66 UColo LRevi

Citation: 66 U. Colo. L. Rev. i 1994-1995 Content downloaded/printed from HeinOnline (http://heinonline.
org) Mon Apr 23 00:41:54 2012 -- Your use of this HeinOnline PDF indicates your acceptance of HeinOnline's Terms and Conditions of the license agreement available at http://heinonline.org/HOL/License -- The search text of this PDF is generated from uncorrected OCR text.
University of Colorado Law Review
Volume 66 1994-1995
HeinOnline -- 66 U. Colo. L. Rev. [i] 1994-1995
HeinOnline -- 66 U. Colo. L. Rev. [ii] 1994-1995
UNIVERSITY OF
COLORADO LAW REVIEW

Volume 66, Number 1 CONTENTS DEDICATION OF THE BYRON WHITE UNITED STATES COURTHOUSE Gene R . N ichol .............. :............................ Ruth Bader Ginsburg David M. Ebel GOVERNING BY INITIATIVE LET THE VOTERS DECIDE? AN ASSESSMENT OF THE INITIATIVE AND REFERENDUM PROCESS David B. Magleby ........................................ STRUCTURING THE BALLOT INITIATIVE: PROCEDURES THAT DO AND DON'T WORK Richard B. Collins ....................................... Dale Oesterle INITIATIVE PETITION REFORMS AND THE FIRST AMENDMENT Em ily Calhoun .......................................... CONSTITUTIONAL CHANGE AND DIRECT DEMOCRACY Lynn A. Baker ........................................... BIBLIOGRAPHY CRITICAL RACE THEORY: AN ANNOTATED BIBLIOGRAPHY 1993, A YEAR OF TRANSITION Richard Delgado ......................................... Jean Stefancic COMMENTS SETTLEMENTS WITH NONPARTIES: A CLOSER LOOK AT COLORADO'S COLLATERAL SOURCE AND CONTRIBUTION STATUTES Kam rin Evans ........................................... THE ROLE OF "COMMUNITY" IN THE PACIFIC NORTHWEST LOGGING DEBATE Daniel S. Reimer ......................................... 1995
13
47
129 143
159
197
225
HeinOnline -- 66 U. Colo. L. Rev. [iii] 1994-1995
HeinOnline -- 66 U. Colo. L. Rev. [iv] 1994-1995
UNIVERSITY OF
COLORADO LAW REVIEW

Volume 66, Number 2 CONTENTS ARTICLES ANTITRUST, IDEOLOGY, AND THE ARABESQUES OF ECONOMIC THEORY Walter Adams James W. Brock .......................................... AM I MY PARTNER'S KEEPER? PEER REVIEW IN LAW FIRMS Susan Saab Fortney ..................................... UNCONSTITUTIONAL? DON'T ASK; IF IT IS, DON'T TELL: ON DEFERENCE, RATIONALITY AND THE CONSTITUTION Mark Strasser ........................................... ESSAY IS GOD UNCONSTITUTIONAL? Phillip E. Johnson ....................................... COMMENTS REPRESSION, MEMORY, AND SUGGESTIBILITY: A CALL FOR LIMITATIONS ON THE ADMISSIBILITY OF REPRESSED MEMORY TESTIMONY IN SEXUAL ABUSE TRIALS Julie M. Kosmond Murray ............................... DIVERGING VIEWS ON THE MERGER OF CRIMINAL OFFENSES: COLORADO HAS VEERED OFF COURSE Mark E. Nolan .................................. 461 1995
257
329
375
477
523
HeinOnline -- 66 U. Colo. L. Rev. [v] 1994-1995
HeinOnline -- 66 U. Colo. L. Rev. [vi] 1994-1995
UNIVERSITY OF
COLORADO LAW REVIEW

Volume 66, Number 3 CONTENTS ARTICLES BIODIVERSITY CONSERVATION AND THE EVEREXPANDING WEB OF FEDERAL LAWS REGULATING NONFEDERAL LANDS: TIME FOR SOMETHING COMPLETELY DIFFERENT? J.B . R uhl ................................................ THE NEW ECONOMIC EFFICIENCY IN NATURAL RESOURCE DAMAGE ASSESSMENTS Robert F. Copple ......................................... RETHINKING THE PARTIAL SETTLEMENT CREDIT RULE IN PRIVATE PARTY CERCLA ACTIONS: AN ARGUMENT IN SUPPORT OF THE PRO TANTO CREDIT RULE Marc L. Frohman ........................................ COMMENTS HOUSE BILL 1041 AND TRANSBASIN WATER DIVERSIONS: EQUITY TO THE WESTERN SLOPE OR UNDUE POWER TO LOCAL GOVERNMENT? Geoffrey M. Craig ........................................ CARROTS FOR CONSERVATION: OREGON'S WATER CONSERVATION STATUTE OFFERS INCENTIVES TO INVEST IN EFFICIENCY Mark H onhart ........................................... 1995
555
675
711
791
827
HeinOnline -- 66 U. Colo. L. Rev. [vii] 1994-1995
HeinOnline -- 66 U. Colo. L. Rev. [viii] 1994-1995
UNIVERSITY OF
COLORADO LAW REVIEW

Volume 66, Number 4 CONTENTS LLCS, LLPS AND THE EVOLVING CORPORATE FORM LIMITED LIABILITY COMPANIES: ORIGINS AND ANTECEDENTS W illiam J. Carney ....................................... SUBCURRENTS IN LLC STATUTES: LIMITING THE DISCRETION OF STATE COURTS TO RESTRUCTURE THE INTERNAL AFFAIRS OF SMALL BUSINESS Dale A. Oesterle .......................................... THE TAMING OF LIMITED LIABILITY COMPANIES Robert B. Thompson ..................................... UNIFORM LAWS, MODEL LAWS AND LIMITED LIABILITY COMPANIES Larry E. Ribstein ........................................ Bruce H. Kobayashi BUSINESS FORM, LIMITED LIABILITY, AND TAX REGIMES: LURCHING TOWARD A COHERENT OUTCOME? W illiam A. Klein ......................................... Eric M. Zolt FIDUCIARY PRELUDES: LIKELY ISSUES FOR LLCs DeborahA. DeMott ....................................... REGISTERED LIMITED LIABILITY PARTNERSHIPS: PRESENT AT THE BIRTH (NEARLY) Robert W. Hamilton ...................................... COMMENT THE CHINESE STUDENT PROTECTION ACT AND "ENHANCED CONSIDERATION" FOR PRC NATIONALS: LEGITIMIZING FOREIGN POLICY WHILE AVERTING FALSE POSITIVES IN ASYLUM LAW John D. Griffin .......................................... 1995
855
881 921
947
1001
1043
1065
1105
HeinOnline -- 66 U. Colo. L. Rev. [ix] 1994-1995
RECENT DEVELOPMENTS THE COLORADO HEMP PRODUCTION ACT OF 1995: FARMS AND FORESTS WITHOUT MARIJUANA Thomas J. Ballanco ...................................... GALLENSTEIN: A WINDFALL IN THE SIXTH CIRCUIT MAY STILL BE A GAMBLE ELSEWHERE D ianne D . Miller .........................................
1165
1175
HeinOnline -- 66 U. Colo. L. Rev. [x] 1994-1995
UNIVERSITY OF
COLORADO LAW REVIEW

