Historical Development of Voting Rights in the US: An Unenumerated Fundamental Right, Study notes of Reasoning

Download Historical Development of Voting Rights in the US: An Unenumerated Fundamental Right and more Study notes Reasoning in PDF only on Docsity! UNENUMERATED DEMOCRACY: LESSONS FROM THE RIGHT TO VOTE Jane S. SchacteW I would venture to guess that, if most constitutional law types were asked to free associate the first case they think of when hearing the term unenumerated rights, Roe or Lochner would be blurted out the most frequently. Or maybe Griswold. But probably not cases like Reynolds v. Sims' or Harper v. Virginia Board of Elections, which recognized an equal right to vote in the absence of explicit constitutional language about voting. Although Reynolds and Harper were controversial deci- sions in their own time and still occasionally kick up a little dust, they have not been nearly as hotly disputed or as publicly controversial as, say, the modern privacy cases on procreative autonomy. In this Essay, I suggest that straying from more familiar terrain and examining the right to vote through the lens of the unenumer- ated-rights debate can generate some valuable insights about that de- bate. In particular, I stress two points: First, looking at the right to vote from this vantage point suggests the sharp limitations of enu- meration as a guiding concept in constitutional law. Enumeration as a norm is plagued by significant uncertainties, including what is ar- guably the crucial question: What counts as enumeration? That is, just how specific does textual enumeration need to be to satisfy the requirement? I will suggest that "enumerationism" itself cannot an- swer this key question, and that it therefore does not-and cannot- do the conceptual heavy lifting on its own. In fact, the interpreter must rely on values extrinsic to enumeration itself in order to imple- ment and give meaning to the enumeration norm. And the selection of the relevant extrinsic values will-no surprise-end up giving the interpreter the very kind of discretion that enumeration itself aspires to cut off. The second point I will explore relates to the normative justifica- tion for an enumeration requirement. Restricting constitutional rights to those with a textual basis is conventionally defended as pro- moting democracy by leaving more questions to the political process. Professor of Law, Stanford Law School. Thanks to Pam Karlan and Kathleen Sullivan for helpful comments on an early draft and to Meredith Nikkel for excellent research assistance. 377 U.S. 533 (1964). 383 U.S. 663 (1966). JOURNAL OF CONSTITUTIONAL LAW Democratic ideals are thus likely to be among the extrinsic values that shape the working contours of enumeration. But democracy turns out to be a problematic justification for enumerationism. It is para- doxical, I will argue, to invoke democracy to object to a right that, like the right to vote, is claimed to be precisely necessary for democ- racy itself. I will argue, moreover, that just as the meaning of enu- meration is contestable, so is the meaning of democracy itself. All of this suggests, in turn, that there is considerable give on both sides of the familiar, if crude, equation: enumeration = democracy. I. THE EVOLUTION OF AN EQUAL PROTECTION BASED RIGHT TO VOTE: FROM HAPPERSETTTO HARPER A. Constitutional Text The constitutional text on voting is sparse, but is practically boun- tiful in the contemporary Constitution as compared to the original document. The original Constitution provided no explicit right to vote in state elections. The Constitution said more, but still relatively little, about federal elections. Under Article I, Section 3, the Senate was originally to be chosen by state legislatures, not voters, so there was clearly no voting right implicated there.3 Article II, Section 1, similarly affords no basis for finding a right to vote for the president because it provides for the Electoral College and for each state legis- lature to determine how that state's electors are selected.4 That leaves the House of Representatives, which, under Article I, Section 2, is to be "composed of Members chosen every second Year by the People of the several States."5 Language requiring representatives to be "chosen... by the People" contemplates a vote, and so might be seen as conferring an implied right to vote, but is notably silent both on who is to be included in the "People" and on how elections are otherwise to be run. On this point, the Section says only that the qualifications for "the Electors" of United States House members are to be the same as those for the "Electors of the most numerous Branch of the State Legislature. 