Can We Relax Now? An Essay About Ballot Measures and Lesbian, Gay and Bisexual Rights after Romer v. Evans William E. Adams, Jr. Associate Professor of Law, NOVA Southeastern University ------------National Journal of Sexual Orientation Law, Vol. 2, Issue 2 ------------ ---------------------------------END PAGE 188--------------------------------------- Introduction As academics and practitioners begin to debate the ultimate impact of the United States Supreme Court's decision in Romer v. Evans1, it is important to consider its immediate effect on ballot measures concerning the rights of lesbians, gays and bisexuals around the country. Other commentators in this colloquium will analyze the constitutional arguments involved in Romer so this essay will focus instead on some of the remaining anti-gay ballot initiatives that may survive constitutional attack, and it wi ll also address problems with measures that may be raised in the future. It is important to recognize that many laypersons may falsely believe that all anti-gay ballot measures have been declared invalid pursuant to Romer. This broad reading of the case is unlikely to be adopted by the courts in regard to at least some of the measures, particularly those that attempt a more "subtle" discriminatory approach or merely offer to repeal existing laws. For a minority group that some would argue already tends to be more divided and complacent than is in its best interest, allowing this false sense of security to exist unchallenged could have disastrous consequences. If opponents of equal rights for gays, lesbians, and bisexuals continue to propose anti-gay ballot measures, the lesbian, gay and bisexual community must be prepared to counter them. Part I of this essay will very briefly review the holding in Romer. Pa rt II will discuss the ballot measure process, explaining the differences between the different type of measures, the history of the process, and the advantages and disadvantages of it. Part III will discuss the impact of Romer on these ballot proposals. Finally, the essay will consider the debate about how the lesbian, gay, and bisexual movement should devote its resources in dealing with ballot measures. Some believe too much time and money is being spent fighting these measures rather than dealing wit h other political and legal battles. In addition to fighting anti-gay measures, some advocates argue that lesbians, gays, and bisexuals should affirmatively use the plebiscite process for passage of positive measures which could muster popular support. However, such an undertaking should not be attempted without a realistic evaluation of success. Recent research indicates that the likelihood of electoral success on matters concerning sexual orientation is small. Part IV of this essay will discuss some of this important research and its implications for future political and legal strategies. I. Procedural History of Romer In the last two national elections, groups hostile to laws that prohibited discrimination on the basis of sexual orientation mounted efforts to repeal existing protections and prohibit similar measures from being passed in the future. In 1992, voters in Oregon rejected such a measure while the voters in Colorado approved theirs2, a measure known as amendment Two.3 An action to declare the initiative invalid and enjoin its enforcement was commenced in the District Court for the City and County of Denver. 4 The trial court granted the preliminary injunction, which was appealed to the Supreme Court of Colorado.5 Relying upon a series of United States Supreme Court voting rights cases, the Colorado Supreme Court upheld the injunction and remanded the matte r for trial.6 On remand, the trial court enjoined enforcement of Amendment 2 and the Colorado Supreme Court affirmed the ruling. In addition to other arguments, the State asserted that it was simply denying special rights to homosexuals.7 The United States Supreme Court found this reading of the amendment to be "implausible."8 The majority opinion affirmed the judgment of the C olorado Supreme Court, but did not adopt its rationale.9 Instead, it found that the Amendment lacked a rational relationship to legitimate state interests.10 Justice Scalia, joined by the Chief Justice and Justice Thomas, wrote a scathing dissent, incl uding an assertion that the decision contradicted the holding by the Court in Bowers v. Hardwick.11 As noted, other authors in this issue will critically analyze this opinion so this essay will not comment further about it except to briefly address in Pa rt III whether it will sweep aside all other anti-gay ballot measures. II. Ballot Measures--Democratic Ideal or Dangerous Weapon of Intolerance? A. DIFFERENCES BETWEEN REFERENDUMS AND INITIATIVES In twenty-three states and the District of Columbia, voters may approve a variety of statutory or constitutional provisions directly through the electoral process by voting upon the specific legislation or constitutional amendment.12 In addition, some ci ty and local governments permit voters to introduce or approve local laws through the ballot box. Although some countries consider all such measures to be referendums, with initiatives considered a special type of referendum, this country usually makes a distinction between the two.13 