus," places a high value on textual determinacy but also recognizes both limits to the constraining power of text and other, sometimes competing, values. Equal protection incorporation plus uses the voting rights amendments and like provisions as a very important starting point, but it also allows a role for heightened scrutiny of unenumerated classifications based on a process of interpolation, extrapolation, and analogy. The choice between these two versions of equal protection incorporation closely parallels the choice between the jot-for-jot version of due process incorporation favored by Justice Hugo Black and the version ultimately adopted by a majority of the Court. Although this Article defends equal protection incorporation plus, I hope that much of what I have to say will also be of interest to those whose temperaments incline them to favor the more formalist version.
A. The Easy Cases
What forms of discrimination does equal protection incorporation render presumptively invalid? Let us begin with the easy cases.
Incorporation of the Fifteenth Amendment and the Religion Clauses of the First Amendment yields no doctrinal innovation. Race or color discrimination is the paradigmatic instance of a presumptively invalid action under the Equal Protection Clause, and while cases involving discrimination based on previous condition of servitude rarely arise today, if they did it is difficult to imagine the Court upholding such discrimination.
[*989] Similarly, strict scrutiny of classifications based on religion is already a feature of First Amendment law itself, although the question of whether the Free Exercise and Establishment Clauses of the First Amendment require anything more than religious equality is contested. Under Employment Division v. Smith, 107
the Free Exercise Clause permits the most severe restrictions on religious practice so long as they result from the application of formally religion-neutral government policies. 108
Many commentators have criticized this ruling, 109
and even the Smith rule's most articulate academic defenders have suggested that the Court has sometimes applied its notion of neutrality in too stingy a fashion. 110
Meanwhile, Congress sought to displace Smith in enacting the Religious Freedom Restoration Act, which was subsequently struck down by the Court. 111
With respect to the Establishment Clause, some Justices believe that government can run afoul of the First Amendment by providing certain kinds of direct financial support to religious institutions, even if the aid is dispensed on an equal basis to religious and secular organizations. Other Justices believe that neutrality in the allocation of funds is the touchstone of the Establishment Clause. 112
Thus, there remains deep division over whether the Religion Clauses of the First Amendment require more than equality, and over precisely what constitutes religious equality. There is no disagreement, however, with the proposition that the Religion [*990] Clauses require at least some relatively robust equality principle. If the Religion Clauses are understood as stating no more than an equality principle, then the effect of equal protection incorporation is to shift the work of making them applicable to the states from the Fourteenth Amendment's Due Process Clause to its Equal Protection Clause. In any event, equal protection incorporation would not change the substance of religion doctrine.
Incorporation of the Nineteenth Amendment
is slightly more complicated but it still presents a fairly easy case. Gender classifications are currently subject to intermediate scrutiny. 113
Incorporation need not entail applying precisely the same level of scrutiny to each of the classifications singled out by the constitutional text. Its principal virtue is in distinguishing the myriad classifications that trigger only rational basis scrutiny from the few that require more exacting scrutiny because the Constitution marks them as especially problematic. Accordingly, incorporation would not necessarily require a change in existing equal protection doctrine regarding sex discrimination.
B. Age Discrimination
Under current doctrine, distinctions based on age do not trigger heightened scrutiny of any sort. 114
As noted above in Part III, the equal protection incorporationist case for treating age discrimination as presumptively invalid is weaker on historical grounds than the case for race and sex. This Section argues that the textual basis for treating age discrimination as presumptively invalid is likewise weaker than the textual basis for treating race and sex discrimination that way. Nonetheless, I tentatively conclude that age discrimination should be treated as presumptively invalid.
First, consider the most obvious textual weakness in the claim that the Twenty-Sixth Amendment supports a general constitutional principle condemning age discrimination: The Amendment itself, even while condemning most age discrimination in voting, authorizes denying the franchise to citizens under the age of eighteen. This fact distinguishes the Twenty-Sixth from the Fifteenth and Nineteenth Amendments.
[*991] Yet it is possible to read too much into the peculiarity that the Twenty-Sixth Amendment itself discriminates on the basis of age. One can think that age discrimination is generally quite invidious, but still recognize that there is some minimum age below which a person does not have the capacity to rationally exercise his or her rights. Perhaps an argument could be made that this age is as low as fourteen or as high as twenty-one, but not that it is two or fifty. For this reason, any age discrimination norm will need to contain some necessarily arbitrary minimum point. 115
Therefore, standing alone, the fact that the Twenty-Sixth Amendment contains an arbitrary age minimum does not disqualify the Amendment as a source of a general constitutional norm against age discrimination.
This brings us to another set of textual difficulties, however, because the Twenty-Sixth Amendment does not stand alone. Three constitutional provisions set different minimum ages for holding federal offices: twenty-five for the House of Representatives, 116
thirty for the Senate, 117
and thirty-five for the Presidency. 118
These provisions are not minima for triggering some anti-age-discrimination norm. Rather, they are arbitrary age distinctions of just the sort that we might think a general anti-age discrimination norm would condemn. They thus tend to validate rather than undermine the practice of selecting a somewhat arbitrary minimum age as a qualification for some government position, benefit, or burden.
Is it significant that each of the four arbitrary ages given by the Constitution is a minimum rather than a maximum? Perhaps this fact can be invoked in favor of reading the Twenty-Sixth Amendment as a generalizable source of value condemning age discrimination against the old, but not the young. Yet that approach would be at least a little bit ironic, given that the primary motivation for the Twenty-Sixth Amendment was to protect the young, while the old benefit only because the language encompasses them as well. Accordingly, if one wanted to use the Twenty-Sixth Amendment as a basis for treating age discrimination as presumptively invalid, one would almost certainly have to read it as a [*992] symmetrical antidiscrimination principle. Clearly, the Constitution's four age minima 119
stand as an obstacle to that reading. Thus, upon reflection, the Twenty-Sixth Amendment supplies at best weak textual support for a general constitutional norm condemning age discrimination.
Perhaps, however, in light of other considerations, weak support will suffice. Looking beyond the text of the Constitution, federal and state statutes arguably provide a basis for applying heightened scrutiny to age discrimination. A recent article by Professors Eskridge and Ferejohn argues that some federal statutes are so deeply embedded in the fabric of the law that they have a quasi-constitutional status. 120
A "super-statute is one of the baselines against which other sources of law - sometimes including the Constitution itself - are read." 121
"Occasionally," they continue, "super-statutes can reshape constitutional understandings." 122
The basic argument of Eskridge and Ferejohn parallels my argument for equal protection incorporation. Just as I argue that provisions like the Fifteenth and Nineteenth Amendments
properly have an impact beyond their application to voting, so Eskridge and Ferejohn contend that super-statutes exert influence beyond their literal terms. This contention suggests difficult questions about how to identify a super-statute, but the core examples provided by Eskridge and Ferejohn appear incontrovertible. Among those core super-statutes is the body of federal antidiscrimination law, 123
which now includes, as one of its key provisions, the Age Discrimination in Employment Act ("ADEA"). 124
[*993] Yet, like the Twenty-Sixth Amendment, the ADEA provides a somewhat tenuous basis from which to infer a general, symmetrical constitutional norm against official age discrimination. The Act categorically prohibits certain forms of discrimination "because of [an] individual's age." 125
However, the Act's purpose section manifests a concern with "older workers," 126
and, like the Twenty-Sixth Amendment, the Act uses an arbitrary age as a criterion for triggering its protections. The ADEA only applies to those over the age of forty. 127
Making matters more complicated, it is unclear whether the Act applies symmetrically to those covered by it. Under the literal language of the Act, a forty-two-year-old has a claim against an employer who promotes a less qualified fifty-two-year-old based on a preference for older workers. Furthermore, a regulation promulgated by the Equal Employment Opportunity Commission so interprets the Act. 128
However, with one notable exception, the lower courts have concluded that, in light of the ADEA's purpose and minimum age threshold, it does not bar "reverse age discrimination." 129
No doubt this is a deep puzzle in the theory and practice of statutory interpretation, but for my purposes it suffices to show that, like the Twenty-Sixth Amendment, the ADEA provides an awkward ground from which to infer a general constitutional norm against age discrimination.
Beyond the ambivalent instruction we may take from constitutional and statutory text, there are normative reasons for questioning the Court's failure to apply heightened scrutiny to age discrimination. Here is the Court's fundamental argument against applying heightened scrutiny: "Old age ... does not define a discrete [*994] and insular minority because all persons, if they live out their normal life spans, will experience it." 130
The argument ignores reality, not to mention proverbial wisdom. Bromides like "youth is wasted on the young" reflect the fact that people do not act as though they recognize that they will eventually experience old age. Denial is a powerful force, whether it be youthful denial of the inevitability of aging or judges' denial about society at large. Insistence that age discrimination is not widespread is closely related to insistence that discrimination in general does not occur because it is irrational. 131
The assumption of rationality, however, is just that, an assumption, and in any event, the law frequently condemns discrimination regardless of whether it is rational or irrational. Moreover, the sorts of stigma that attach to age are similar, though of course not identical, to the stigmas that attach to race and gender hierarchies.
A complete normative argument for treating most age discrimination as presumptively invalid would include, among other things, an account of the fact that it occurs simultaneously in both directions. Even while cultural stereotypes devalue the old, many young workers complain that the government unfairly redistributes resources to the old. 132
Is this discrimination against the young, made possible by the lobbying power of the AARP and the disenfranchisement of those under eighteen? Does the disparity reflect a value judgment that the needs of the old are greater than those of the young? Is that a legitimate value judgment? Such questions have a disturbingly open-ended quality to them, reminiscent of the sort of ad hoc inquiry the Court currently undertakes in deciding which forms of discrimination are presumptively invalid. But because of the constitutional text's ambivalent commitments regarding age discrimination, we cannot entirely avoid these difficult questions.
[*995] For what it is worth, my own tentative judgment is that, notwithstanding the arbitrary minimum ages in the Constitution and the ADEA, as well as the historical circumstances under which the Twenty-Sixth Amendment was adopted, age discrimination should be presumptively invalid - and the Twenty-Sixth Amendment is at least a small part of what leads me to this conclusion. Perhaps the best way to give weight to the Constitution's ambivalence regarding age discrimination is through a doctrinal rule that would validate reasonable arbitrary minimum (and possibly maximum) age limits in circumstances where there is a substantial need for drawing a clear line. 133
Others may reasonably conclude, however, that the textual, historical, and normative differences between, on the one hand, the Fifteenth and Nineteenth Amendments,
and on the other hand, the Twenty-Sixth Amendment, imply that age discrimination should not be treated as presumptively invalid. My aim in this Part as a whole is only to show how the practice of looking to other provisions of the Constitution can be useful in guiding interpretation of the Equal Protection Clause. In this Section on age discrimination, I have also aimed to show that such guidance is not always very constraining.
C. Other Grounds of Discrimination
Are there any other grounds of discrimination that the constitutional text identifies as potentially suspect? Numerous provisions of the original Constitution establish state of origin (of persons and [*996] goods) as an illegitimate basis for state discrimination. 134
Like the Religion Clauses of the First Amendment, however, these provisions already operate of their own force, so that incorporating them via the Equal Protection Clause probably would not have any appreciable practical effect.
More controversially, one might derive a general economic equality principle from the Twenty-Fourth Amendment, which prohibits denials of the right to vote in federal elections based upon "failure to pay any poll tax or other tax." 135
Such a principle would not necessarily entail a general government obligation to redress inequalities in wealth. 136
Indeed, the anti-redistributionist mood of the last two decades, as well as the Court's perception of its own limited institutional capacity, mean that such a suggestion would be considered "off-the-wall." 137
Even in this parsimonious age, however, constitutional doctrine still prohibits the government from charging the indigent unaffordable user fees for access to the courts. 138
A plausible argument could be constructed that, under the influence of the Twenty-Fourth Amendment, the category of impermissible user fees ought to be larger than current doctrine [*997] recognizes. Such an argument would take aim at decisions like United States v. Kras, 139
upholding a filing fee for bankruptcy, and San Antonio Independent School District v. Rodriguez, 140
upholding local funding - analogous to a user fee - of public schools. Admittedly, however, the applicability of the Twenty-Fourth Amendment to general equal protection doctrine is even less clear-cut than the applicability of the Fifteenth, Nineteenth, and Twenty-Sixth Amendments outside the realm of voting. By its terms, the Twenty-Fourth Amendment is limited to federal elections, 141
and it was enacted in circumstances that belie a more general commitment to protecting the poor. 142
In sum, equal protection incorporation would certainly treat as presumptively suspect discrimination based on religion, state origin, race, color, previous condition of servitude, and sex. 143
The version of equal protection incorporation that I favor would also treat age discrimination as suspect, and might apply to some laws that disadvantage the poor as well. A remaining question is whether adopting equal protection incorporation means that no categories beyond those expressly singled out by the constitutional text are presumptively invalid. In answering that question, the due process incorporation experience is instructive.
D. Jot-for-Jot Incorporation or Selective Incorporation Plus?
Justice Black favored jot-for-jot incorporation of the Bill of Rights because of its apparent ability to limit judicial discretion. 144
That is certainly a virtue. Indeed, (relative) determinacy is the principal attraction of due process and equal protection incorporation. Accordingly, scholars and judges who place a very high value on determinacy often favor abandoning substantive due process beyond its incorporation of the Bill of Rights. 145
Presumably, such scholars and judges would prefer a version of equal protection incorporation that is limited to the classifications singled out by the constitutional text itself.
In my view, however, the better approach - in both due process and equal protection - is to treat enumeration or non-enumeration as an important factor, but not the sole factor, in fashioning doctrine. To begin with, even if one accepts enumeration as talismanic, there is the difficulty of deciding what falls within the scope of a given enumeration. On the due process side, for example, despite Justice Black's vaunted free speech absolutism, in his later years he was willing to use the speech/conduct distinction to deny constitutional protection to profanity. 146
More recently, despite Justice Antonin Scalia's disavowal of freestanding substantive due process, 147
he has been willing to find in the First Amendment's Free Speech and Freedom of Assembly provisions a penumbral right to [*999] freedom of expressive association. 148
Thus, jot-for-jot due process incorporation itself leaves substantial room for judicial discretion.