6 The text of Article I thus crucially relies on state law to determine who is permitted to vote in House elections. Finally, Article I, Section 4 states that state legislatures shall 3 U.S. CONST. art. I, § 3, cl. 1 ("The Senate of the United States shall be composed of two Senators from each State, chosen by the Legislature thereof. . . ."), amended by U.S. CONST. amend. XVII, § 1. 4 Id. art. II, § 1, cl. 2 ("Each State shall appoint, in such Manner as the Legislature thereof may direct, a Number of Electors, equal to the whole Number of Senators and Representatives to which the State may be entitled in the Congress... "). 5 Id. art. I, § 2, cl. 1. 6 Id. [Vol. 9:2 UNENUMERA TED DEMOCRACY built upon three earlier cases. The first was Baker v. Carr, which re- nounced the rule that malapportionment claims were nonjusticiable political questions. 2 The next was Gray v. Sanders, where the Court rejected a "county unit" system that aggregated votes on a per-county basis.2 That system diluted the power of populous urban counties and enhanced the power of their more sparsely populated rural counterparts. Invoking the Equal Protection Clause, the Court said in Gray that "all who participate in the election are to have an equal vote."2 Notably, however, Justice Douglas's opinion in Gray did not restrict itself to equal protection. Perhaps foreshadowing the pen- umbral approach that he would soon famously advance in Griswold v. Connecticut,14 Douglas instead opted for something that, if properly called enumeration at all, was more like panenumeration: "[t]he conception of political equality from the Declaration of Independ- ence, to Lincoln's Gettysburg Address, to the Fifteenth, Seventeenth, and Nineteenth Amendments can mean only one thing-one person, one vote."25 Finally, in Wesberry v. Sanders, the Court struck down malapportioned congressional districts and sounded a similar theme about equality.2 6 This time, however, the Court read into the text of Article I, Section 2, a strong-if textually questionable-equality principle.27 Reynolds followed closely on the heels of these three cases and im- posed a one-person-one-vote requirement on state elections through the Equal Protection Clause. In doing so, the Court used robust rhetoric about the right to vote, casting it as central based on its abil- ity to protect other rights: A predominant consideration in determining whether a State's legislative apportionment scheme constitutes an invidious discrimination violative of rights asserted under the Equal Protection Clause is that the rights al- legedly impaired are individual and personal in nature.... Undoubtedly, the right of suffrage is a fundamental matter in a free and democratic so- ciety. Especially since the right to exercise the franchise in a free and un- impaired manner is preservative of other basic civil and political rights, any alleged infringement of the right of citizens to vote must be carefully and meticulously scrutinized.2 s 2 369 U.S. 186, 231 (1962). 372 U.S. 368, 380 (1963). 23 Id. at 379. 24 381 U.S. 479, 484 (1965) (asserting that "specific guarantees in the Bill of Rights have pe- numbras, formed by emanations from those guarantees that help give them life and sub- stance"). 25 Gray, 372 U.S. at 381. 26 376 U.S. 1, 7-8 (1964). 27 Article I, Section 2, Clause 3 refers to apportionment "among the several States," not within individual states. See Wesbeny, 376 U.S. at 25-45 (Harlan,J., dissenting). 28 Reynolds v. Sims, 377 U.S. 533, 561-62 (1964). Jan. 2007] JOURNAL OF CONSTITUTIONAL LA W The strong language characterizing the right to vote as preservative of other rights was derived from dicta in the Court's decision in Yick Wo v. Hopkins,2 another equal protection case, and has become one of Reynolds's signature phrases. Rhetoric like this, focusing as it does on the instrumental central- ity of voting to politics, would suggest that Reynolds recognized an un- qualified constitutional right to vote. But in terms of both its holding and its justification, the case was about voting equality, not voting per se. The plaintiffs' claim was one of voting dilution, not deprivation, so the focal point was voters' equality in relation to one another. Consider the Court's framing of the right: [E]ach and every citizen has an inalienable right to full and effective par- ticipation in the political processes of his State's legislative bodies. Most citizens can achieve this participation only as qualified voters through the election of legislators to represent them. Full and effective participation by all citizens in state government requires, therefore, that each citizen have an equally effective voice in the election of members of his state leg- islature. Modern and viable state government needs, and the Constitu- tion demands, no less.30 The emphasis on citizens