Volume 66, Number 1 1995
DEDICATION OF THE BYRON WHITE UNITED STATES COURTHOUSE DENVER, AUGUST 1994
GENE R. NICHOL* On August 10, 1994, the Byron White United States Courthouse, new home of the United States Court of Appeals for the Tenth Circuit, was dedicated in downtown Denver, Colorado. The dedication ceremony was one of many Colorado events celebrating the life and legal career of Justice Byron R. White-by any standards this state's favorite son, and again by any standards, one of the most remarkable figures of twentieth century American life. The central feature of the courthouse dedication was the following tribute to Justice White delivered by Justice Ruth Bader Ginsburg. Her words again reminded us that Byron White is not only larger than life, accomplished in a way that is difficult to even seriously contemplate, but real, solid, and as aspiring as the Rocky Mountain West that is his true home. The day after the dedication, a portrait of Justice White was unveiled at the Tenth Circuit Judicial Conference. This time a former clerk of Justice White, Judge David M. Ebel, did the honors-speaking of his mentor and friend with both head and heart. We asked Justice Ginsburg and Judge Ebel to permit us to print their tributes to Justice White in the pages of this Review. It is difficult to describe the impact that Byron White has had on the lawyers of this state and of this law school. He is a walking human achievement. Scholar, athlete, lawyer's lawyer,
* Dean and professor, University of Colorado School of Law.
HeinOnline -- 66 U. Colo. L. Rev. 1 1994-1995
UNIVERSITY OF COLORADO LAW REVIEW
[Vol. 66
lawyer's justice, statesman, and to so many of us, friend-Byron White is the best that the legal profession has to offer. We are honored to salute him.
RUTH BADER GINSBURG* I am so pleased that Judge Tacha, on behalf of Chief Judge Seymour, asked me to speak at this ceremony, to dedicate the United States Federal Courthouse in honor of Justice Byron R. White. The building fits the man whose name it bears, for they are both "national treasures."' The brochure announcing this event aptly called the restored building a "magnificent example of preserving architectural and social heritage in this region." Justice White, as his former law clerk, now Dean of Columbia Law School Lance Liebman, wrote, is himself a "magnificent example," "[h]e has set a magnificent example for those who will follow."2 Among Justice White's many "firsts," he was first Justice to have served the Supreme Court earlier as a law clerk. Chief Justice Fred M. Vinson, for whom Byron White clerked in the 1946 Term, wrote on the photograph he gave Byron at the end of that Term that his future would be as brilliant as his past. That prediction was altogether accurate. Let me quickly recall that brilliant past-Byron White's gallant youth. . Justice White grew up in Wellington, Colorado. He was a teenager in hard Depression days. Ranking first in his high school class, he won a scholarship to the University of Colorado, from which he graduated in 1938, again first in his class. Lettered in three sports, football eclipsed his basketball and baseball fame. These words appear in the cheer for White as All-American, no doubt the only such accolade ever to be recorded in the HarvardLaw Review:
* Associate Justice, United States Supreme Court.
1. Justice White, so described by his former law clerk, now Fifth Circuit Judge Rhesa H. Barksdale, in Rhesa H. Barksdale, A Tribute to Justice Byron R. White, 107 HARv. L. REV. 3, 7 (1993). 2. Lance Liebman, A Tribute to Justice Byron R. White, 107 HARV. L. REV. 13, 19 (1993).
1995]
JUSTICE BYRON R. WHITE
How good is Colorado, We hear the home crowd call, How good is Colorado, When Whizzer has the ball?' Oxford deferred Byron White's Rhodes Scholarship until 1939, which allowed him to play professional football for the Pittsburgh Pirates (now the Steelers) at the then astronomical guaranteed salary of $15,000-well earned, it appears, for he led the National Football League in rushing, first rookie ever to accomplish that feat. At Oxford, the next year, Byron White met young John F. Kennedy, whose father was then our Ambassador to the Court of St. James. Returning home at the outbreak of World War II, the future Justice entered Yale Law School, one year behind his Court colleague-to-be, the late Justice Potter Stewart. Byron White managed, along with his law classes, two more seasons of professional football, this time playing for the Detroit Lions, and again leading the League in rushing. He also earned Yale's prize for the highest grades. To Justice Stewart, Byron White, the student and game-winner, was in reality "both Clark Kent and Superman."4 When the United States entered the war, Byron White volunteered for the Navy. He served in the Pacific as an intelligence officer, earning two Bronze Stars and a Presidential Unit Citation. News reports tell of his courage as a fire fighter and a saver of lives when his ship, Bunker Hill, was hit in South Pacific waters. During his Navy service, Byron White again encountered John F. Kennedy, when Byron wrote the report on the sinking of Kennedy's PT 109. Byron returned to Yale in 1946 and graduated first in his class. That summer he married Marion Lloyd Stearns, to whom this city is home, and that life partnership continues to flourish. After the clerkship with Chief Justice Vinson, Byron and Marion White returned to Denver, where he practiced law until 1961.
3.
Barksdale, supra note 1, at 4.
4. Louis F. Oberdorfer, Justice White and the Yale Legal Realists, 103 YALE L.J. 5, 6 n.6 (1993) (quoting BOB WOODWARD & ScoTT ARMSTRONG, THE
BRETHREN: INSIDE THE SUPREME COURT 65
(1979)).
[Vol. 66
Active in the 1960 presidential campaign, Byron White became deputy to Attorney General Robert Kennedy. He superintended the day-to-day work of Kennedy's highly respected Justice Department and headed-up the. federal task force sent to protect Freedom Riders from mob violence in Alabama in May 1961, at the dawn of the civil rights era. When Justice Whittaker retired, President Kennedy appointed then Deputy Attorney General White to fill the vacancy. Characteristically, White, the nominee, said the 5 President was "putting him out to pasture mighty early," but the President had it right. He said simply: "[Mr. White] has excelled in everything he has attempted-in his academic life, in his military service, in his career before the bar and in the federal government-and I know that he will excel on the highest court in the land."6 And so he did. As his friend and Yale Law School classmate, Senior District Judge Louis F. Oberdorfer wrote: Justice White brought to the Court ... sheer intellectual power, ... battle-tested courage, the ability of a world-class team player to be both competitive and collegial, exquisitely good judgment, [enormous] capacity to work and concentrate, ... and a rich experience of living and working at7the cutting edge of his generation's rendezvous with destiny. Skeptical of abstractions, conscious of the historical and practical limits on the Court's role, averse to rote, rule-bound decisionmaking, Justice White constantly reminded the Court to consider the consequences and common sense of the legal rules it announced. 8 Of all the top grades he received, Justice White may prize most the recognition accorded him by Justice William 0. Douglas, a colleague with whom he often disagreed. Justice Douglas included Byron White on the "All-American team" of justices with whom Douglas had served, a list Justice Douglas intended to reflect "not an ideological line, but basic ability and judicial attitude," justices who "would have adorned any Court 9 in our history."
Barksdale, supra note 1, at 5. PUB. PAPERS: JOHN F. KENNEDY, 1962, at 283 (1962). Oberdorfer, supra note 4, at 16-17. Id. at 14. 9. Id. at 17 (quoting WILLIAM0. DOUGLAS, THE COURT YEARS: 1939-1975, at 42 (1980)). 5. 6. 7. 8.
1995]
May all who serve justice iilJ thistfine building, and in all of the federal courts, strive to be like Justice Byron R. White in their devotion to their jobs and their willingness to work hard to do those jobs well. And, as we dedicate this building, may I now ask all in this audience to join me in applauding Justice White, for his work and deeds, with a resounding BRAVO!
DAVID M. EBEL*
It would be a daunting task to attempt to describe anyone in the time allotted me, and it is clearly an impossible task when the subject is Justice Byron R. White. Justice Ginsburg, at the courthouse dedication ceremonies on Wednesday, eloquently recounted some of Justice White's remarkable accomplishments, and I dare repeat them here simply because some of you were not there to hear her speech and, frankly, his story would not be complete without their retelling. I also venture a few personal observations from the vantage point as his former law clerk and as his friend for the last twenty-eight years.' I do so, however, with some trepidation because Justice White also recalls a few stories from our time together, and he will have the final word today-as is his custom, I might add. The man I describe has perhaps the most sterling r6sum6 that one can imagine, and yet he is a genuinely modest and humble man. He has captivated all of America with his athletic prowess, and yet his most lasting contributions have been intellectual. He is a fierce competitor who loves to win, and yet he values fair play more. He has held awesome power in his hands for the last thirty-one years, and yet he is an apostle of restraint. He has lived in the gossamer world of Constitutional theory, yet grounds his most significant decisions on common sense. He is a man of the world, but he is also a man of faith.
* Judge, United States Court of Appeals for the Tenth Circuit and former law clerk to Justice Byron R. White. 1. Much of this description is drawn from a tribute that I wrote to Justice White that appeared in David M. Ebel, The HonorableByron R. White, 33 WASHBURN L.J. 1 (1993). An occasional story was also first told in a tribute that I wrote about the Justice that appeared in David M. Ebel, A Tribute to Justice Byron R. White, 107 HARV. L. REV. 8 (1993).
[Vol. 66
This is Justice Byron R. White, the 93rd Justice of the United States Supreme Court. Justice White was born on June 8, 1917, and, as most of you know, he grew up in the small farming community of Wellington, Colorado during the Depression. As a youth, the future Justice worked in the sugar beet fields, doing dusty, hard labor. You also know that he was valedictorian of his high school class, but, to quote Paul Harvey, you may not know the rest of the story, which is that his graduating class consisted ofjust six students. I am here just guessing, but I have a feeling that his school may have had a better basketball team than it did a football team. In any event, he never forgot the roots from which he came. Although he regularly declines to give interviews to the New York Times, the Washington Post, and the like, he has granted interviews to the Boulder Daily Camera, the Wellington Triangle Review, and the Colorado Daily! At the University of Colorado, the Justice was, or did, just about everything. He was president of the student body and valedictorian of his class. And, of course, he was an athlete of national renown. He received a total of ten varsity letters in football, basketball, and baseball. Ultimately, however, it was football that earned him national fame as a tailback, punter and passer. By the way, he also played defense. In 1937, White was selected as an All-American football player. He was, probably, the most famous college athlete of his day. You also know that Justice White went on to play three seasons of professional football in the National Football League, and that he was selected Rookie of the Year his first year and twice he led the league in rushing. You know that he was the highest paid player in the entire league-receiving $15,000 for his first year. That may not seem like much by today's standards (except perhaps to our law clerks), but that salary was three times what the owner had paid for the entire professional football team just a few years earlier. You know that Justice White received a Rhodes Scholarship to study in Oxford, England, and that the Rhodes Scholarship is the most prestigious and competitive academic scholarship available to a college graduate. I believe only thirty-two are awarded throughout the entire country. But what you may not know as the rest of the story is that his older brother, Sam, previously had also won the exact same scholarship. I have often wondered how many, if any, families from a small rural
1995]
community like Wellington have ever been able to boast of two Rhodes scholars in the same family. Ultimately, the war interrupted his studies abroad and he returned to enroll at Yale Law School. There, he was influenced by the doctrine of legal realism as it was taught by Professor Myres McDougal and others, and the influence of that school of thought can be seen in many of his later opinions. At Yale, all he managed to do was to serve as an editor of the Yale Law Journaland to graduate at the top of his class academically (again), all the while, I should add, continuing to play football in the National Football League, this time for the Detroit Lions. During the dark years of World War II, Justice White served in the Pacific Theatre as a naval intelligence officer, and he played a critical intelligence role in locating a previously undetected portion of the enemy's fleet, which resulted in the destruction of a number of Japanese vessels. He served on an aircraft carrier, the Bunker Hill, when it was struck by two kamikaze pilots, and he performed what has been described as an heroic role in rescuing survivors from that attack. He was transferred to another ship and five days later it too was struck by kamikazes. It was also during this time that he renewed an acquaintance with another young naval officer by the name of John Fitzgerald Kennedy, as he helped investigate the sinking of PT Boat 109. I Following the war, Justice White married his college sweetheart, Marion Stearns. Marion is one of the beautiful and intelligent daughters of Robert Stearns, who at the time was the President of the University of Colorado. Marion is a remarkable individual in her own right. She has been active in many community affairs and, in addition to being an obviously successful wife and mother, she has been an extended mother to the nearly one hundred law clerks who have served with the Justice over the years. The Whites have two equally remarkable children. Their daughter, Nancy, was an Olympian selected to compete for the United States on the women's field hockey team until we canceled our participation in the games that year because of the Soviet invasion in Afghanistan. Their son, Barney, is one of the most prominent water lawyers in Colorado. When you put together the genetic pool from the Whites and the Stearns, it should be no surprise that the result is outstanding progeny,
[Vol. 66
and I suppose for their six grandchildren-Amy, Hannah, Emily, Max, Alexander, and Darcy-the sky is the limit. In any event, after graduating from Yale Law School, Justice White clerked for Chief Justice Vinson on the United States Supreme Court. Although he helped the Chief Justice grapple with some very significant issues during that term, as is typical of him, he did not neglect his athletics. Indeed, on at least one occasion, the sitting Justices had to send a note up to him during oral argument requesting that he stop playing badminton in the gymnasium which was located immediately above the courtroom. He was making so much noise with his characteristic exuberant play that it was hard for the Justices to hear the lawyers making their arguments in court below. The rest of the story, once again, is that a mere fifteen years later he became the first law clerk to a Supreme Court Justice to take his own seat on that Court. Justice White then returned to his roots in Colorado and practiced law in Denver for fourteen years with the firm that ultimately became known as Davis, Graham & Stubbs. In the fateful 1960 Presidential election, White was suddenly thrust into national politics. Robert Kennedy selected him to head the National Citizens for Kennedy Committee. After the election, White went to Washington as the Deputy Attorney General of the United States. Perhaps his most dramatic incident in that position came when the President sent him to Birmingham, Alabama to lead the federal marshals in protecting the civil rights of the Alabama Freedom Riders, and he had a tense confrontation with Governor Patterson. Just fourteen months into his service as Deputy Attorney General, President Kennedy called Justice White and offered him a seat on the Supreme Court. Although Justice White protested with typical modesty, and reputably asked the President why in the world the President would want to appoint him, he was persuaded to accept the appointment. It took the Senate-this seems incredible by today's standards-just eleven days to confirm him after his nomination. At the tender age of forty-four, Byron R. White became a Justice of the United States Supreme Court. He served in that capacity for thirty-one years, longer than all but a handful of other Justices in our nation's history. During that time, he has authored more than 450 opinions affecting the lives of all Americans. However, his contributions on the Supreme Court
1995]
cannot be measured strictly byIlongevity or number of opinions authored. His tenure must be placed in the context of the magnitude of the issues that came before the Court during that time. It was on his watch that the Court addressed remedies for school segregation, prayer in schools, affirmative action, abortion, privacy rights, legislative reapportionment, the Watergate tapes, the Pe*ntagon papers, the First Amendment rights of the press, the constitutionality of the death penalty, and a wide array of criminal procedural issues including Miranda, search and seizure, and jury selection. Seldom, if ever, in the history of our country, has the Supreme Court faced such a momentous supply of difficult and politically-charged issues, in addition to the less notorious, but every bit as important, diet of cases involving statutory construction in areas such as antitrust, tax, labor, environment, and the like. It is beyond the pale of this presentation to attempt to describe the judicial legacy that Justice White has left for the nation. However, it has been a recurring theme of his that the federal judiciary should exercise considerable self-restraint. He believed that the basic political and social agenda of the country should be established by the people and the elected branches of government rather than by the federal judiciary, which is an unelected, isolated, and life-tenured branch. He had particular faith in the federal government, perhaps as a result of his participation in the Camelot years under President Kennedy. However, he was vigilant to make sure that the processes of government were fair and accountable and that the powers of government were exercised in good faith. His belief in accountability extended to the private sphere as well, and particularly to the concept that the press should have some accountability for its own speech. Justice White placed a high value on education, and his cases reflect a strong belief that education should be available in equal measure to all. In the area of criminal law, he espoused the doctrine of reasonableness. Once again, he insisted on a fair process for judging guilt or innocence, but then he was prepared to allow law breakers to suffer the consequences of their actions. Like all good lawyers, he felt deeply bound by precedent. Reasonableness, accountability, fair processes, judicial restraint, and a belief that the nation's agendas should be set by the political branches of government-these are among the
10
[Vol. 66
judicial legacies that Justice White has left for us as a result of his tenure on the Supreme Court. Although Justice Ginsburg quoted from the same paragraph, I know of no better way to summarize the Justice than to quote a paragraph written by one of his old-time friends, Judge Lou Oberdorfer, Senior Judge for the United States District Court for the District of Columbia, writing in the Yale Law Journal. Judge Oberdorfer summed up the Justice as well as is humanly possible to do in a single paragraph. I'd like to read that paragraph to you again and have you listen carefully to every word, because every word has significance: Justice White brought to the Court the very best that his generation produced by way of sheer intellectual power, a superb and sophisticated education, battle-tested courage, the ability of a world-class team player to be both competitive and collegial, exquisitely good judgment, indefatigable capacity to work and concentrate, a perfectionist's attention to detail, a habit of successful achievement, and a rich experience of living and working at the cutting edge of his generation's 2 rendezvous with destiny. The only thing that I could add-and this is from my heart rather than my head-is that the Justice has also been a loyal friend and mentor, not only to me but to countless people across this land who have been lucky enough to have had him grab them by the elbow and call them "friend." He was never one for flowery expressions of sentiment but I'm here to tell you that that one word-friend-coming from him, has changed a whole lot of lives. He will deny it of course, but there are hundreds of people across this country who walk not only taller but straighter because Justice White has called them friend. I might add, the list of people he calls friend goes far beyond Presidents, Justices, Congressmen, and the like. While clerking for him, we would observe an eclectic assortment of people from his past parade into his office to visit, and the Justice welcomed them all with warmth. I recall once we heard some gruff sounds of affection coming from his office and out walked an older woman who we were advised later had worked in the food services offices at the University of Colorado way back when the Justice had been a student there. She felt free
2.
Louis F. Oberdorfer, Justice White and the Yale Legal Realists, 103 YALE
L.J. 5, 16-17 (1993).
1995]
11
to just walk into the Justice's chambers when she was in Washington those many years later, and he obviously concurred in that judgment. For those who have been fortunate enough to know him as a friend, his personal legacy is every bit as impressive as is his public one. All of us in the Tenth Circuit are immensely proud that our newly-renovated courthouse now bears his name as the Byron White United States Courthouse. Hanging in the main corridor will be his picture, which I would like to ask his wife, Marion, to unveil for us. We will then invite Justice White to make a few responsive remarks. Ladies and gentlemen, I give you Justice and Mrs. Byron R. White.
LET THE VOTERS DECIDE? AN ASSESSMENT OF THE INITIATIVE AND REFERENDUM PROCESS
DAVID B. MAGLEBY*
The initiative and popular referendum permit citizens to set the political agenda by placing statutes and constitutional amendments on the ballot, and by vetoing actions taken by legislative bodies at the state and local levels. Using the initiative, voters may write statutes, and in some states constitutional amendments, which will go to the ballot if sufficient valid petition signatures are gathered. Initiative sponsors in states that provide both the constitutional and statutory initiative will often submit their measure as a constitutional initiative because of its "more secure" legal standing,' a trend that is growing in at least one state.2 Constitutional initiatives can typically be changed only by a subsequent vote of the people, although statutory initiatives in most states may be amended by the legislature.' Initiatives that meet the signature requirement go either directly to the voters (the direct initiative) or are placed on the state legislature's agenda (the indirect initiative), where, if the legislature does not enact the initiative within a specified time, the proponents have the option to gather additional signatures and place the measure on the ballot. Five states permit both the
* Professor of Political Science and Department Chair, Brigham Young University. Research funding for this article was provided by the College of Family, Home and Social Sciences research committee. Research assistance was provided by David Barlow, Scott Baxter, Steven Davis, Ariane Holtkamp, and Quin Monson. Don Norton read and provided helpful comments. In the article I rely heavily on research I have published or presented elsewhere, which is cited throughout the article. 1. DAVID B. MAGLEBY, DIRECT LEGISLATION: VOTING ON BALLOT PROPOSITIONS IN THE UNITED STATES 72 (1984). However, because of that greater legal significance, voters are less likely to approve constitutional amendments. Id. 2. See Eugene C. Lee, RepresentativeGovernment and the InitiativeProcess, in 6 CALIFORNIA POLICY CHOICES 228 (John Kirlin & Donald Winkler eds., 1990) (discussing the growing trend in California). 3. Only five states do not allow the legislature to amend or repeal statutory initiatives. See PHILIP L. DuBOIS & FLOYD F. FEENEY, IMPROVING THE CALIFORNIA
INITIATIVE PROCESS: OPTIONS FOR CHANGE
70
(1992).
14
[Vol. 66
indirect and direct initiative, and not surprisingly when they do, initiative sponsors prefer taking their issue directly to the voters.4 In contrast, citizens may use the popular referendum to place laws previously enacted by local or state legislative bodies before the voters for approval. As with the initiative, citizen petitions qualify a popular referendum for the ballot. Initiatives are sometimes used as referendums, as in the case of a 1964 California measure that overturned an open-housing law passed by the state legislature.5 Sponsors may choose the initiative rather than the referendum because initiatives provide greater latitude in the time petitions may be in circulation. Referendums can also be called by local or state legislatures. For instance, all states except Delaware require that constitutional changes go before the voters in a referendum, and many state legislatures place measures on the ballot for voter approval.6 While there are some similarities in voting behavior on all referendums, the subject of this article is direct legislation: the initiative and popular referendum. In the United States, the initiative and popular referendum were most frequently adopted in western states during the Progressive Era,' but many other states have experienced renewed interest in direct legislation since the mid-1970s.8
4. The five states are Michigan, Nevada, Ohio, Utah, and Washington. 5. Raymond E. Wolfinger & Fred I. Greenstein, The Repeal of FairHousing in California: An Analysis of Referendum Voting, 62 AM. POL. Sci. REV. 753, 753 (1968); see also Reitman v. Mulkey, 387 U.S. 369 (1967) (overturning the constitutional provision approved by the initiative as violative of equal protection). 6. Referendums placed on the ballot by state legislatures in the United States are similar to the facultative referendums in Western European nations. For a discussion of this use of referendums, see Vernon Bogdanor, Western Europe, in REFERENDUMS AROUND THE WORLD 24, 24-97 (David Butler & Austin Ranney eds., 1994). 7. Charles M. Price, The Initiative: A Comparative State Analysis and Reassessment of a Western Phenomenon, 28 W. POL. Q. 243, 247 (1975). 8. The politics of the initiative in Mississippi reflect the changing perspectives on direct legislation from the Progressive Era to today. Mississippi adopted the full slate of direct legislation-direct initiative, both statutory and constitutional, and popular referendum-in 1916. Six years later the state supreme court declared the provisions unconstitutional because they had been adopted as a single measure rather than as separate constitutional changes. Power v. Robertson, 93 So. 769, 775-77 (Miss. 1922). In 1990, two state legislators tested this ruling by submitting petitions to establish a state lottery. The Mississippi Supreme Court upheld the 1922 ruling. Moore v. Molpus, 578 So. 2d 624 (Miss. 1991) (forcing supporters of the initiative to enact a new law permitting direct legislation). This time, however, the state will only have the direct
1995]
INITIATIVE AND REFERENDUM PROCESS
15
Western states were strongholds for the Progressives, and these states were in their formative political years during the peak of the movement. Only six states west of the Mississippi River do not have some form of initiative or popular referendum, while only eight states east of the Mississippi have the process in some form. Moreover, states in the East and South that have the process are more likely to have only part of it, like Kentucky and Maryland's use of the popular referendum only, and Florida and Mississippi's use of the initiative only.9 Periodically, interest has been expressed in direct legislation at the national level, but proposals have not advanced very far in the legislative process.'
I.
DIRECT LEGISLATION AND DEMOCRATIC THEORY
Petition circulators frequently use the argument, "the voters should decide," to persuade people to sign petitions that place initiatives on the ballot. Popular decision-making on issues is consistent with the expanded view of democracy that has come to characterize American government. The Progressive ideology toward government is effectively summarized by Benjamin Parke DeWitt, who, writing in 1915, said: The first of these tendencies is found in the insistence by the best men in all political parties that special, minority, and corrupt influence in government-national, state, and city-be removed; the second tendency is found in the demand that the structure or machinery of government, which has hitherto been admirably adapted to control by the few, be so changed and modified that it will be more difficult for the few, and
constitutional initiative. To quiet fears of black legislators, there is a geographic distribution requirement for petition signatures. The new law also permits legislators to place on the ballot their alternative for each initiative that qualifies, allowing the voters to choose between the alternatives. Finally, to pass an initiative, supporters must get at least 40% of the total of all voters participating in that election, as well as a majority of those voting on the measure. For a discussion of these developments, see David B. Magleby, Direct Legislation in the American States, in REFERENDUMS AROUND THE WORLD 218, 223-24 (David Butler & Austin Ranney eds., 1994); Sarah C., Campbell, House Kills Voter Voice Amendment, CLARION LEDGER (Jackson, Miss.), Mar. 21, 1992, at 1A, 11A; Jay Eubank & Sarah C. Campbell, Lawmakers Vote for Initiative and Referendum Measure, CLARION LEDGER (Jackson, Miss.), May 9, 1992, at 1A, 15A; Cathy
Hayden, Initiative Shown, Amendment Passes, CLARION LEDGER (Jackson, Miss.),
Nov. 4, 1992, at llA. 9. Magleby, supra note 8, at 220. 10. See infra text accompanying notes 110-16.
16
[Vol. 66
easier for the many, to control; and finally, the third tendency is found in the rapidly growing conviction that the functions of government are too restricted and that they must be increased and extended to relieve social and economic distress.1 The initiative incorporates this ideology in its attempt to: replace "special, minority, and corrupt" influences with the people; to replace the structure of government run "by the few" with a structure in which citizens set the agenda and then voters decide the direction; and finally, in its early years, at least, to enact a social reform agenda. 12 Defenders of the initiative frequently argue that the initiative and referendum are "merely ...attempt[s] to get back to the basic idea of the old town meeting, where local measures were directly proposed 3 and adopted or rejected by the people."" However, the town meeting analogy is misplaced because the practice only works in small places with relatively homogeneous populations. 4 Other direct democratic devices that expanded the role of citizens include direct election of U.S. senators, the direct primary, the recall, and women's suffrage.' 5 Advocates of these devices wanted to limit and reduce the power of political parties, state legislatures, governors, city councils, and mayors, which they thought were dominated by a corrupt party system. To reduce the influence of state legislatures on national government, reformers enacted the direct election of U.S. Senators; to reduce the power of political parties generally, they enacted the direct primary; to remove party politics from local government, they enacted nonpartisan local elections; to remove officials from office, they enacted the recall; to limit the power
BENJAMIN P. DEWITT, THE PROGRESSIVE MOVEMENT 4-5 (1915). 12. Up through the 1940s the initiative was used occasionally to propose measures to "relieve social and economic distress." Perhaps the best known of these measures was a series of California initiatives to establish pension 11. programs. MAGLEBY, supra note 1, at 45. Social and economic reforms have been much less common in recent decades. For instance, no states have voted on
universal health care, nor have working conditions or improved social welfare programs been on the ballot in recent elections.
13.
14. AUSTIN RANNEY & WILLMOORE KENDALL, DEMOCRACY AND AMERICAN PARTY SYSTEM 60-61 (Earl Latham ed., 1956).
J. ALLEN SMITH, THE SPIRIT OF AMERICAN GOVERNMENT 352 (1965). THE
15. MAGLEBY, supra note 1, at 23. The Progressive reformers attacked political machines in other ways as well by instituting merit systems, substituting
nonpartisan local elections for partisan ones, permitting home rule, and instituting
the commission form of local government. Id.
1995]
17
of state legislatures, they enacted the popular referendum; and to put the voters in the position of enacting reforms and other legislation, they passed the initiative. This sweeping reform agenda changed the nature of American politics by elevating the role of individual citizens and sharply reducing the role of political parties and elected officials. 6 A second major wave of "progressive" reforms came to the United States in the 1970s and remains a force today. Hostility to party officials during the 1968 Democratic National Convention, and in the months prior, led to the proliferation of direct primaries for selection of presidential candidates. 17 As a consequence, the number of states with primaries rose dramatically. In 1968, 47% of the delegates to the Republican convention and 49% to the Democratic convention were chosen in direct primary elections; by 1972, the proportions had risen to 58% for the Republicans and two-thirds for the Democrats; in 1992, over 85% of the delegates in both parties were elected in
primaries. 18
States that had not yet adopted the initiative and referendum seriously considered adopting them, especially after California's tax-cutting Proposition Thirteen in 1978. States like New Jersey, Hawaii, Texas and Minnesota all gave serious consideration to adopting the process.' 9 Republican governors in Texas and Minnesota became the main advocates of the process and ran on the issue during their gubernatorial campaigns.2 Those states that already had the initiative or popular referendum experienced a dramatic increase in initiative activity, especially at the petition-circulation stage.21 Some contemporary reformers have proposed expanding direct democracy further to include such devices as electronic town
16.
RICHARD HOFSTADTER, THE AGE OF REFORM: FROM BRYAN TO F.D.R.
257-71 (1955).
17. LARRY M. BARTELS, PRESIDENTIAL PRIMARIES AND THE DYNAMICS OF PUBLIC CHOICE 17-21 (1988). See also JAMES I. LENGLE, REPRESENTATION AND PRESIDENTIAL PRIMARIES: THE DEMOCRATIC PARTY IN THE POST-REFORM ERA (1981); NELSON W. POLSBY & AARON WILDAVSKY, PRESIDENTIAL ELECTIONS: CONTEMPORARY STRATEGIES OF AMERICAN ELECTORAL POLITICS (8th ed. 1991). 18. HAROLD W. STANLEY & RICHARD G. NiEMI, VITAL STATISTICS ON AMERICAN POLITICS 148 (4th ed. 1994). 19. Legislative Update, INITIATIVE Q. (National Ctr. for Initiative Rev.,
Englewood, Colo.), Oct. 1982, at 8-9. 20. Magleby, supra note 8,at 234. 21. Id.
18
[Vol. 66
meetings, a randomly selected national caucus in which citizens interact with presidential candidates,22 and a national initiative and referendum process.2 3 Expanding the role of voters in our democracy has remained a constant in U.S. history. But political reforms, like the initiative and popular referendum, do not always live up to the promises of their proponents, and the arguments for letting the voters decide need to be critically examined. For instance, which voters actually decide ballot questions-all voters or only a much-narrower subset of voters? Does the direct legislation issue agenda represent the concerns of a broad cross-section of the public, or does it reflect the concerns of an active minority? How do those persons who actually vote on initiatives decide their vote? Does the process of voter decision-making raise concerns about the reliability of the initiative as a measure of public opinion? Finally, what implications do the initiative and popular referendum have for the operation of democratic values like political equality, liberty, and freedom of speech? Does a democratic device like the initiative conflict with fundamental rights and liberties? These are the issues I will examine in this article in light of recent experience with direct legislation. The Progressive reformers who advocate direct legislation typically distrust politicians, elected officials, and institutions like legislatures, governors, mayors, and courts. They remove these processes from the normal checks and balances of American government. There are, for instance: no provisions for amendment or modification of the initiative once it begins petition circulation; no provisions for gubernatorial veto; and no compromise or consensus building like that fostered by bicameral legislatures. Direct legislation is a departure from the "republican form of government" provided in the U.S. Constitu24 tion.
22. Ross Perot promoted the idea of an electronic town meeting in his 1992 campaign. ROSS PEROT, UNITED WE STAND 32 (1992). For a discussion of the idea of a "deliberative opinion poll" in presidential nominations, see JAMES S. FISHKIN,
DEMOCRACY AND DELIBERATION: NEW DIRECTIONS FOR DEMOCRATIC REFORM
(1991). 23. Bruce Ackerman calls for national referendums on major constitutional BRUCE ACKERMAN, WE THE PEOPLE: or system-transforming occasions. FOUNDATIONS 54-56 (1991). For a discussion of the dynamics of a national
referendum more generally, see THOMAS E. CRONIN, DIRECT DEMOCRACY: THE
POLITICS OF INITIATIVE, REFERENDUM, AND RECALL (1989).
24. U.S. CONST. art. IV, 4.
1995]
19
The core theoretical issues in the debate over direct democracy have to do with the proper scope of voter decisionmaking. Advocates of an expanded or plebiscitary democracy assert that government should follow Aristotle's definition of 2 democracy as the "rule of the many." 5 Citizens were permitted to propose and decide legislation directly in Greek citystates of the fifth century, in Swiss cantons of the Middle Ages, and again in Switzerland during the nineteenth and twentieth centuries.26 In the United States, the New England town meeting is frequently cited as a successful use of expanded voter decision-making. But this latter example, while success2 ful in some instances, has worked only in small polities. While contemporary examples of direct democracy exist in a wide variety of locations, they are all subordinate parts of 2 governments which operate within a representative system. 8 Another line of criticism holds that voters lack the interest, time, inclination, or expertise to act as legislators. Proponents of direct legislation have an idealized view of voters' interest, knowledge, and attentiveness. This idealized view of popular decision-makingbecomes even more problematic, given the long and complex ballots that now regularly go before voters in initiative-prone states. 29 As I will discuss later in the article, the ability of voters to cope with direct legislation is strongly correlated with high levels of education and interest, and many voters thus remain disenfranchised. . A final objection to initiatives focuses on the implications of voting on issues that weaken the fabric of society. Direct legislation, instead of resolving an issue, may result in greater political tension or difficulty. This was clearly the view of James Madison, who wrote in a Federalist paper that "such Democracies have ever been spectacles of turbulence and contention; have ever been found incompatible with personal security, or the rights of property; and have in general been as
25. ARISTOTLE, THE POLITICS 89 (Stephen Everson trans., Cambridge Press 1988). 26. Jean-Francois Aubert, Switzerland, in REFERENDUMS: A COMPARATIVE STUDY OF PRACTICE AND THEORY 39, 39-40 (David Butler & Austin Ranney eds., 1978). 27. Peter Laslett, The Face to Face Society, in PHILOSOPHY, POLITICS AND 'SOCIETY 157, 157-73 (Peter Laslett ed., 1956); FISHKIN, supra note 22, at 14-20. 28. ROBERT A. DAHL & EDWARD R. TuFTE, SIZE AND DEMOCRACY 22-23
(1973). 29.
See infra notes 98-101 and accompanying text.
20
[Vol. 66
short in their lives, as they have been violent in their deaths."" Madison's alternative would be a republican form of government, which he believed would "refine and enlarge the public views, by passing them through the medium of a chosen body of citizens, whose wisdom may best discern the true interest of their country, and whose patriotism and love of justice, will be least likely to sacrifice it to temporary or partial considerations."3 The very nature of the initiative is to ask voters to make "temporary" and "partial" considerations. A voter is asked, for instance, to raise or lower taxes without regard for what programs will be cut as a result.32 Legislators, on the other hand, are more likely to confront the tradeoffs and implications of any given policy question. The primary alternative model to direct democracy is representativedemocracy. Joseph Schumpeter defined representative democracy as "that institutional arrangement for arriving at political decisions in which individuals acquire the power to decide by means of a competitive struggle for the people's vote."3 3 In this system, most, if not all, decisions are made by government officials who stand for election in free and open contests set at regular intervals. Voters and political parties are to hold these officials accountable, and, through the mechanism of candidate elections, orient the government in terms of policy. Early state adoptions of the initiative were challenged on the basis that they violated the U.S. Constitution's guarantee that states would have a "republican form of government."34 The U.S. Supreme Court decided in 1912 that direct legislation does not violate this guarantee.3 5 This decision has recently come under renewed criticism by a former Oregon Supreme Court justice who asserts that recent experience with direct
30. THE FEDERALIST No. 10, at 61 (James Madison) (Jacob E. Cooke ed., 1961). 31. Id. at 62.
32.
The tax-cutting initiatives of the late 1970s and early 1980s fit this
description. See JAMES R. ADAMS, SECRETS OF THE TAX REVOLT (1984); ROBERT
KUTTNER, REVOLT OF THE HAVES: TAX REBELLIONS AND HARD TIMES (1980); DAVID 0. SEARS & JACK CITRIN, TAX REVOLT: SOMETHING FOR NOTHING IN CALIFORNIA
(1985). 33. JOSEPH A. SCHUMPETER, CAPITALISM, SOCIALISM, AND DEMOCRACY 269 (3d. ed. 1950). 34. U.S. CONST. art. IV, 4. 35. Pacific States Tel. & Tel. Co. v. Oregon, 223 U.S. 118 (1912).
1995]
21
legislation has confirmed the foiindets' fear that direct democracy is prone "to unbridled interest and passion."" II. THE IMPORTANCE OF ELECTION LAWS States that permit direct legislation have a wide range of procedural rules for the process. Among the most important rules are those that relate to ballot access. States that have high signature thresholds have lower use, while states with low signature requirements generally have the most extensive use of the process. State signature requirements range from a low of two percent in North Dakota for statutory initiatives to a high of fifteen percent in Wyoming for statutory initiatives and popular referendums. Table One presents the signature thresholds and number of initiatives and popular referendums that have appeared on the ballot since 1950. As I have written elsewhere, "It]he stringency of a state's signature threshold is inversely related to the frequency of 37 Requiring that signameasures qualifying for the ballot." tures be geographically distributed throughout the state is another hurdle imposed on proponents in some states. The intent of geographic distribution requirements is to require minimal support for an initiative across most of the state. "Since 1950, the states without a distribution requirement have averaged forty-eight propositions per state (initiatives and popular referendums), while the states with the requirement 38 have averaged twenty-two per state."
36. Hans A. Linde, When Initiative Lawmaking Is Not "Republican Government": The CampaignAgainst Homosexuality, 72 OR. L. REV. 19, 32 (1993); see also Hans A. Linde, Who is Responsible for Republican Government?, 65 U. COLO. L. REV. 709 (1994). 37. Magleby, supra note 8, at 225. To test for the relationship between signature requirements and whether the number of direct legislation options open to citizens (constitutional, statutory, or both) are related to the frequency of initiative use the following regression was run: y = a + BIx I + 132x. Where y = number of 2 measures on the ballot, x, = number of direct legislation options, and x 2 = average signature threshold. The results of the model being tested are y = 18.55 + 21.34 (options) - 4.22 (threshold). The R2 for the model is a statistically significant .479. Significant at a P value of .01. A model that included the independent variable geographic distribution was also computed, but the coefficient was not significant; however, the negative sign of the coefficient indicates that the presence of geographic distribution requirements and the number of measures making it to the ballot are inversely related. Id. at 225 n.23. 38. Id. at 227.
22
TABLE