The same problem confronts the equal protection incorporationist who would limit heightened scrutiny to the enumerated categories of discrimination. We have already seen how the Constitution can sometimes speak out of both sides of its mouth regarding an equality norm: even as the Twenty-Sixth Amendment condemns age discrimination in voting, it - along with other constitutional provisions - draws arbitrary age lines.
There are other questions about the scope of an enumeration as well. Consider discrimination on the basis of sexual orientation. There is a substantial body of literature arguing that sexual orientation discrimination is sex discrimination, both formally - because, by analogy to Loving v. Virginia, 149
it turns on the sex of the romantic object of the person targeted by the discrimination - and substantively - because gays, lesbians, bisexuals, and transgendered people pose a threat to traditional sex-role stereotypes. 150
I find these arguments persuasive, and would accordingly treat sexual orientation discrimination as a species of sex discrimination, but I am not now especially interested in how one resolves the question. My point is that jot-for-jot-ism does not resolve the issue. There remains the matter of what each jot covers.
Moreover, even where a form of discrimination (such as disability discrimination) does not readily fit into one of the enumerated categories, there may be reasons to treat it as presumptively unconstitutional. Here we might reconsider Professor Strauss's argument about the accidental nature of the constitutional text. As I noted above in Part III, Strauss's observation that constitutional change occurs without formal amendments to the text does not [*1000] provide a reason for a court to ignore duly enacted text. 151
The difficulty of the amendment process ensures that values that do not enjoy widespread acceptance will not find their way into the Constitution.
The converse, however, does not hold. That a value is not expressed in so many words in constitutional text does not necessarily mean that the value is not widely and deeply held. Because the amendment process is so difficult, and overvalues the views of residents of small states (first through their equal representation in the Senate, and again through their equal role in ratification), it prevents some fundamental commitments from finding their way into the Constitution's text. In addition, as noted above, 152
the Supreme Court can sap the strength of a constitutional reform movement by adjusting doctrine to accomplish what otherwise might have been achieved through a formal amendment. For that reason and others, some precedents appropriately take on the status of constitutional text itself.
To be sure, concerns about judicial discretion mean that courts should be more willing to give presumptive weight to those values mentioned in the text. Moreover, at least as a formal matter, all constitutional interpretation must ultimately rest on text. 153
Even precedents that have quasi-textual status are themselves formally based on real texts. Thus, the "accidental" exclusion of some values will inevitably mean that the case for recognizing an unenumerated ground of discrimination as presumptively invalid is [*1001] more difficult than the case for recognizing an enumerated ground - in exactly the same way that the courts are more skeptical of claims of unenumerated rights under the Due Process Clause than claims under the express provisions of the Bill of Rights.
But skepticism need not be nihilism. With respect to due process, we ended up with selective incorporation plus, rather than jot-for-jot incorporation because a majority of Justices understood that enumeration of a right in the Bill of Rights provides a strong but not irrefutable basis for concluding that the right should be deemed fundamental, and therefore applicable to the states. In effect, they concluded that enumeration creates a rebuttable presumption in favor of a right's application against the states. Conversely, the Court was unwilling to treat non-enumeration as conclusively ruling out recognition for an asserted right under the Due Process Clause. Even absent the presumption created by enumeration, the Justices concluded, the Bill of Rights and common sense can provide interpretive guidance. Justice Harlan's oft-quoted observation that the enumerated rights lie on a "rational continuum" 154
that also includes unenumerated rights has become the classic statement of the relation between the incorporation doctrine and freestanding substantive due process. That observation should apply with equal force to equal protection incorporation.
What would this mean in practice? One could rationally extrapolate from the explicit categories of religion, race, color, previous condition of servitude, sex, and perhaps age, to other categories. Most obviously, and consistent with current doctrine, it is a very small - perhaps even nonexistent - step from race or color to ancestry or ethnicity. To put it more concretely, imagine a state policy [*1002] that facially discriminates against "Japanese," "Irish," and "Mexicans," and which applies not only to citizens of Japan, Ireland, and Mexico, but also to native-born U.S. citizens whose ancestors immigrated from these countries. 155
Today, almost no one thinks that there is a "Japanese race," an "Irish race," or a "Mexican race." But is there any doubt that the Fifteenth Amendment's textual marker for race discrimination - either in itself or as the basis for a small extrapolation - should apply to such categories?
The Supreme Court confronted a similar question in Saint Francis College v. Al-Khazraji. 156
In that case, the respondent brought suit under 42 U.S.C. 1981, alleging that his employer had discriminated [*1003] against him because he was an Arab. Under the Court's prior precedent, the suit was permissible only if this amounted to an allegation of race discrimination. 157
The petitioner argued that modern usage classifies Arabs as Caucasians, but the Court deemed this fact irrelevant. 158
When the statute was adopted in 1870 (the same year that the Fifteenth Amendment was ratified), the notion of race was much broader. In the mid-to late-nineteenth century, separate races were sometimes assigned to several varieties of Arabs, as well as to Basques, Germans, Greeks, Gypsies, Hebrews or Jews (also sometimes considered a subset of the Semitic race), Hungarians, Italians, Mongolians, Russians, Scandinavians (including separate racial classifications for Finns, Norwegians, and Swedes), Spaniards, and others. 159
Accordingly, the Saint Francis College Court concluded "that Congress intended to protect from discrimination identifiable classes of persons who are subjected to intentional discrimination solely because of their ancestry or ethnic characteristics. Such discrimination is racial discrimination ... whether or not it would be classified as racial in terms of modern scientific theory." 160
The Court's opinion in Saint Francis College is originalist. Because race was a broader concept in 1870 than it is today, the Court reasoned, a statute enacted in 1870 refers to the broader idea. 161
Yet equal protection incorporation, like any tenable account of equal protection, departs from narrow originalism. Accordingly, one might think it inappropriate to use the exact understanding of race that prevailed when the Fifteenth Amendment was adopted as the basis for applying the Fifteenth Amendment to the Fourteenth in a way that the adopters of neither amendment would have approved. If so, then Saint Francis College would not provide a satisfactory foundation on which to build an incorporationist explanation for why discrimination on the basis of ancestry or ethnicity should be presumed invalid.
[*1004] Nonetheless, there is a deeper truth in Saint Francis College that does appropriately inform equal protection incorporation. The point is not that we should emulate late-nineteenth-and early-twentieth-century racists in giving meaning to our prohibition on race discrimination. Instead, the question for one who would extrapolate from or analogize to the enumerated grounds of presumptively invalid discrimination is: What other grounds of discrimination are, in relevant respects, sufficiently similar to the enumerated grounds to warrant (more or less) the same presumptive invalidation?
As the number and variety of categories proliferates, this question may be thought to invite too much of the sort of subjectivity that incorporation is supposed to cabin. Even if that objection is valid with respect to some categories, though, there remain other unenumerated categories that are so closely linked to the enumerated ones that they pose very little risk in this direction. Ancestry or ethnicity is such a category because our history reveals that discrimination on such a basis not only paralleled race discrimination; as Saint Francis College reveals, typically it was indistinguishable from race discrimination.
The relevant history here begins before the adoption of the Fourteenth Amendment and extends well into the twentieth century. By the middle of the nineteenth century, "scientific" racism had supplemented (and partly supplanted) religious argument as the primary justification for the enslavement of African-Americans and the extermination of Native Americans, 162
and by the twentieth century it had become a complete ideology. 163
Southern white prejudice against African-Americans, prejudice against the Chinese and Japanese in the western United States, and general resentment of European immigrants (especially those from southern and eastern Europe) may have had distinct histories and causes, but they eventually became parts of a single phenomenon [*1005] of racism/nativism. 164
Even Supreme Court opinions well into the twentieth century refer to members of the "Chinese race," 165
the "Japanese race," 166
and the "Jewish race." 167
In light of this history, it is an almost purely theoretical question whether discrimination on the basis of ancestry or ethnicity ought to be considered as race discrimination per se, or as a very slight extrapolation from the enumerated prohibition on race discrimination. Either way, text, history, and common sense point towards treating ancestry or ethnicity discrimination as presumptively invalid.
Are there other presumptively invalid forms of discrimination that, while not enumerated as such in the constitutional text, can be extrapolated from or analogized to the enumerated forms? As noted above, sexual orientation discrimination is so closely entwined with sex discrimination that even if the former does not literally count as a version of the latter, only a very small conceptual extrapolation is required to treat sexual orientation discrimination as presumptively invalid.
In an earlier era, illegitimacy was also considered a candidate for recognition as a suspect classification, even if the leading Supreme Court decisions were somewhat inconsistent with one another. 168
A case for consistent application of heightened scrutiny to illegitimacy classifications might be built on the observation that it is similar to other "accidents of birth" such as race, ancestry, sex, and perhaps sexual orientation (to the extent it is an inherited trait). Accordingly, one might argue that most accidents of birth are a presumptively impermissible basis for the distribution of burdens and benefits. Such an argument, of course, would have to confront [*1006] the fact that numerous accidents of birth, such as physical strength, intelligence, and the disposition to work hard, are typically deemed entirely proper bases for government decisions. The kernel of a response to this objection could begin with the observation that illegitimacy, unlike these other attributes, is defined solely by reference to parentage, in the same way that ancestry is. Additional textual support could be found in the Bill of Attainder Clauses of Article I, Sections 9 and 10, 169
although this is indirect textual support at best, substantially weaker than the parallel textual support found in the voting rights amendments.
Another candidate for presumptively invalid official discrimination is disability discrimination, even though there is no plausible reading of the Constitution's text that would locate disability discrimination within a specific enumerated category. The only textual reference to disability appears in the Twenty-Fifth Amendment, which authorizes the transfer of power from the President to the Vice President upon a determination that the former "is unable to discharge the powers and duties of his office." 170
One might infer from this provision a principle that disability is not by itself a sufficient basis for presuming inability to perform. After all, a President's powers cannot be transferred for just any disability; it must be one that makes him "unable" to perform his job. But such an inference stretches textual interpretation to the breaking point. The Twenty-Fifth Amendment is not in any meaningful sense an equality provision. Accordingly, any argument for treating disability discrimination as presumptively invalid must proceed by analogy and from normative principles.
Whether such an argument should prevail is a difficult question for exactly the sorts of reasons that the question whether any form of discrimination should be presumptively invalid is difficult under the current, multi-factored approach. It counts in favor of such a presumption that, as Congress recognized in enacting the Americans With Disabilities Act ("ADA"), "historically, society has tended to isolate and segregate individuals with disabilities, and, despite some improvements, such forms of discrimination against [*1007] individuals with disabilities continue to be a serious and pervasive social problem." 171
Moreover, the very enactment of the ADA, as well as the enactment of parallel provisions in all fifty states, 172
indicates a substantial national consensus that discrimination on the basis of disability is considered odious.
Nevertheless, courts may be ill-equipped to fashion a freestanding constitutional principle prohibiting disability discrimination. Here again, the ADA is instructive. For example, its employment provisions do not merely proscribe decisions not to hire or promote the disabled; they also define as discrimination the failure to make reasonable accommodations. 173
The Act goes on to list examples of reasonable accommodations. 174
Courts interpreting the ADA and similar statutes can use these examples as useful starting points, but a court recognizing a freestanding equal protection principle barring disability discrimination would find its judgment in this regard substantially unconstrained. Whether that factor is enough to warrant denying the presumption of invalidity to disability discrimination is a difficult question. How it should be resolved is not my concern here, except to note that the absence of a textual marker in the Constitution is part of what renders the inquiry difficult, and thus part of what makes equal protection incorporation attractive.
Finally, equal protection incorporation may have some further utility outside the heightened scrutiny framework. Harms caused by legislation challenged as irrational could be usefully compared with the harms caused by laws that discriminate on an enumerated basis. 175
Wherever one draws the line on the plus side of selective incorporation plus, using multiple textual markers to inform the general concept of equality provides considerably greater guidance than trying to reason outward from the single example of the Black [*1008] Codes 176
or by reference to completely open-ended multi-factor tests.
V. Equal Protection Incorporation and Antidiscrimination
Equal protection incorporation does not merely aid in the determination of which classifications are presumptively invalid; it also bears on the question of whether equal protection analysis properly concerns itself with classifications rather than, or at least as well as, classes. Except for the Twenty-Fourth Amendment, each of the voting rights amendments is written expressly in symmetrical antidiscrimination terms, barring discrimination in voting on account of various criteria. These constitutional provisions prohibit more than denials of the vote to African-Americans, former slaves, women, or eighteen-to twenty-year-olds, even though they were clearly adopted to rectify the prior disenfranchisement of these groups. Rather, they enact general prohibitions. Laws barring whites, men, or persons over forty from voting are literal violations of, respectively, the Fifteenth, Nineteenth, and Twenty-Sixth Amendments. If equal protection incorporation is to be true to text, then laws discriminating against whites, men, or persons over forty in other domains should be presumptively invalid.
Of course, one could make a textual argument for the exact opposite reading. Unlike the voting rights amendments, the Equal Protection Clause is not expressly written in the form of an antidiscrimination principle. The contrast, one might think, means that the Equal Protection Clause should not be interpreted as an antidiscrimination principle, but instead as what is sometimes termed an antisubordination principle. 177
There is nothing illogical about juxtaposing the Equal Protection Clause and the voting rights amendments in this way. As I argued in Part III, however, the voting rights amendments are best read as carrying forward the egalitarian project of the Equal Protection [*1009] Clause. If one accepts that argument for purposes of deciding which classes or classifications trigger heightened scrutiny under the Equal Protection Clause, it is difficult to see why the same argument does not bear on whether heightened scrutiny is triggered by subordination of protected classes or discrimination on the basis of illicit classifications.