having an "equally effective voice" in elec- tions, if utopian, signals that Reynolds was grounded in ideas of politi- cal equality, not unfettered political liberty. Where Reynolds concerned vote dilution, the Court's 1966 deci- sion in Harper striking down the poll tax in state elections more di- rectly implicated the right to vote itself. Those failing to pay the Vir- ginia tax were turned away at the polls. The Court began its analysis with the constitutional text, noting in passing and as if it had never been controversial that "the right to vote in federal elections is con- ferred by Art. I, § 2, of the Constitution."" More pertinent to the Virginia poll tax, the opinion proceeded to say that, while "the right to vote in state elections is nowhere expressly mentioned [in the Con- stitution]," it was nevertheless protected.3 2 The Court noted, but did not engage, the argument that the First Amendment might give rise to an implicit right to vote in state elections." It instead saw the case as crucially about inequality, saying that "once the franchise is granted to the electorate, lines may not be drawn which are inconsis- tent with the Equal Protection Clause."34 Justice Douglas distin- 118 U.S. 356, 370 (1886) ("Though not regarded strictly as a natural right, but as a privi- lege merely conceded by society according to its will, under certain conditions, nevertheless it is regarded as a fundamental political right, because preservative of all rights."). 30 Reynolds, 377 U.S. at 565. 31 Harper v. Va. Bd. of Elections, 383 U.S. 663, 665 (1966). 52 Id. 33 Id. 34 Id. [Vol. 9:2 UNENUMERA TED DEMOCRACY guished the decision in Lassiter v. Northhampton County Board of Elec- tions, decided seven years earlier, which had upheld the use of literacy tests in North Carolina.3 5 The opinion argued that literacy had some reasonable relation to the right to vote, but wealth did not.36 The Court in Harper thus characterized the case as one of "invidious dis- crimination." T Had enumeration been its central preoccupation, the Harper Court might have pursued a different doctrinal path. The year be- fore Harper was decided, the Supreme Court had decided Harman v. Forssenius, a case that struck down an attempt by Virginia to evade the Twenty-fourth Amendment's ban on poll taxes in federal elections. 38 Once it became clear that the Twenty-fourth Amendment would pass, Virginia abolished the poll tax as an absolute prerequisite for federal elections, while maintaining the tax for state elections. In response to the new federal amendment, Virginia enacted a law requiring vot- ers in federal contests either to pay the poll tax or to file proof of resi- dence six months before the election. ° The Court found the law to be a sophisticated means of evasion and struck it down as violative of the Twenty-fourth Amendment.4' In the course of its analysis, how- ever, the Court alluded to what might qualify as the mother of all smoking guns-a statement from the 1902 Virginia Constitutional Convention, at which the state first wrote a poll tax into its constitu- tion: Discrimination! Why, that is precisely what we propose; that, exactly, is what this Convention was elected for-to discriminate to the very extrem- ity of permissible action under the limitations of the Federal Constitu- tion, with a view to the elimination of every negro voter who can be got- ten rid of, legally, without materially impairing the numerical strength of the white electorate.42 Recall that when Virginia passed the law struck down in Harman, it retained the poll tax for state elections that was subsequently struck 35 360 U.S. 45, 53-54 (1959). For subsequent statutory elimination of literacy tests, see Vot- ing Rights Act of 1965, Pub. L. No. 89-110, § 4, 79 Stat. 437, 438 (codified as amended at 42 U.S.C. § 1973b (2000)). 36 Lassiter, 360 U.S. at 53-54. 37 Harper, 383 U.S. at 668 (internal quotation marks omitted). 38 380 U.S. 528, 544 (1965). 1 am indebted to Pam Karlan for a conversation about this case. :9 Id. at 531. 0 Id. at 532. 41 Id. at 544. 42 Id. at 543 (quoting 2 VIRGINIA CONSTITUTIONAL CONVENTION (PROCEEDINGS AND DEBATES, 1901-1902), at 3076-77 (statement of Rep. Glass)). The opinion went on to assert that "[t]his statement was characteristic of the entire debate on the suffrage issue; the only real controversy was whether the provisions eventually adopted were sufficient to accomplish the disenfranchisement of the Negro." Id. at 543 n.23 (citing 2 VIRGINIA CONSTITUTIONAL CONVENTION (PROCEEDINGS AND DEBATES, 1901-1902), at 2937-3080). Jan. 2007] JOURNAL OF CONSTITUTIONAL LAW Equal protection rhetoric aside, casting the right as "fundamen- tal" has proven to be a bit of an attractive nuisance for critics of un- enumerated rights because such