1
[Vol. 66
STRINGENCY OF SIGNATURE REQUIREMENTS AND NUMBER OF INITIATIVES AND REFERENDUMS QUALIFYING FOR BALLOT, Signature Requirements Statutory Initiative Constitutional Initiative 4% 5 5 8 8 8 10 10 10 10 10 10 8 8 15 15 -
1950-92
Initiatives & Referendums 1950-1992 95 14 32 63 0 127 97 24 69 49 31 28 24 48 40 8 . 10 38 22 54 .13 22 2 24 14 0 3
Popular Referendum 2% 3 2 5 5 5 4 5 4 5 5 6 5 -5 6
-
Average 2.7% 3.0 4.0 5.0 5.0 6.0 6.0 6.0 6.0 6.7 6.7 7.3 7.3 7.7 8.0 8.0 9.0 9.3 10.0 10.0 10.0 10.0 10.0 10.0 10.0 12.0 15.0 7.5
North Dakota 2% Maryland* Massachusetts* 5 Colorado ,5 Kentucky California 5 Oregon 6 Missouri* 5 Washington 8 Montana* 5 South Dakota 5 Ohio* 6 Nebraska* 7 Michigan 8 Arkansas* 8 Florida* Illinois Oklahoma 8 Alaska* 10 Arizona 10 Idaho 10 Maine 10 New Mexico Nevada* 10 Utah* 10 Mississippi* Wyoming* 15 Median 8 Geographic Requirement ** Non-Binding, Advisory SOURCE: The numbers of initiatives and
from
VIRGINIA GRAHAM, CONG.
10 12 10
10** 5 10 5 10 10 10 10 10 15 5
popular referendums on the ballot since 1950 were obtained
REs. SERV., A COMPILATION OF STATEWIDE INITIATIVE PROPOSALS
ON BALLOTS THROUGH 1976 (1978); DAVID BALLOT PROPOSmONS IN THE UNITED STATES
B. MAGLEBY, DIRECT LEOISLATION: VOTING ON 43 (1984), as well as from interviews with election officials in all twenty-seven initiative and popular referendum states.
What determines a valid signature is also subject to differing standards. Richmond, Virginia has the most stringent standards. Each signature must be independently notarized, effectively requiring that circulators be notaries. At the other end of the stringency continuum is North Dakota, which presumes each signature to be valid, a necessary consequence of the fact that North Dakota does not have voter registration. Colorado validates signatures at the state level by drawing a
1994]
23
random sample of signatures and verifying those against the address at which the voter is registered to vote.39 Until recently, states like Colorado had imposed limitations on paid signature solicitation, in part to keep the process more citizen oriented and less a tool of well-funded interest groups. However, in 1988 the U.S. Supreme Court declared the limitation on paid signature collection unconstitutional as a violation of the First Amendment. 4' Although means other than paid signature solicitation are available to proponents, the Colorado ban was overturned because it was seen as a significant impediment to the proponents. Colorado now routinely has paid signature solicitation, and on these measures the state finds "more fraud than on petitions circulated entirely by volunteers."4 ' California has liberalized its rules on signature collection to include signature solicitation by mail, and some firms now routinely combine signature collection with requests for campaign contributions, in order to finance the signature solicitation and the subsequent campaign. One recent study found that direct mail signature solicitation has spread to most of the states that do not require petition circulators to notarize their petitions." Another recent development is to have one circulator gather signatures for several propositions at the same time, asking voters to choose from a list of petitions, all being circulated by the same paid solicitor. Ballot access for initiatives is not easy and the vast majority of proponents fall short of the signature threshold." Because signature thresholds are typically set as a percent of the vote for governor in the preceding election, they rise with the state population. To qualify an initiative for the ballot in most states requires an intensely committed set of volunteers,
39. Telephone Interview with Donetta Davidson, Elections Division, Office of the Colorado Secretary of State (Sept. 19, 1994). 40. Meyer v. Grant, 486 U.S. 414 (1988). 41. Telephone Interview with Donetta Davidson, supra note 39.
42. CALIFORNIA COMM'N ON CAMPAIGN FIN., DEMOCRACY BY INITIATIVE:
SHAPING CALIFORNIA'S FOURTH BRANCH OF GOVERNMENT 153 (1992) [hereinafter
CALIFORNIA COMM'N]. 43. Charles M. Price, Signingfor Fun and Profit:the Business of Gathering Petition Signatures, CAL. J., Nov. 1992, at 545, 547. 44. In California during the 1970s and 1980s, there Was a substantial increase over preceding decades in the number of direct initiatives in the titling and petition-circulation phase, but not such a significant increase in the number actually adopted. See Figure Two infra p. 28. During the 1970s, 12% of all measures titled made it to the ballot, and in the 1980s that number rose to 20%.
24
[Vol. 66
or hiring the initiative industry to help an initiative qualify for the ballot.45 The success of the initiative industry in qualifying measures has encouraged greater initiative activity. Many groups facing an initiative that they think they cannot defeat have begun to sponsor their own counter-initiative, a measure that goes part way in addressing the concerns of the original initiative but is more acceptable to the opponents of the original initiative. These groups not only pay to get this initiative qualified, they then spend disproportionately large amounts of money to persuade the voters that their counter-initiative is better, or at least to confuse voters enough that they vote
against all initiatives on the same topic. 46 In the counter-
initiative one sees the next step in the growing professionalization of the initiative process. The process has moved from groups that hire consultants, pollsters and media consultants to defeat an initiative to the use of petition circulation firms to sponsor counter-initiatives that have the effect of defeating the original initiative by getting more votes for the counterinitiative. Many other statutory and constitutional provisions play an important part in how direct legislation is practiced. Examples include pre-circulation provisions for titling, petition form, financial disclosure, limitations on petition circulation, how the measure is given its official title and summary, voter education via a statewide voter pamphlet, specification of the majority needed for passage, provisions for legislative amendment or repeal, and a cooling-off period. Initiatives are often lengthy and technically worded. States regularly provide a short summary of the measure printed on the ballot, and some states provide a voter information pamphlet that offers a more detailed analysis. The official summary and title are often challenged in court, as is the substantive constitutionality of the measure, whether the measure has met
45. For a discussion of the tactics of paid and volunteer petition circulators, see David B. Magleby, Ballot Access for Initiatives and PopularReferendums: The Importance of Petition Circulationand Signature Validation Procedures, 2 J.L. & POL. 287, 297-306 (1985).
46. E.g., Susan A. Banducci, Voter Confusion and Voter Rationality: The Use of Counter Proposals in the Direct Democracy Process (Sept. 4-6, 1992) (unpublished manuscript, on file with the University of ColoradoLaw Review) (originally
prepared for the 1992 Annual Meeting of the American Political Science Association).
1995]
25
the procedural requirements for ballot certification, and whether the measure falls within the permissible subject matter for initiatives or popular referendums.4 7 States often limit the subject matter of initiatives and popular referendums first in whether or not they provide for a constitutional initiative, and next by excluding some subjects from the process. Colorado enacted a subject matter limitation in 1994 by passing Initiative A. There is a rough parity in the country as a whole in the use of statutory as opposed to constitutional initiatives. Table Two summarizes the initiative activity by state since 1898. As I have argued elsewhere, "[i]n the aggregate there is... little difference in the passage rates for statutory and constitutional initiatives . "..."48 Given this reality, it is not surprising that in states, like Colorado, which permit both the constitutional and statutory initiative, sponsors will choose to use the constitutional initiative. As found in Table Two, there have been twice as many constitutional as statutory initiatives in Colorado. Moreover, the fact that Colorado sets its signature threshold at five percent for both statutory and constitutional initiatives provides no disincentive for those who might otherwise construe their measure as constitutional rather than statutory in nature. There is some variability by state in how voters perceive initiatives. For instance, all three of Wyoming's initiatives have passed while only two of Utah's sixteen initiatives have. Subjects excluded from the ballot in some states include naming a person to office by initiative, emergency legislation, and using the referendum to block appropriations. Some states "require that a measure may not encompass more than a single subject."49 How the single subject rule came into effect in California is an illustration of how procedural rules have been developed to correct for direct legislation abuses. A 1948 California initiative sought to build a winning coalition of voters by including several subjects in one initiative; the subjects included gambling, reapportionment of the state legislature, surface mining, and retirement pensions. To avoid
47.
James D. Gordon III & David B. Magleby, Pre-ElectionJudicialReview
of Initiatives and Referendums, 64 NOTRE DAME L. REV. 298, 302-03 (1989).
48. Magleby, supra note 8,at 229. 49. Gordon & Magleby, supra note 47, at 303. See, e.g., CAL. CONST. art. 2, 8(d); FLA. CONST. art. XI, 3; OR. CONST. art IV, 1(2)(d).
26
[Vol. 66
such omnibus initiatives in the future, the legislature adopted the single-subject rule.5 TABLE 2 INITIATIVE PROPOSITIONS SUBMITTED TO THE VOTERS, 1898-1992
Statutory Initiatives State Alaska Arizona Arkansas California Colorado Florida Idaho Illinois Maine Massachusetts Michigan Missouri Montana Nebraska Nevada North Dakota Ohio Oklahoma Oregon South Dakota Utah Washington Wyoming Total Constitutional Initiatives Totals # # % # # % # # % Proposed Approved Approved Proposed Approved Approved Proposed Approved Approved 20 78 26 121 48 17 I 30 40 II 20 50 22 16 151 10 31 158 35 16 96 3 1000 9 28 13 42 21 45.0 35.9 50.0 34.7 43.8 64.7 100.0 33.3 47.5 92.9 30.0 62.0 18.2 43.8 39.7 30.0 25.8 34.8 25.7 12.5 47.9 100.0 39.6 55 58 115 99 8 10 2 54 38 8 16 18 34 48 47 116 6 732 24 28 36 32 5 6 2 17 11 3 8 9 19 12 14 34 3 223 43.6 48.3 31.3 32.3 62.5 60.0 100.0 61.5 28.9 37.5 50.0 50.0 . 55.9 25.0 29.8 29.3 50.0 36.0 20 133 84 236 147 8 17 11 30 42 65 58 58 38 34 185 58 78 274 41 16 96 3 1732 9 52 41 78 53 5 45.0 39.1 48.8 33.1 36.1 62.5 64.7 63.6 33.3 50.0 38.5 29.3 50.0 31.6 47.1 42.7 25.9 28.2 32.5 29.3 12.5 47.9 100.0 38.1
II
I
II
7 10 21 25 17 34 12 16 79' 15 22 89 12 2 46 3 659
10
19 8 6 31 4 7 60 3 8 55 9 2 46 3 396
SouRcE: 1900-1976: VIRGINIA GRAHAM, CONo. Ra.s. SERv., A COMPILATION OF STATEWIDE INITIATIE PROPOSALS APPEARING ON BALLOTS THROUGH 1976 (1978); 1977-1991: interviews with election officials in each state, respectively; 1992: ERIC KENNERK, FREE CONGRESS FOUNDATION, 1992 STATE BALLOT LISTING (1992).
A. The Increased Use of Direct Legislation The initiative has experienced renewed use in several states since the mid-1970s. Figure One presents the incidence and passage rate of initiatives and popular referendums in the United States since 1900. Use of direct legislation was immediate and substantial from 1910 to 1919. Use remained relatively high through the
1930s and then dropped systematically from the 1940s through