Accordingly, the argument for equal protection incorporation points in the direction of an antidiscrimination principle. On this coherentist account, it is no accident that nearly all statutory efforts to give effect to the concept of equality do so in the express language of an antidiscrimination principle. 178
In short, the voting rights amendments and most antidiscrimination statutes are best understood as interpretations rather than repudiations of the constitutional value of equality.
The choice between an antidiscrimination principle and an antisubordination principle has the greatest potential doctrinal impact in two categories of cases: challenges to laws that advantage rather than disadvantage a traditionally subordinated group, that is, affirmative action cases, and challenges to neutral laws that have a disparate impact on a protected class. Current Supreme Court doctrine understands equal protection as an antidiscrimination principle rather than an antisubordination principle by subjecting affirmative action programs to the same level of scrutiny as government policies that disadvantage traditionally subordinated groups, 179
and by requiring a showing of purposeful discrimination [*1010] before subjecting a neutral law with a disparate impact on a protected class to heightened scrutiny. 180
Although equal protection incorporation supports reading the Equal Protection Clause as an antidiscrimination principle, it does not necessarily follow that antidiscrimination must be understood in the highly formal way in which the Supreme Court currently understands it. The constitutionality of affirmative action and neutral laws that have a disparate impact on a protected class are difficult questions, but they are difficult whether or not one accepts antidiscrimination as the best conception of the Equal Protection Clause.
Consider a federal statute like Title VII of the Civil Rights Act of 1964. It is expressly worded as an antidiscrimination principle - it tells employers not "to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual's race, color, religion, sex, or national origin." 181
A parallel provision applies the identical principle to job training programs. 182
Nonetheless, in United Steelworkers of America v. Weber, 183
the Supreme Court ruled, 5-2, that Title VII does not prohibit an employer's voluntary efforts to redress a racial imbalance in its workforce by giving a hiring preference to African-American applicants for an on-the-job training program. 184
Although Chief Justice Warren Burger and then-Justice William Rehnquist found it impossible to square this result with the express language of the Act, 185
the majority had no difficulty finding, based on the statute's legislative history and the context in which it arose, that the antidiscrimination law in question permitted at least some forms of affirmative action. 186
[*1011] Scholars have long debated whether the Court correctly characterized Congress's purpose in enacting Title VII, and if so, whether that should have been dispositive in Weber. 187
I am not now much interested in that question as such. Rather, my point here is that the debate over the validity of affirmative action is not resolved by construing equal protection as an antidiscrimination principle. Many or most of those who read antidiscrimination language as categorically barring affirmative action also read the Equal Protection Clause as having exactly the same effect. 188
By the same token, as Weber shows, those Justices and scholars who find the words "equal protection" malleable enough to permit some preferences for members of traditionally disadvantaged groups typically reach the same conclusion when interpreting statutory antidiscrimination provisions.
The actual decision in Weber conceded that a "literal" reading of the antidiscrimination norm would forbid race-based affirmative action, but the Court rejected that reading based on the statute's purpose and spirit. 189
Yet the concession was not obviously necessary. One could plausibly argue that where a program of affirmative action is sufficiently justified, the denial of some opportunity to non-minority applicants is not, in the end, based on race (or whatever the forbidden characteristic is), but is instead based on whatever reason justifies the affirmative action program. Alternatively, one might think that as used in the statute and everyday language, the word "discriminate" means something like "invidiously distinguish," so that distinctions drawn to benefit [*1012] traditionally disadvantaged groups fall outside its ambit. 190
Which argument one finds persuasive is likely to have less to do with one's views on language than with one's views on the moral and practical implications of affirmative action.
The same broad point also applies to the question of whether laws that have a disparate impact on a traditionally disadvantaged group should be understood to trigger any special judicial scrutiny. Here too, the disagreement over particulars persists even after one moves from equal protection to antidiscrimination.
Consider another statutory provision expressed as an antidiscrimination norm, Title VI of the Civil Rights Act of 1964, which provides that no person shall, "on the ground of race, color, or national origin, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any [federally funded] program or activity." 191
In Alexander v. Sandoval, 192
a 5-4 Supreme Court ruled that there is no private right of action to enforce a Department of Justice regulation prohibiting recipients of federal funds from using neutral criteria that "have the effect of subjecting individuals to discrimination because of their race, color, or national origin." 193
The core prohibition of Title VI is on express or purposeful discrimination, the Court reasoned, and prohibition of disparate impact was too far removed from that core to permit the assumption that Congress intended to create a private right of action to enforce the regulation. 194
Although the Sandoval majority asserted that it is "beyond dispute ... that [Title VI] prohibits only intentional discrimination," 195
the four dissenters thought that the disparate impact "regulations are inspired by, at the service of, and inseparably intertwined with [the statute's] anti-discrimination mandate. Contrary to the majority's suggestion, they "apply' [the statutory] prohibition on [*1013] discrimination just as surely as the intentional discrimination regulations." 196
The proper judicial treatment of disparate impact is a hard question because there are difficulties on either side of the debate. Current doctrine requires plaintiffs challenging neutral laws or policies with a disparate impact to prove illicit motive, but such laws or policies are typically generated by multi-member bodies whose motives are always a somewhat artificial construct. The best that a court can do is to ask whether some policy or decision would have been adopted even absent the illicit factor. Because this inquiry is invariably speculative, the existing illicit motive test may be more difficult to administer than a disparate impact test would be. 197
A disparate impact approach has its own difficulties, however. For example, proponents of treating disparate impact alone as constitutionally objectionable seldom address an equally challenging puzzle - the question whether a policy's disparate impact voids the policy in toto or only for members of the disadvantaged group. Consider, for example, Test 21, the civil service examination used by the Washington, D.C. police force and challenged in Washington v. Davis. 198
Test 21 had a disparate racial impact on African-Americans. 199
Under a disparate impact regime, could the test nonetheless be used in screening white applicants? Doing so would create an explicit racial classification, in which white applicants take Test 21 but African-Americans are screened by some other mechanism. If that is unacceptable, does this mean that an unsuccessful white applicant should have standing to challenge the use of Test 21 on the grounds that it has a disparate impact on African-Americans? 200
Answering yes would open the floodgates even more [*1014] because a great many laws and policies have a disparate impact on some protected class.
As with the affirmative action example, so here too my concern is not with who has the better of the argument, but with the fact that the argument persists in nearly all of its intensity once one moves from the general concept of equal protection to the supposedly more determinate conception of antidiscrimination. The practical difficulties I have just raised are inherent in the fine-grained question of whether to treat disparate impact as itself constitutionally problematic. They are not conceptual problems that arise out of the choice between an antidiscrimination principle and an antisubordination principle.
Incorporation leads to an antidiscrimination conception of equal protection, but such a conception does not necessarily rule out affirmative action or immunize neutral laws and policies that have a disparate impact. What then is the point of saying that equal protection incorporation suggests an antidiscrimination rather than an antisubordination principle?
Besides aiding in the identification of those forms of discrimination that are presumptively invalid, equal protection incorporation gives effect to the intuition that discrimination against whites on the basis of race or discrimination against men on the basis of sex, if not quite as harmful as discrimination against African-Americans and women, nonetheless poses a more substantial constitutional problem than, say, discrimination against people who rent rather than own their homes 201
or against people who drive red cars. 202
An antisubordination principle would lead to a different conclusion.
In an important 1976 article, Professor Fiss championed the antisubordination principle as the best understanding of the Equal Protection Clause. Professor Fiss complained that under the antidiscrimination principle, "the permissibility of preferential treatment [for blacks and, by extension, other disadvantaged groups] is tied to the permissibility of hostile treatment against [*1015] blacks [and other such groups]." 203
This is a fair argument against colorblindness in its most extreme form, but not against antidiscrimination as such, nor even against a doctrinal structure that subjects all uses of some suspect or quasi-suspect classification to the same level of scrutiny. An antidiscrimination principle does not necessarily say that whites and blacks, men and women, or heterosexuals and homosexuals are identically situated. What it does say is, "that the government's use of race [or sex or sexual orientation, etc.] is frequently inconsistent with notions of human dignity" 204
in a way that the use of other classifications - like owning versus renting or the color of one's automobile - is not.
To summarize, this Part has argued that a holistic interpretation of the equality provisions of the Constitution has implications beyond identifying the presumptively illegitimate grounds for government decisionmaking - it also leads to an understanding of the Equal Protection Clause as an antidiscrimination principle. As a practical matter, this principally means that even laws advantaging members of traditionally disadvantaged groups should be subject to heightened judicial scrutiny. Interpreting equal protection as an antidiscrimination principle does not, however, have many further doctrinal implications because the meaning of the antidiscrimination principle is itself deeply contested. Disputes over affirmative action and disparate impact reemerge within the antidiscrimination principle.
VI. The Section Five Power
Equal protection incorporation is mainly a strategy for judicial interpretation of the Equal Protection Clause, one that uses constitutional text in a manner that is broadly consistent with the text's historical underpinnings, for the purpose of constraining what would otherwise be an open-ended interpretive enterprise. Unconstrained judicial interpretation of the Constitution is problematic because federal judges are not democratically accountable - at least not after they are confirmed. The extent to which the so- [*1016] called countermajoritarian difficulty 205
is a substantial problem may be debated, 206
but regardless of the extent to which judicial review is an aberrant institution in American democracy, surely Congress may act pursuant to the subjectively held values of its members and their constituents without the requirement that these values be tied to constitutional text or any other external constraint. That, after all, is what an elected legislature is for. Accordingly, it would seem that Congress, acting pursuant to its power to enforce the Fourteenth Amendment, should have considerably wider latitude than the Supreme Court to define equal protection without reference to other provisions of the constitutional text. 207
The Rehnquist Court has concluded, nonetheless, that Congress has no power under the Fourteenth Amendment to define substantive rights differently from the Court. For example, under the Court's precedents, discrimination on the basis of age or disability is subject only to rational basis scrutiny. 208
The point of the rational basis test, of course, is to avoid excessive judicial interference with decisions by politically accountable actors. In other words, it is a principle of deference. That, at any rate, was the (perfectly sound) argument made in defense of Acts of Congress that barred many forms of age and disability discrimination, and it was accepted by four Justices in Board of Trustees of the University of Alabama v. Garrett. 209
A majority of the Court, however, disagreed, treating the rational basis test as the meaning of the Equal Protection Clause, rather than a mere judicial tool. Applying a standard it first fashioned [*1017] in City of Boerne v. Flores, 210
the Court held that laws proscribing age and disability discrimination are not "congruent and proportionate" responses to what the Justices themselves would deem violations of the Fourteenth Amendment under their own precedents. 211
Given the Court's premises, this conclusion meant that the Acts were not authorized by the Section Five power.
In assessing the Court's recent Section Five jurisprudence it is worth distinguishing two lines of cases. Several of the Section Five cases do not concern congressional power per se, but only congressional power to subject unwilling states to lawsuits for money damages. Even after United States v. Lopez, 212
there is little doubt that Congress has the authority under the Commerce Clause to proscribe whatever forms of employment discrimination it wishes. 213
The Court, however, has held that Congress may abrogate state sovereign immunity when acting pursuant to the Reconstruction Amendments, but not when acting pursuant to its Article I powers. 214
That is how the Section Five issue arises in state sovereign immunity cases such as Kimel and Garrett. After those two cases, plaintiffs could still sue private actors under the ADEA and the ADA. They could even seek injunctive relief against individual state officials pursuant to Ex parte Young. 215
Not all of the Court's Section Five cases have arisen in the context of state sovereign immunity, however. For example, the Religious Freedom Restoration Act ("RFRA") never applied to private actors, and it is quite clear that after City of Boerne, an Ex parte Young action against state officials pursuant to RFRA is unavailable. City of Boerne invalidated RFRA; it did not merely grant states immunity to damages actions. 216
Likewise, the Court's [*1018] decision in United States v. Morrison 217
was about power as such, rather than state sovereign immunity. There, the Court held that Congress was without power to enact the civil remedy provision of the Violence Against Women Act. 218
The Court rejected both the Commerce Clause and Section Five of the Fourteenth Amendment as sources of congressional authority. 219
How, if at all, does equal protection incorporation bear on the scope of Congress's power under Section Five of the Fourteenth Amendment? The answer depends on whether one accepts the logic of the Court's recent federalism decisions, which have nearly all been decided by the same 5-4 margin. If one thinks that the Court's extension of state sovereign immunity beyond the language of the Eleventh Amendment is unwarranted, 220
then one need not puzzle much over whether the ADEA or the ADA is authorized by the Section Five power. The Commerce Clause, according to this view, provides all the necessary authority.
Likewise, one might think that the Court's efforts to circumscribe Congress's affirmative powers are unsound in light of the Constitution's political safeguards for state sovereignty. 221
If so, [*1019] then equal protection incorporation would have no implications for the scope of Congress's power under Section Five, which would be beyond the Court's purview. Somewhat more narrowly, but to the same ultimate effect, one might think that judicial review of congressional action under Section Five is misguided because the Fourteenth Amendment's primary purpose was to empower Congress, not the Court. 222
If one accepts such critiques of the Court's Section Five jurisprudence, then equal protection incorporation, as an approach to judicial interpretation of the Equal Protection Clause, places no limits on Congress's ability to act pursuant to Section Five. Although I have some sympathy for the criticisms of the Court's federalism decisions - especially criticisms of the state sovereign immunity decisions - it is nonetheless worth asking whether equal protection incorporation has any implications if one accepts the basic premises of the recent decisions. For absent a dramatic change of heart or personnel, those decisions are likely here to stay, at least in the medium term.
The driving force behind both strands of the Court's Section Five jurisprudence is a concern about circumvention. Having held that Congress may not abrogate state sovereign immunity under Article I, 223
the Court does not want to see Congress use the Section Five power to abrogate state sovereign immunity, except in a limited domain. Likewise, having insisted in Lopez that the principle of enumerated powers means that some activities are beyond Congress's authority to regulate, 224
five Justices believe that they must deny Congress a de facto plenary power under Section Five. 225
[*1020] As Professor Friedman and I have noted, "virtually any law, indeed, any human action, can, on some rational understanding, be seen to deprive someone of life, liberty, property, or equality; thus, a congressional power to enforce Congress's own definition of the substantive provisions of the Fourteenth Amendment could well become [a de facto] plenary congressional power." 226
The congruence and proportionality test is the Court's response. By closely mooring Congress's Section Five power to the Court's own definition of rights under Section One of the Fourteenth Amendment, the Court ensures that there are limits to Congress's authority under the former.