critics often object to unenumerated and fundamental rights on the same grounds. Robert Bork and Lino Graglia, for example, criticize both unenumerated and fundamental rights as antidemocratic. s Their argument is the familiar one that charges judges with finding textually unsupported, bogus "rights" that have the effect of removing from the ordinary political process questions that should be decided by the electorate's chosen represen- tatives. Despite this common critique of unenumerated and funda- mental rights, however, the two categories are distinct in certain ways. Fundamentality is a very old idea in constitutional interpretation, 56 and it goes to importance, while enumeration goes to textual specific- ity. The perceived importance of a right may, of course, be relevant to a court's willingness to find it within constitutional text, or despite the absence of clear constitutional text. But the two adjectives never- theless describe different things. Moreover, not all fundamental rights are unenumerated. The Sixth Amendment, to name one of many examples, is said to create a "fundamental" right to a jury trial in federal criminal cases through its explicit text.57 And if it is in fact less objectionable to call a right "fundamental" if it is enumerated, then the two categories may mutu- ally shape and modify one another in interesting ways. The broader a view one takes about what constitutes enumeration, the smaller the set of assertedly illegitimate fundamental rights. That takes us to the key question: Is the Reynolds-Harper funda- mental voting right necessarily unenumerated? It is sometimes char- acterized in those terms,8 but one might well argue the contrary proposition. The voting right might, instead, be seen as enumerated because it falls within the broadly worded terms of, and the textually state policy choices to promote a set of values responsive to the Justices' vision of political and social ideals."); see alsoJames A. Gardner, Liberty, Community and the Constitutional Structure of Po- litical Influence: A Reconsideration of the Right to Vote, 145 U. PA. L. REv. 893, 972 (1997) (arguing that the Warren Court "could not feasibly rely on the doctrine of substantive due process" after Lochner); Karlan, supra note 52, at 479 (characterizing the Harper decision's use of equal protec- tion as "largely an artifact of the Warren Court's decision to avoid the then-discredited idea of substantive due process"). 5 See ROBERT H. BORK, THE TEMPTING OF AMERICA: THE POLITICAL SEDUCTION OF THE LAW 352-53 (1990) (criticizing interpretive theories that empower courts as impairing "the full right of self-government"); Lino A. Graglia, The Constitution and "Fundamental Rights", in THE FRAMERS AND FUNDAMENTAL RIGHTS 86, 97-101 (Robert A. Licht ed., 1991) (arguing that judicial activ- ism is usurping the established authority of majority rule). SeeMarburyv. Madison, 5 U.S. (1 Cranch) 137, 176 (1803) ("The principles, therefore, so established, are deemed fundamental."). 57 See, e.g., Duncan v. Louisiana, 391 U.S. 145, 157-58 (1968) (deeming the Sixth Amend- mentjury-trial right a "fundamental right"). 58 See supra note 51. (Vol. 9:2 0UNENUMERA TED DEMOCRACY unqualified equality norm created by, the Equal Protection Clause. This interpretive hypothesis would proceed from the fact that the enumerated text of the Clause fails to exclude voting laws from the class of laws subject to the equal protection requirement. Does this hypothesis reflect a sound understanding of enumera- tion? Confining our attention to constitutional text for the moment, there are various interpretive counterarguments that might defeat this suggestion that the Reynolds-Harper right should be seen as enu- merated. One objection might flow from the Representation Reduc- tion Clause in Section 2 of the Fourteenth Amendment, which spe- cifically provides a representational sanction for states that deny the vote to male citizens twenty-one or older, except in specified circum- stances. 9 Echoing Justice Harlan's dissent in Reynolds, Raoul Berger inferred that Section 2 was the only provision in the Fourteenth Amendment that was intended to deal with voting.60 Yet, as a textual matter alone, this argument is problematic for several reasons. First, the text of Section 2 is considerably less significant than it once might have been because subsequent amendments have made much of that Section obsolete by specifically barring race, sex, and age discrimina- tion in voting.6' Second, Section 2's representational sanction might reasonably be read to address only the denial of voting to males over twenty-one, and not other voting-related matters like reapportion- ment or the poll tax. Indeed, many scholars parsing the legislative history of the Fourteenth