In the United States, referendums are measures which allow voters to approve or remove statutes or ordinances that have been proposed or enacted by a legislative body.14 By contrast, voter initiatives are proposals for l egislation or constitutional amendments which are brought directly to the ballot through some type of petition process.15 The referendum and initiative processes are thus sometimes called "direct democracy" as opposed to "representative democracy" becau se they bypass the legislative and executive branches.16 This article will utilize the same distinction in discussing referendums and initiatives. When discussing both types of measures together, the terms "ballot measures" and "plebiscites" will be use d. B. HISTORY The process for ballot measures was initially proposed by progressive movements which believed that it offered the public an opportunity to counter the undue influence of special interest groups.17 Of the twenty-three states utilizing the process, eleven are in the west and eight are either Midwest or Plains states.18 The idealistic notion that this process is democracy in its purest f ------------National Journal of Sexual Orientation Law, Vol. 2, Issue 2 ------------ ---------------------------------END PAGE 189--------------------------------------- orm, a form in which the people directly express their preference without the mediating influence of intermediaries, has been recognized by a number of courts.19 Recent public disenchantment with elected officials has caused a surge of growth in using th e process.20 There are several problems with this romantic notion of plebiscites. First, it ignores that the country was founded upon the principle of representative, not direct, democracy, and that the ballot measure process now seems more, not less, s usceptible to the manipulations of special interest groups. In addition, the difference between a special interest group and an unfairly oppressed minority may be one of perception, depending upon one's political philosophy. Members of the religious rig ht and gay and lesbian organizations thus trade charges that the other side represents special interests. Thus, the extent to which the legislative process should assist or deter minority interests is one that generates considerable debate. However, the tendency in this country to use the ballot measure process against persons from racial, ethnic, and sexual minority groups has been noted and analyzed elsewhere.21 C. ADVANTAGES/DISADVANTAGES The utility and impact of the ballot measure process have been discussed by both political science and legal scholars. Although the latter have debated it as a philosophical matter, the former have conducted some research to determine its practical effec t, which will be discussed in Part IV below. In spite of the criticism of the process, polls in California, where usage of the process has grown dramatically, and Colorado indicate that voters still like the process.22 Commentators who support the pro cess note that it permits individuals to express their decisions directly without dependency upon a third party to sponsor and introduce the legislative proposal. By utilizing this process, the measure circumvents the checks and balances inherent in the representative legislative process, which some believe permit multiple opportunities for special interest groups to exert pressure.23 It is this same circumvention of checks and balances which critics of the process point to as problematic. They view these procedures as necessary to ensure that the majority does not unfairly disadvantage minority groups.24 Although few critics would c all for an abolition of plebiscites, they criticize the tendency of courts to give more deference to ballot measures than they do to laws passed through the normal legislative process. They argue that the fact that these measures avoid the checks and bal ances should subject them to heightened scrutiny because the plebiscite process avoids the fact-finding and amendment process existent in legislative bodies. They argue further that because legislation normally must be approved by both the legislative an d executive branches, the ballot legislation concentrates power instead of distributing it as the representative process does. Ballot measures cause other problems by eliminating the deliberative process.25 They reduce decisions on issues, some of whi ch are complex, to simplistic yes or no resolutions. Some commentators have noted that such a polarized reduction of issues fails to recognize the problem of resolving majority sentiment when there are three or more alternative solutions to a problem.26 Further, the proliferation of these ballot measures has resulted in increased confusion as voters are confronted with a growing number of propositions on each ballot, some of which are difficult to understand. In 1990, California's statewide ballot inclu ded seventeen initiatives and constitutional amendments, including a "16,000-word environmental measure."27 Perhaps as a result, more than ten percent of the persons who voted in California did not vote on any of the ballot measures.28 Additionally, res earchers found that the readability of the ballot descriptions in California and Oregon were at the eighteenth-grade level.29 Based upon the