Even if one thinks, as I do, that the Court has applied the congruence and proportionality test in an overly restrictive manner, given the Court's premises, some attempt to draw an outer boundary around Congressional power is necessary. Within these premises, an account of Section Five that permits Congress to define the Fourteenth Amendment independently of the Court's Section One jurisprudence must, at a minimum, include a limit on the Section Five power.
Professors Post and Siegel think that such a limit is self-evident in the case of antidiscrimination law. They write: "Whatever might be said about Section 5 power generally, the use of Section 5 power to combat unconstitutional discrimination cannot be conceived as a potential threat to the legitimate balance of the federal system so long as the history [of the Civil Rights Movement] retains its normative force." 227
Yet the category of antidiscrimination law is hardly self-limiting. Consider four plausible ways in which Congress might expand the scope of equal protection beyond the Court's own interpretation: First, Congress might bar state and federal laws and policies having a disparate impact on some set of protected classes; second, Congress might lift the state action requirement; third, Congress might expand the list of suspect and quasi-suspect classifications beyond those recognized by the Court; and fourth, Congress might bar any [*1021] number of governmental or private activities as "arbitrary" and thus unfairly discriminatory, independent of their use of any proscribed classification or their impact upon any designated class. If the Court permitted this fourth move, Congress could enact virtually any law as an "antidiscrimination" or "equal protection" measure.
Given political realities, this worry is almost certainly unrealistic in practice. That rejoinder, however, in effect revives the argument that the political safeguards of federalism will prevent Congress from acting as though it has anything like plenary power, but for present purposes we are accepting the Court's premises, which reject exclusive reliance on political safeguards. Operating within the Court's premises, we must look for some formal limit.
Although antidiscrimination law is not, in the abstract, self-limiting, there is by now a long-established understanding of the term that provides a genuine limit. Antidiscrimination laws typically proscribe disparate treatment of persons based on specified classifications and/or disparate impact upon specified classes. They need not be limited to state action, and in some circumstances they may reach a category of "arbitrary" or "irrational" action even absent an effect on a discernible class or classification, at least so long as terms like "arbitrary" and "irrational" are used to signify only extreme cases.
It is immediately apparent that this definition of "antidiscrimination law" would permit Congress great latitude in interpreting the Equal Protection Clause independently of the Court's own interpretation, and in that sense, this approach differs sharply from the Court's recent Section Five jurisprudence. Nonetheless, the features of antidiscrimination law that I have just set forth would serve the underlying aim of the Court's Section Five jurisprudence: They would prevent Section Five from becoming a de facto plenary power.
Suppose, for example, that under the guise of enforcing the Equal Protection Clause, Congress were to re-enact the Gun Free School Zones Act invalidated in Lopez. Guns near schools, it might be argued, deprive those students who are exposed to the risk of violence of an equal opportunity to receive an education. This is an extension of the Court's account of equal protection, but is it a greater extension than a statute that expands the Court's list [*1022] of presumptively invalid forms of discrimination to include a new category?
The answer I would give - and the one suggested by equal protection incorporation - is yes. Equal protection incorporation uses the voting rights amendments to shape our understanding of the Equal Protection Clause. On the one hand, where judicial interpretation of the Equal Protection Clause is at stake, the need to constrain judges may require that equal protection doctrine track the form and content of the textual markers rather closely. Congress, on the other hand, should only be subject to a considerably weaker constraint requiring that there be some discernible limit to its powers. Accordingly, it should be permitted to enact antidiscrimination laws that substantially expand upon the Constitution's text. At some point, however, even legislation that may be plausibly understood as serving egalitarian values - such as the Gun Free School Zones Act - ceases to be antidiscrimination legislation as that term is commonly understood.
We can now see why Professors Post and Siegel are correct in thinking that the category of antidiscrimination law is a bounded one. Although the category is theoretically limitless, it is bounded by a set of cultural understandings of what is and what is not an antidiscrimination measure. A statute like the Civil Rights Act of 1964, the core legislative accomplishment of the Civil Rights Movement, perfectly fits the paradigm of the voting rights amendments. Thus, although "antidiscrimination law" may not be a self-defining term, a court charged with enforcing outer limits on Congress's power will not have an overly difficult task distinguishing statutes that fit within that general paradigm from those, like the Gun Free School Zones Act, that do not.
Finally, note that any effort to circumscribe Congress's power under Section Five of the Fourteenth Amendment must limit what can be done under the Due Process and Privileges or Immunities Clauses as well as the Equal Protection Clause. Exactly how that task should be accomplished is beyond the scope of this Article. I would observe, however, that the Boerne Court's refusal to accept Congress's substitution of an effects test for the Court's purpose test tethers Congressional action under Section Five to the Court's Section One jurisprudence more tightly than necessary. A decision upholding the Religious Freedom Restoration Act would not have [*1023] threatened to convert Section Five into anything like a plenary power.
In sum, equal protection incorporation is mainly a strategy for judicial interpretation of the Equal Protection Clause. It thus has no necessary implications for Congressional interpretation under Section Five. If, however, one takes seriously the current Supreme Court's goal of circumscribing the Section Five power, equal protection incorporation suggests an outer boundary that, while more deferential than the Court's strict interpretation of the congruence and proportionality test, nonetheless serves as a real limit. Like equal protection incorporation generally, that limit has the further virtue that it is rooted in the Constitution's text.
During his ill-fated Supreme Court confirmation hearings, Judge Robert Bork told the Senate Judiciary Committee that the Ninth Amendment, with its mysterious reference to "other" 228
rights, cannot be a source of judicially enforceable norms, because it provides judges with no more guidance than a provision whose key terms are obscured by an ink blot. 229
The same can be said, and has been said repeatedly, about the Equal Protection Clause of the Fourteenth Amendment.
The Supreme Court eventually hit upon a successful strategy for constraining its own discretion in interpreting the Constitution's liberty provisions - incorporation of most of the Bill of Rights plus textually constrained interpolation, extrapolation, and reasoning by analogy from and beyond those provisions. However, the Court's equal protection jurisprudence remains largely unconnected to the Constitution's text. Equal protection incorporation can fill this gap.
Should it be objected, finally, that equal protection incorporation is a gimmick, an effort to grasp whatever tools are at hand and utilize them for purposes for which they were never intended, we can respond that the objection overstates the point. There is solid [*1024] evidence that the voting rights amendments were understood as continuing the project of the Equal Protection Clause. Moreover, the "gimmick" accusation can easily be leveled against due process incorporation as well. The success of due process incorporation has next to nothing to do with its fidelity (or infidelity) to the original understanding of the Fourteenth Amendment. It results from the important role that constitutional text plays in our constitutional culture. Equal protection incorporation respects the constitutional text without making a fetish of it.Legal Topics:
For related research and practice materials, see the following legal topics:
Constitutional LawBill of RightsFundamental RightsProcedural Due ProcessGeneral OverviewConstitutional LawBill of RightsState ApplicationConstitutional LawEqual ProtectionGeneral Overview FOOTNOTES:
n1. Except where otherwise noted, in this Article I include in the term "Equal Protection Clause" the principle of equal protection applicable to the federal government via the Fifth Amendment Due Process Clause. See Bolling v. Sharpe, 347 U.S. 497, 500 (1954).
n2. The process by which the United States Supreme Court used the specific provisions of the Bill of Rights to inform its interpretation of the Fourteenth Amendment Due Process Clause is commonly called incorporation. Constitutional lawyers sometimes refer to the application of equal protection principles to federal action via the Fifth Amendment Due Process Clause, which was enacted long before the Equal Protection Clause, as "reverse incorporation." See, e.g., Michael C. Dorf, What Does the Second Amendment Mean Today?, 76 Chi.-Kent L. Rev. 291, 326 (2000) (examining the argument that the Fourteenth Amendment altered the meaning of the Second Amendment). Equal protection incorporation as I defend it here is atemporal. It draws guidance from previously enacted provisions, like the Religion Clauses of the First Amendment, as well as subsequently enacted provisions, like those prohibiting voting discrimination on a variety of grounds.
n3. See FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 132-33 (2000); United States v. Fausto, 484 U.S. 439, 453 (1988).
n4. See Charles L. Black, Jr., Structure and Relationship in Constitutional Law 7 (1969).
n5. Akhil Reed Amar, Intratextualism, 112 Harv. L. Rev. 747, 748 (1999).
n7. Black, supra note 4, at 7.
n8. See generally John Hart Ely, Democracy and Distrust (1980).
n9. Professor Jackson makes a similar point when she writes:
I believe I am using the term "holistic interpretation" somewhat differently from both Professors Ackerman and Amar, who each also have used that term. Professor Ackerman's focus is on the meaning invested in the document as a whole by political paradigm-shifting constitutional moments that may be represented either in (some of the) actual amendments to the constitution or in implied amendments expressed by voting (under some circumstances) and ratified by judicial construction. Professor Amar's special emphasis is on the technique he calls "intratextualism," in which uses of similar text in different parts of the document are brought into focus together to give meaning to each, and more generally on the textual document itself (though with attention to the text's chronology). My focus is somewhere in between, emphasizing both the benefits of reasoning from overall structure and values and the need to synthesize more recent amendments into an understanding of what the Constitution as a whole requires.
Vicki C. Jackson, Holistic Interpretation: Fitzpatrick v. Bitzer and Our Bifurcated Constitution, 53 Stan. L. Rev. 1259, 1281 n.95 (2001) (citations omitted).
n10. But see Reva B. Siegel,
She the People: The Nineteenth Amendment,
Sex Equality, Federalism, and the Family, 115 Harv. L. Rev. 947, 951 (2002) (arguing that the framers of the Nineteenth Amendment
viewed the question of women's suffrage as having broad implications for social relations between men and women).
n11. 367 U.S. 497, 543 (1961) (Harlan, J., dissenting) ("The liberty guaranteed by the Due Process Clause cannot be found in or limited by the precise terms of the specific guarantees elsewhere provided in the Constitution.").
n12. 426 U.S. 229 (1976).
n13. Id. at 239.
n14. 488 U.S. 469 (1989).
n15. Id. at 493-95.
n16. See Howard Gillman, The Constitution Besieged: The Rise and Demise of Lochner Era Police Powers Jurisprudence (1993); Michael Les Benedict, Laissez-Faire and Liberty: A Re-Evaluation of the Meaning and Origins of Laissez-Faire Constitutionalism, 3 Law & Hist. Rev. 293, 297 (1985).
n17. It may seem hard to take seriously the Declaration's egalitarian language in light of its coexistence with slavery. See Scott v. Sandford, 60 U.S. (19 How.) 393, 410 (1856) ("The enslaved African race were not intended to be included, and formed no part of the people who framed and adopted this declaration ... ."). Even before the Civil War, however, there were those who did argue that the Declaration could be taken as applying to all persons. See id. at 574-75 (Curtis, J., dissenting); cf. Frederick Douglass, The Constitution of the United States: Is It Pro-Slavery or Anti-Slavery?, in 2 Life and Writings of Frederick Douglass, 467-80 (P. Foner ed., 1950), reprinted in Paul Brest & Sanford Levinson, Process of Constitutional Decisionmaking 207-11 (3d ed. 1992) (arguing that slavery was inconsistent with the antebellum Constitution).
n18. Compact Made on Board the Mayflower (Nov. 11, 1620), in Contexts of the Constitution 1 (Neil H. Cogan ed., 1999) (undertaking to "enact, constitute and frame such just and equal laws, ordinances, acts, constitutions and officers, from time to time, as shall be thought most meet and convenient for the general good of the colony").
n19. Durkee v. City of Janesville, 28 Wis. 464, 470 (1871) (approving the Tennessee Supreme Court's derivation of this principle from the "law of the land" clause of the Tennessee Constitution (citing Wally's Heirs v. Kennedy, 10 Tenn. (2 Yer.) 554, 555 (1831)).
n20. See Peter Westen, The Empty Idea of Equality, 95 Harv. L. Rev. 537 (1982).
n21. See Paul Brest, The Supreme Court, 1975 Term - Foreword: In Defense of the Antidiscrimination Principle, 90 Harv. L. Rev. 1, 5 (1976) ("The text and history of the clause are vague and ambiguous ... .").
n22. See The Slaughter-House Cases, 83 U.S. (16 Wall.) 36, 67-71 (1872); Eric Foner, Reconstruction 257 (1989); Akhil Reed Amar, The Supreme Court, 1999 Term - Foreword: The Document and the Doctrine, 114 Harv. L. Rev. 26, 64 (2000) ("The Fourteenth Amendment framers clearly aimed to prohibit [the Black Codes] as a paradigm case of impermissible legislation ... ."); Michael W. McConnell, The Importance of Humility in Judicial Review: A Comment on Ronald Dworkin's "Moral Reading" of the Constitution, 65 Fordham L. Rev. 1269, 1281 (1997) ("The clearest and most indisputable purpose of the Fourteenth Amendment was to provide constitutional authority for the Civil Rights Act of 1866, which outlawed the Black Codes.").
n23. 347 U.S. 483 (1954).
n24. See, e.g., Michael W. McConnell, Originalism and the Desegregation Decisions, 81 Va. L. Rev. 947, 984-1078 (1995) (invoking Republican support in the 1870s for a bill outlawing segregated schools as evidence of the original understanding of the Fourteenth Amendment). But see Michael J. Klarman, Brown, Originalism, and Constitutional Theory: A Response to Professor McConnell, 81 Va. L. Rev. 1881, 1884-1914 (1995) (arguing that Brown contradicts the original understanding of the Fourteenth Amendment); Earl M. Maltz, Originalism and the Desegregation Decisions - A Response to Professor McConnell, 13 Const. Comment. 223, 228-31 (1996) (same); Herbert Hovenkamp, The Cultural Crises of the Fuller Court, 104 Yale L.J. 2309, 2337-43 (1995) (reviewing Owen M. Fiss, Troubled Beginnings of the Modern State, 1888-1910 (1994)) (observing, inter alia, that in its day, the separate-but-equal doctrine of Plessy v. Ferguson, 163 U.S. 537 (1896), was largely uncontested).