Amendment have concluded, contrary to 59 The Clause allows states to exclude only those male citizens over twenty-one who par- ticipated in crimes: But when the right to vote at any election for the choice of electors for President and Vice-President of the United States, Representatives in Congress, the Executive and Judi- cial officers of a State, or the members of the Legislature thereof, is denied to any of the male inhabitants of such State, being twenty-one years of age, and citizens of the United States, or in any way abridged, except for participation in rebellion, or other crime, the basis of representation therein shall be reduced .... U.S. CONST. amend. XIV, § 2, amended by U.S. CONST. amend. XXVI, § 1. On the Reduction of Representation Clause, see Pamela S. Karlan, Unduly Partial: The Supreme Court and the Fourteenth Amendment in Bush v. Gore, 29 FLA. ST. U. L. REv. 587, 589-93 (2001). 60 See RAOUL BERGER, GOVERNMENT BY JUDICIARY: THE TRANSFORMATION OF THE FOURTEENTH AMENDMENT 70-89 (1977) (arguing that Section 1 did not provide Congress with control over voting rights); see also Reynolds v. Sims, 377 U.S. 533, 594 (1964) (Harlan, J., dis- senting) ("The comprehensive scope of the second section and its particular reference to the state legislatures preclude the suggestion that the first section was intended to have the result reached by the Court today."); cf Richardson v. Ramirez, 418 U.S. 24, 54-55 (1974) (relying on Section 2's specific language about criminals to reject a challenge to felony disenfranchisement under Section 1). 61 See U.S. CONST. amend. XV, § 1 (race); id. amend. XIX, cl. 1 (sex); id. amend. XXVI, § 1 (age). For a particularly aggressive form of this argument, see Gabriel J. Chin, Reconstruction, Felon Disenfranchisement, and the Right to Vote: Did the Fifteenth Amendment Repeal Section 2 of the Fourteenth Amendment?, 92 GEO. L.J. 259, 291-92 (2004). One need not necessarily go as far as Chin does to say, more modestly, that later amendments have diminished the relevance of Sec- tion 2 to voting-rights jurisprudence. Jan. 2007] JOURNAL OF CONSTI'UTIONAL LAW Harlan's and Berger's. conclusions, that the legislative history is in- conclusive and does not necessarily support the idea that Section 2 bars reading Section 1 as an independent source of protection for voting rights.62 As Neil Komesar has pointed out, the fact that there may have been insufficient votes to include language protecting suf- frage does not mean that there were necessarily sufficient votes to ex- clude such protection.63 And, of course, no such exclusion does ap- pear in the text, so it seems at least plausible to conclude that the drafters intended to leave to later interpreters the meaning of Sec- tion 1 in relation to voting equality. Consider a different textual objection to viewing the Equal Protec- tion Clause as sufficient to establish the Reynolds-Harper right as enu- merated: its relative vagueness compared to the specificity of the Fif- teenth, Nineteenth, Twenty-fourth, and Twenty-sixth Amendments, which ban voting restrictions based on race, sex, poll taxes in federal elections, and age, respectively.64 The specificity of these provisions might be enlisted to support the negative inference that nothing else in the Constitution protects voting equality.65 This is a version of the specific-trumps-the-general notion in interpretation. But this reading also has its shortcomings. The words in the voting-rights Amend- ments can be sensibly parsed in the opposite direction. One might, indeed, take a textual tack and read the language in the voting-rights Amendments referring to the "right to vote" not being denied or abridged on the textually specified bases as evidence that some "right to vote," in fact, predated these Amendments. And, since each of these Amendments was enacted after the Equal Protection Clause, the general equality command in the Fourteenth Amendment might 62 The most extensive argument of this kind appears in William W. Van Alstyne, The Four- teenth Amendment, the "Right" to Vote, and the Understanding of the Thirty-ninth Congress, 1965 Sup. CF. REV. 33. For other analyses pointing in a similar direction, see Oregon v. Mitchell, 400 U.S. 112, 276-78 (1970) (plurality) (Brennan, White, and Marshall, JJ., dissenting in part and con- curring in part), WILLIAM E. NELSON, THE FOURTEENTH AMENDMENT: FROM POLITICAL PRINCIPLE TO JUDICIAL DOCTRINE 49-63 (1988), and Michael Kent Curtis, John A. Bingham and the Story of American Liberty: The Lost Cause Meets the "Lost Clause," 36 AKRON L. REV. 617, 644-45 (2003). Neil K. Komesar, Back to the Future-An Institutional View of Making and Interpreting Constitu- tions, 81 Nw. U. L. REV. 191, 204-08 (1987) (arguing that the Fourteenth Amendment may not have passed had its drafters made explicit either a right to vote or an exclusion of that right). This question is especially relevant in relation to the Twenty-fourth Amendment, banning poll taxes in federal elections, and Harper, which invalidated poll taxes in state elections. 