the 1960s. It started to rise again in the 1970s and has
50. The amended constitution now reads, "An initiative measure embracing more than one subject may not be submitted to the electors or have any effect.".
CAL. CONST. art. 2, 8(d).
1995]
27
increased in use since that time. Figure Two examines the number of California measures titled, qualified, and adopted, to show initiative and referendum activity, including petitions that did not make it to the ballot. Figure 1 History of Initiatives and Referendums In the United States, 1900-1992 E] Passed 1900-091910-191920-291930-391940-49 1950-59 1960-691970-791980-891990-99*I I I I I I I
I
Did Not Pass
-11
---
0
*Projected
50
100
150
200
250
300
350
400
In California, with the exception of the placid 1950s, there have always been more measures titled that did not make it to the ballot than measures that did. But since the 1970s, the number of unsuccessful petitions has grown dramatically. Some of these measures are frivolous and never gather much support, but the frequency of petition activity indicates an increase in interest and attempted use of the initiative. The limited data available in Colorado is consistent with the picture of initiative activity in California. The Colorado Secretary of State has some data on number of "titles set," or measures which began the petition circulation process dating back to 1972. During each two year election cycle, about thirteen initiatives began the petition circulation process. In 1992 and 1994 the number
28
[Vol. 66
of "titles set" climbed to an average of forty-six. The number of measures that made it to the ballot has also risen, but not as much as the number of petitions in circulation. Colorado averaged six initiatives per two year election cycle in the 197282 period and that number rose to ten initiatives on the ballot in 1992 and 1994."'
Figure 2
California Direct Initiatives Titled, Qualified, and Adopted, 1912-1992 E] Adopted 1912-191920-29-
E] Qualified
Titled
1930-39 1940-49 1950-591960-69-
1970-791980-89-
z~mi
0 50 100 150 200 250 300 350
1990-99*
*Projected
What explains this resurgence in initiative activity? Issue activists have learned that placing an initiative on the ballot, regardless of the outcome of the election, generates widespread media attention for their issue. The fact that a vote in a single state will often be interpreted as setting a national mandate on gay rights, nuclear waste, or limiting where people may smoke
51. Memorandum from Catharyn A. Baird, Colorado Secretary of State's Office, to David B. Magleby (Dec. 15, 1994) (on file with the University of Colorado Law Review).
1995]
29
has meant that the stakes of campaigns are higher for affected groups, and the intensity of campaigns increases. Politicians in several initiative-prone states have discovered the same dynamic. Candidates for governor in California now regularly sponsor their own initiatives during their gubernatorial campaign, as Pete Wilson's strong support of Proposition 187, Illegal Aliens. Ineligibility for Public Services. Verification and Reporting, illustrates.52 Jerry Brown's involvement in the political reform initiative of 1974 helped him win the Democratic nomination in a crowded field,53 his successor George Deukmejian sponsored the "Victim's Bill of Rights,"5 4 and Pete Wilson promoted an initiative on criminal justice reform.5 5 Two of Wilson's Democratic opponents in 1990, Diane Feinstein and John Van de Kamp, sponsored initiatives.5 6 In 1992, the governors of Colorado and Michigan actively supported ballot initiatives.5 7 State legislators have also turned to the initiative to promote issues they cannot get passed in the legislature. The temptation to pursue legislation in the public arena not only diverts legislators from the work of the legislature, but encourages legislators to duck tough issues and "let the voters decide." The media's fascination with the initiative is also part of the reason interest and single-issue groups have used the process. Initiatives and initiative activists make great copy. They are often more interesting than, or at least different from, major party nominees for elected office. Proponents are often outside the political mainstream, and can define themselves as underdogs fighting entrenched interests or powerful groups. Initiatives are often controversial issues that generate conflict, which in turn fosters news coverage. Finally, the outcome of an
52. In California in 1990, over half (11 of 18) of the initiatives on the ballot were sponsored by candidates. CALIFORNIA COMM'N, supra note 42, at 62-63. 53. DAVID D. SCHMIDT, CITIZEN LAWMAKERS: THE BALLOT INITIATIVE REVOLUTION 28 (1989). 54. Cf., Douglas Shuit & Leo C. Wolinsky, Governor Accuses Bird of Ignoring Public's Will, L.A. TIMES, June 6, 1986, at 3 (noting Deukmejian's criticism of California Chief Justice Rose Bird for her vote to strike down the "Victim's Bill of Rights"). 55. CALIFORNIA COMMN, supra note 42, at 62-63. 56. Id. 57. Colorado Governor Roy Romer actively supported an initiative concerning public education finance. Michigan Governor John Engler actively supported a tax cut initiative.
30
[Vol. 66
initiative has consequences that are immediate and often important-lower taxes, changing the status of one group's rights, limiting nuclear power-and that is very different from the more difficult-to-predict consequences of candidate elections. Increased interest in the initiative process also arises because of the large and increasing sums of money spent on campaigns, with several states setting new records for spending in recent elections. "In 1992, over 117.3 million dollars were 8 spent on ballot measures in 21 states."" Examples of heavy spending on single measures would include the $6.8 million spent in 1988 by the National Rifle Association and others on a gun control referendum in Maryland, the $21 million spent in California in 1988 against an increase in the tobacco tax, and in the same election the more than $101 million spent in California on five initiatives relating to automobile insur59 ance. An initiative industry which specializes in direct-legislation politics has grown in several states. Professionals help draft measures, circulate petitions, manage campaigns, provide polling, and produce media. The business is highly profitable and has become an increasingly common part of the process. Some of the lessons learned in initiative campaigns have become part of the dynamics of issue advertising generally. Advertising on NAFTA and health care reform reflects some of the strategies of initiative campaigns with emotional appeals that often play on people's fears. "The strategy is often to raise doubts, create confusion, and trigger a negative reaction to the proposition." Part of the increase in initiative activity is attributable to voter frustration with government, especially elected officials. Voter hostility to Congress has been growing, and the movement to limit legislative terms has clearly tapped into this
58. David B. Magleby, Campaign Spending and Referendum Voting 4 (Mar. 10-12, 1994) (unpublished manuscript, on file with the Universityof ColoradoLaw Review) (originally prepared for the 1994 Annual Meeting of the Western Political Science Association). 59. Id. 60. David B. Magleby, Opinion Formation and Opinion Change in Ballot PropositionCampaigns, in MANIPULATING PUBLIC OPINION: ESSAYS ON PUBLIC OPINION AS A DEPENDENT VARIABLE 95, 113 (Michael Margolis & Gary A. Mauser eds., 1989).
1995]
31
view.6 ' The national movement to cut taxes and limit government spending taught conservatives an important lesson. They could set the political agenda even if they did not control the state legislature. One political scientist has gone even further to argue that the conservative agenda of several recent initiatives is a reaction of middle-class, white voters to increased representation for minorities in legislatures." . Finally, there has been an increase in initiative use because issue activists have found the process a good way to advance their political careers. For some this has meant running for elective office as in the case of Proposition Thirteen co-sponsor Paul Gann. For others it has generated a following and much attention in the press. Examples of this sort would include Howard Jarvis and Colorado's Douglas Bruce.
III. WHICH VOTERS DECIDE BALLOT QUESTIONS?
The setting in which voters may decide direct legislation varies from state to state. Some states permit initiatives to appear on primary election ballots; most do not." California and Maine provide for special statewide elections to decide 6 ballot questions, including initiatives. 4 Turnout in these special elections or primary elections is often much lower than turnout in general elections and may be so low as to call into question the legitimacy of decisions reached in such low-turnout elections. Turnout in primary elections is often much lower than in the general election held later that year.65 Not only is turnout lower but it may be skewed in one direction or another by the nature of contests in the respective parties. Howard Jarvis timed his Proposition Thirteen for a primary in which the Democratic party had no contested statewide party race for Governor, while the Republicans did. Jarvis assumed correctly
61. Kelly D. Patterson & David B. Magleby, PublicSupport for Congress, 56 PUB. OPINION Q. 539-51 (1992); David B. Magleby & Kelly D. Patterson,
CongressionalReform, 59 PUB. OPINION Q. 419-28 (1994).
62. Bruce Cain, The Contemporary Context of Ethnic and Racial Politicsin California, in RACIAL AND ETHNIC POLITICS IN CALIFORNIA 23-24 (Byron 0. Jackson & Michael B. Preston eds., 1991). 63. States that do permit initiatives to appear on primary election ballots are California, CAL. CONST. art. 2, 8, Nebraska, NEB. CONST. art. III, 2, and North Dakota, N.D. CONST. art. III, 5, ch. 40-12. 64. Magleby, supra, note 8, at 245. 65. STANLEY & NIEMI, supranote 18, at 85; Walter D. Burham, The Turnout
Problem, in ELECTIONS AMERICAN STYLE 97-131 (A. James Reichley ed., 1987).
32
[Vol. 66
that Republicans would give stronger support for his tax cutting proposition.66 Strategic timing can also affect initiatives slated for general elections, but turnout varies more dramatically in primaries and presents more important strategic opportunities for proponents. In terms of rates of participation and the representativeness of those who vote, primary voters are better educated, older, better off, and more ideological than voters in general elections. 7 The biased composition of primary voters calls into question the' legitimacy of deciding ballot questions in primary elections. Turnout in special elections called to decide ballot questions is even lower than in primaries. In 1993, California Governor Pete Wilson helped to place a school voucher initiative on the ballot, and turnout was 36% of registered voters, up from the 24% to 30% of registered voters who voted in special elections in the 1970s. Other special elections have been called but canceled by the California Supreme Court. A reapportionment initiative scheduled for a special election called by Governor George Deukmejian was declared unconstitutional, and the election not held, because it would violate a state constitutional statute permitting redistricting "only once within ten-year period following a federal census." 8 Since the legislature had already made districting changes after the 1980 census, the court ruled that no other redistricting by legislative act, initiative, or referendum would be constitutionally permissible.6 9 While special elections are comparatively rare in the United States, they are more common in Europe. In response to problems of low turnout, Denmark, Scotland, and Wales have established qualified majority requirements whereby thirty to sixty-seven percent of the electorate (all citizens) must vote affirmatively if a measure is to pass. Some states have already moved to establish electoral rules to insure majority or extraordinary majority opinion in support of ballot questions. For an initiative to take effect, some states require a majority of the voters voting in the election rather
66. 67. 68. (1983). 69. 70.
See SEARS & CITRIN, supra note 32, at 166. LENGLE, supra note 17, at 15-26; BARTELS, supra note 17, at 140-48. Legislature of the State of California v. Deukmejian, 194 Cal. Rptr. 781 Id. Bogdanor, supra note 6, at 29-30.
1995]
33
than a majority of those voting on the particular measure. When Minnesota voted on adopting the initiative process in 1980, 53.2% of those voting on the question voted to adopt the initiative, but only 46.7% of all persons voting in the election voted affirmatively. 71 Minnesota law requires a majority of persons voting in the election to enact constitutional change, and so the initiative was not adopted.72 Other states, like Nevada, require that a majority of voters approve a constitu7 tional initiative in two consecutive elections. 3 One argument for voting on initiatives in primary and special elections is to shorten the ballot in general elections. This is a real concern in several states where ballot length now approaches fifty or more state and local measures in a general election. The California voter pamphlet in 1988 was 159 pages long, a record broken in 1990 when the length stretched to 224 7 pages. 4 Few would argue that even the most committed voters are willing to invest time in understanding this many measures; the political behavior literature certainly gives good evidence that they are not. 75 Two scholars who have recently studied the process in California recommend no more than six statewide initiatives on any single ballot." However, this would likely not reduce the pressure on voters, because most ballot questions are placed on the ballot by the legislature, often because changes are needed to update or modify previously enacted initiatives. It is well documented that, turnout is lower for less educated, poorer, and younger voters.7 7 Voting on ballot propositions only amplifies the social class bias in participation,
71. Charles Backstrom, Popular Vote on Populist Amendments 3 (Sept. 2, 1982) (unpublished manuscript, on file with the University of Colorado Law Review) (originally prepared for the 1982 Annual Convention of the American Political Science Association). 72. Id. 73. MAGLEBY, supra note 1, at 38-40 tbl. 3.1. 74. Charles Price & Robert Waste, Initiatives:Too Much of a Good Thing?, CAL. J., Mar. 1991, at 117. 75. E.g., Arthur Lupia, Political Information, Political Behavior and Policy Outcomes in Direct Legislation: An Experimental Study (Aug. 13, 1994) (unpublished manuscript, on file with the University of Colorado Law Review) (originally prepared for the 1991 Annual Meeting of the American Political Science Association). 76. DUBOIS & FEENEY, supra note 3, at 3. 77. E.g., RAYMOND E. WOLFINGER & STEVEN ROSENSTONE, WHO VOTES? (1980).
34
[Vol. 66
because those with lower incomes or less education tend to skip voting on ballot questions at much higher rates. '" Those who do participate are more likely to confuse their vote on the ballot question from their position on the issue being decided.79 Even on issues of direct relevance, such as a graduated income tax in Massachusetts, one-third of the voters in the lowest quartile of income reported that they would skip voting on the measure because the ballot was too long and difficult to 0 understand." Progressive reformers expressed high hopes that direct legislation would increase turnout and reduce alienation because voters would have more meaningful choices with more direct consequences than those offered by candidate contests. However, initiatives do not increase turnout across states and over time.8 ' Some comparisons of turnout in initiative and non-initiative states are problematic. One problem with some comparisons of turnout across states is that the South, which has lower turnout, also has only two states with the, initiative; one of these, Mississippi, adopted the process in 1992. A comparison that avoids this sampling problem is to look at the state of Maine, where voters have initiatives in elections in even-numbered years (with candidates) and in odd-numbered years (without candidates). "Turnout in Maine is much lower in odd-numbered years... even when there are highly controversial issues on the ballot." 2 These findings of no systematic increase in turnout due to the initiative process make all the more interesting the widespread support for adopting the initiative among nonvoters, an attitude that appears to have more to do with democratic values than voting behavior.
78. MAGLEBY, supra note 1, at 105. 79. Id. at 116.
80. Id. at 117.

81. Id. See also David H. Everson, initiatives and Voter Turnout (Sept. 5, 1981) (unpublished manuscript, on file with the University of Colorado Law Review) (originally prepared for the 1981 Annual Meeting of the American Political Science Association). 82. Dave Kehler et al., Initiativesand Voter Turnout, INITIATIVE & REFERENDUM ANALYsIs (Public Affairs Res. Inst. of N.J., Princeton, N.J.), July 1992, at 1, 3.
1995]
35
IV. WHAT ISSUES ARE DECIDED BY VOTERS IN THE INITIATIVE
PROCESS? The idea that the voters should decide important issues strikes a responsive chord for most Americans because it is consistent with our notions of government by the people. But the most powerful people in the initiative and referendum process are the people who set the agenda for the voters to decide at the next election, and that agenda typically reflects the narrow goals of the proponents of initiatives and referendums. The claim that more democratic government results from direct legislation rests on the assumption that the issues placed on the ballot by initiative or popular referendum reflect the issues that people think are important. Plebiscites on secondary or peripheral issues may provide for a measure of public opinion, but it remains opinion on a secondary or peripheral issue. In fact, the issue agenda of direct legislation rarely reflects the most important political problems or issues on the minds of voters. In 1992, voters indicated that the economy, unemployment and the deficit were the most important problems facing the nation, 3 and yet the 1992 direct legislation agenda rarely focused on these issues. Most voters cannot name a particular issue on which they would like to see the public vote." Absent from the initiative agenda are issues of concern to the poor, the less educated, and those who lack political organization or financial resources. Instead, issues tend to reflect the concerns of ideological or reform groups that have been unsuccessful in getting their way with the legislature, or who desire to elevate their issue as a result of the media attention that comes with getting a measure on the ballot. Because all but a few voters are passive with respect to setting the political agenda via the initiative, the agenda of
83. WARREN E. MILLER ET AL., INTER-UNIVERSrrY CONSORTIUM FOR POL. AND

Soc. RES., AMERICAN NATIONAL ELECTION STUDY, 1992 (1993).
84. David B. Magleby, Legislatures and The Initiative: The Politics of Direct Democracy, J. STATE GOV'T, Spring 1986, at 31, 35; see also CENTER FOR
POL. STUDIES, INTER-UNIvERSITY CONSORTIUM FOR POL. AND Soc. RES., AMERICAN NATIONAL ELECTION STUDY 1970 (1972); WARREN E. MILLER, ET AL., INTERUNIVERSITY CONSORTIUM FOR POL. AND Soc. RES., AMERICAN NATIONAL ELECTION STUDY 1972 (1975); WARREN E. MILLER, ET AL., INTER-UNIVERSITY CONSORTIUM
FOR POL. AND Soc. RES., AMERICAN NATIONAL ELECTION STUDY 1974 (1975).
36
[Vol. 66
issues decided by voters is determined by the proponents' capacity to hire professional signature-gathering firms or by the dedication of issue activists or single-issue groups who desire to promote their narrow issue agenda. Most citizens and "grassroots" groups lack the resources or organizational commitment to get their issues on statewide ballots. Some groups have been successful: antinuclear power, antismoking, utility-reform, term limits and antihomosexuality are examples; but these issues do not represent the issue concerns of the general public, or even most voters.85 Voters are skeptical of most initiatives placed on the ballot by citizen groups. This skepticism is evidenced by the high rate of rejection of "citizen-placed" initiatives, compared to the high rate of approval of measures placed on the ballot by the legislature. Looking at twelve states over time, "[a]bout three of every five measures put on the ballot by state legislatures were approved by the voters ....In contrast, only one-third of measures placed on the ballot by popular petitions in these 8 same states were approved by voters." 6 This difference may be the result of the more intense negative campaigns waged against initiatives, but it may also reflect an underlying distrust of groups trying to sell something to the voters by way of the initiative. What does the initiative process do to the electoral system more generally? Does it alter the agenda or dynamics of elections for governor, state legislature, or other offices? The answer is clearly yes. In several states, initiatives spill over into candidate contests; politicians are pushed to address the content of initiatives or their possible implementation, should the initiative be passed and they be elected. In California,
85. See, e.g., CENTER FOR POL. STUDIES, INTER-UNIVERSITY CONSORTIUM FOR POL. AND Soc. RES., AMERICAN NATIONAL ELECTION STUDY 1970 (1972); WARREN E. MILLER, ET AL., INTER-UNIVERSITY CONSORTIUM FOR POL. AND Soc. RES.,
AMERICAN NATIONAL ELECTION STUDY 1972 (1975); WARREN E. MILLER, ET AL., INTER-UNIVERSITY CONSORTIUM FOR POL. AND SOC. RES., AMERICAN NATIONAL ELECTION STUDY 1974 (1975). 86. Magleby, supra note 8, at 252. In California, as one example, local initiatives are more likely to pass than statewide initiatives. Over 60% of local
initiatives on the ballot between 1983 and 1988 in California passed. When the election was a local election that percentage dropped to 39%, when the election coincided with a statewide election. See David Hadwiger, The InitiativeComes to
Town: CaliforniaCities and Citizen-Sponsored Ballot Measures, WESTERN CITY,
Oct. 1989, at 60-65.
1995]
37
candidates for governor now regularly sponsor their own initiatives, and in 1990 John K. Van de Kamp sponsored three initiatives.8 7 The three initiatives, numbers 128, 129, and 130 on the 1990 ballot dealt with environmental reform, bonds for crime prevention, and campaign finance and ethics reform.8" This level of candidate sponsorship is based on the expectation that identification with a popular measure-say, limiting toxins-will generate additional media attention and a positive impression on the voters. But the issue agenda of initiatives is often diversionary, directing attention away from pressing political issues in the state or nation toward topics like automobile insurance rates, overturning statutes that protect the rights of homosexuals, whether nondentists can prescribe dentures, or restrictions on the hunting and trapping of black bears. The initiative now sets too much of the political agenda, making it increasingly hard for candidates to set their own agendas and independently orient their campaigns. As Daniel Lowenstein and Robert Stern have argued, recent court rulings intended to make the process more democratic-for instance, by declaring bans on paid signature gathering unconstitutional-may actually make the process less democratic and subject to wealthy interests who are able to manipulate the system. 89 A. What Explains Voting Behavior on Initiatives and PopularReferendums? Voting behavior in most candidate contests is explained in terms of three major factors: partisan identification of the voter, candidate appeal or the lack of it, and the issue positions of the candidates and parties. 90 Voters use the political party as a simplifying or economizing device in most partisan candidate elections. Voters also evaluate what they like and dislike about candidates, and the more high-profile the race, the
87. CALIFORNIA COMM'N, supra note 42, at 63. 88. Id. 89. Daniel H. Lowenstein & Robert M. Stern, The First Amendment and Paid Initiative Petition Circulators: A Dissenting View and a Proposal, 17 HASTINGS CONST. L.Q. 175, 176-77 (1989). 90. See ANGUS CAMPBELL ET AL., THE AMERICAN VOTER (1960); WILLIAM H. FLANIGAN & NANCY H. ZINGALE, POLITICAL BEHAVIOR OF THE AMERICAN ELECTORATE (1994); NORMAN H. NIE ET AL., THE CHANGING AMERICAN VOTER (enlarged ed. 1979).
38
[Vol. 66
more important candidate appeal is likely to be. Issues tend to be of lesser significance in candidate voting in part because candidates and parties often avoid clear positions and because voters themselves may not have firm opinions. On ballot propositions, voters must decide their vote without the benefit of party cues, candidate appeal, an incumbent's record of successes or failures, or other, candidate comparisons. The vote choice becomes what the voter perceives the ballot question to be about and how that fits into the voter's understanding of the broader issues and implications. I described the best available data on voting choice and initiatives a decade ago as follows: The individual voter faces a sizable decision-making challenge when confronted by ballot propositions. On the noncontroversial measures, most voters face an informational vacuum, and it is difficult to explain why people vote as they do. It appears that most voters make snap judgments on the measure in the voting place. Even on the contested and controversial measures-typically initiatives-voters can rarely rely on a party cue, and the only remaining decision strategy is to attempt to translate the issue into an ideological predisposition. On some issues the question is poorly focused for such an analysis-how can one translate basic problems with the drafting of a measure into ideological terms? On occasion, however, ideology appears to orient voters, and if one variable is going to work across measures, it will be this
one.
9
1
This summary applies to voting choices over the past decade as well. How campaigns have defined issues before the voters is probably the most important explanation of voting behavior on ballot questions. Not surprisingly, campaigns seek to define the issue in terms favorable to their side. This strategy may mean that the proponents and opponents contend with each other openly about what the measure means, how much it will cost, and its implications. As a result, voters are sometimes confused about the meaning of the proposition they are voting on. Previous research has demonstrated that significant numbers vote in ways inconsistent with their preferences on the issue
91.
MAGLEBY, supra note 1, at 179.
1995]
39
generally. 92 Take the case of a 1980 California rent control initiative. The measure's title implied that a "yes" vote was a vote in favor of rent control and a "no" vote against it. It was just the opposite. Citing its exit poll, the Los Angeles Times concluded that "voters apparently ended up confused and suspicious of the proposition."93 Analysis of this exit poll demonstrates that more than half of all California voters were confused about what a "yes" and "no" vote meant. Over threefourths of California voters did not match their views on rent control with their vote on the proposition: twenty-three percent wanted to protect rent control but incorrectly voted "yes," and fifty-four percent were opposed to rent control but incorrectly voted "no."9 " Because opinions on many initiatives are superficial, they can change during the course of a campaign. "Voters tend to decide later how they will vote on ballot propositions and in some cases larger proportions of voters have not seen or heard about statewide ballot questions until the last few days of the campaign. Voters' choices about candidates are more stable over time."" One reason for this greater electoral volatility is the lack of party and candidate cues to help voters decide how to vote on initiatives. Well-funded campaigns make a difference in initiative outcomes, generally by fostering more "no" voting. Campaign spending is significantly related to voting outcomes in referendum elections. An analysis of campaign expenditures on ballot questions in twenty-four initiative and popular referendum states finds that substantial spending increases the likelihood of "no" voting, that one-sided spending against the measure is significant at lower spending levels, and that more than one96 third of all ballot questions have no spending whatsoever. Eight percent of initiatives had no spending at all; thirteen percent had no spending by proponents; and twenty-five percent had no spending by opponents. 97
92. Id. at 170-73; Magleby, supra note 60, at 106-12. 93. Roger Smith & Dorothy Townsend, Proposition10: Its Defeat Hailedand Lamented, L.A. TIMES, June 5, 1980, at 1, 22. 94. MAGLEBY, supra note 1, at 143-44. 95. Magleby, supra note 60, at 113. 96. Magleby, supra note 58. 97. Id.
40
[Vol. 66
In an effort to assist voters with the complexity of ballot propositions and to give them more neutral information, some states have instituted voter information pamphlets.9 8 These pamphlets provide an official summary of the ballot questions, and often provide opposing arguments by the respective campaigns. Unfortunately, despite the considerable expense, these pamphlets are read only by a small fraction of those who receive them.9 One reason for this is the frequent reliance on technical language to describe the measures. When official summaries and voter handbooks are examined in terms of their readability levels, they fare poorly. My examination of these materials in four states found readability levels ranging from the fifteenth to eighteenth grade levels. ' Some academics have called for a maximum readability score of twelfth grade for these materials. 1 1 V. CHECKS AND BALANCES ON DIRECT LEGISLATION The only institutional checks on the excesses of direct legislation are the courts. The proponents of the initiative and popular referendum so distrusted the traditional system of checks and balances that they largely isolated the initiative and popular referendum process from such restraints. Not surprisingly, the courts now frequently rule on activities of initiative proponents, opponents, and election officials during the preelection period. The courts are also called on to decide the constitutionality of measures enacted by voters. Between 1960 and 1980, only two successful California initiatives were not declared unconstitutional in whole or in part by state or federal courts.0 2 The willingness of the California Supreme Court to declare popular initiatives unconstitutional became an issue in gubernatorial campaigns in California and ultimately became an issue in the judicial retention elections of 1986, when three 03 California Supreme Court justices were defeated. 1 The need
98.
MAGLEBY, supra note 1, at 136.
99. Id. at 136-37. 100. Id. at 118-19.