n25. See generally Jed Rubenfeld, Affirmative Action, 107 Yale L.J. 427, 429-32 (1997) (arguing that race-conscious Reconstruction programs show that the framers of the Fourteenth Amendment could not have intended it to act as a ban on affirmative action programs); Eric Schnapper, Affirmative Action and the Legislative History of the Fourteenth Amendment, 71 Va. L. Rev. 753, 754-83 (1985) (arguing that those who claim to be orginalists cannot condemn affirmative action because of the many statutes conferring benefits on the basis of race enacted by the framers of the Fourteenth Amendment).
n26. See Cass R. Sunstein, Legal Reasoning and Political Conflict (1996).
n27. Indeed, even those who doubt the ability of courts to provide principled interpretations of constitutional norms such as equality argue forcefully for some conception of equality as a political principle. See, e.g., Mark Tushnet, Taking the Constitution Away From the Courts (1999); Jeremy Waldron, Law and Disagreement (1999).
n28. For a defense of conventionalist judicial review, see Harry H. Wellington, Interpreting the Constitution: The Supreme Court and the Process of Adjudication 20-40 (1990).
n29. Brown v. Bd. of Educ., 347 U.S. 483, 495 (1954) ("Separate educational facilities are inherently unequal.").
n30. As Professor Black explained, what made separate unequal was the social meaning of segregation in the real world. See Charles L. Black, Jr., The Lawfulness of the Segregation Decisions, 69 Yale L.J. 421 (1959).
n31. 163 U.S. 537 (1896).
n32. The choice may also turn partly on empirical judgments concerning questions like whether affirmative action reinforces racial prejudice (as some of its critics contend), how difficult it is for plaintiffs to prove illicit motive in discrimination cases, and how expensive it would be for courts to prohibit government conduct that has a disparate racial impact. I have argued elsewhere that courts can improve the methods by which they learn about the consequences of their decisions. See Michael C. Dorf, The Supreme Court, 1997 Term - Foreword: The Limits of Socratic Deliberation, 112 Harv. L. Rev. 4 (1998). I have also argued that in some contexts, courts are already beginning to adopt methods that encourage other actors to create a record that greatly reduces the guesswork required of judges. See Michael C. Dorf & Charles F. Sabel, Drug Treatment Courts and Emergent Experimentalist Government, 53 Vand. L. Rev. 831 (2000); Michael C. Dorf & Charles F. Sabel, A Constitution of Democratic Experimentalism, 98 Colum. L. Rev. 267 (1998); Michael C. Dorf, An Institutional Approach to Legal Indeterminacy (unpublished manuscript, on file with the Virginia Law Review Association). At present, however, most courts - including the Supreme Court - still decide legislative facts through a combination of hunch and ideology, no more bound by external factors than when they decide purely normative questions.
n33. See, e.g., Pers. Adm'r of Mass. v. Feeney, 442 U.S. 256, 273 (1979) (invoking "the settled rule that the Fourteenth Amendment guarantees equal laws, not equal results").
n34. The classic statement of the antisubordination principle was given by Professor Fiss, who argued that the Equal Protection Clause prohibits laws or official practices that "aggravate (or perpetuate?) the subordinate position of a specially disadvantaged group." Owen M. Fiss, Groups and the Equal Protection Clause, 5 Phil. & Pub. Aff. 107, 157 (1976).
n35. With two exceptions, to which I come momentarily, when courts subject inequalities with respect to fundamental rights to strict scrutiny, the relevant fundamental right is independently protected. For that reason, it might be thought that the fundamental rights prong of equal protection analysis is redundant with due process and could accordingly be jettisoned. See, e.g., Gerald Gunther, The Supreme Court, 1971 Term - Foreword: A Model for a Newer Equal Protection, 86 Harv. L. Rev. 1, 21 (1972) (advocating a version of "equal protection scrutiny [that] would not involve adjudication on the basis of fundamental interests with shaky constitutional roots").
However, as noted, the Court has recognized two fundamental rights for equal protection purposes that are not independently protected: the right to travel and the right to vote. Although a state has no obligation to provide welfare (or other) benefits, it cannot deny those benefits based on a person's having exercised the right to interstate travel. See Shapiro v. Thompson, 394 U.S. 618 (1969). The fundamental right to vote (which is integral to my purposes in this Article), is an even clearer exception to the putative principle that only independently protected rights trigger heightened scurtiny under the Equal Protection Clause. Although there is no substantive right to vote for any particular office, when a state makes a particular office elective, inequalities in the distribution of the franchise are subject to strict scrutiny. See Dunn v. Blumstein, 405 U.S. 330, 336 (1972) (citing cases that establish "a constitutionally protected right to participate in elections on an equal basis with other citizens in the jurisdiction"). This line of cases is hardly free of controversy - see, for example, Bush v. Gore, 531 U.S. 98 (2000) - but the intuition at its core is widely shared. A state need not hold elections for attorney general or comptroller, but if it does, it cannot give substantially greater weight to the votes of rural residents than urban residents (or vice-versa), even if doing so would satisfy rational basis scrutiny.
The Supreme Court has not provided an adequate explanation for why it applies heightened scrutiny to inequalities bearing on the rights to travel and to vote but not to inequalities bearing on other rights that are not independently recognized (explicitly or implicitly) by the Constitution. The leading case is San Antonio Independent School District v. Rodriguez, 411 U.S. 1, 30-39 (1973) (finding no equal protection fundamental right to education). Rodriguez does not explain why an implicit fundamental right to vote can be inferred from the Constitution's silence, while a right to education cannot, stating simply that by 1973 the former had become doctrinally entrenched, while the latter had not. See id. at 34 n.74 ("The constitutional underpinnings of the right to equal treatment in the voting process can no longer be doubted ... ."). Earlier in the opinion, the Rodriguez Court makes the same unconvincing move in distinguishing the unenumerated right to travel, citing its long pedigree but little else. See id. at 31-32.
Whether the Rodriguez Court was right to reject a fundamental right to education is beyond the scope of this Article. This much, however, I can say: Given the Court's entirely appropriate recognition of an unenumerated fundamental right to vote that is protected for equal protection purposes without conferring a general right to have any office be elective, the general proposition for which Rodriguez stands - that the only equal protection fundamental rights are those that "independently enjoy full-blown constitutional protection," id. at 100 (Marshall, J., dissenting) - is false. Perhaps the best that can be said for Rodriguez is that the Court's approach was based on the same impulse that underlies this Article - namely, that equal protection doctrine should be rooted in the Constitution's text. Yet as the voting and travel examples vainly invoked in Rodriguez illustrate, and as I argue throughout this Article, rooting equal protection doctrine in the text need not and should not entail a crabbed literalism.
n36. According to the conventional wisdom, sex or gender is a quasi-suspect classification that triggers intermediate scrutiny. However, in United States v. Virginia, 518 U.S. 515 (1996), the Court appeared to apply something like strict scrutiny to a gender classification, requiring an "exceedingly persuasive justification" for the Virginia Military Institute's all-male student body. Id. at 531. Then, in Nguyen v. INS, 533 U.S. 53 (2001), the Court reverted to a more deferential form of scrutiny for gender classifications. Id at 70. Although the difference between strict and intermediate scrutiny will no doubt sometimes be dispositive, I disregard that difference in this Article. In my view, the dominant question is whether a set of government policies will be subject to any form of heightened scrutiny. Cf. Michael C. Dorf, Incidental Burdens on Fundamental Rights, 109 Harv. L. Rev. 1175, 1200 (1996) (making the same point with respect to heightened scrutiny of laws infringing constitutionally protected liberties).
n37. See Vill. of Willowbrook v. Olech, 528 U.S. 562, 564 (2000) (holding that arbitrary or irrational discrimination against a "class of one" violates the Equal Protection Clause); Romer v. Evans, 517 U.S. 620, 632 (1996) (invalidating a Colorado constitutional amendment that could only be explained by animus toward homosexuals); City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 450 (1985) (holding that a permit requirement for a group home was invalid because it "appears to us to rest on an irrational prejudice against the mentally retarded"); U.S. Dep't of Agric. v. Moreno, 413 U.S. 528, 534 (1973) (striking a food stamp eligibility provision because "a bare congressional desire to harm [hippies] cannot constitute a legitimate governmental interest") (emphasis in original).
n38. Note, Justice Stevens' Equal Protection Jurisprudence, 100 Harv. L. Rev. 1146, 1151 (1987).
n39. See Daniel Farber & Suzanna Sherry, The Pariah Principle, 13 Const. Comment. 257, 260 (1996) (noting the use of this term by various commentators, but declining to join those who classify Romer as such a case).
n40. See Romer, 517 U.S. at 632; Ashutosh Bhagwat, Purpose Scrutiny in Constitutional Analysis, 85 Cal. L. Rev. 297, 314 (1997). I agree with Professor Bhagwat's main thesis that much constitutional law turns on an analysis of government purposes. Legislation, however, often plausibly serves multiple purposes, and courts can only discern the "true" purpose by subjecting challenged laws to heightened scrutiny. Yet in cases like Romer and Cleburne, the Supreme Court denies that it is applying heightened scrutiny.
n41. 304 U.S. 144 (1938).
n42. Id. at 152 n.4.
n43. James Madison contended that in an extended republic, no faction could permanently oppress the others, because factions would need to form varying coalitions on particular issues. See The Federalist No. 10, at 83 (James Madison) (Clinton Rossiter ed., 1961). In justifying the special solicitude that the Court affords to so-called "discrete and insular minorities," Professor Ely explained how persistent, deep-seated prejudice against a group can exclude that group from lawmaking coalitions. See Ely, supra note 8, at 75-77.
n44. See Bruce A. Ackerman, Beyond Carolene Products, 98 Harv. L. Rev. 713, 718-40 (1985).
n45. See id. at 729; Ely, supra note 8, at 174 (observing nonetheless that "there remains something that seems right in the claim that women have been operating at an unfair disadvantage in the political process").
n46. See, e.g., Frontiero v. Richardson, 411 U.S. 677, 686 (1973) ("Since sex, like race and national origin, is an immutable characteristic determined solely by the accident of birth, the imposition of special disabilities upon the members of a particular sex because of their sex would seem to violate "the basic concept of our system that legal burdens should bear some relationship to individual responsibility ... .'") (quoting Weber v. Aetna Casualty & Surety Co., 406 U.S. 164, 175 (1972)).
n47. Immutability is not a sufficient condition for suspectness either. See City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 472 n.24 (Marshall, J., concurring in part, dissenting in part, and concurring in the judgment) ("Many immutable characteristics, such as height or blindness, are valid bases of governmental action and classifications under a variety of circumstances.").
n48. See Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 359 (1978) (Brennan, White, Marshall, & Blackmun, JJ., concurring in part, dissenting in part, and concurring in the judgment).
n49. Several constitutional scholars with whom I shared a draft of this Article objected, arguing that the Twenty-Sixth Amendment only applies to age discrimination against young adults. Yet the text is plainly to the contrary. It states: "The 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." U.S. Const. amend. XXVI, 1. On its face, the Amendment would forbid a maximum voting age of eighty along with a minimum voting age of twenty-one. Nonetheless, for reasons I explore below, the case for treating age discrimination as presumptively invalid is weaker than the case for the other enumerated criteria. See infra Section IV.B.
n50. I do not address whether constitutional doctrine ought to treat each of these (or other) classifications as triggering exactly the same form of scrutiny. See supra note 36; infra text accompanying note 113.
n51. See Pacific Mut. Life Ins. Co. v. Haslip, 499 U.S. 1, 34 (1991) (Scalia, J., concurring) (observing that the Fourteenth Amendment incorporates "virtually all" of the protections of the Bill of Rights); Robert H. Bork, The Tempting of America: The Political Seduction of the Law 94, 236 (1990) (reading the Fourteenth Amendment as entirely procedural, except for its incorporation of most of the Bill of Rights); Ely, supra note 8, at 20-28 (1980) (accepting incorporation via the Fourteenth Amendment's Privileges or Immunities Clause).
n52. Letter from Oliver Wendell Holmes to Learned Hand (June 24, 1918), in Gerald Gunther, Learned Hand and the Origins of Modern First Amendment Doctrine: Some Fragments of History, 27 Stan. L. Rev. 719, 757 (1975).
n53. 250 U.S. 616 (1919).
n54. See id. at 630 (Holmes, J., dissenting) ("But when men have realized that time has upset many fighting faiths, they may come to believe even more than they believe the very foundations of their own conduct that the ultimate good desired is better reached by free trade in ideas - that the best test of truth is the power of the thought to get itself accepted in the competition of the market, and that truth is the only ground upon which their wishes safely can be carried out. That at any rate is the theory of our Constitution."). A few years later, Justice Holmes endorsed incorporation of the Free Speech Clause of the First Amendment against the states. See Gitlow v. New York, 268 U.S. 652, 672 (1925) (Holmes, J., dissenting).
n55. See Ely, supra note 8, at 18 (comparing substantive due process to "green pastel redness," a "contradiction in terms").
n56. See Duncan v. Louisiana, 391 U.S. 145, 166 (1968) (Black, J., concurring) ("[The Privileges or Immunities Clause] seems to me an eminently reasonable way of expressing the idea that henceforth the Bill of Rights shall apply to the States."); Laurence H. Tribe, Taking Text and Structure Seriously: Reflections on Free-Form Method in Constitutional Interpretation, 108 Harv. L. Rev. 1221, 1297 n.247 (1995) (advocating a move to the Privileges or Immunities Clause and calling for the overruling of the Slaughter-House Cases, 83 U.S. 36 (1872), to accomplish this goal); cf. Troxel v. Granville, 530 U.S. 57, 80 n. (2000) (Thomas, J., concurring) (arguably hinting that the "Privileges and Immunities Clause" encompasses rights not thought to be within the scope of the Due Process Clause). (Presumably, Justice Thomas meant to cite the Privileges or Immunities Clause of the Fourteenth Amendment, rather than the Privileges and Immunities Clause of Article IV.)