readability levels of ballot descriptions and the records concerning years of schooling, researchers found that l ess than one-fifth of the adults in California, Massachusetts, Oregon, and Rhode Island had the ability to read and understand the ballot measures and descriptions in their states.30 Research also indicates that voters will take "information shortcuts" r ather than gather all of the possible information about a proposition.31 When you add to these problems the number of persons who do not vote in any particular election, the arguments that these measures actually reflect the majority viewpoint of the pu blic become weaker. The plebiscite process is better geared to "small, single-minded, well-organized interest groups" than "large dispersed groups with diverse agendas."32 Contrary to the romantic notion that ballot measures are populist-driven measures devoid of interest g roup manipulation, money has become increasingly important to the process as paid signature-gathering has become increasingly common.33 As one commentator has noted in discussing the process in California, "It was a $130 million bonanza for political con sultants, advertising agencies, and television stations in 1988."34 In addition to ensuring ballot access, research has also shown that funding can be a primary factor in determining initiative outcomes.35 Why proponents believe that this electoral pr ocess is less vulnerable to the types of pressures and manipulations that they believe make elections for public officials suspect seems to be more a matter of wishful thinking than an analysis of what happens in ballot measure campaigns. Finally, critic s note the tendency for such measures to be used by the majority to mistreat unpopular minority groups. It is this latter concern that is particularly critical to lesbians and gays because of the long history of oppression suffered by members of sexual m inorities in this country. III. Impact of Romer on other Anti-gay Ballot Measures A. TO WHAT EXTENT WILL MEASURES MOTIVATED BY ANTI-GAY ANIMUS RESULT IN THEIR INVALIDATION BY COURTS? Although it is not the primary purpose of this essay to exhaustively analyze the potential reach of the Romer decision, it must be recognized that a very expansive interpretation of the decision could result in the nullification of many of the anti-gay me asures. If the courts should provide expansive interpretations to the discussion in Romer about the illegitimacy of animus toward homosexuality as a rational basis for disparate treatment,36 then it is possible that a broad range of ballot measures that express discriminatory sentiments about lesbians and gays could be struck. ------------National Journal of Sexual Orientation Law, Vol. 2, Issue 2 ------------ ---------------------------------END PAGE 190--------------------------------------- Because the present United States Supreme Court is not one that appears anxious to provide an expansion of individual rights, it seems unlikely, however, that such a broad interpretation will occur in the immediate future. The opinion itself does not ind icate that it invites such an extension. In criticizing Amendment Two, Justice Kennedy mentions in several places that it is unusually broad and sweeping in its scope.37 Its breadth is problematic in both the number and scope of laws struck as well as t he fact that it invaded legislative and executive powers. Further, courts have generally declined to apply heightened scrutiny to ballot measures resulting in the fact that they are rarely entirely invalidated.38 This includes a relatively recent case w hen the Court declined to apply heightened scrutiny to California's Proposition 13, which provided property tax relief.39 B. REPEAL INITIATIVES The type of anti-gay ballot measure that is perhaps most likely to survive an application of the Romer case is the repeal measure that seeks to void legislation or ordinances that prohibit sexual orientation discrimination. Because jurisdictions are not constitutionally required to have anti-discrimination laws protecting sexual orientation, it would seem difficult to argue that a jurisdiction could not repeal such an ordinance. It could be argued that if the repeal is by initiative, the process should undergo heightened scrutiny, but it seems unlikely that a complete ban on such measures would be enforced by the courts. C. FAMILY LAW AND OTHER INITIATIVES The passage of the misnamed "Defense of Marriage Act"40 and the continued mistreatment of gays and lesbians in the family law area41 bode ill for the potential use of ballot measures against lesbians, gays, and bisexuals in the family law area. Polling d ata concerning gay and lesbian rights seems to indicate that it is in this area where the American public still seems to be most uneasy. This unease about lesbian and gay families could make initiatives which target those issues new weapons in the arsen al of religious conservatives. Similarly, reviving sodomy laws where they have been repealed or proposing other measures that can help anti-gay groups organize and recruit could also cause such groups to continue to offer anti-gay measures. On the other hand, the success which such groups have had in legislative bodies on family law issues may make it unnecessary to utilize ballot measures. However, the