101. DuBoIs & FEENEY, supra note 3, at 170. 102. The two initiatives not declared unconstitutional during this period are the 1964 measure to control railroad train crews and the 1972 measure regarding coastal zone conservation. MAGLEBY, supra note 1, at 203 app. a. 103. Julian N. Eule, Judicial Review of Direct Democracy, 99 YALE L.J. 1503, 1581-82 (1990).
1995]
41
to rule on the constitutionality of initiatives places courts, the least democratic branch of our government, in the position of overturning a majority vote, perhaps an extraordinary majority. The United States Supreme Court will again face this issue 0 1 4 when it decides on the very popular term limits initiatives. The issue of checks and balances on direct democracy is most relevant when initiatives target racial or other minority groups. As noted, critics of direct democracy from James Madison to the present day have warned of this majoritarian tendency. Perhaps the best known example of this tendency is 0 the 1964 California open housing initiative. 1 5 In declaring this vote of the people unconstitutional, the United States Supreme Court essentially said that voters do not have the right to remove or limit the constitutional rights of minorities.0 6 Such judicial doctrine has been frequently cited in court cases which overturned successful initiatives requiring poll taxes and literacy tests for voters. 0 7 It is less clear what the courts will do with initiatives that ban public funding of abortions, a policy that has had a disproportionate effect on racial minorities. Other minorities have also been the target of initiatives. "English as the official language" initiatives have been targeted against Hispanics and Asians in at least three states. 0 8 Both anti-homosexual groups, seeking to limit gay rights, and homosexuals have sponsored initiatives addressing the rights of gays and lesbians. The 1992 Colorado vote which overturned local ordinances that protected gays from discrimination is the best known recent example of an initiative targeted against homosexuals.' 9 In 1978, Californians voted on an initiative
104. U.S. Term Limits, Inc. v. Hill, 872 S.W.2d 349 (Ark. 1994), cert. granted sub nom. U.S. Term Limits, Inc. v. Thornton, 114 S. Ct. 2703 (1994). 105. Wolfinger & Greenstein, supra note 5. 106. MAGLEBY, supra note 1, at 53 (citing Reitman v. Mulkey, 387 U.S. 369 (1967)). 107. Derrick A. Bell, Jr., The Referendum: Democracy's Barrierto Racial Equality, 54 WASH. L. REV. 1, 11 (1978-79). 108. Michele Arington, English-OnlyLaws and DirectLegislation:the Battle in the States Over LanguageMinority Rights, 7 J. L. & POL. 325, 343 n.121 (1991). 109. See Craig C. Burke, FencingOut PoliticallyUnpopularGroupsfrom the NormalPoliticalProcesses:The EqualProtectionConcerns of ColoradoAmendment Two, 69 IND. L.J. 275 (1993); John F. Niblock, Anti-Gay Initiatives:A Call for Heightened JudicialScrutiny, 41 UCLA L. REV. 153 (1993-94); Robert J. Wagner, Evans v. Romer: ColoradoAmendment 2 and the Searchfor a FundamentalRight for Groups to ParticipateEqually in the PoliticalProcess, 38 ST. LOUIS U. L.J. 523
42
[Vol. 66
that provided for the dismissal of gay and lesbian public school teachers; Californians voted in 1986 on an initiative that would have quarantined all AIDS victims; and in 1992, Oregon voted on a measure requiring state and local governments to discourage homosexuality. The California and Oregon measures were defeated by large margins, but the politics of these campaigns were divisive, reinforcing hostility between opposing groups: heterosexuals and homosexuals, liberals and conservatives.
VI. A NATIONAL INITIATIVE?

The United States is one of only five democracies which has never held a national referendum," but interest in a national initiative grew during the 1970s, its principal advocates being Senator James Abourezk (D-SD) and Representative Jack Kemp (R-NY). Kemp also made the initiative a topic in his campaign for the presidency and referred to the process in his book An American Renaissance: There are stronger democratic processes in the United States than almost anywhere else in the world, but I think we have learned in this last decade that we could have used more. ... The time is right, I think, for the United States to take the lead in a fresh global wave of democratization that demonstrates the efficiency of government forms that rest on ,the wisdom of ordinary citizens. The most fundamental change we could make, I think, is to provide for a national initiative, through an amendment to the Constitution.... Kemp, like other advocates of expanded use of direct legislation at the state level, goes on to assert that use of the process will reduce the high levels of alienation and distrust found in many citizens." 2 He argues, "I feel as stronglyas I do about this reform because I believe it goes to the heart of our national malaise." 3 A proposal for a national referendum to permit popular consideration of fundamental constitutional change has recently
(1993); Note, ConstitutionalLimits on Anti-Gay-Rights Initiatives, 106 HARV. L. REV. 1905 (1993). 110. Magleby, supra note 8, at 218.
111. JACK KEMP, AN AMERICAN RENAISSANCE: A STRATEGY FOR THE 1980S,
at 187-89 (1979). 112. Id. at 190-92. 113. Id. at 190.
1995]
43
been advocated. Bruce Ackerman gives as examples of these kinds of "constitutional moments": the Founding, Reconstruction, and the New Deal.1 4 Ackerman bases his recommendation on his theory of the Constitution rather than on an examination of actual experience with the process in the United States or elsewhere." 5 How to limit the scope of such a device, once established as a tool, is a real concern. As states have learned, many statutory initiatives are presented as constitutional ones, and the tendency will be to expand the scope of the process and to "let the voters decide." Given the appeal of direct democracy, a national initiative proposal will likely return to the national agenda." 6 CONCLUSION Direct legislation, and direct democracy more generally, differ from representative democracy not only in the types of institutions and processes they prescribe, but in the political values and ends they foster. The initiative and popular referendum value widespread participation, open access to the political agenda, and political equality. Representative democracy and the legislative process foster compromise, continuity, and consensus. Moreover, legislatures and representative institutions generally are more insulated from fluctuations in public opinion and from groups that would seek to limit fundamental freedoms or liberties. This paper has examined the initiative and referendum process as it now operates in the United States in terms of which voters dictate the process, which voters decide ballot questions, how they decide their vote, and what limits, if any, exist on the outcomes of these plebiscites. In the initiative process, the voter is only partially legislator. Voters generally are not permitted to participate in the drafting of initiatives, nor may they amend the measure, as
114. ACKERMAN, supra note 23, at 58. 115. Philip J. Weiser, Ackerman's Proposal for Popular Constitutional Lawmaking: Can It Realize His Aspirationsfor DualistDemocracy?, 68 N.Y.U. L. REV. 907 (1993). 116. The Abourezk and Kemp proposals were not the first legislation introduced for a national initiative. During the 1920s, a constitutional amendment called the Ludlow Amendment was introduced which would have required a national referendum before the president could declare war. MAGLEBY, supra note 1, at 23-24.
44
[Vol. 66.
legislators can. There are no hearings, markup, floor debate, or conference between the two houses to work out technical issues or modify the bill to make it more acceptable. Sponsors of initiatives rarely circulate their proposals before the petition phase, and once this phase begins, the language of the measure cannot change. Voters instead face an initiative crafted entirely by the sponsors on which they may only cast a "yes" or "no" vote. They may in fact favor the concept behind the initiative but object to some specific parts of the proposition. The lack of prior consultation may explain why initiatives are often extreme in their approach and prone to defeat by voters, even though a majority of voters may favor the broad issue concern of the initiative. Direct legislation also does not permit an assessment of the intensity of opinion held by those who vote "yes" or "no." All votes are counted equally, but some voters may feel much more strongly about their opinion than others. In the legislative process, legislators routinely calculate and communicate the intensity of each others' opinions. Moreover, minorities, while not able to prevail on their overall decision on many matters, can influence the content of measures during the process itself. Lynn Baker has used public choice theory to explore the question of whether minorities do better in representative institutions than in direct democracy." 7 She refutes "the claims that racial minorities are better served by representative than direct law-making processes" and urges minorities to focus on ways to improve direct democracy rather than abolish it." 8 Accommodation is a norm of importance to the successful operation of a legislature as well as to that of a political system generally. By their nature, referendum campaigns appeal to passions and prejudices, spotlight tensions, and may foster even greater conflict and disagreement. This has been the case in referendums involving race relations and the rights of homosexuals. Writing during a time when race relations were a more frequent topic of referendums, Derrick Bell concluded that referendums were democracy's barrier to racial equality."9 The tendency for majority passions to be excited against
A 117. Lynn A. Baker, Direct Democracyand Discrimination: Public Choice Perspective, 67 CHI.-KENT L. REV. 707 (1991). 118. Id. at 710. 119. Bell, supra note 107.
1995]
45
unpopular minorities is a real danger of expanded use of the initiative, as evidenced by Colorado's 1992 Amendment Two, which called for "no protected status" to be given to homosexu0 als. 12 Moreover, direct legislation may only worsen the problems of representative government. Lucinda Simon has argued that "[i]n many states, the legislature is actually being ignored-and circumvented. While futurists romanticize the virtues of direct democracy, the key institution of representative government-the legislature-is under attack through the use of the initiative, referendum and recall."' 2 ' The initiative has made important contributions to government, with political reform one of its most important contributions. Women's suffrage, for instance, was an early and frequent topic of direct legislation; Colorado, Oregon, Wyoming, and Arizona granted the right to vote to women as the result of 22 an initiative. As Ackerman has argued, referendums generally can be important legitimizing devices for new structural or constitutional arrangements of a significant nature.'23 They have been useful for this purpose in ratifying entry into the European Economic Community for European countries, and when American states consider fundamental structural changes, they would enjoy that benefit as well. But these occasions are infrequent, and as most American states have learned, the constitutional initiative once established as a tool will be frequently abused.
120. Amendment Two reads as follows: 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. COLO. CONST. art II, 30(b). This amendment was declared unconstitutional and the state district court's permanent injunction against enforcement of the provision was affirmed in Evans v. Romer, 882 P.2d 1335 (Colo. 1994), petitionfor cert. filed, 63 U.S.L.W. 3477 (U.S. Dec. 12, 1994) (No. 94-1039). 121. Lucinda Simon, Representative Democracy Challenged, STATE LEGISLATURES, Aug. 1994, at 10. 122. CRONIN, supra note 23, at 97. 123. ACKERMAN, supra note 23.
46
[Vol. 66
Initiatives can also be a way around legislative log jams or inaction because of organized interests. The California taxcutting Proposition Thirteen was in part a reaction to years of legislative inaction on escalating property taxes and had the effect of altering the agenda of tax policy in several states. But many initiatives that are on topics blocked by powerful interests in the legislature run into the same phalanx of interest groups in the election campaign, and as we have seen, they are often unsuccessful in enacting legislation via initiative as well. A discussion of how the initiative could be improved is itself I have written a topic worthy of sustained attention. 2 4 previously about the benefit that would arise if our plebiscites were more straightforward and easy to understand, and then left to elected officials to implement. I also think that the length of the ballot, in some states at least, needs to be addressed. When voters in some counties in California face fifty state and local ballot questions in a single election, we have gone beyond the ability of all but a few voters to participate meaningfully and effectively. Direct legislation is at best a supplement to representative democracy. When it works well it can orient elected officials to the will of the people and ratify fundamental structural changes. But carried to an extreme it has negative consequences for the political system and can undermine the very structure it is intended to supplement. Institutions of representative democracy-legislatures, city councils, mayors, governors-are by nature required to compromise and will be seen by the public as slow to act. Sustained efforts to improve accommodative institutions are vital to the maintenance of democracy. Finally, as this article has argued, the courts play a vital role in the direct legislation process. They not only balance competing rights and liberties but are the "traffic cops" over the procedures and practices of direct legislation. Their vital role and independence must be understood and reinforced.
124. See Richard B. Collins & Dale Oesterle, Structuring the Ballot Initiative:Procedures That Do and Don't Work, 66 U. COLO. L. REV. 47 (1995).
STRUCTURING THE BALLOT INITIATIVE: Procedures That Do and Don't Work

RICHARD B. COLLINS* DALE OESTERLE** I. Initiatives and Constitutional Change .............. II. History of Initiatives ........................... III. Assessing the Aims of Direct Democrats ............ A. Empowerment ............................ B. Good Government .......................... C. Majority Rule ............................. D. Efficiency ............................. . IV. "The Devil's in the Details": The Design of Initiative Procedures with a Special Focus on Colorado ........ A. Introduction ............................. 1. Four Basic Procedural Issues .............. 2. The Colorado Experience With Initiatives .... B. The Petition Process as a Gate to Ballot Access ... 1. Introduction to the Petition Process ......... 2. The Colorado Petition Procedure ........... 3. Comments on the Colorado Petition Process . C. Drafting Initiatives: Amateurism, Ardency, and Inadequate Review ........................ 1. Introduction to the Drafting Problem ........ 2. The Non-Binding, Consultative Procedures in Colorado ............................. 3. Comments on the Colorado Procedures ....... D. Scope of Initiatives ........................ 1. Introduction to the Problem of Multi-Issue Initiatives ............................ 2. The Colorado Single-Subject Rule .......... E. Informing Initiative Voters .................. 1. Introduction to the Problem of Voter Education on Initiatives ............ .............. 2. The Influence of Money .................. 3. The Colorado Notice Procedure: The Ballot Title, the Ballot Information Booklet and Newspaper Notices ..................... 49 53 55 56 56 58 6C 64 64 64 6 7( 7C 71 74 7E 7E 81 K 84 84 & 91 91 9C
9k
A preliminary version of this paper was presented at a conference on Governing by Initiative, sponsored by the University of Colorado School of Law, September 23, 1994.
* Professor of Law, University of Colorado School of Law. ** Monfort Professor of Law, University of Colorado School of Law.
48
[Vol. 66
4. Comments on the Colorado Notice Procedures . 101 104 V. Recommendations ............................ A. Our Primary Conclusion: Procedures Not Only Determine the Mix of Representative and Direct Democracy, They Also Determine Whether the Initiative Process Works to Effect Direct 104 Dem ocracy ............................... 107 B. Writing on a Clean Slate ..................... 107 1. Use the Indirect Initiative Process .......... 2. Preserve the Integrity of the American Constitution: Require Minimum Votes and Word Limits 107 for Constitutional Amendments ............ 3. Treat Initiated Laws No Differently Than 110 Legislatively Passed Statutes ............. 111 4. Enforce a Single-Subject Rule ............. 5. Provide a Qualifying Period Long Enough to Allow Discussion, Deliberation, and Amendment 112 of Proposed Initiatives ................... C. The Role of Implementing Legislation in Colorado . 113 1. Extend the Qualifying Periods, and Establish Fair and Effective Procedures for Ballot Title Protests, Single-Subject Review, and 114 Negotiations About Draft Initiatives ........ 2. Put the Full Text of Initiatives on the Ballot . 115 D. The Role of the Colorado Courts in Policing 116 Initiative Procedures ....................... 1. Necessary Implications of the Initiative Grant . 118 2. Comparing "Amendments" to "Revisions and Alterations": Initiatives Cannot Fundamentally Alter the Existing Constitutional Structure ... 121 3. The Due Process Clause and Initiative 122 Procedures ............................ 124 4. Enforcing the Single-Subject Requirement .... 126 VI. Conclusion ..................................
Ballot initiatives that add provisions to state constitutions are rapidly coming to dominate state government in California, Colorado, and Oregon, and other states will soon join the list. The initiative device has become a major outlet for citizens dissatisfied with their governments, and its use is likely to increase, both in frequency and geography. There is significant support for amending the United States Constitution to
1995]
STRUCTURING BALLOT INITIATIVES
authorize a national initiative.' Given this trend, the structure and procedures for initiatives have ever greater consequences. Public discussion of the initiative process focuses almost exclusively on the content of controversial measures.2 If one favors term limits for elected officials, tax limits, or limitations on gay rights, then initiatives are a meaningful instrument of democracy. If one does not support most of the successful measures, then initiatives are tyranny of the majority and a threat to the republican form of government. The error is the assumption that all initiative processes are indistinguishable. The discussion overlooks the importance of the details of initiative procedures. Some initiative processes work substantially better than others.3 The adequacy (or inadequacy) of procedure is central to the debate over whether initiatives are useful or harmful. This article explores the principal procedural differences among forms of state-wide initiatives, both generally and with particular attention to Colorado. It assesses the goals of initiators and reviews the most frequent criticisms of the device. The article then analyzes the structure of Colorado's initiative laws in detail, both in the state constitution and statutes. The concluding section recommends ways to structure the initiative to avoid its practical shortcomings and better realize its potential for genuine reform.
I. INITIATIVES AND CONSTITUTIONAL CHANGE
Initiatives are among several forms of "direct democracy" in common use in American states. The initiative empowers citizens, by petition, to require a popular vote on whether to adopt a statute or constitutional amendment they have drawn up. 4 Twenty-four states authorize state-wide initiatives, and
1. See infra note 28 and accompanying text. 2. See, e.g., Julian N. Eule, JudicialReview of Direct Democracy, 99 YALE L.J. 1503 (1990); Hans A. Linde, When is Initiative Lawmaking Not "Republican
Government," 17 HASTINGS CONST. L.Q. 159 (1989).
3. For commentary on initiative procedures see Nick Brestoff, Comment, The California Initiative Process: A Suggestion for Reform, 48 S. CAL. L. REV. 922 (1975).
4. DAVID B. MAGLEBY, DIRECT LEGISLATION: VOTING ON BALLOT PROPOSITIONS IN THE UNITED STATES 1, 35-36 (1984). Some writers, especially outside the
United States, define the initiative as a particular kind of referendum, using the latter term to mean any popular vote to adopt or reject a law. Others limit referendums to popular votes on measures originating in legislatures, in contrast
50
[Vol. 66
many local governments have them, including some in states lacking state-wide initiatives.' Our concern in this article is with state-wide initiatives. We refer to other forms of direct 6 democracy only incidentally. Initiatives can be direct or indirect. The indirect initiative requires that citizens first petition the legislature. Only if rebuffed may they force a popular vote, although at that point the citizens' proposal has the same right to popular enactment as under the direct initiative.' Thus the terms are unfortunate, suggesting that the indirect version is an inferior form of "direct democracy." More accurate terms would be immediate and delayed initiative. But the direct/indirect locution is in wide use, so we adhere to it in this paper. So defined, the most important example of the indirect initiative is the Swiss national initiative.' Most initiative states in the United States allow only direct initiatives to compel popular votes without any
to initiatives. In any case, there is general agreement that initiatives are laws drafted by citizens and enacted by popular vote. Compare KRIS W. KOBACH, THE REFERENDUM: DIRECT DEMOCRACY IN SWITZERLAND 12 n.10 (1993) (initiative a kind of referendum) with MAGLEBY, supra, at 1, 35-36 (initiative and referendum distinguished). We use referendums as the plural rather than referenda on advice of the OXFORD ENGLISH DICTIONARY 466 (2d ed. 1989). 5. See THOMAS E. CRONIN, DIRECT DEMOCRACY: THE POLITICS OF INITIATIVE, REFERENDUM, AND RECALL 51 (1989) (listing Alaska, Arizona, Arkansas, California, Colorado, District of Columbia, Florida, Idaho, Illinois, Maine, Massachusetts, Michigan, Missouri, Montana, Nebraska, Nevada, North Dakota, Ohio, Oklahoma, Oregon, South Dakota, Utah, Washington, and Wyoming); MISS. CODE ANN. 23-17-1 to -61 (Supp. 1994); MAGLEBY, supra note 4, at 38-39; Austin Ranney, The United States of America, in REFERENDUMS: A COMPARATIVE STUDY OF PRACTICE AND THEORY 71-75 (David Butler & Austin Ranney eds., 1978). 6. See generally CRONIN, supra note 5, at 38-59. The other form of direct democracy most relevant to the issues discussed in this article is the device that allows citizens, by petition, to force a popular vote whether to rescind or block a statute newly enacted by the legislature. Everyone calls this device a referendum, but modifiers to distinguish it from other referendums, such as those put on the ballot by the legislature, vary chaotically. See, e.g., KOBACH, supra note 4, at 15 ("optional legislative referendum"); MAGLEBY, supra note 4, at 36 ("popular referendum"); GEOFFREY DE Q. WALKER, INITIATIVE AND REFERENDUM: THE PEOPLE'S LAW 13 (1987) ("legislative referendum"); CRONIN, supra note 5, at 2 ("popular referendum" or "petition referendum"); Ranney, supra note 5, at 70 ("statutory referendum"). In this article, we have chosen the term "rescinding referendum" to identify its function in rescinding a legislative enactment. Kentucky, Maryland, and New Mexico allow rescinding referendums but not initiatives.. All but one of the initiative states (Florida) also allow rescinding referendums. MAGLEBY, supra note 4, at 38-39. 7. See, e.g., MAGLEBY, supra note 4, at 35-36. 8. KOBACH, supra note 4, at 15.
1995]
involvement of the legislature. The indirect initiative is important in the five states that allow it but do not have the direct initiative, notably Massachusetts. 9 Most initiative states, including Colorado, allow initiatives 0 either to enact statutes or to amend their constitutions. 1 However, the distinction between statutory and constitutional initiatives is blurred because many states that allow statutory initiatives restrict the power of their legislatures to amend or repeal initiated statutes." In California, an initiated statute can be changed only by another popular Vote.' 2 Under the California rule, the statutory initiative is nearly as entrenched as the constitutional. The principal difference is that the statutory initiative is subject to judicial review under the state constitution, while the constitutional initiative overrides inconsistent provisions of the state constitution. 3 ! In most states that allow it, the constitutional initiative is easy enough to qualify for the ballot that it is frequently used. Our received concept of the U.S. Constitution is fundamental law that is seldom changed and only with substantial consen-
9.
CALIFORNIA COMM'N ON CAMPAIGN FIN., DEMOCRACY BY INITIATIvE 359-60
(1992) [hereinafter CALIFORNIA COMM'N] (Alaska, Maine, Massachusetts, and Wyoming allow only the indirect initiative. Utah and Washington allow both. Michigan, Nevada, and Ohio allow only direct for constitutional amendments and only indirect for statutes); MISS. CONST. art. 15, 273 (providing for indirect initiative only). Colorado's provision for local-government initiatives authorizes only the indirect initiative, while its state-wide initiative is only the direct form. See infra note 76. 10. Six states (Alaska, Idaho, Maine, Utah, Washington, and Wyoming) allow initiatives only to enact statutes, and three (Florida, Illinois and Mississippi) allow them only to amend their constitutions. MAGLEBY, supra note 4, at 38-39; MISS. CODE ANN. 23-17-1 to -61 (Supp. 1994). 11. CALIFORNIA COMM'N, supra note 9, at 366-67 (half the states allowing initiated statutes impose some restrictions on legislative repeal; four impose severe restrictions). 12. Id. at 366 (unless the initiated measure itself authorizes legislative amendment). Arizona imposes the same limit when an initiative is approved by a majority of registered voters. Id. at 367. Colorado allows amendment or repeal of initiated state statutes by the normal legislative process, but requires that changes in laws initiated in local governments be made only by another initiative. See infra note 76. 13. E.g., Floridians Against Casino Takeover v. Let's Help Fla., 363 So. 2d 337,341-42 (Fla. 1978). But see MISS. CONST. art. 15, 273(5)(a), (d) (constitutional initiative cannot be used to amend state's bill of rights or initiative provision). All state and local initiatives are, of course, subject to review under the U.S. Constitution.
52
[Vol. 66
sus. By that standard, many state constitutional initiatives look more like ordinary legislation than fundamental law. 4 When initiators have a choice between enacting a statute
or amending the state constitution, they incline strongly toward entrenching their handiwork in the constitution. For example,
Colorado's initiators proposed fifty-five amendments and thirtyeight statutes between 1912 and 1958.15 This is not surprising because initiators tend to be very strong advocates of their proposals. Moderate views usually do not generate the incentive to conduct an initiative campaign. In recent years, dominance of the constitutional form has increased in the states that allow it.16 For example, in Colorado twenty-six of the thirty measures on the ballot since 1984 have sought to amend the state constitution. 7 The constitutional initiative is the dominant form of state-wide initiative today, and it presents the most controversial issues about the
initiative and its uses.