n57. See Cong. Globe, 39th Cong., 1st Sess. 1088-89 (1866) (statement of Rep. Bingham); Cong. Globe, 39th Cong., 1st Sess. 2764-65 (1866) (statement of Sen. Howard). There is also substantial evidence that the Congress that proposed the Fourteenth Amendment believed, contra Barron v. Baltimore, 32 U.S. (7 Pet.) 243 (1833), that the Privileges and Immunities Clause of Article IV already made the Bill of Rights applicable to the states. See Richard L. Aynes, On Misreading John Bingham and the Fourteenth Amendment, 103 Yale L.J. 57, 69-74 (1993). For an interesting attempt to uncover the prevailing view of the Bill of Rights during Reconstruction, see generally Akhil Reed Amar, The Bill of Rights: Creation and Reconstruction (1998). Professor Amar criticizes the manner in which the Supreme Court has incorporated the Bill of Rights, but agrees that some form of incorporation is justified on originalist grounds.
n58. See Raoul Berger, The Fourteenth Amendment and the Bill of Rights (1989) (arguing against incorporation); cf. Earl M. Maltz, Commentary on Akhil Reed Amar's The Bill of Rights: Creation and Reconstruction: The Concept of Incorporation, 33 U. Rich. L. Rev. 525, 525 (1999) ("Although the idea that the Privileges and Immunities Clause incorporated the Bill of Rights was discussed extensively during the early Reconstruction Era, the potential legal implications of the incorporation doctrine were not explored in detail until the debates over what was ultimately to become the Civil Rights Act of 1875.").
n59. See Hurtado v. California, 110 U.S. 516, 538 (1884) (holding that the grand jury right is not incorporated); Minneapolis & St. Louis R. Co. v. Bombolis, 241 U.S. 211, 217 (1916) (refusing to incorporate the civil jury trial right). Although Hurtado and Bombolis were decided prior to the Court's incorporation of most of the provisions of the Bill of Rights, they have both been cited as good law in the post-incorporation period. See, e.g., Apprendi v. New Jersey, 530 U.S. 466, 499 (2000) (Thomas, J., concurring) (citing Hurtado); Georgia v. McCollum, 505 U.S. 42, 52 (1992) (citing Bombolis).
n60. 198 U.S. 45 (1905).
n61. 410 U.S. 113 (1973).
n62. David Yassky, The Second Amendment: Structure, History, and Constitutional Change, 99 Mich. L. Rev. 588, 652 n.265 (2000). Justices Murphy and Rutledge advocated what is sometimes called "incorporation plus." See, e.g., Adamson v. California, 332 U.S. 46, 124 (1947) (Murphy, J., dissenting) (arguing that the rights protected by the Fourteenth Amendment are not "entirely and necessarily limited by the Bill of Rights").
n63. See Bolling v. Sharpe, 347 U.S. 497, 500 (1954) (reasoning that it would be "unthinkable" to impose a duty on the federal government under the Fifth Amendment's Due Process Clause that is less than the one imposed on the states under the Fourteenth Amendment's Equal Protection Clause).
n64. Note that according weight to this fact does not necessarily entail originalism. For example, Justice Story argued that originalism is inconsistent with a written constitution. If the meaning of a written text must be discerned from obscure sources, Justice Story contended, it cannot serve as a popular charter. See Joseph Story, Commentaries on the Constitution of the United States 184-85, 210 (Ronald D. Rotunda & John E. Nowak eds., 1987) (originally published as an abridgement, 1833).
n65. In a recent poll that I devised, a representative sample of a thousand Americans were asked whether "the fact that the Supreme Court can strike down unconstitutional laws is a good feature of the American system of government." See Lawyers and the Legal Profession: A Columbia Law Survey (April 2002), at http://www.law.columbia.edu/news/surveys/survey<uscore>opinion<uscore>questionnaire.htm (last visited Aug. 20, 2002). Eighty-three percent said yes. See Detailed Tabulations at 17, at http://www.law.columbia.edu/news/surveys/Survey.pdf (last visited Aug. 20, 2002).
n66. Merlo J. Pusey, 1 Charles Evans Hughes 204 (1951) (quoting an extemporaneous 1907 speech by Justice Hughes).
n67. See Minor v. Happersett, 88 U.S. (21 Wall.) 162 (1875) (rejecting an Equal Protection Clause challenge to Missouri's restriction of the franchise to men); Vikram David Amar, Jury Service as Political Participation Akin to Voting, 80 Cornell L. Rev. 203, 204-05 (1995). A third category, consisting of social rights, was sometimes said to be entirely outside the purview of law. See Plessy v. Ferguson, 163 U.S. 537, 544 (1896) ("In the nature of things [the Fourteenth Amendment] could not have been intended ... to enforce social, as distinguished from political equality ... .").
n68. See Siegel, supra note 10, at 976-87.
n69. See Barry Friedman & Scott B. Smith, The Sedimentary Constitution, 147 U. Pa. L. Rev. 1, 5-7 (1998); Larry Kramer, Fidelity to History - And Through It, 65 Fordham L. Rev. 1627, 1630-32 (1997).
n70. Amar, supra note 57, at 48.
n71. 100 U.S. 303 (1880).
n72. See Amar, supra note 67, at 204.
n73. Here is the Court's explanation:
That the West Virginia statute respecting juries - the statute that controlled the selection of the grand and petit jury in the case of [Strauder] - is such a discrimination ought not to be doubted. Nor would it be if the persons excluded by it were white men. If in those States where the colored people constitute a majority of the entire population a law should be enacted excluding all white men from jury service, thus denying to them the privilege of participating equally with the blacks in the administration of justice, we apprehend no one would be heard to claim that it would not be a denial to white men of the equal protection of the laws. Nor if a law should be passed excluding all naturalized Celtic Irishmen, would there be any doubt of its inconsistency with the spirit of the amendment.
Strauder, 100 U.S. at 308. Earlier in the opinion, the Court expressly associates the Fourteenth and Fifteenth Amendments. See id. at 306 (describing the Fourteenth Amendment as "one of a series of constitutional provisions having a common purpose").
n74. Richard Briffault, Bush v. Gore as an Equal Protection Case, Fla. St. U. L. Rev. 325, 347 (2001) (quoting Reynolds v. Sims, 377 U.S. 533, 568 (1964)).
n75. 531 U.S. 98 (2000).
n76. 509 U.S. 630 (1993).
n77. See Foley v. Connelie, 435 U.S. 291, 295 (1978) (holding that state classifications based on alienage are suspect only to the extent that they strike "at the noncitizens' ability to exist in the community, a position seemingly inconsistent with the congressional determination to admit the alien to permanent residence").
n78. But see Gerald M. Rosberg, Aliens and Equal Protection: Why Not the Right to Vote?, 75 Mich. L. Rev. 1092 (1977). I do not mean to suggest that alien disenfranchisement is constitutionally required. It is not. See Gerald L. Neuman, "We Are the People": Alien Suffrage in German and American Perspective, 13 Mich. J. Int'l L. 259 (1992); Jamin B. Raskin, Legal Aliens, Local Citizens: The Historical, Constitutional and Theoretical Meanings of Alien Suffrage, 141 U. Pa. L. Rev. 1391 (1993).
n79. See Ely, supra note 8, at 135.
n80. Admittedly, in the absence of the Twenty-Sixth Amendment, one could readily imagine the Court upholding a minimum voting age of nineteen, twenty, or twenty-one, but this may be only a matter of line drawing. For example, even absent the Twenty-Sixth Amendment, certainly a state law restricting the franchise to those between the ages of 45 and 65 would be subject to, and likely fail, heightened scrutiny.
n81. Siegel, supra note 10, at 1045 (emphasis in original).
n82. See id. at 976-87.
n83. See Mary Frances Berry, Why ERA Failed 102-04 (1986); Jane J. Mansbridge, Why We Lost the ERA 67-89 (1986).
n84. I have been making this point with respect to the Nineteenth Amendment
for some time now. See Michael C. Dorf, Integrating Normative and Descriptive Constitutional Theory: The Case of Original Meaning, 85 Geo. L.J. 1765, 1778-79 (1997) [hereinafter "Integrating Constitutional Theory"]. I have also discussed the interaction between the Nineteenth and Fourteenth Amendments in Michael C. Dorf, A Nonoriginalist Perspective on the Lessons of History, 19 Harv. J.L. & Pub. Pol'y 351, 356-58 (1996).
n85. For an argument that it does, see Jackson, supra note 9, at 1290-91. Professor Jackson writes:
I view the Fourteenth Amendment as having had constitutional progeny in each of the four Amendments subsequent to the Fourteenth that affirmed the basic constitutional value of equality in the exercise of rights of citizenship - notably, voting. The Fifteenth (1870), Nineteenth (1920), Twenty-Fourth (1964), and Twenty-Sixth (1971) Amendments all sought to expand and secure the right to vote. They thus can be read as elaborations of the basic message of equality of national citizenship founded in Section 1 of the Fourteenth Amendment. Their frequency and relative currency vis-a-vis 1789 makes plausible the argument that these amendments should be understood as importantly redefining what the basic values of this Constitution are ... .
n86. Professor David Strauss asserts that the Twenty-Sixth Amendment does not even reflect "a decision by the People during the Vietnam War and in response to the baby boom generation, that eighteen-year-olds should have the franchise. The People may have made such a decision," he writes, "but if so, they made it before the formal amendment process began." David A. Strauss, The Irrelevance of Constitutional Amendments, 114 Harv. L. Rev. 1457, 1489 (2001). Professor Strauss correctly points out that the Twenty-Sixth Amendment was enacted in response to a Supreme Court ruling sustaining the Voting Rights Act's establishment of an eighteen-year-old voting age for federal elections, but invalidating it as to state elections. See id. Professor Strauss is also correct in supposing that the People, through Congress, did in fact decide that people old enough to be drafted are old enough to vote. See id. Congress stated in the Voting Rights Act Amendments that denying the franchise to eighteen-year olds was ""a particularly unfair treatment of such citizens in view of the national defense responsibilities imposed' on them." Oregon v. Mitchell, 400 U.S. 112, 142 (1970) (Douglas, J., concurring in part and dissenting in part) (quoting Voting Rights Act Amendments of 1970, Pub. L. No. 91-285, 301(a)(1), 84 Stat. 314, 318 (1970)). It is thus difficult to understand why Professor Strauss thinks that the Twenty-Sixth Amendment did not also enact at least that principle when it was proposed and ratified closely on the heels of the Court's decision in Oregon v. Mitchell. See infra text accompanying notes 94-95.
n87. 29 U.S.C. 621-33(a) (1994).
n88. 347 U.S. 483 (1954).
n89. 411 U.S. 677 (1973). The idea that the Nineteenth Amendment
might affect the interpretation of other provisions of the Constitution, however, was accepted almost immediately. Although the holding of Adkins v. Children's Hospital, 261 U.S. 525 (1923), has been rejected because of its expansive view of constitutionally protected economic liberty, surely there was nothing wrong with the Adkins Court's use of the Nineteenth Amendment
to discredit the sex-discriminatory nature of the minimum wage law that applied to children and women but not to men. The Court recognized "the great - not to say revolutionary - changes which [had occurred since decisions like Muller v. Oregon, 208 U.S. 412 (1908)] ... in the contractual, political and civil status of women, culminating in the Nineteenth Amendment
... ." Id. at 553.
n90. See supra note 49.
n91. Such an inference would have two principal doctrinal consequences. First, it would authorize Congress to create private rights of action against states for age discrimination pursuant to its power to enforce the Fourteenth Amendment. Contra Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 82-83 (2000). Second, it would require heightened scrutiny of laws setting nineteen, twenty, or twenty-one as the minimum age for certain entitlements, such as the right to purchase alcohol. Whether such laws could be justified because of a compelling interest in prohibiting drunk driving or in preventing minors (who can more readily pass for eighteen than for twenty-one) from obtaining alcohol are questions I do not consider here.
n92. See Pollock v. Farmers' Loan & Trust Co., 157 U.S. 429 (1895).
n93. Even before 1937, when the Court abandoned heightened scrutiny of most economic and social legislation - see, for example, West Coast Hotel Co. v. Parrish, 300 U.S. 379 (1937) - and accepted a broad view of Congress's power to regulate interstate commerce - see, for example, NLRB v. Jones & Laughlin Steel Corp., 301 U.S. 1 (1937) - the Court accepted that an executive agency exercising quasi-legislative and quasi-judicial powers could be insulated from Presidential oversight. See Humphrey's Ex'r v. United States, 295 U.S. 602 (1935).
n94. See Oregon v. Mitchell, 400 U.S. 112, 117-18 (1970).
n95. See S. Rep. No. 92-26, at 11-18 (1971); H.R. Rep. No. 92-37, at 6 (1971).
n96. See Michael C. Dorf, The Paths to Legal Equality: A Reply to Dean Sullivan, 90 Cal. L. Rev. 791, 807 (2002).
n97. Cf. Frontiero v. Richardson, 411 U.S. 677, 692 (1973) (Powell, J., concurring) (stating that the Court should avoid deciding whether sex is a suspect classification while the ERA is pending in state legislatures).
n98. See Strauss, supra note 86.
n99. Professor Strauss makes this further claim in rejecting my argument that the Constitution should be read as a whole. See id. at 1466-67 & n.22 (criticizing a claim I made in Dorf, Integrating Constitutional Theory, supra note 84, at 1778-79).
n100. See H.L.A. Hart, The Concept of Law v (2d ed. 1994).
n101. See id. (describing his method as "descriptive sociology").