65 See, e.g., Reynolds v. Sims, 377 U.S. 533, 611-12 (1964) (Harlan, J., dissenting) (arguing that the Fifteenth and Nineteenth Amendments would not have been necessary had the Four- teenth protected voting rights); BERGER, supra note 60, at 104-05 (same); cf Minor v. Happer- sett, 88 U.S. (21 Wall.) 162, 175 (1875) ("If suffrage was one of [the] privileges or immunities [protected by the Fourteenth Amendment], why amend the Constitution to prevent its being denied on account of race, &c.?"). [Vol. 9:2 UNENUMERA TED DEMOCRACY Moreover, there is a significant potential tension between enu- merationism and originalism. The two are often embraced by fellow travelers, but there may well be more friction than harmony between the two ideas. This is because enumeration of broad values like equality (or free speech, liberty, or republican government) might itself be understood to militate against a search for any specific origi- nal intent. The drafters' very choice to employ broad text would seem to support the idea that such clauses were not intended to be tightly tethered to any originally contemplated set of particular prob- lems. The character of the enumerated text, in other words, might well be seen to undercut the legitimacy of searching for the unenu- merated, specific, original intent. If originalism does not work, perhaps a better place to look to re- solve the level of specificity necessary to constitute enumeration is to the values driving the enumeration norm-that is, to the conceptual justification for insisting upon enumerated text in the first instance. There is a range of plausible answers to that question, but the most familiar and important one flows from ideas about democracy.72 The majoritarian would say that we should insist upon enumeration be- cause it leaves more decisions to the political process and fewer to the courts.73 The crude calculation is that the more textual specificity that is required before a court can find a right, the more democracy the polity gets. For several reasons, democracy alone won't work as the relevant rule on specificity. One reason is that it is too general to be of much help in determining the quantum of specificity necessary in particu- lar cases. Consider the voting example: With the normative pull of democracy firmly in mind, we still know that the Fourteenth Amendment contains a general equality norm that does not exclude voting, and we still don't know if that is sufficiently specific to satisfy the demands of enumeration. And, as I have suggested, there are simply too many generalized constitutional commands to support any categorical, democracy-driven requirement of exacting specificity. More importantly, perhaps, there is a deeper paradox here: the democratic case against unenumerated rights is tautological when applied to a right that is precisely claimed to be demanded by de- mocracy. How, in other words, can democracy be the grounds to deny the equal-voting right said to be vital to supporting democracy 72 Other obvious candidates include arguments relating to institutional competence and rule-of-law values like certainty, notice, and predictability. 73 See BERGER, supra note 60, at 85-90 (arguing that the legislative history of the Equal Pro- tection Clause favors leaving more decision to the political process); BORK, supra note 55, at 352-53 (arguing for "untidy" political responses to moral problems rather than courts' "abstract generalizations"); Graglia, supra note 55, at 97-101 (characterizing fundamental rights deci- sions as illegitimately wresting control from the political process). Jan. 2007] JOURNAL OF CONSTITUTIONAL LAW itself? Here we begin to see a central point: the Constitution plays a necessary, if contested, role in constituting democracy. To put it another way, democracy is itself unenumerated. There are many textual provisions that might be understood to articulate democratic norms and values, but they don't all point in the same di- rection or correspond to the same underlying view of democracy. The skeletal outlines of representative democracy are traced in the Constitution, but much about the meaning and requirements of de- mocracy is left to be decided. Broadly worded structural provisions creating majoritarian political institutions coexist with broadly worded norms about such core democratic values as free expression, equality, liberty, and citizenship. Sometimes, applying these different provisions to the same set of facts produces