The state-wide, constitutional initiative works a significant
change in the traditional American concept of governmental

structure. Differences between constitution and legislation are much reduced. The notion of a constitution as fundamental law that is seldom changed, and only with substantial consensus, is gone. Constitutions are amended often and by voting majorities that are typically a minority of a state's adult population."
14. See Ranney, supra note 5, at 76. Switzerland's national initiative allows amendments to the constitution but does not allow adoption of ordinary legislation. Hence, the subjects of many Swiss initiatives seem more statutory than constitutional in character. See Charles B. Blankart, A Public-ChoiceView of Swiss Liberty, 23 PUBLIUS: THE JOURNAL OF FEDERALISM, Spring 1993, at 88. 15. See Paul D. Starr, The Initiative and Referendum in Colorado 37 (1958) (unpublished M.A. thesis, University of Colorado). 16. CALIFORNIA COMM'N, supra note 9, at 359. 17. See infra note 77; see also CALIFORNIA COMM'N, supra note 9, at 174-82. 18. For example, two prominent amendments that passed in Colorado in 1992 each received about 32.5% of the voting age population, or about 40.6% of registered voters. See Submission of Interrogatories on Senate Bill 93-74,852 P.2d 1, 4 (Colo. 1993) (812,308 votes for 1992 Amendment One, limiting taxing and spending); Evans v. Romer, 854 P.2d 1270, 1272 (Colo. 1993) (813,966 votes for 1992 Amendment Two, restricting gay rights); THE ELECTION DATA BOOK: A STATISTICAL PORTRAIT OF VOTING IN AMERICA 1992, at 132 (Kimball W. Brace ed., 1992) (voting age population of Colorado in 1992 estimated at 2,501,000, with 2,003,375 registered voters). California's best known ballot measure, tax-cutting Proposition 13, was adopted at a primary election by about 26.9% of the voting age population, or 43.9% of registered voters in 1978. Votes for the measure are computed from MAGLEBY, supra note 4, at 88 (44% of voting age turnout); CALIFORNIA COMM'N, supra note 9, at 186 (96.5% of turnout voted on Proposition
1995]
The role of the state's judiciary is reduced, and the executive veto is bypassed. This is not necessarily bad. Many democracies in the world function well without our concepts of a fundamental constitution and separation of powers. In traditional parliamentary governments, such as Britain's, constitutional change can be enacted by majority vote of Parliament. 9 In Switzerland, the constitutional initiative is used frequently.2 However, those governments have traditional sources of stability that ours lack. State constitutions filled with provisions that look like ordinary legislation, and easily amended by a small percentage of the voters, diverge greatly from what most Americans think a constitution is. By our traditions, a constitution establishes the structure for government to act subject to basic and general prohibitions in a bill of rights. It is supported by consensus and should seldom need amending. Constitutional language is broad and general, destined to endure. II
HISTORY OF INITIATIVES
Americans of the generation that framed the Constitution were familiar with accounts of Athenian and other ancient examples of direct democracy, and the New England town meeting was established well before independence. 2 ' But many of the Framers expressed hostility to direct democracy, and during the founding period, constitutional and ratifying conventions were the broadest form of citizen participation in
general use22
13); DAVID D. SCHMIDT, CITIZEN LAWMAKERS: THE BALLOT INITIATIVE REVOLUTION 132 (1989) (64.8% yes votes). California had 10,129,986 registered voters. THE
ELECTION DATA BOOK, supra, at 108. These are the most prominent ballot measures; votes on others are lower. See MAGLEBY, supra note 4, at 77-99.
19. See generally A. V. DICEY, INTRODUCTION TO THE STUDY OF THE LAW OF
THE CONSTITUTION 39-85 (10th ed. 1959) (discussing the nature of parliamentary sovereignty). 20. See KOBACH, supra note 4, at 70-78.
21. CHARLES S. LOBINGIER, THE PEOPLE'S LAw 99-102 (1909). See generally 1 ALEXIS DE TOCQUEVILLE, DEMOCRACY IN AMERICA 69-79 (Phillips Bradley ed.,
1945) (discussing New England townships).

22. CRONIN, supra note 5, at 12-20, 22, 41; see also MAGLEBY, supra note 4, at 31. See generally LOBINGIER, supranote 21, at 137-87 (discussing constitutional
origin and development during Revolutionary Era); THE FEDERALIST Nos. 10, 55, 63 (James Madison).
54
[Vol. 66
Referendums were prominent in Europe during the French Revolution and its Napoleonic aftermath." They gradually took hold in this country as the means to ratify state constitutions and amendments.24 But these were measures referred to the voters by legislatures or constitutional conventions. The concept of the initiative was a Swiss innovation, copied here during the Progressive Era. Several Swiss cantons adopted the constitutional initiative during the 1830s, and the legislative initiative during the 1860s.25 In 1891, Switzerland added the constitutional initiative for amending its national constitution. American interest in broader forms of direct democracy grew during the 1880s and 1890s, propelled by Swiss practice.26 Americans who had visited that nation led the movement to promote the initiative and referendum in this country. Most states that allow the initiative adopted it during the Progressive Era.2 ' Beginning with South Dakota, nineteen states authorized state-wide initiatives between 1898 and 1918. The next state to do so was Alaska in its statehood constitution in 1959, and four states have adopted it since. Several states have rejected the initiative at various times. However, the initiative is very popular, and it is far more likely that more states will adopt it than any will repeal it. There is also a movement to establish a national right to initiative. It does not seem close to success, although opinion polls back the concept.28 Other movements seek to promote various forms of
23. See generally MARKKU
SUKSI, BRINGING IN THE PEOPLE
41-47 (1993)
(discussing French Revolution's influence on national decision-making). 24. See CRONIN, supra note 5, at 41. See generally LOBINGIER, supra note
21, at 188-291 (overview of the extension of "popular ratification" throughout the United States). 25. See generally KOBACH, supra note 4, at 18-30 (examining development of the initiative in Switzerland). Some Swiss cantons and towns had the townmeeting form of direct democracy long before the nineteenth century. Id. at 16-18. 26. CALIFORNIA COMM'N, supra note 9, at 35-36. See generally CRONIN, supra note 5, at 43-54 (overview of populist movement and state adoptions of initiative and referendum). Lobingier claims that a colonial Rhode Island law and another in Georgia's first constitution were the first American initiatives. LOBINGIER, supra note 21, at 358. However, the Rhode Island practice seems more like a town meeting and Georgia's a form of convention. 27. See CRONIN, supra note 5, at 51; see also id. at 3-4 (discussing states that have rejected the initiative). 28. MAGLEBY, supra note 4, at 7, 12-14; see also CRONIN, supra note 5, at 45. See generally id. at 157-95 (examining the desirability of a national initiative and referendum).
1995]
electronic voting that might make initiatives and referendums cheaper to conduct.29 Colorado adopted the initiative in 1910.30 The right was added to the state's constitution by a measure proposed by the legislature in a year when Democrats controlled both houses.3 The governor, John Shafroth, was of Swiss descent and sought to promote the concept as a bastion of Swiss liberty. As an amendment to the constitution, the proposal required approval in a popular referendum. It had the backing of most major newspapers and swept the state, receiving over seventy-five percent of the votes cast on the measure.32 III.
ASSESSING THE AIMS OF DIRECT DEMOCRATS
Proponents often claim that the initiative is a more democratic way of governing than representative elections, that it is a way of perfecting democracy. We sometimes say that representative democracy is indirect democracy, while initiatives are a form of direct democracy. It is natural to assume that direct is better, more nearly perfect, than indirect-that the ideal of consent of the governed is better achieved by consenting to the laws themselves, rather than to representative lawmakers. This argument from the logic of democracy 33 surely has much to do with the initiative's popularity. When discussion turns from the abstract to messy reality, proponents' claims become more particular, and some range beyond the appeal to democracy into the realm of governmental efficiency. We have chosen to group concrete claims into four categories: empowerment of ordinary citizens; "good government" aims such as increased participation, involvement, and public discussion; majority rule as an ideal; and countering governmental inefficiency.
29. CRONIN, supra note 5, at 220-22. 30. SCHMIDT, supra note 18, at 225-26. For an overview of Colorado's adoption of the initiative, see generally CRONIN, supra note 5, at 52; Starr, supra note 15, at 9-20. 31. See CRONIN, supra note 5, at 52. 32. Id. 33. William Jennings Bryan lent his oratorical gifts to this proposition during the Progressive Era. See CRONIN, supra note 5, at 165-68. See also MAGLEBY, supra note 4, at 21; WALKER, supra note 6, at 50-52; Starr, supra note 15, at 11:
56
UNIVERSITY OF COLORADO LAW REVIEW A. Empowerment
[Vol. 66
Money was central to Progressive Era debates about the initiative. The most common claim was that the initiative was a way around monied influence on elected representatives.34 Corruption, especially of The Machine, had pride of place in every debate. The initiative would empower citizens to override legislatures held in thrall of wealthy patrons; it would "thrust 3 from power the Captains of Greed." 5 Direct democracy would allow citizens to enact measures to curb legal privileges of wealth. Populists saw the initiative as a means to enact redistributive measures.36 As originally envisioned, these aims have been very little realized. While states have enacted some redistributive measures, virtually all have been the work of legislatures. 7 Some anti-corruption campaigns have succeeded, but the initiative was not a prominent weapon in them.3 8 Moreover, it is generally agreed that money is at least as important in initiative campaigns as in elections. In California, money is now essential simply to get enough signatures to force a vote.39 On the other hand, the initiative tends to benefit social groups that are different from those served by representative democracy on an important class of issues. These arise where self-interest of the governing class is challenged. Such challenges usually relate to the efficiency of government, so we discuss them under that heading below. But they have a distant kinship with the original, populist aims of initiative proponents. B. Good Government In addition to the general appeal to the logic of democracy, proponents of the initiative tout good government aims of promoting greater citizen involvement and participation in government, of increasing and broadening public discussion of
34. See CRONIN, supra note 5, at 53-57. 35. CALIFORNIA COMM'N, supra note 9, at 33. 36. See CRONIN, supra note 5, at 54-57, 200-01. For general discussions of reformers' distrust of the political machine, see RIcHARD HOFSTADTER, THE AGE OF REFORM 257-67 (1955); MAGLEBY, supra note 4, at 21-25. 37. See HOFSTADTER, supra note 36, at 242-43, 268. 38. See generally id. 257-71. at 39. See CRONIN, supra note 5, at 196, 202; MAGLEBY, supra note 4, at 64, 76.
1995]
issues, and of reforming the procedures of representative government. Polls show a sizeable number of citizens who say that being able to vote directly makes them more likely to vote at all.4 Good government aims have been realized to some extent. Initiative campaigns do involve large numbers of people, many of whom are not involved in ordinary politics. Campaigns over controversial subjects attract media attention, probably more than the same issues before a legislature. 4 ' And some initiatives, particularly in Colorado, have improved government procedures in ways that representatives would not have done.42 However, circulating petitions has become a business in California, belying any image of amateur lawmaking.43 If this tendency spreads to smaller states, participation will diminish. Moreover, the total vote cast on initiative measures is usually somewhat lower than for representatives. That is, some who vote for representatives skip initiatives. Thus the promise of inspiring more citizens to vote may not work out in practice, with the exception of a few, very controversial issues." The falloff in votes cast on initiatives from those cast for representatives tends to increase with the length of the ballot, so one irony is that the more the initiative is used, the less effectively it involves more voters.4 Another good government claim is that availability of the initiative keeps representatives in line, that fear of the initiative induces better representation. However, there is little
40. 41.
See CRONIN, supra note 5, at 4, 11, 198, 202; MAGLEBY, supra note 4, at See SCHMIDT, supra note 18, at 26-29; CRONIN, supranote 5, at 198, 202.
2, 21-25, 28, 181-84; WALKER, supra note 6, at 52-54. 42. See, e.g., COLO. CONST. art. V, 20, 22a, 22b (1988 amendment requiring legislative committee votes on bill's merits and barring binding party
caucus); id. art. V, 46-48 (1974 amendment providing for legislative redistricting by reapportionment commission); id. art. VI, 20, 23-26 (1966 amendment
providing for merit selection of judges); Act effective Jan. 14, 1985, 1985 Colo. Sess. Laws 1793 (initiated statute to liberalize voting registration). 43. CALIFORNIA COMM'N, supra note 9, at 9, 12-13. For an examination of how the practice of gathering signatures has lived up to its original purpose, see
generally id. at 125-55. 44. On average the falloff is not large. CRONIN, supra note 5, at 67-68, 20910. See generally MAGLEBY, supra note 4, at 77-121 (analyzing who votes on ballot
propositions).
45. See MAGLEBY, supra note 4, at 90-95 ("voter fatigue"). But see CRONIN,
supra note 5, at 68-70 (phenomenon of ballot fatigue is exaggerated).
58
[Vol. 66
evidence that this is achieved by the direct initiative. It may result from the indirect initiative, where the first phase elicits consideration by the legislature. Studies in Switzerland, where direct democracy rights are most extensive and most often used, suggest that the legislature is more influenced by the threat of a referendum to rescind enacted statutes than by the initiative. This is because those who lose a vote in the legislature often begin the adoption of a rescinding referendum.4 6 C. Majority Rule An important variant of the concept that the initiative is a perfection of democracy is the claim that the initiative allows expression of pure majoritarian will, and that this is a virtue.4" The premise is true relative to laws passed by legislatures. Representative government in America is designed to avoid pure majority rule, for purposes of protecting minorities, promoting compromises, and insuring deliberation of issues. The initiative is a way around these barriers for at least some majorities. The initiative hardly enacts all majority preferences. To the extent that money is necessary for a successful initiative campaign, the initiative is not available to unorganized interests lacking financial backing. And those who actually vote on initiatives tend to be better educated and more affluent than those who vote for representatives or who do not vote at all.48 Some issues are passed based on genuine voter confusion.4 9 Nevertheless, initiatives surely enact majority will more effectively than do legislatures. Whether direct majority rule is a virtue is fiercely contested. It clearly was not for Madison and other framers; who spoke against it and deliberately designed the Constitution to avoid it. 0 But majority rule at some level remains the fundamental principle of democracy, as Madison often acknowl-
46. See CRONIN, supra note 5, at 10, 53; Blankart, supra note 14, at 90-91. 47. CRONIN, supra note 5, at 9. See also Dennis W. Arrow, Representative Government and Popular Distrust: The Obstruction/Facilitation Conundrum Regarding State ConstitutionalAmendment by Initiative Petition, 17 OKLA. CITY U. L. REV. 3 (1992). 48. MAGLEBY, supra note 4, at 103-09, 183; see also CRONIN, supra note 5,
at 67. 49. 50.
CALIFORNIA COMM'N, supra note 9,at 87-89. See text accompanying note 22; see also CRONIN, supra note 5, at 17-19.
1995]
edged. 5 Its opposite, minority rule, is plainly undemocratic. The question is how ultimate majority rule should be tempered to protect minority interests and ensure deliberation. Every successful initiative defines the minority who voted against it, as does every successful enactment by a legislature. Concern for all such minorities would put into question any majoritarian act and thus be contrary to democracy. Rather, concern for minority interests arises in those cases when a majority targets, distinct groups that are unpopular generally, groups that are consistent losers in the majoritarian process or are the subject of disabilities that are hard to explain in terms other than dislike. America's classic example was the Jim Crow system. Three examples adopted by initiative are Colorado's amendment barring legal protection for gays, California's law disabling Japanese farmers from owning land, and Oregon's law requiring Catholic children to attend public schools.52 Compare the initiative with the complex relationship between legislatures and minority interests. Recall the many times one has thought a legislature too responsive to a distinct electoral ("special") interest, such as farmers, or oil companies, or veterans, or environmentalists, or defense contractors, or a civil rights group. We can all think of legislation enacted or blocked because of intense support of a well-organized group whose preference would probably not be approved by a referendum of the whole electorate. There are two reasons why organized interests can, influence legislatures in this way. The more familiar is the corruption of financial support. When the initiative overcomes that, it is performing as its idealistic backers intend. The other reason, less often acknowledged in common discourse, is that legislatures reflect not only the number of a measure's proponents, but also the intensity of their preferences. In this way, minority interests are able to get their most strongly desired legislation passed, even when that legislation would not achieve majority support in a referendum. More important still, minorities can persuade legislatures to amend the parts of majoritar-
51. See, e.g., THE FEDERALIST Nos. 39, 49 (James Madison). 52. Each of the three initiatives was subsequently subject to judicial review. See Pierce v. Society of Sisters, 268 U.S. 510 (1925) (striking down Oregon initiative); Porterfield v. Webb, 263 U.S. 225 (1923) (sustaining California initiative); Evans v. Romer, 854 P.2d 1270 (Colo. 1993) (striking down Colorado initiative). See generally CRONIN, supra note 5, at 90-98, 212-14 (minority rights).
60
[Vol. 66
ian bills they find most objectionable, even if a referendum would give majority backing to those parts.53 That legislatures are able to reflect intensity of preferences as well as their popularity may make legislatures less efficient, a point taken up below. But it is also an important feature of mature democracy. Madison's solution to tyranny of the majority was layered representative government, in which no minority should consistently lose out.54 Legislatures have failed to achieve his aim in some circumstances, such as the long reign of Jim Crow. But initiatives can much more readily isolate minorities, as recent anti-gay measures show. Initiatives that target unpopular minorities are not a large percentage of citizens' initiatives, 5 but they are some of the best known. Many of these have been overturned by the courts on constitutional grounds, and proponents of the initiative sometimes argue that judicial review is sufficient protection." But there are several reasons for concern about relying solely on judicial review. First, the constitutional initiative neutralizes judicial review under the state's constitution, which would otherwise preserve more legal diversity, thereby nationalizing more issues. Second, judges are troubled by judicial review of initiatives. Some of them personally believe the initiative to be more democratic than the legislature and give it more deference. Others sense that overturning an initiative is more likely to incur popular wrath than overturning a legislative act. This inhibits the timid, particularly in states where recall of judges is allowed.5 7 D. Efficiency When the initiative was adopted during the Progressive Era, government in America was much smaller than it is today. Government internally was not a major target of reformers; rather, it was thought to be the tool of powerful private interests. Proponents of the initiative tended to be on the political left, more concerned about private power than public, and indeed, wanting to increase the ability of public power to
53. 54. 55. 56. 57.
See MAGLEBY, supra note 4, at 184-86. See THE FEDERALIST. Nos. 10, 51 (James Madison). See, e.g., CRONIN, supra note 5, at 92. See SCHMIDT, supra note 18, at 38. See Eule, supra note 2, at 1579-84.
1995]
improve the lot of have-nots.58 Thus attacking governmental inefficiency was not a prominent aim of original promoters of the initiative, although it had occasional mention. 9 In recent years, attacking the size and inefficiency of government has become a major purpose of initiators.6" Also, the initiative arguably reduces what economists call agency costs-the costs of supervising agents, here elected representatives and appointed bureaucrats, by a principal, here the electorate. More controversially, some economists, dubbed "Leviathan theorists," have argued that modern representative government has an inherent bias toward excessive budget size.6 Fully assessing all arguments about the relationship of the initiative to governmental efficiency presents many difficulties. Formidable questions include what definition of efficiency to apply and whether efficiency is a coherent concept in various contexts. However, the most serious and obvious class of problems is rather clearly connected to governmental efficiency. These are problems arising from self-interest of the governing classes, both elected and appointed, in their compensation, jobsecurity, power, and prestige. Agency costs are very likely to be high on this class of issues. The attentiveness of legislatures to minority interests has the virtue, already noted, of giving effect to the intensity of preferences, to which majorities are deaf. But that virtue comes at a cost because legislatures try to satisfy all wellorganized interests at their door, often by spending or protectionism at the expense of unorganized taxpayers or consumers. And legislators do this in large part for the self-interested purpose of garnering reelection support from organized groups. The initiative is deployed as a remedy for these costs, at least on behalf of taxpayers. In modern times, initiatives to limit governmental borrowing and taxing have become prominent, curbing legislators' ability to buy favor with organized groups. The job-security interest of legislators has also been
58. See MAGLEBY, supra note 4, at 21-25; CRONIN, supra note 5, at 43-59. 59. E.g., ROCKY MTN. NEWS (Denver), Nov. 3, 1910, at 1 (cartoon). 60. See CRONIN, supra note 5, at 203-07; SCHMIDT, supranote 18, at 125-45. 61. See Geoffrey Brennan, Tax Limits and the Logic of Constitutional Restrictions, in TAX AND EXPENDITURE LIMITATIONS 121 (Helen F. Ladd & T. Nicolaus Tideman eds., 1981); Gordon Tullock, Problemsof Majority Voting, 67 J. POL. ECON. 571 (1959).
62
[Vol. 66
the target of initiatives on reapportionment, term limits, and campaign finance regulation.62 Claims to improve governmental efficiency must reckon the cost of direct democracy itself. Initiatives incur administrative costs to review, verify, and print petitions, print ballots, and
administer more elaborate, and sometimes more frequent, elec6 tions. Moreover, there are gains from agency that can be lost when laws are made by initiative. It is cheaper for a few officials to become fully informed about an issue, or expert in an activity, than for the electorate as a whole. To sustain a claim of efficiency for the initiative, these costs must be exceeded by gains in governmental efficiency flowing from initiated measures. Gains seem likely in the discrete class of issues directly related to the self-interest of legislators and bureaucrats, but are more doubtful for other questions. The Leviathan theorists make a theoretical case for greater efficiency under direct democracy than under representative government." We have found no good studies to test whether their claim has empirical support. There is abundant political commentary, particularly about California. All agree that the state's tax-cutting initiatives have succeeded in reducing taxes and government spending. But some argue that the crudeness of initiatives, and their use to mandate spending priorities as well as tax cuts, have led to less efficient government.65 One very crude question about efficiency of government is its per capita cost. Higher cost does not equate with inefficiency, because it may correspond with better services. Yet the suspicion that higher average costs mean less efficiency is widespread.66 Thus we compared the per capita cost of state government in states with and without the initiative. Initiative
62. See Ranney, supra note 5, at 78; CALIFORNIA COMM'N, supra note 9, at .72-73; SCHMIDT, supra note 18, at 15, 30-34, 51-54, 108. 63. See CALIFORNIA COMM'N, supra note 9, at 121, 126, 243. In every state,
initiative proponents bear the costs of circulation, which can be considerable. In 1994, an unsuccessful Colorado initiative, Amendment 12, sought to shift this cost to the state. 1994 Amendment 12 (rejected by Colorado voters Nov. 8, 1994) (on file with the Colorado Secretary of State). 64. See Arthur T. Denzau et al., On the Initiative-Referendum Option and
the Control of Monopoly Government, in TAX AND EXPENDITURE LIMITATIONS, supra
note 61, at 191. 65. See, e.g., Peter Schrag, California'sElected Anarchy, HARPER'S MAG., Nov. 1994, at 50. 66. See, e.g., Helen F. Ladd, Introductionto TAX AND EXPENDITURE LIMITATIONS, supra note 61, at 2.
1995]
states are somewhat above average in taxing and spending, so the initiative does not seem to curb taxing and spending better than the representative process in any absolute sense.6 7 Studies of states that have initiated tax limits show significant reductions in government spending over time,6 8 but we have not found any study that compares efforts to control spending through the initiative, such as in California, with efforts to do so legislatively, such as that presently underway in New Jersey. 9 Other claims that initiatives improve efficiency are occasionally made. There are issues that deadlock the representative process, and the initiative can be a way to resolve them. The most obvious and dramatic examples are tax revolt initiatives in California and Colorado, which followed periods of legislative stalemate.7 Some argue that the threat of an initiative deters legislators and bureaucrats from self-dealing. It is hard to substantiate this claim. Swiss practice shows that the rescinding referendum has an important effect on the legislative process, deterring passage of bills by thin majorities because of the threat that the legislators who voted against the bill will promote a rescinding referendum. 7 ' However, it is difficult to know whether that threat promotes more efficient government. It could have the opposite effect by inducing the majority to buy off more legislative minorities.
67. See U.S. BUREAU OF THE CENSUS, U.S. DEP'T OF COMMERCE, PUB. No. GF/91-5, GOVERNMENT FINANCES: 1990-91, at 101-09 (1993) (computations on file with authors). 68. See, e.g., DEAN STANSEL, CATO INSTITUTE, TAMING LEVIATHAN:
TAXING AND SPENDING LIMITS THE ANSWER? (1994). ARE
69. On California, see Schrag, supra note 65, at 50. On New Jersey, see Christine Scissorhands, THE ECONOMIST, Apr. 23, 1994, at 32; Jim Florio's Toughest Fight, THE ECONOMIST, May 2, 1992, at 38. 70. On California, see CALIFORNIA COMM'N, supra note 9, at 9. On Colorado, see S. Con. Res. 2, S. Con. Res. 6, S. Con. Res. 7, S. Con. Res. 8, S. Con. Res. 9, H. Con. Res. 1002, 58th Colo. Gen. Assembly, 1st Sess. (1991) (failed tax limiting measures). 71. See Blankart, supra note 14, at 91.
64
[Vol. 66
IV. "THE DEVIL'S IN THE DETAILS": 72 THE DESIGN OF INITIATIVE PROCEDURES WITH A SPECIAL Focus ON COLORADO
A. Introduction 1. Four Basic Procedural Issues States with initiatives must specify procedures for initiating a popular measure. The procedures selected have their roots in
state constitutions, enabling legislation, and judicial decisions.
The ease or difficulty of qualifying an initiative for the ballot is