n102. See Barry Friedman, Modeling Judicial Review (forthcoming) (on file with the Virginia Law Review Association).
n103. See Michael C. Dorf, Prediction and the Rule of Law, 42 UCLA L. Rev. 651, 685-89 (1995); Hart, supra note 100, at 138-44.
n104. Professor Mark Tushnet is an instructive example. When explaining how he would interpret the Constitution if he were on the Supreme Court, Professor Tushnet said he would vote so as "to advance the cause of socialism" and then write an opinion "in some currently favored version of Grand Theory." Mark Tushnet, The Dilemmas of Liberal Constitutionalism, 42 Ohio St. L.J. 411, 424 (1981). More recently, he has advocated making the Constitution nonjusticiable. See Tushnet, supra note 27.
n105. Professor Strauss's views on these questions are unclear. On the one hand, he criticizes scholars like Professor Amar and me for interpreting the Equal Protection Clause in light of the Nineteenth Amendment,
adding that "the fact that women's suffrage was formally recognized by the Nineteenth Amendment
... should not carry great weight." See Strauss, supra note 86, at 1466-67 & n.22. At the same time, however, he assumes that there is such a thing as a "correct" interpretation of the Fourteenth Amendment, implying that sometimes judges are supposed to interpret amendments rather than treat them as irrelevancies. Id.
n106. See generally Jed Rubenfeld, Freedom and Time: A Theory of Constitutional Self-Government (2001) (placing the Constitution's written status at the center of an account in which only fundamental commitments are enacted).
n107. 494 U.S. 872 (1990).
n108. See id. at 878-79.
n109. See, e.g., Dorf, supra note 36, at 1210-19; Douglas Laycock, The Remnants of Free Exercise, 1990 Sup. Ct. Rev. 1, 1-2; Michael W. McConnell, Free Exercise Revisionism and the Smith Decision, 57 U. Chi. L. Rev. 1109, 1111 (1990).
n110. See Christopher L. Eisgruber & Lawrence G. Sager, The Vulnerability of Conscience: The Constitutional Basis for Protecting Religious Conduct, 61 U. Chi. L. Rev. 1245, 1247-48 (1994).
n111. See City of Boerne v. Flores, 521 U.S. 507, 536 (1997).
n112. Compare Mitchell v. Helms, 530 U.S. 793, 837-44 (2000) (O'Connor, J., concurring in the judgment) (applying a multi-factor test that treats formal neutrality as a factor in determining whether a no-aid principle is violated), and id. at 867-913 (Souter, J., dissenting) (applying the same test, but reaching a different result), with id. at 809-10 (plurality opinion) (treating formal neutrality as dispositive). By joining the majority opinion in Zelman v. Simmons-Harris, 122 S. Ct. 2460 (2002), Justice O'Connor appeared to come closer than she had in Mitchell to the view that the formal neutrality of a program of government aid insulates it against challenges under the Establishment Clause.
n113. See supra note 36.
n114. See Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 83 (2000).
n115. See supra note 80.
n116. U.S. Const. art. I, 2, cl. 2.
n117. U.S. Const. art. I, 3, cl. 3.
n118. U.S. Const. art. II, 1, cl. 5.
n119. Although superseded by the Fifteenth Amendment, Section Two of the Fourteenth Amendment set a fifth age minimum (of twenty-one) for purposes of calculating the diminution in a state's representation in the national government that would result from disenfranchising male citizens. See U.S. Const. amend. XIV, 2.
n120. See William N. Eskridge, Jr. & John Ferejohn, Super-Statutes, 50 Duke L.J. 1215 (2001).
n121. Id. at 1216.
n123. See id. at 1237-1242 (discussing the Civil Rights Act of 1964, Pub. L. No. 88-352, 78 Stat. 241 (codified as amended in scattered sections of 42 U.S.C.)).
n124. 29 U.S.C. 621-634(a) (1994). As further evidence of the national consensus, one could point to the fact that nearly every state has enacted similar legislation. See Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 91 n. (listing various state age discrimination statutes).
n125. 29 U.S.C. 623(a)(1) (1994).
n126. 29 U.S.C. 621(a)(1).
n127. 29 U.S.C.S. 631(a) (2002).
n128. See 29 C.F.R. 1625.2 (2001).
n129. Hamilton v. Caterpillar, Inc., 966 F.2d 1226, 1227 (7th Cir. 1992). See also Dittman v. Gen. Motors Corp.-Delco Chassis Div., 941 F. Supp. 284, 287 (D. Conn. 1996), aff'd 116 F.3d 465 (2d Cir. 1997); Stone v. Travelers Corp., 58 F.3d 434, 437 (9th Cir. 1995) (affirming, on other grounds, a district court decision that the ADEA does not bar discrimination in favor of the old). The exception is Cline v. General Dynamics Land Systems, Inc., 296 F.3d 466 (6th Cir. Jul. 22, 2002), which relied on the plain language of the ADEA to permit a reverse age discrimination case to go forward.
n130. Kimel, 528 U.S. at 83 (citing Massachusetts Board of Retirement v. Murgia, 427 U.S. 307, 313-14 (1976)).
n131. See generally Richard A. Epstein, Forbidden Grounds: The Case Against Employment Discrimination Laws 445-46 (1992) (arguing that antidiscrimination laws are a "major social and intellectual mistake" because they erect unnecessary barriers in the labor marketplace).
n132. For a balanced rendition of both sides' arguments, see The Generational Equity Debate (John B. Williamson et al. eds., 1999).
n133. Such a need is apparent in the case of voter qualifications, the subject of the Twenty-Sixth Amendment itself. There is no such need for members of the House and Senate and the President, because voters could be entrusted with the task of rejecting candidates they thought too young or immature. Is it an embarrassment that specific constitutional provisions contravene the test I would erect for other government action? Perhaps, but that objection is hardly fatal. After all, current doctrine requires close judicial scrutiny of some distinctions between naturalized and natural-born citizens, even though the Constitution expressly approves such a distinction in its requirement that the President be a natural-born citizen. Arbitrary distinctions in the Constitution's text may have to be swallowed, but they need not serve as a basis for arbitrary generalizations. The tricky part is knowing when to cabin and when to generalize. See generally John Hart Ely, Interclausal Immunity, 87 Va. L. Rev. 1185 (2001) (exploring the questions of whether and when the mention of a practice in one clause of the Constitution immunizes that practice from invalidation under other clauses of the Constitution).
n134. See U.S. Const. art. I, 9, cl. 5 (prohibiting state export taxes); U.S. Const. art. I, 9, cl. 6 (prohibiting discrimination against out-of-state ships); U.S. Const. art. I, 10, cl. 2 (limiting state authority to collect duties on imports or exports to those "absolutely necessary" to enforcing inspection laws and those authorized by Congress); U.S. Const. art. IV, 1 (Full Faith and Credit Clause); U.S. Const. art. IV, 2 (Privileges and Immunities Clause). In terms of doctrinal importance, the most substantial such provision is the Dormant Commerce Clause, which is not expressly enumerated, but an inference from the grant of power to Congress in Article I, Section 8, Clause 3.
n135. U.S. Const. amend. XXIV.
n136. Cf. Frank I. Michelman, The Supreme Court, 1968 Term - Foreword: On Protecting the Poor Through the Fourteenth Amendment, 83 Harv. L. Rev. 7 (1969) (reading the Court's decisions protecting the poor in their access to various government services, opportunities, and benefits as vindicating a duty to ameliorate some of the hazards of poverty).
n137. J.M. Balkin, Agreements With Hell and Other Objects of Our Faith, 65 Fordham L. Rev. 1703, 1733 (1997); see also William E. Forbath, Constitutional Welfare Rights: A History, Critique and Reconstruction, 69 Fordham L. Rev. 1821 (2001) (explaining that the absence of welfare rights on the constitutional agenda is a product of contingent historical circumstances).
n138. See Boddie v. Connecticut, 401 U.S. 371, 382-83 (1971) (mandating affordable access to courts for divorce proceedings); Griffin v. Illinois, 351 U.S. 12, 18-19 (1956) (requiring states to provide criminal trial transcripts to defendants for appeal).
n139. 409 U.S. 434 (1973).
n140. 411 U.S. 1 (1973).
n141. However, in Harper v. Virginia Board of Elections, 383 U.S. 663, 666 (1966), the Court, relying on the Equal Protection Clause, held poll taxes in state elections unconstitutional. Although the majority opinion in Harper did not cite the Twenty-Fourth Amendment, it is plausible to surmise that, at least in the limited domain of election law, the Court was reading the Fourteenth Amendment in light of the Twenty-Fourth. Cf. id. at 685-86 (Harlan, J., dissenting) (citing the "fact that Congress and three-quarters of the States quickly ratified the Twenty-Fourth Amendment" as evidence of widespread disapproval of poll taxes, which the dissenters nonetheless believed should be left as a constitutional option for states that chose them).
n142. See Strauss, supra note 86, at 1481-82.
n143. This closely parallels current doctrine, leaving off the list only national origin and alienage, which is sometimes said to be suspect. Discrimination on these grounds could be readily understood as falling in the categories of "race" or "color." See infra text accompanying notes 155-167.
n144. See Duncan v. Louisiana, 391 U.S. 145, 171 (1968) (Black, J., concurring) (grudgingly accepting selective incorporation as preferable to open-ended interpretation of due process because the latter permits judges to roam "at will in their own notions of what policies outside the Bill of Rights are desirable and what are not").
n145. See Troxel v. Granville, 530 U.S. 57, 92 (2000) (Scalia, J., dissenting) ("I do not believe that the power which the Constitution confers upon me as a judge entitles me to deny legal effect to laws that (in my view) infringe upon what is (in my view) that unenumerated right."); City of Chicago v. Morales, 527 U.S. 41, 85 (1999) (Scalia, J., dissenting) ("The entire practice of using the Due Process Clause to add judicially favored rights to the limitations upon democracy set forth in the Bill of Rights (usually under the rubric of so-called "substantive due process') is in my view judicial usurpation."); Bork, supra note 51, at 236.
n146. See Cohen v. California, 403 U.S. 15, 27 (1971) (Black, J., dissenting).
n147. See supra note 145.
n148. See Boy Scouts of Am. v. Dale, 530 U.S. 640, 647-48 (2000); Cal. Democratic Party v. Jones, 530 U.S. 567, 574-75 (2000).
n149. 388 U.S. 1, 11 (1967).
n150. See, e.g., Janet E. Halley, The Politics of the Closet: Towards Equal Protection for Gay, Lesbian, and Bisexual Identity, 36 UCLA L. Rev. 915, 915-23 (1989); Andrew Koppelman, Why Discrimination Against Lesbians and Gay Men is Sex Discrimination, 69 N.Y.U. L. Rev. 197, 201 (1994); Cass R. Sunstein, Homosexuality and the Constitution, 70 Ind. L.J. 1, 18-23 (1994). But see George W. Dent, Jr., The Defense of Traditional Marriage, 15 J.L. & Pol. 581, 608-15 (1999); Lynn D. Wardle, A Critical Analysis of Constitutional Claims for Same-Sex Marriage, 1996 BYU L. Rev. 1, 83-88.
n151. Strauss notes that the Fourteenth and Fifteenth Amendments were almost completely ignored within a generation of their adoption. See Strauss, supra note 86, at 1463, 1482-84. I do not understand him to be applauding this phenomenon, however, suggesting that, once again, his insights are useful primarily to one who takes an external perspective.
n152. See supra text accompanying note 97.
n153. This last proposition is now almost universally accepted, although it was controversial at the founding. See Calder v. Bull, 3 U.S. (1 Dall.) 386 (1798). The leading dissenter among current constitutional scholars is Professor Ackerman, whose theory of "constitutional moments" allows that the People, acting through mechanisms other than the procedure set forth in Article V, can informally amend the Constitution. See Bruce A. Ackerman, 1 We the People: Foundations 44 (1991). Among the problems with Ackerman's theory is the difficulty of reading the implicit subtext of a constitutional moment that produces no explicit text. See Dorf, Integrating Constitutional Theory, supra note 84, at 1781-83 (arguing that Ackerman's approach is indeterminate).
n154. Poe v. Ullman, 367 U.S. 497, 543 (1961) (Harlan, J., dissenting). Justice Harlan wrote:
This "liberty" [protected by the Due Process Clause of the Fourteenth Amendment] is not a series of isolated points pricked out in terms of the taking of property; the freedom of speech, press, and religion; the right to keep and bear arms; the freedom from unreasonable searches and seizures; and so on. It is a rational continuum which, broadly speaking, includes a freedom from all substantial arbitrary impositions and purposeless restraints ... and which also recognizes, what a reasonable and sensitive judgment must, that certain interests require particularly careful scrutiny of the state needs asserted to justify their abridgment.
n155. Constitutional doctrine sometimes refers to this cluster of characteristics under the heading of "national origin." See, e.g., United States v. Virginia, 518 U.S. 515, 532 n.6 (1996) (observing that the "most stringent judicial scrutiny [applies to] classifications based on race or national origin"). National origin can refer to the nationality of a current citizen's ancestors, in which case it is equivalent to "ancestry or ethnicity" as I use that unitary term in the text. National origin, however, might also be taken to refer to the nationality, in the sense of citizenship, of an alien. As Professor Neuman explains:
Distinctions in federal law among aliens on the basis of their country of current nationality are not constitutionally suspect. Bilateral and multilateral treaties frequently create reciprocal privileges for U.S. citizens and citizens of selected foreign countries, and some federal legislation extends specific favored treatment to particular nationalities independent of treaties. If these distinctions are not defined in terms of race and are not motivated by racial prejudice - unlike the notorious Chinese exclusion laws, which had both of these characteristics - then they would not elicit heightened scrutiny under ordinary equal protection analysis.