importance of ballot measures as organizing and fundraising tools could make these types of measures popular in the future, particularly in areas where lesbians and gays are able to convince legislative bodies to not pass hostile legislation. Again, it would seem unlikely that all such measures would necessarily be struck pursuant to the reasoning off ered by the Court in Romer. IV. Fighting Anti-gay Initiatives or Developing Our Own A. CAN WE EFFECTIVELY USE THE BALLOT MEASURE PROCESS FOR OURSELVES? Before utilizing or revising the ballot measure process, advocates must first decide whether the process is more detrimental than beneficial for lesbians and gays. If advocates make it difficult to get a proposition on the ballot, it will create barrie rs for positive proposals as well as negative. With polls showing large percentages of persons opposing discrimination on the basis of sexual orientation in employment, it could be argued that anti-discrimination laws should be proposed through the ball ot process. Further, proposing measures as opposed to responding to those offered by opponents helps prevent the agenda from being controlled by the opponents of lesbian and gay rights. Before proposing ballot initiatives, however, activists need to be aware of the very real barriers to getting such proposals approved at the ballot box. Those who have studied the success rates for gay and lesbian measures at the ballot box do not have good news for supporters of gay rights. Research by political scientists Haider-Markel and Meier indicates that gays and lesbians are likely to lose such elections and are even more likely to do so when they are the ones to bring the matter to a vote.42 In looking at policies which reflect one group's values over another's, referred to as morality policies, they note that compromise is unlikely and intere st groups have little influence because their prime resource, information, is of little value in determining the outcome.43 They point out that interest groups like gays and lesbians are most successful when they "discretely lobby sympathetic policy make rs" and suggest incremental policy changes.44 Haider-Markel and Meier note that of the 69 cities, counties, and states that have held initiatives and referenda attempting to repeal sexual orientation anti-discrimination laws, 77% have resulted in wins fo r the anti-gay forces.45 Of the thirteen initiatives attempting to enact laws protecting lesbian and gay rights, more than 84% were unsuccessful.46 These results are consistent with other research showing that minority groups in general are likely to lo se in the initiative process.47 On a more positive note, Haider-Markel and Meier say that lesbians and gays have been more successful in blocking repeals of existing anti-discrimination laws or preventing bans on laws protecting gay rights, than on enacting new pro-gay laws through the use of the electoral process.48 They also note that the fact that five of seven statewide anti-gay ballot initiatives were defeated may indicate that such measures are different from those held on the local level.49 Ultimately, though, they conclude tha t gays and lesbians "fare better with political institutions than with elections and better in relatively quiescent elections than they do in highly salient and contentious ballot initiatives . . . "50 Additionally, it has been suggested by another commentator that an effective strategy may be to offer alternative initiatives to combat those which are already on the ballot.51 Alternative initiatives propose different variations or choices on the same issue to other measures on the same ballot. Thus, in California, the 1990 ballot offered three alternative proposals on the environment and the 1988 ballot offered four alternative measures on reforming automobile insurance rates.52 All of the environme ntal proposals failed and only one of the automobile insurance measures passed.53 This alternative process can help address the problem of ------------National Journal of Sexual Orientation Law, Vol. 2, Issue 2 ------------ ---------------------------------END PAGE 191--------------------------------------- the fact that the plebiscite process normally only presents yes or no options without the normal amendment and compromise of the representative legislative process. Observers debate, however, whether offering alternative proposals simplify or further com plicate understanding of the underlying issue.54 Even if advocates want to use the referendum/initiative process for pro-gay measures, there are philosophical reasons for those advocates to want to ensure that the process provides adequate safeguards against initiative use by small, but well-financed or zealous groups that are able to take advantage of voter disinterest to place dangerous measures on the ballot. Obtaining electoral victory through fraudulent or other unsavory means does little to advance public confidence in the passed measure. In fac t, it is arguable that the public is even more skeptical when minority groups get positive legislation passed since such groups are accused of being "special interest groups." Arguably, procedural requirements should be in place to guarantee that the me asure has demonstrable public support. Advocates need to know what