based, at least in part, on an important policy question regarding the seriousness of ballot measures. There is a general assumption that initiators ought to be able to demonstrate substantial public support before the state is put to the expense of conducting a vote. Also, proliferation of ballot measures reduces citizens' understanding. States have chosen the petition process as the primary method of restricting ballot access. This process requires a minimum number of registered voters to sign petitions requesting state officials to place any given measure on the ballot. The process is supplemented by procedures to detect obvious petition fraud. States must also decide on judicial review of these procedures, particularly for review that might delay a vote. Beyond qualifying procedures, regulation of initiatives is concerned with three related issues: minimizing voter confusion, defining the extent of legal change that can be made in a single initiative, and informing voters on the issues. A related subject on the last issue is the delicate matter of controlling the influence of money. The first two questions are closely related because voter confusion increases with the length and complexity of an initiative. For an extreme example, in 1948 California citizens proposed an initiative that exceeded 21,000 words and would have rewritten much of the state's constitution.7 3
72. H. Ross Perot. See Steve Whitworth, Clinton, Cabinet Takes Economic Message to the People, UPI, Feb. 18, 1993, available in LEXIS, Nexis Library,
CURNWS File (Perot's comment on President Clinton's economic recovery plan).

73. See McFadden v. Jordan, 196 P.2d 787 (Cal. 1948).
1995]
The judiciaries of initiative jurisdictions have parallel concerns and issues of their own. Can substantive questions about initiatives be reviewed before a vote or only after? How are courts to determine legislative intent of initiated measures? In constitutional review, should courts treat initiated measures with greater or less deference than given to ordinary legislation? Are initiatives a unique threat to minority rights?74 An issue that has become prominent in California is how courts should interpret conflicting initiatives that are both enacted, v the so-called counter-initiative question." Our interest here is regulation of initiatives. We are not directly concerned with substantive judicial review. Therefore, some of the judicial questions stated above are beyond the scope of this article. 2. The Colorado Experience with Initiatives In 1910 the citizens of Colorado voted to amend the Colorado Constitution, adding a procedure empowering citizens to initiate and pass constitutional amendments and laws.76
74. See generally Eule, supra note 2, at 1503. 75. See CALIFORNIA COMM'N, supra note 9, at 306-12. See generally, K. K. DuVivier, By Going Wrong All Things Come Right: Using Alternate Citizen Initiatives to Improve Citizen Lawmaking, 63 U. Cin. L. Rev. (forthcoming 1995). 76. Colorado citizens may, by petition, initiate laws or constitutional amendments that become effective if passed by majority vote. See COLO. CONST. art. V, 1; COLO. REV. STAT. 1-40-101 to -133 (Supp. 1994). [T]he people reserve to themselves the power to propose laws and amendments to the constitution and to enact or reject the same at the polls independent of the general assembly. The ... power hereby reserved by the people is the initiative.
... All elections on measures initiated by... the people ... shall become the law or a part of the constitution, when approved by a majority of the votes cast thereon, and not otherwise, and shall take effect from and after the date of the official declaration of the vote thereon by proclamation of the governor, but not later than thirty days after the vote has been canvassed. COLO. CONST. art. V, 1(1), 1(2), 1(4). The Colorado Constitution also reserves to its citizens the power to submit, by petition, referendums on laws enacted by the General Assembly. Id. art. V, 1(3). The power, however, does not apply to laws that the General Assembly has declared "necessary for the immediate preservation of the public peace, health or safety" (an emergency clause) and to laws that contain "appropriations for the support and maintenance of the departments of state and state institutions." Id. Since the General Assembly routinely attaches an emergency clause to its bills invoking the public safety exception, and the Colorado courts have refused to look behind the declaration, referendum by citizen petition has not occurred since 1932, when the electorate rejected a legislative measure increasing the tax on
66
[Vol. 66
Since that time there have been over 160 ballot initiatives to amend the constitution and approximately sixty ballot initiatives to enact laws." The clear trend is towards the increasing use of the initiative power, with an emphasis on proposals to amend the constitution.78 Since 1976 there have been over eighty constitutional initiatives (over one half of the total of all
oleomargarine. See, e.g., In re Interrogatories by Governor, 181 P. 197 (Colo. 1919); Van Kleeck v. Ramer, 156 P. 1108 (Colo. 1916); In re Senate Resolution No. 4, 130 P. 333 (Colo. 1913); see also American Constitutional Law Found. v. Meyer, 33 F.3d 62 (10th Cir. 1994) (unpublished disposition), reported at 1994 U.S. App. LEXIS 19136 (10th Cir. July 27, 1994) (use of emergency clause does not violate the First Amendment). Colorado also sports a complex system of initiatives and referendums at the local level. The Colorado Constitution provides that the initiative and referendum powers are also "reserved to the registered electors of every city, town, and municipality as to all local, special, and municipal legislation of every character in or for their respective municipalities." COLO. CONST. art. V., 1(9). The Colorado Constitution also provides for the amendment of home rule municipal charters and for adoption of "any measure" by initiative. Id. art. XX, 5. All home rule charters must include referendum and initiative provisions. Burks v. City of Lafayette, 349 P.2d 692, 694 (Colo. 1960). Interestingly, the procedures for local initiatives deviate in substantial respects from the procedures for state wide initiatives, demonstrating that there is substantial disagreement among even initiative proponents in Colorado on the best procedures for implementing the process. The General Assembly has established an indirect initiative system for local governments. COLO. REV. STAT. 1-40-127 to -129 (Supp. 1994). Five percent of the registered electors of a city or town jurisdiction can, by petition, submit a proposed ordinance to the local legislature. If the legislature refuses to adopt the proposal "without alteration" within twenty days (or refuses to override an executive veto), id. 1-40-128, the proposed ordinance is put to the electorate at a regular or special election within one hundred and fifty days. Id. 1-40-129. A second difference is the treatment of initiatives that generate local ordinances. Once a measure is adopted by initiative, it can be amended or repealed only by initiative. COLO. CONST. art. XX, 5. 77.. A tally of all initiatives, successful and unsuccessful, is on file with the authors. This paper cites Colorado initiatives by ballot number and year (e.g., 1994 Amendment 12). The text of Colorado initiatives is available through the Colorado Secretary of State's office. Where initiatives have been formally incorporated into the Colorado Constitution, the appropriate article and section numbers are also cited. 78. The reason for the modern popularity of constitutional initiatives and the declining popularity of statutory initiatives is easy to understand; the petition and vote requirements are the same for both, but a constitutional amendment by initiative is more difficult to alter or repeal. It may be altered or repealed only through a subsequent constitutional amendment (either initiated by the people or referred by the General Assembly or by a constitutional convention). COLO. CONST. art. V, 1(1) (initiative and referendum power), art. XIX, 1 (constitutional convention). Initiatives that enact laws may not be vetoed by the Governor but may be altered or repealed by subsequent legislation passed by the General Assembly and signed by the Governor as well as by subsequent initiated laws. Id. art. V, 1(4).
1995]
amendment initiatives proposed since 1910) and only ten statutory initiatives.7 9 Of these modern initiatives, about half of the amendment proposals have passed, but only three statutory initiatives have passed. 0 The more important of the successful initiatives to amend the Colorado Constitution include amendments changing procedures for recalling state officials, providing for appointed rather than elected judges, setting home rule for cities and towns, establishing a- state civil service, repealing state Prohibition, creating a reapportionment commission to draw state legislative districts, limiting the power of cities to annex territories without a favorable vote of those in the territory annexed, setting eight-year term limits for all state elected officials, and establishing taxing and spending limitations on all government districts. The popularity of constitutional initiatives has caused the Colorado Constitution to be lengthened substantially with language one would not expect to find in the constitution of a sovereign government. Examples of unusual constitutional subjects added by initiatives include participation in the 1976 Winter Olympics,"' gambling for selected cities, 2 using selected tax revenues only for roads, 3 old age pensions of one hundred dollars a month for Colorado citizens, 4 nuclear detonations, 5 preference for veterans on civil service exams, public funding for abortion, use of lottery funds for state parks, 8 busing to achieve racial balance, 9 and English
79. See supra note 77. 80. In 1980 a proposal to create a Regional Transportation District passed; in 1984 a voter registration act passed; and in 1992 an act prohibiting the hunting of black bears by the use of bait or dogs or in the spring passed. COLO. REV. STAT. 32-9-103(3.5), -109.5, -111, -112, -117 (Supp. 1994) (Regional Transportation District); id. 33-4-101.3 (Supp. 1994) (black bear hunting); Act effective Jan. 14, 1985, 1985 Colo. Sess. Laws 1793 (voter registration). See also supra note 77 (since 1984, 26 of 30 initiatives have attempted to amend the Colorado Constitution). 81. COLO. CONST. art. X, 20, art. XI, 10 (both repealed). 82. Id. art. XVIII, 9. 83. Id. art. XI, 3. The section also contains very technical limitations on the state's bonding authority. 84. Id. art. XXIV, 6. 85. Id. art. XXVI. 86. COLO. CONST. art. XII, 15. 87. Id. art. V, 50. 88. Id. art. XXVII (creating the "Great Outdoors Colorado Program"). 89. Id. art. IX, 8.
68
[Vol. 66
as the official language for the State. 90 In the latest election, proposals failed that would have added provisions on workers compensation and a special tobacco tax.9' Recently, however, initiative petitioners, recognizing the power and permanency of amendments on the structure of
government, have proposed constitutional amendments that