Gerald L. Neuman, Terrorism, Selective Deportation and the First Amendment After Reno v. AADC, 14 Geo. Immigr. L.J. 313, 339-40 (2000). I agree with Professor Neuman that U.S. law may confer benefits on aliens of some, but not all, nationalities in exchange for reciprocal benefits for U.S. citizens abroad. To the extent, however, that he distinguishes between racial prejudice and national prejudice, I question his conclusion. Imagine a law that expressly relies on stereotypes of drunken Russians, cheap Scots, or stupid Poles. I would read our commitments to equality as inconsistent with national stereotypes even if they are not racial stereotypes. Or perhaps this is another way of saying, as I conclude below in the text, that national stereotypes should not be distinguished from racial stereotypes. This line of inquiry may have implications for the federal government's decision, following the events of September 11, 2001, to target Middle Eastern men for investigation, although those implications would depend, inter alia, on what counts as stereotyping and on whether the Constitution forbids selective enforcement in the deportation context. See Reno v. American-Arab Antidiscrimination Comm., 525 U.S. 471, 488 (1999) (stating that generally, "an alien unlawfully in this country has no constitutional right to assert selective enforcement as a defense against his deportation").
n156. 481 U.S. 604 (1987).
n157. See id. at 609 (citing Runyon v. McCrary, 427 U.S. 160, 168, 174-75 (1976)).
n158. Saint Francis College, 481 U.S. at 609-10.
n159. See id. at 611-12.
n160. Id. at 613; accord Rice v. Cayetano, 528 U.S. 495, 515 (2000).
n161. See also Shaare Tefila Congregation v. Cobb, 481 U.S. 615, 617 (1987) (relying on Saint Francis College to find that discrimination against Jews is race discrimination under 42 U.S.C. 1982).
n162. See Reginald Horsman, Race and Manifest Destiny: The Origins of American Racial Anglo-Saxonism 1-6 (1981). Racism was hardly confined to apologists, however. For example, Scottish phrenologist George Combe thought slavery could safely be abolished because of the "scientific" evidence that black Africans were naturally docile. See id. at 144.
n163. See John Higham, Strangers in the Land: Patterns of American Nativism 1860-1925, at 131-33 (2d ed. 1988).
n164. See id. at 165-75.
n165. Nagle v. Loi Hoa, 275 U.S. 475, 477-78 (1928).
n166. Morrison v. California, 291 U.S. 82, 85 (1934).
n167. Hirabayashi v. United States, 320 U.S. 81, 111 (1943) (Murphy, J., concurring) (remarking on the "melancholy resemblance" between Nazi treatment of "members of the Jewish race" and a curfew ordered by the U.S. military applicable to all persons of Japanese descent on the Pacific coast, but nonetheless joining the Court's unanimous decision sustaining the curfew order).
n168. See, e.g., Levy v. Louisiana, 391 U.S. 68 (1968) (nominally applying rational basis scrutiny to invalidate a state statute that did not permit illegitimate children to sue for the wrongful death of their parents); see also Laurence H. Tribe, American Constitutional Law 1553-58 (2d ed. 1988) (describing the uncertain application of covert heightened scrutiny to illegitimacy classifications).
n169. See King v. Smith, 392 U.S. 309, 336 n.5 (1968) (Douglas, J., concurring) ("Penalizing the children for the sins of their mother is reminiscent of the archaic corruption of the blood, a form of bill of attainder ... .").
n170. U.S. Const. amend. XXV, 3.
n171. 42 U.S.C. 12,101(a)(2) (1994).
n172. See Bd. of Trs. of the Univ. of Ala. v. Garrett, 531 U.S. 356, 368 n.5 (2001).
n173. See 42 U.S.C. 12,112(b)(5)(A).
n174. See id. 12,111(9).
n175. See Vill. of Willowbrook v. Olech, 528 U.S. 562, 564 (2000); Romer v. Evans, 517 U.S. 620, 632 (1996); City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 450 (1985); U.S. Dep't of Agric. v. Moreno, 413 U.S. 528, 534 (1973).
n176. Cf. Kenji Yoshino, Covering, 111 Yale L.J. 769 (2002) (exploring the uses and limits of analogies from race and sex to sexual orientation).
n177. See Fiss, supra note 34, at 157 (arguing that the Equal Protection Clause prohibits a "state law or practice [that] aggravates (or perpetuates?) the subordinate position of a specially disadvantaged group").
n178. See Brest, supra note 21, at 1 ("The antidiscrimination principle lies at the core of most state and federal civil rights legislation ... ."). Even the Americans with Disabilities Act of 1990, Pub. L. No. 101-336, 104 Stat. 327 (codified as amended at 42 U.S.C. 12,101-12,213 (1994) and in scattered sections of 47 U.S.C.), which requires employers and others to make reasonable accommodations to disabilities, expresses its core norm as an antidiscrimination principle (albeit an asymmetrical one). See 42 U.S.C. 12112(a) (1994) ("No covered entity shall discriminate against a qualified individual with a disability because of the disability of such individual in regard to job application procedures, the hiring, advancement, or discharge of employees, employee compensation, job training, and other terms, conditions, and privileges of employment."). See generally Samuel Issacharoff & Justin Nelson, Discrimination With a Difference: Can Employment Discrimination Law Accommodate the Americans With Disabilities Act?, 79 N.C. L. Rev. 307 (2001) (using the ADA to illustrate the redistributive function of antidiscrimination law).
n179. See Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 226 (1995); City of Richmond v. J. A. Croson Co., 488 U.S. 469, 493-94 (1989).
n180. See Washington v. Davis, 426 U.S. 229, 248 (1976).
n181. 42 U.S.C. 2000e-2(a)(1) (1994).
n182. See id. 2000e-2(d) ("It shall be an unlawful employment practice for any employer, labor organization, or joint labor-management committee controlling apprenticeship or other training or retraining, including on-the-job training programs to discriminate against any individual because of his race, color, religion, sex, or national origin in admission to, or employment in, any program established to provide apprenticeship or other training.").
n183. 443 U.S. 193 (1979).
n184. See id. at 197. The particular program at issue in Weber was jointly administered by an employer and a union. See id.
n185. See id. at 216 (Burger, C.J., dissenting); id. at 220-21 (Rehnquist, J., dissenting).
n186. See id. at 202-08.
n187. See, e.g., Ronald Dworkin, A Matter Of Principle 316-31 (1985); William N. Eskridge, Jr., Dynamic Statutory Interpretation 14-47, 80, 135, 173, 303-06 (1994); Philip P. Frickey, From the Big Sleep to the Big Heat: The Revival of Theory in Statutory Interpretation, 77 Minn. L. Rev. 241, 245-60 (1992) (describing Weber as the catalyst for theories, like those of Dworkin and Eskridge, that defend judicial inquiry into Congress's purpose in enacting a statute, as well as for public choice theory, which attacks such inquiry).
n188. For example, Justice Scalia, concurring in Adarand Constructors, Inc. v. Pena, 515 U.S. 200 (1995), laid emphasis on the fact that the Equal Protection Clause applies to "any person." Id. at 239 (Scalia, J., concurring in part and concurring in the judgment) (citations omitted).
n189. See Weber, 443 U.S. at 201 (citing Church of the Holy Trinity v. United States, 143 U.S. 457, 459 (1892)).
n190. See Dworkin, supra note 187, at 318 (1985) (explaining how the term "discriminate" can be used neutrally - connoting any racial classification - or normatively - connoting invidious classifications); William N. Eskridge, Jr., Dynamic Statutory Interpretation, 135 U. Pa. L. Rev. 1479, 1489-90 (1987) (same).
n191. 42 U.S.C. 2000d (1994).
n192. 532 U.S. 275 (2001).
n193. 28 C.F.R. 42.104(b)(2) (2001).
n194. See Sandoval, 532 U.S. at 283-92.
n195. Id. at 280.
n196. Id. at 307 (Stevens, J., dissenting).
n197. See Brest, supra note 21, at 19. But see Laurence H. Tribe, The Mystery of Motive, Private and Public: Some Notes Inspired by the Problems of Hate Crime and Animal Sacrifice, 1993 Sup. Ct. Rev 1 (1994) (arguing that the motive underlying private or public conduct is often a legitimate factor for consideration in assessing how to treat the conduct).
n198. 426 U.S. 229 (1976).
n199. See id. at 235.
n200. For a discussion of this kind of standing inquiry in equal protection cases, see Michael C. Dorf, Facial Challenges to State and Federal Statutes, 46 Stan. L. Rev. 235, 251-61 (1994); Michael C. Dorf, The Heterogeneity of Rights, 6 Legal Theory 269, 278 (2000).
n201. See I.R.C. 163(h)(3) (2001) (permitting deduction for interest on a home mortgage).
n202. I have heard claims that this is a real phenomenon, although I have been unable to discover any reliable empirical study.
n203. Fiss, supra note 34, at 129.
n204. Peter J. Rubin, Reconnecting Doctrine and Purpose: A Comprehensive Approach to Strict Scrutiny after Adarand and Shaw, 149 U. Pa. L. Rev. 1, 19 (2000).
N205. See Alexander M. Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of Politics 16 (1962).
n206. For example, Barry Friedman argues that concerns about the countermajoritarian nature of judicial review are an (unhealthy) academic obsession of the last fifty years. See Barry S. Friedman, The Birth of an Academic Obsession, Yale L.J. (forthcoming 2002).
n207. Indeed, one might even understand Congressional action pursuant to Section Five of the Fourteenth Amendment as satisfying a duty to implement constitutional norms that courts are ill-situated to implement themselves. See Lawrence Gene Sager, Fair Measure: The Legal Status of Underenforced Constitutional Norms, 91 Harv. L. Rev. 1212, 1213 (1978); Lawrence G. Sager, Justice in Plain Clothes: Reflections on the Thinness of Constitutional Law, 88 Nw. U. L. Rev. 410, 419 (1993).
n208. See City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 435 (1985) (mental retardation as a species of disability); Mass. Bd. of Ret. v. Murgia, 427 U.S. 307, 313-14 (1976) (per curiam) (age).
n209. 531 U.S. 356, 377 (2001) (Breyer, J., dissenting).
n210. 521 U.S. 507, 520 (1997).
n211. See Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 82-83 (2000); Bd. of Trs. of the Univ. of Ala. v. Garrett, 531 U.S. 356, 357-58 (2001).
n212. 514 U.S. 549 (1995).
n213. Even if an employment relationship is purely intrastate, it is nonetheless economic activity that substantially affects interstate commerce. See United States v. Morrison, 529 U.S. 598, 610 (2000) (distinguishing economic from non-economic activity affecting interstate commerce).
n214. See Seminole Tribe of Florida v. Florida, 517 U.S. 44 (1996).
n215. 209 U.S. 123 (1908).
n216. To be more precise, City of Boerne invalidated RFRA in all of its applications to state actors. Although some commentators take the view that RFRA was held unconstitutional in toto - see, for example, Edward J.W. Blatnik, Note, No RFRAF Allowed: The Status of the Religious Freedom Restoration Act's Federal Application in the Wake of City of Boerne v. Flores, 98 Colum. L. Rev. 1410 (1998); and Marci A. Hamilton, City of Boerne v. Flores: A Landmark for Structural Analysis, 39 Wm. & Mary L. Rev. 699, 718 (1998) - it is at least an open question whether RFRA is valid as applied to the federal government. In my view, it is valid. "As a restraint on the federal government, RFRA ... "is both a rule of interpretation for future [and past] federal legislation and an exercise of general legislative supervision over federal agencies, enacted pursuant to each of the federal powers that gives rise to legislation or agencies in the first place.'" Dorf, supra note 36, at 1212 n.162 (quoting Douglas Laycock & Oliver S. Thomas, Interpreting the Religious Freedom Restoration Act, 73 Tex. L. Rev. 209, 211 (1994)). Accordingly, if RFRA is invalid as applied to the federal government, that must be because of separation of powers concerns. Although there is language in Boerne suggesting that RFRA trenched on the judicial province of constitutional adjudication, it is difficult to understand why a statute that would otherwise be within Congress's power to enact would be beyond its authority simply because the Supreme Court happened to rule, in an earlier case, that what the statute aims to accomplish is not independently required as a matter of constitutional law.
n217. 529 U.S. 598 (2000).
n218. See id. at 617.
n219. See id. at 627.
n220. See Seminole Tribe of Florida v. Florida, 517 U.S. 44, 75-100 (1996) (Stevens, J., dissenting); id. at 100-85 (Souter, J., dissenting).
n221. See Jesse H. Choper, Judicial Review and the National Political Process 171-259 (1980); Larry D. Kramer, Putting the Politics Back Into the Political Safeguards of Federalism, 100 Colum. L. Rev. 215, 287-93 (2000); Herbert Wechsler, The Political Safeguards of Federalism: The Role of the States in the Composition and Selection of the National Government, 54 Colum. L. Rev. 543, 559 (1954). A majority of the Court accepted this proposition for some purposes in Garcia v. San Antonio Metropolitan Transit Authority, 469 U.S. 528 (1985), but the Court's recent cases have clearly backed away from it.
n222. A student Note succinctly collects evidence for this position. See Steven A. Engel, Note, The McCulloch Theory of the Fourteenth Amendment: City of Boerne v. Flores and the Original Understanding of Section 5, 109 Yale L.J. 115 (1999).
n223. See Seminole Tribe of Florida v. Florida, 517 U.S. 44, 57-73 (1996).
n224. See United States v. Lopez, 514 U.S. 549, 566 (1995) ("The Constitution ... withholds from Congress a plenary police power.").
n225. See United States v. Morrison, 529 U.S. 598, 619 (2000) ("As broad as the congressional enforcement power is, it is not unlimited.").
n226. Michael C. Dorf & Barry Friedman, Shared Constitutional Interpretation, 2000 Sup. Ct. Rev. 61, 89.
n227. Robert C. Post & Reva B. Siegel,
Equal Protection by Law: Federal Antidiscrimination Legislation After Morrison and Kimel, 110 Yale L.J. 441, 508 (2000).
n228. U.S. Const. amend. IX.
n229. See Nomination of Robert H. Bork to be Associate Justice of the Supreme Court of the United States: Hearings before the Senate Comm. on the Judiciary, 100th Cong. 249 (1989) (testimony of Robert Bork).