the procedural requirements for initiatives and referendums are in their jurisdictions. How many signatures are required? Who may sign the petitions? How are they verified? What kind of access does the public have to the signatures, and what procedures are in place to make sure that the verification process is conducted properly? Who has standing to challenge the signatures or verification process? What are the geographic distributi on requirements for the signatures? How long do the proponents of an initiative have to collect signatures? Advocates should also look for answers to the following questions to ensure that the measure is clear and provides the necessary information for voters to make informed choices. What are the textual requirements for the proposition? Are the measures, ti tles, and descriptions prepared by a state or local official or may the proponents prepare them? Is there a single subject requirement and how is it enforced by the courts in the state? Are there restrictions on the length or readability of the measure? Are there restrictions on the titles and descriptions of measures to ensure that they are not misleading or blatantly misrepresentative of the proposition? How strictly are these provisions enforced by the courts? Do the textual requirements apply on ly to what is placed on the ballot or to both the ballot and petition language(i.e., do the petitions have to explain exactly what will be on the ballot)? Which, if any, of the above changes to the process will ensure more fairness is subject to debate.55 Requiring simplification of measures so that they cover only a single subject or are made more readable are goals that many would support.56 However, ma king it more difficult to access the ballot may "place agenda control more squarely in the hands of a few moneyed interests."57 Certainly it could be argued that increased judicial scrutiny of ballot measures should be applied, particularly where they at tempt to limit the rights of minority groups. It was the fear of majoritarian tyranny which helped lead the framers of the Constitution to choose a representative form of government, although some commentators argue that the motivation was based upon a f ear that the minority propertied interests would not be sufficiently protected under direct democracy as opposed to an interest in protecting disadvantaged minorities.58 Because the plebiscite process eliminates protections built into our representative system for minority interests, some commentators believe heightened scrutiny of these measures is necessary, however.59 On the other hand, some commentators are concerned that determining which initiatives warrant heightened scrutiny would be difficult to apply in a consistent and reasoned way. In addition, at least one commentator has argued that to attack the process be cause of the failures of equal protection jurisprudence is attacking the wrong source of the problem. Under this analysis, rather than eliminate or restrict plebiscites, equal protection jurisprudence needs to be revised.60 The problem with this conclus ion of course is that the latter is a daunting task that those minority groups not currently protected are unlikely to see happen in the near future. One of the objections to changing the scrutiny for certain groups is that there is insufficient empirica l data to demonstrate that plebiscites are more dangerous than other forms of legislative enactment.61 However, as noted above, there is empirical data to support the argument that the plebiscite process does in fact disadvantage lesbians and gays, as we ll as other minority groups.62 Whether it is sufficient to justify a change in judicial review may be a matter for debate, but it supports rather than undercuts the conclusions by the founders of the Constitution and the legal commentators that majoritar ian groups can tyrannize members of minority groups. Conclusion The Romer case is a significant case, both substantively and symbolically, for lesbians and gay persons. It would probably be too optimistic, however, to believe that the initiative and referendum processes will not continue to be used against us. The b allot measure has become a potent tool against members of various minority groups in this society. It is also unlikely that the scope of Romer will be broadened so dramatically in the immediate future that all ballot measures aimed at curbing the rights of lesbians and gays will automatically be invalidated by the courts. In addition, the religious right seems to have identified issues where public opinion is still hostile to the rights of lesbians and gays such as in the family law area. Thus, it woul d appear that lesbians and gays must continue to be prepared for hostile ballot measures. Further, it would be wise to develop strategies that ensure that the process is fair and minimizes the damage caused by these harmful measures. 1. ---U.S.---, 116 S.Ct. 1620 (1996). 