different results. And we must consider, not only these multiple pieces of democracy-shaping constitutional text, but also the structural inferences-inferences that further complicate the concept of enumeration. All of this makes it problematic to rely on simple appeals to something thought self- evidently to be "democracy" as a way to defeat constitutional rights claimed to lack a sufficient basis in text. Consider an example of the plural meanings of democracy that can be distilled from the Constitution. In Romer v. Evans, the Su- preme Court struck down a state constitutional initiative passed by Colorado voters that would have eliminated existing laws banning discrimination based on sexual orientation and made ay persons categorically ineligible for the protection of such laws. Different understandings of democracy, each plausibly grounded in the Fed- eral Constitution, might point in very different directions in analyz- ing the constitutionality of the anti-gay rights initiative.75 A majori- tarian democrat would presumably favor allowing the electorate to decide, by voting, whether this kind of discrimination should be barred or allowed. A republican democrat, by contrast, might well object to giving the mechanisms of direct democracy such broad lati- tude to trump and proactively preempt policy decisions made by elected representatives. And an advocate of cultural democracy might object, on different grounds, to the sweeping exclusion of gay persons from what the Romer Court called the domain of "ordinary civic life."76 Democracy, in other words, has multiple meanings that are some- times sharply opposed. This makes it hard to determine both what the Constitution says about democracy and what it means to use de- 74 517 U.S. 620, 631-36 (1996). 75 I explore Romer through the lens of democracy in Jane S. Schacter, Romer v. Evans and Democracy's Domain, 50 VAND. L. REV. 361 (1997). 76 517 U.S. at 631. [Vol. 9:2 UNENUMERA TED DEMOCRA CY mocracy as the value shaping the contours of the enumeration norm. One response to these quandaries from an enumeration enthusiast might be to suggest that there is no reason not to let the democratic process decide questions of democracy. Avoid the confusing welter of potential constitutional ideas about democracy, such a person might say, by tightly linking constitutional provisions to their explicit textual meaning, and thereby letting the political process-not the Constitution-define democracy's meaning and its requirements. But if we take Reynolds and Harper as the examples, we can quickly see the difficulty with the let-democracy-decide-what's-democratic solu- tion. It is not hard to see why it is objectionable on democratic grounds to leave significant policy questions about democracy to the vote-diluted polity of Reynolds or to the skewed-against-the-poor polity of Harper. Moreover, as John Hart Ely lucidly saw, it is problematic to expect incumbent elected officials to change a system that benefits them, whether that system might be said to violate constitutionally grounded democratic precepts or not." In short, once we focus closely on the role of democracy, we can see the limits of the enumeration principle, because it becomes clear how little work the idea of enumeration itself actually does in the analysis. Because we need to look beyond the idea of enumeration to draw the essential lines, it begins to look like a weak principle on both the descriptive and normative levels. It is weak as a descriptive matter because it crucially fails to tell us how much specificity is re- quired to constitute enumeration. It is weak as a normative matter because it, in fact, relies on independent normative principles like democracy to make sense of enumeration itself. All of this leaves us needing to work out the paradox of resolving what democracy requires when the very right claimed is said to be necessary for democracy. This presents a new set of line-drawing is- sues. On the one hand, for the structural reasons that are so well il- lustrated by Reynolds and Harper themselves, we can't sustainably pre- fer that democratic institutions resolve all rules of democracy. But we also ought not expect courts to design every aspect of democracy through the vehicle of constitutional interpretation. That would be unworkable, given the vast array of macro- and micro-institutional choices involved in a democratic political process. And it would also be undesirable to cut the populace out of shaping any aspects of de- mocratic institutions. This dilemma might thus be restated as a question: Which sorts of democratic questions should be resolved at the constitutional level, and which at the political level? We might draw the line where the 7 ELY, supra note 69, at 105-34. Jan. 2007]