2. William E. Adams, Jr., Pre-Election Anti-Gay Ballot Initiative Challenges: Issues of Electoral Fairness, Majoritarian Tyranny, and Direct Democracy, 55 OHIO ST. L. J. 583, 585, 589 (1994). ------------National Journal of Sexual Orientation Law, Vol. 2, Issue 2 ------------ ---------------------------------END PAGE 192--------------------------------------- 3. Amendment Two reads: "No Protected Status Based on Homosexual, Lesbian, or Bisexual Orientation. Neither the State of Colorado, through any of its branches or departments, nor any of its agencies, political subdivisions, municipalities or school districts, shall enact, adopt or enforce any statute, regulation, ordinance or policy whereby homosexual, lesbian or bisexual orientation, conduct, practices or relationships shall constitute or otherwise be the basis of or entitle any person or class of persons to have or claim any minority status, quota preferences, protected status or claim of discrimination. This Section of the Constitution shall be in all respects self-executing." Romer, at 1623. 4. Id. at 1624. 5. Id. 6. Evans v. Romer, 854 P.2d 1270 (Colo. 1993). 7. Romer at 1624. 8. Id. 9. Id. 10. Id. at 1627. 11. 478 U.S. 186 (1986). 12. K.K. DuVivier, By Going Wrong All Things Come Right: Using Alternative Initiatives to Improve Citizen Lawmaking, 63 U..CIN.L. REV. 1185, 1186 (1995). 13. Daniel H. Lowenstein, Campaign Spending and Ballot Propositions: Recent Experience, Public Choice Theory and the First Amendment, 29 UCLA L. REV. 505, 508 N.4 (1982). 14. REFERENDUMS: A COMPARATIVE STUDY OF PRACTICE AND THEORY (David Butler & Austin Ramney eds., 1978) [hereinafter REFERENDUMS]; James D. Gordon III & David B. Magleby, Pre-Election Judicial Review of Initiatives and Referendums, 64 NOTRE DAME L. REV. 29 8, 299 (1989). 15. Id. 16. Kovis J. Sirico, Jr., The Constitutionality of the Initiative and Referendum, 65 IOWA L. REV. 637 (1980). 17. REFERENDUMS, supra note, at 18. 18. DuVivier, supra note 12, at 1187. 19. See, e.g., Associated Home Builders of the Greater Eastbay, Inc. v. City of Livermore, 557 P.2d 473, 477(Cal. 1976) (footnotes omitted), in which the court states the "initiative and referendum signifies one of the outstanding achievements of the pro gressive movement"; James v. Valetierra, 402 U.S. 137, 141 (1971), where Justice Black stated that such measures "demonstrate devotion to democracy, not to bias, discrimination, or prejudice."; Mervyne v. Acker, 11 Cal. Rptr. 340, 343, (Cal. Ct. App. 1961 ), where the court states that the right of initiative or referendum is "one of the most precious rights of our democratic process." 20. DuVivier, supra note 12, at 1189. 21. Adams, supra note 2, 604-7. 22. DuVivier, supra note 12, at 1192-3. 23. Adams, supra note 2, at 592-594. 24. Id. at 596-97. 25. Duvivier, supra note 12, at 1205. 26. Id. at 1201. 27. Id. at 1190. 28. Id. at 1199. 29. Id. at 1195. 30. Id. 31. Donald P. Haid Donaler-Markel and Kenneth J. Meier, Legislative Victory, Electoral Uncertainty: Explaining Outcomes in the Battles Over Lesbian and Gay Rights (Manuscript presented at 1995 annual meeting of the Midwest Political Science Associatio n, Publication forthcoming in POLICY STUDIES JOURNAL) 9 (1995). 32. DuVivier, supra note 12, at 1204-5. 33. Id. at 1206. 34. Id. 35. Id. 36. Romer, at 1627. 37. See, e.g., Romer at 1625 ("Sweeping and comprehensive is the change. . . The change that Amendment 2 works in the legal status of gays and lesbians in the private sphere is far-reaching, . . .), at 1626( If this consequence follows from Amendment 2, as its broad language suggests,. . .), 1627 (First, the amendment has the peculiar property of imposing a broad and undifferentiated disability on a single group, . . . Second, its sheer breadth is so discontinuous with the reasons offered for it. . .), 1 628 (It is at once too narrow and too broad.), and 1629 (The breadth of the Amendment is so far removed from these particular justifications ------------National Journal of Sexual Orientation Law, Vol. 2, Issue 2 ------------ ---------------------------------END PAGE 193--------------------------------------- that we find it impossible to credit them.). 38. DuVivier, supra note 12, at 1210. 39. Nordlinger v. Hahn, 112 S.Ct. 2326 (1992). 40. 110 Stat 2419. The statute attempts to permit the United States and individual states and territories from recognizing marriages between members of the same sex authorized by any state other than the one where the marriage is authorized. 41. For a discussion of continuing problems encountered by lesbians and gays in the courts concerning rights in the family law context, see, William E. Adams, Jr., Whose Family Is It Anyway? The Continuing Struggle for Lesbians and Gay Men Seeking to Ad opt Children, 30 NEW ENG. L.REV. 579, 583-587 (1996). 42. Haider-Markel and Meier, supra note 31, at 21. 43. Id. at 3. 44. Id. 45. Id. at 8. 46. Id. 47. Id. 48. Id at 9. 49. Id. at 27. 50. Id. at 22. 51. DuVivier, supra note 12, at 1186. 52. Id. at 1190-91. 53. Id. at 1991. 54. Id. at 1198. 55. Most jurisdictions have only some of these procedural protections. 56. See, e.g., Id. at 1197. 57. Id. at 1197. 58. Robin Charlow, Judicial Review, Equal Protection and the Problem with Plebiscites, 79 CORNELL L. REV. 527, 533 (1994). 59. See, e.g., Derrick a. Bell, Jr., The Referendum: Democracy's Barrier to Racial Equality, 54 WASH. L. REV. 1 (1978); Julian N. Eule, Judicial Review of Direct Democracy, 99 YALE L.J. 1503 (1990); Hans a. Linde, When Is Initiative Lawmaking Not "Repub lican Government"?, 17 HASTINGS CONST. L.Q. 159 (1989). 60. Charlow, supra note 57, at 630. 61. Id. at 626. 62. Supra, note 42.