relate to the election process and the general powers and responsibilities of government officials. In 1992 and 1994, state ballots included constitutional initiative proposals on limiting taxing and spending powers of government absent a ratifying vote,9 2 on election reform,9 3 on recall of state judges,94 on initiative and referendum procedures,9 5 and on term limits for elected officials.96 The Colorado Constitution contains selected procedural requirements, 97 but most of the details of Colorado initiative procedure are found in legislation supplementing the constitutional language. 98 Recently, the General Assembly passed two
90. Id. art. II, 30a. 91. 1994 Amendments 11 and 1 (both rejected by Colorado voters Nov. 8, 1994) (on file with Colorado Secretary of State). 92. COLO. CONST. art. X, 20 (Taxpayer's Bill of Rights). This provision itself requires a vote of the people on every measure to increase taxes, borrowing, or government spending. The required ballot measure can be referred by a legislative body or initiated by citizens.
93.
LEGISLATIVE COUNCIL OF THE COLO. GEN. ASSEMBLY, AN ANALYSIS OF
1994 BALLOT PROPOSALS, RESEARCH PUB. No. 392, at 20 (1994 Amendment 12), 41 (1994 Amendment 15) [hereinafter LEGISLATIVE COUNCIL] (both amendments were rejected by Colorado voters Nov. 8, 1994) (full text of amendments on file with Colorado Secretary of State). 94. Id. at 20 (1994 Amendment 12). 95. Id. 96. COLO. CONST. art. XVIII, 9(a), 11 (1994 Amendment 17) (adopted by Colorado voters Nov. 8, 1994). 97. Article V, 1, in several separate bits of language, generally empowers the General Assembly to enact implementing legislation, specifically empowers the Secretary of State of prescribe the form of petitions, and specifically empowers the General Assembly to pass laws on publication. In subsection (2), the General Assembly may prescribe the "form" or, in subsection (7) "all matters pertaining to the form" of initiative petitions and, in an odd bit of language in Subsection (7), may, through "general laws" guide the Secretary of State on "submitting" initiatives "to the people for adoption or rejection at the polls." Subsection (7.3) states that the "form and manner of publication shall be as prescribed by law." Under subsection (6), the Secretary of State can designate or prescribe the "general form" of the "top" of the "printed or written" petition "sheets." Subsection (5.5) impliedly authorizes the ballot title statutes. 98. The courts have interpreted the general grants of legislative authority to authorize the General Assembly to pass statutes that "prevent fraud, mistake or abuse in the initiative process," but the statutes may not "create an undue
1995]
major overhauls of the implementing legislation, one in 198999 and' another in 1993.100 Occasionally, as this year, the Gener-
al Assembly proposes constitutional amendments, complete with contingent legislation, to amend the initiative process. 10 ' The implementing legislation must, of course, be consistent with the constitutional language, and according to the courts, the General Assembly has overstepped its constitutional authority 02 with some regularity over the years. 1 Colorado answers the four basic procedural questions in the following manner. First, in accord with all other initiative jurisdictions, Colorado limits access to the ballot through a petition requirement. Registered voters, equal in number to five percent of the total number of votes cast for all candidates for the office of secretary of state at the previous election, must sign a petition requesting that specified language be put on the ballot to amend the constitution or enact a law. With the petition requirement comes, of necessity, a host of supplement-
burden on the exercise of the initiative process." See, e.g., Committee for Better Health Care for All Colo. Citizens v. Meyer, 830 P.2d 884, 893 (Colo. 1992). The General Assembly has passed Article 40, in its current version a 34 page, 35 section provision on initiative procedures. COLO. REV. STAT. 1-40-101 to -133 (Supp. 1994). The procedures specified are complex and the time deadlines short. Only the most committed or the best advised members of the public are able to negotiate the maze. 99. Act effective June 10, 1989, ch. 42, 1989 Colo. Sess. Laws 319. 100. Act effective May 4, 1993, ch. 183, 1993 Colo. Sess. Laws 676. 101. LEGISLATIVE COUNCIL, supra note 93, at 2 (1994 Referendum A), 5 (1994 Referendum B) (adding subsections (5.5), (7.3) and (7.5) and amending old subsection (7) to art. V, 1 of the Colorado Constitution) (both adopted by Colorado voters Nov. 8, 1994). 102. E.g., American Constitutional Law Found. v. Meyer, No. 93-M-1467, 1994 U.S. Dist. LEXIS 17134 (D. Colo., Nov. 23, 1994) (requirement that paid circulators wear badges and that proponents report their names, addresses, and payments invalid); Meyer v. Grant, 486 U.S. 414, 420 (1988) (prohibition on the use of paid petition circulators violates First Amendment); Urevich v. Woodard, 667 P.2d 760, 763 (Colo. 1983) (dictum that prohibition on paying petition circulators for soliciting contributions would be void); Francis v. Rogers, 514 P.2d 311, 313 (Colo. 1973) (for municipal referendum petitions, qualified electors need not be registered voters to sign a referendum petition); Colorado Project-Common Cause v. Anderson, 495 P.2d 220, 223 (Colo. 1972) (statute requiring petition circulators and petition signers to be registered voters void); Colorado ProjectCommon Cause v. Anderson, 495 P.2d 218, 219 (Colo. 1972) (law requiring publication expense to be borne by petitioners void); Yenter v. Baker, 248 P.2d 311, 317 (Colo. 1952) (requirement that petitions be filed eight months before an election void); Baker v. Bosworth, 222 P.2d 416, 419 (Colo. 1950) (requirement increasing the minimum petition signatures void); In re House Resolution No. 10, 114 P. 293, 296 (Colo. 1911) (legislation on publication void).
70
[Vol. 66
ing rules designed to deter and detect petition fraud and other petition irregularities. Second, to minimize drafting problems, Colorado requires proponents to submit to a non-binding consultation procedure with state officials. Third, Colorado's efforts to notify citizens on the content of initiatives consist of an elaborate ballot title procedure, newspaper publication of the text, an official summary of initiatives, and a pamphlet mailed to all registered voters containing the text and an official summary of initiatives. Fourth, until 1994, Colorado had no regulation of the scope of initiatives, but a successful 1994 referendum amended the constitution to add a single-subject 13 0 limit. B. The Petition Process as a Gate to Ballot Access 1. Introduction to the Petition Process
The process of qualifying an initiative for the ballot varies greatly among jurisdictions, for both policy and practical reasons. States allowing only the indirect initiative require that a measure first be considered by the legislature, necessarily slowing the process.' 4 Even in direct initiative states, the time periods to qualify for the ballot vary substantially.' The most central part of the qualifying process is gathering voters' signatures on petitions, and the number of qualifying signatures has a major effect on the frequency of citizens' resort to the initiative. Some states set high qualifying numbers, or set significantly higher numbers for constitutional than for statutory initiatives.' s In others, the number is low enough 0 that qualifying is not difficult. 1 7 The size of a state's population has a great effect on the qualifying process. California requires a higher percentage of voter signatures to qualify a constitutional initiative than does
103. See LEGISLATIVE COUNCIL, supra note 93, at 2 (1994 Referendum A) (providing summary of proposed referendum); see also infra part IV.D.2. 104. See supra note 7 and accompanying text. 105. See CALIFORNIA COMM'N, supra note 9, at 130, 361 (from 90 days in Oklahoma to unlimited time in five states). 106. See id. at 130, 361-62 (all but two states require more signatures for constitutional than for statutory initiatives). 107. See id. Colorado is the easiest state in which to qualify a constitutional initiative and may also be the easiest for statutory initiatives.
1995]
Colorado.' 8 In California, the actual number needed is over 600,000109 -about eleven times Colorado's. The logistics of gathering over half a million signatures are significantly different from 55,000. Signature gathering is often an amateur enterprise in Colorado, but it is mostly a commercial venture in California. Signatures must be gathered personally in Colorado, 10 but California allows solicitation by mail."' The process of administrative review of a proposed initiative can be very complex and can cause difficulties for those unfamiliar with it. Knowledgeable opponents can seek judicial review at strategic moments to cause substantial and sometimes fatal delay to an initiative." 2 2. The Colorado Petition Procedure As noted above, ballot access for initiatives is limited by a petition requirement. After the fixing of the ballot title, submission clause, and summary, proponents have no more than six months to gather signatures and file their petition with the Secretary of State."' They must then file completed petitions at least three months before the election." 4 The constitution requires "signatures by registered electors in an amount equal to at least five percent of the total number of votes cast for all candidates for the office of secretary of state at the previous general election" for an initiative petition., 5 A series of provisions in the Colorado Constitution deals with the form of the petition itself. Every petition must include the full text of the measure proposed," 6 and signers must be "proper persons" who sign for themselves and the signatures must include a residential address and date." 7 Petition circulators must be registered electors and must attach a notarized affidavit to each petition attesting that "each signature thereon is the signature of the person whose name it purports to be and that, to the best of the knowledge and belief of the
108. Id. at 130. 109. Id. at 126. 110. See infra note 119 and accompanying text.
111. CALIFORNIA COMM'N, supra note 9, at 137-42, 151-55.
112. See infra text accompanying note 238. 113. COLO. REV. STAT. 1-40-108 (Supp. 1994). 114. COLO. CONST. art. V, 1(2). 115. Id. 116. Id. 117. COLO. CONST. art. V, 1(6).
72
[Vol. 66
affiant, each of the persons signing said petition was, at the time of signing, a registered elector."" 8 Supplementing legislation adds that persons circulating petitions must do so in person. 9 Circulators do not have to know the signers personally. 2 ' A provision recently held unconstitutional by a federal district court had required circulators to wear badges identifying themselves and designating whether they were volunteer or paid circulators.' 2 ' Each circulator must sign, date, and have notarized an affidavit containing information about the circulator's address and status as a registered voter.'2 2 The required form for petitions is cumbersome. Each individual petition for a single initiative, defined as a "section," 2 ' must begin with pages containing a specified warning statement, 24 the title, the summary, the ballot title (which in large part repeats the title), and the proposed language. 2 5 The top of each succeeding page, on which signatures are affixed, must repeat the warning statement and the ballot title (or title for a local ballot issue). 26 The final page must contain the affidavit of the petition circulator.'2 7 Upon receiving the petitions, the Secretary of State has thirty days to check the form of the petitions. The Secretary of State verifies that the petitions are prepared on printed forms,
118.
119.
Id.
COLO. REV. STAT. 1-40-111(2), -112(1), -113(1) (Supp. 1994).
120. Brownlow v. Wunsch, 83 P.2d 775, 781 (Colo. 1938). 121. COLO. REV. STAT. 1-40-112(2) (Supp. 1994). This requirement was held to violate the First Amendment. American Constitutional Law Found. v. Meyer, No. 93-M-1467, 1994 U.S. Dist. LEXIS 17134 (D. Colo. Nov. 23, 1994), appeal docketed, No. 94-1576 (10th Cir. Dec. 16, 1994). 122. COLO. REV. STAT. 1-40-111(2) (Supp. 1994). See Committee for Better Health Care for All Colo. Citizens v. Meyer, 830 P.2d 884, 893-94 (Colo. 1992) (requirements do not violate Colorado Constitution). 123. COLO. REV. STAT. 1-40-102(6) (Supp. 1994). 124. Id. 1-40-110. The warning statement contains two parts. First, it warns that it is against the law to forge another's name, to sign more than once, or to sign knowing the signer is not a registered elector. Second, the statement commands that a voter not sign unless the voter has read the proposed initiative, or a summary thereof, and understands its meaning. 125. Id. 1-40-102(6). Petition circulators have reported that no one reads the text or summary of initiatives in their petitions and that the warning would, if enforced, disqualify all signers on all petitions. See JoElyn Newcomb and audience, Tape of Conference on Governing by Initiative, sponsored by the University of Colorado School of Law (Sept. 23, 1994) (on file with authors). 126. COLO. REV. STAT. 1-40-102(6) (Supp. 1994). 127. Id.
1995]
include the warning, contain no extraneous material, show no evidence of disassembly,'28 and have sections bound by the hundreds. The Secretary must also verify that petition circulators have signed, notarized, and dated appropriate affidavits. Finally, the Secretary must sample at least five percent or four thousand signatures, whichever is greater.'29 The Colorado Supreme Court has held that the constitutional requirement that a properly verified petition "shall be prima facie evidence that the signatures thereon are genuine and true and that the persons signing the same are registered electors"' does not prohibit the Secretary of State from investigating the validity of signatures of circulators and petition signers.' 3 ' If the Secretary of State declares that a petition does not have a sufficient number of valid signatures, the proponents have fifteen days to gather additional signatures. 3 2 Parties contesting the Secretary's finding on the sufficiency of signatures may file a protest within thirty days in district court with an appeal to the Colorado Supreme Court. 3 The Colorado Supreme Court has ruled that the Secretary's rejection of petition signatures was arbitrary and capricious3 and has
128. See id. 1-40-113(2); Elkins v. Milliken, 249 P. 655 (Colo. 1926) (evidence of disassembly of sections destroys the integrity of each section). 129. COLO. REV. STAT. 1-40-116(4) (Supp. 1994). Random sample results are extrapolated to the full number of signatures and used to make one of three determinations of the petition's validity. If the random sample establishes that the number of valid signatures is less than or equal to 90% of the number needed, the petition "shall be deemed to be not sufficient." Id. If the sample established that the number of valid signatures is equal to or greater than 110% of the number needed, the petition "shall be deemed sufficient." Id. For anything between, the Secretary "shall order the examination and verification of each signature filed." Id. See also infra text accompanying notes 141-43. 130. COLO. CONST. art. V, 1(6). 131. Committee for Better Health Care for all Colo. Citizens v. Meyer, 830 P.2d at 884, 895 (Colo. 1992). 132. COLO. REV. STAT. 1-40-117(3)(b) (Supp. 1994). See Montero v. Meyer, 795 P.2d 242 (Colo. 1990) (amended petition may be filed within three months of election, but case decided under prior law before 1-40-117(3)(b) was adopted). 133. COLO. REV. STAT. 88 1-40-118, -119 (Supp. 1994). The protestor must plead specific defects in the Secretary of State's signature verification results and has the burden of proof in the hearing. Id. 134. Committee for Better Health Care, 830 P.2d at 884, 896-97, 898-99 (reversing Secretary of State's rejection of petition forms, where Secretary assumed that initiative sections with extra staple holes had been disassembled, and Secretary rejected affidavits that had different dates for the signature and notarization).
74
[Vol. 66
also ruled that the Secretary of State's acceptance of signatures was unreasonable. 1 35 In most cases, however, the supreme 6 13 court has sustained the Secretary's determinations. 3. Comments on the Colorado Petition Process Drafters of the Colorado Constitution limited ballot access by requiring, as a condition to appearing on the ballot, the signatures of a significant number of registered voters. But the petition process is open to fraud and misrepresentation. Those gathering signatures can forge them (copy them from telephone books), or solicit signatures from individuals who lack'capacity to sign (individuals who are not registered voters in the appropriate jurisdiction), or solicit signatures from individuals under false pretenses. The last category includes clearly actionable conduct by those soliciting signatures, such as reattaching signatures solicited for one petition to another, altering petition language after signatures have been solicited, and the like. It also includes conduct that is borderline illegal, like oral misrepresentations or material omissions about the content and effect of the initiative to induce individuals to sign otherwise accurate and legally compliant petition forms. Initiative proponents do not have a monopoly over questionable behavior in the petition process. Opponents of initiatives have been known to offer to pay signature solicitors, particularly those that are paid solicitors, more per signature to destroy already gathered signatures than the solicitors will make per signature if they turn in their petitions."' Moreover, proponents of one initiative, needing paid solicitors to gather signatures, have hired signature solicitors off the street who are doing the bidding of proponents of another initiative. The Secretary of State's office reports that the incidence of 18 petition fraud in Colorado appears to be on the increase.
135. Elkins v. Milliken, 249 P. 655 (Colo. 1926) (invalidating signatures where petition sections separated and altered). 136. E.g., Billings v. Buchanan, 555 P.2d 176 (Colo. 1976) (validating petition forms distributed in newspapers); Haraway v. Armstrong, 36 P.2d 456 (Colo. 1934) (improper actions of one circulator did not invalidate all signatures collected by that circulator). 137. This is illegal if the petition signatures gathered are not returned to the original employer. COLO. REV. STAT. 1-40-131 (Supp. 1994). 138. See Catharyn Baird, Tape of Conference on Governing by Initiative, supra note 125. In 1982, the Secretary of State ordered a controversial casino gambling proposal removed from the ballot. Secretary of State Rejects Petition
19951
The use of circulators paid on a commission basis by signature, illegal in Colorado until a 1988 Supreme Court opinion,'39 is one of the reasons for the trend. Currently a single citizen with an extra $40,000, fifty cents a signature for eighty thousand signatures, has a good chance of putting an initiative on the state-wide ballot using paid solicitors. The state legislature has applied a minor salve by requiring paid solicitors to identify themselves. 14o The State's main protection against forged signatures and signatures of unregistered voters is not signature verification from filed signature cards and the like. There are simply too many signatures on too many petitions to check them individually. Instead, the Secretary of State's office has relied on cross checks of a random sample of signers against local voter rolls. If a signed name and address do not match the name and address on a voter roll, the signature is rejected. Some critics argue that the process is too easily abused.' 4 ' For example, an unscrupulous petition circulator could forge signatures equal in number to 125% of the minimum required, gathering them from a phone book (or other mailing lists that can be bought commercially) and assuming that seventy percent of the people listed were registered voters and that the names and addresses in the phone book match local voter rolls. Only careful attention to ink color and handwriting and the like would detect the fraud. Other critics contend that the random sample process results in the rejection of too many valid signatures for Some signatures technical errors that ought to be excused.' are rejected because there is not a perfect match between the petition signature and address and the voter rolls. There can be
Effort, UPI, Sept. 10, 1982, available in LEXIS, Nexis Library, ARCNWS File. Testimony indicated that some names were forged and others added after petitions were notarized. Id. Petitions were left in bars and many names appeared in alphabetical order. Id. 139. Meyer v. Grant, 486 U.S. 414 (1988). See discussion infra note 311. 140. The rule is currently of questionable constitutional status. See supra note 121 and accompanying text. 141. Catharyn Baird, Tape of Conference on Governing by Initiative, supra note 125. 142. This is a common complaint of Colorado initiative activist Douglas Bruce and led him to propose an initiative in 1994 (Amendment 12) that, among other things, would have stopped the Secretary of State from using the random sample process. The initiative failed. See supra note 93.
76
[Vol. 66
a registration error in the voter rolls or an error in the petition signature, or the signature, although accurate, may not be a perfect match with the form of the voter rolls (when, for example, a signer uses a middle initial and a full name appears on the rolls). Some signatures are rejected because voting rolls have yet to be updated to correspond with a change of address. Proponents can contest the rejection of these signatures, although this is difficult, or simply use the fifteen-day grace period to gather others. The results of the sample are extrapolated to the full number of signatures for an evaluation of whether the initiative meets the minimum signature requirements. To protect their initiatives, petitioners solicit excess signatures (if 55,000 is the minimum, they will file 80,000) so that disqualification of some signatures (in this example, a twenty-nine percent error rate) would leave the petitions with sufficient signatures.' The effect of the random sampling process is to induce proponents to tender substantially more than the minimum number of required petition signatures. With the large numbers involved and the limited staff of the Secretary of State's office, however, the verification process using the random sample seems a reasonable compromise. C. DraftingInitiatives: Amateurism, Ardency, and Inadequate Review 1. Introduction to the Drafting Problem
There are four separate drafting problems. First, initiatives can be ineptly written: internally inconsistent, full of lacunae, incoherent, or leading to unintended consequences. Second, initiative drafting does not allow room for compromise with targeted (and often minority group) opponents. Third, initiatives can be mischievous; they can be intentionally misleading in their complexity or opaqueness. Fourth, when several initiatives pass that contain conflicting individual provisions, the result can be a patchwork monster of regulations. The initiative with the most votes pre-empts contrary
143.
Newcomb, Tape of Conference on Governing by Initiative, supra note
1995]
provisions of any other initiatives that pass with lesser majorities,' but a fragmentary, maimed residue of the other initiatives may survive. 4 5 The problem of inept drafting arises in large part from the way initiatives are drafted. A group of citizens comes up with a draft provision. The group can be small, and they are likely to be very strong partisans of their proposal. They can be inexperienced in the drafting of legislation. They can work in secret. After some procedural steps, their draft is printed on petitions for circulation to voters. From that moment, the draft is frozen.'4 6 Both petition signers and voters must accept or reject the proposal as is. The contrast with representative democracy could not be starker. When a legislative bill is introduced, it may get review from a legislative counsel or administrative department. It will often go to a committee that includes members of both major parties. It will undergo hearings, at which both expert and lay witnesses testify. Opponents will have an opportunity to explain what parts of the bill are most objectionable to them and to make amendments, and this will often result in changes. Unintended consequences of the bill will often be brought to the sponsors' attention. Possible costs of the bill will be estimated and discussed, and more amendments may result. Ambiguities will be pointed out and clarified. The process will be more or
144. COLO. REV. STAT. 1-40-123 (Supp. 1994). "A majority of the votes cast thereon shall adopt any measure so submitted, and, in case of adoption of
conflicting provisions, the one which receives the greatest number of affirmative
votes shall prevail in all particulars to which there is a conflict." Id. (emphasis as added). 145. In the 1994 election an initiative (Amendment 12) had language in
conflict with both another initiative (Amendment 15) and a referendum (Referendum A). Amendments 12 and 15 contained conflicting provisions on campaign finance reform, and Amendment 12 and Referendum A conflicted on the single-subject rule for initiatives. See LEGISLATIVE COUNCIL, supra note 93. If all three had passed with different majorities, we would be left with bits and pieces of the least successful provisions on the books, a result no one wanted. For example, Amendment 12, with multiple issues, could have pre-empted Referendum A, with a single issue, although one could argue that the result should be the opposite because voters were voting for Amendment 12 on other grounds and disagreed with Amendment 12's position on the Referendum A issue.
146. See, e.g., COLO. CONST. art. V, 1(5); COLO. REV. STAT. 1-40-105(2)
(Supp. 1994). But see id. 1-40-120 (allowing later amendment to comply with
federal law). See generally CALIFORNIA COMM'N, supra note 9, at 92-96, 99-107.
On withdrawal of initiatives by proponents, see infra notes 158, 172 and accompanying text.
78
[Vol. 66
less open. When the process works properly, the bill will undergo genuine deliberation by a body that includes opponents and neutral members as well as backers. Bicameralism and the executive veto repeat the process. Representative democracy suffers many instances when these processes do not work as ideally intended, but even in an imperfect world, they are very different from drafting of initiatives. Except for executive veto, the same drafting safeguards are in place for a legislative referendum.' 47 As this discussion suggests, several problems arise from the way initiatives are drafted. The most obvious is that initiatives are more likely than legislation to be drafted badly. Less expertise is brought to bear, and a small group of proponents will overlook unintended effects and ambiguities more often than will the broad range of interests present in the legislative
process. 148
The drafting problem is severely magnified by early freezes of initiatives' wording. The many opportunities for adjustment in the representative process are lacking. Moreover, entrenchment of enacted initiatives makes later adjustment extremely difficult. If the initiative is in the form of a constitutional amendment, it can be altered only by the same process. If it is nominally a statute, many states either disable legislative adjustments or require a super-majority in the legislature to make them. 149 Fine-tuning a measure in the light of experience, a common aspect of the legislative process, is difficult to imp6ssible. A troublesome aspect of the drafting process is that it shuts out minority comment and thus minority interests. Target groups do not have an opportunity to make their case in the drafting process; they can only campaign against the initiative language before the electorate. By contrast, the representative process often accommodates minority concerns during the enactment process. (This is true to a greater extent in parlia147. E.g., COLO. CONST. art. XIX, 2. 148. See generally CALIFORNIA COMM'N, supra note 9, at 79-91. The problem is not new. See Hubert D. Henry, PopularLaw-Making. in Colorado, 26 ROCKY MTN. L. REV. 439, 448 (1954). Drafting problems can surface during the campaign on an initiative that has already qualified for the ballot. For two dramatic examples, see Christopher A. Coury, Note, Direct Democracy Through Initiative and Referendum: Checking the Balance, 8 NOTRE DAME J.L. ETHICS & PUB. POLY 573 (1994). 149. See supra notes 11-12 and accompanying text.
1995]
mentary systems lacking judicial review, where accommodative traditions are stronger than in the United States.) Opponents of a bill can often persuade its backers to modify parts that are most objectionable to opponents and least important to proponents. The legislative process forces opposing sides to listen to one another before wording is final. Two houses and a governor must be satisfied. The initiative process does none of this. Several methods are used to try to improve drafting of citizens' initiatives. Most formal and successful is the indirect initiative. Initiators must first submit their proposal to the legislature for its review and possible action. The legislature can employ its procedures to point out shortcomings of the initiators' draft. It can enact an alternative measure or propose an alternative by referendum. This procedure is one reason for success of the Swiss national initiative. 5 An additional procedure recommended by some critics is a public hearing on the measure during or after petitioning.' 5 ' A second method is to require that a state official, such as the attorney general or legislative counsel, review drafts of all proposed initiatives and point out any shortcomings to the initiators. Some form of this procedure is used in many initiative states, including Colorado. 5 2 However, this method is much less thorough than legislative review. A more drastic reform is to have the actual ballot measure drafted by a public official, who is charged with carrying out the initiators' purpose, and whose draft is subject to judicial review. This form is presently used in New Zealand.' 5 ' A third approach involves efforts to postpone freezing of the initiative's text. One variation is to lengthen the period between the initial submission and petitioning. It is very short
150. The Swiss system enacts few initiatives as such; it is much more common for voters to enact legislative counterproposals or substitutes. See KOBACH, supra note 4, at 87, 110. See generally id. at 70-121 (trends in the exercise of direct democracy in Switzerland). See also CALIFORNIA COMM'N, supra note 9, at 104-07 (advisory legislative hearings after ballot qualification). 151. See CALIFORNIA COMM'N, supra note 9, at 112-19. 152. Id. at 99-103. On Colorado, see also supra part IV.A.2. For the view that this method is too expensive for California because of the number of proposals, see CALIFORNIA COMM'N, supra note 9, at 121. 153. Citizens' Initiated Referenda Act 1993 (N.Z. Stat. 1993 no. 101). This statute provides for advisory referendums only.
80
[Vol. 66
in some states, including Colorado."" Another is to allow amendments by the initiators later in the process.1 55 The difficulty is how to allow alterations to the petitions citizens have signed without undermining the integrity of petitioning. However, if changes do not alter the essence of a proposal, they can be allowed to clarify it or to correct unintended defects. One purpose of each of the procedures discussed above is to promote discussions and bargaining between initiators and public officials to get better-drafted and more practicable initiatives. An alternative to amendment of an initiative, discussed in the last paragraph, is withdrawal of an initiative after its proponents are satisfied by a legislative substitute. The latter can, of course, be either a statute or a measure submitted by the legislature for referendum. This practice is well-established in Switzerland but rare in the United States.' To make it work, it is important to know who has the right to withdraw an initiative and how late in the process this can be done. State laws are vague on these points. Statutes use indefinite terms such as "proponents" to refer to the persons who submit initiative proposals,' 5 7 and it is implicit that whoever controls signed petitions can withdraw an initiative simply by failing to submit them to elections officials. ' But there are no rules to govern disputes among persons claiming to be proponents. If bargaining with officials becomes common, disputes are likely to arise. Finally, drafting problems can be alleviated by lowering the barriers to post-enactment amendment. If procedures for statu-
154.
CALIFORNIA COMM'N, supra note 9, at 130. On Colorado, see infra text
accompanying note 300. 155. See generally CALIFORNIA COMM'N, supra note 9, at 109-19 (amendability of initiatives). 156. Swiss legislation allows withdrawal of an initiative by majority decision of the sponsoring initiative committee. See KOBACH, supra note 4, at 104. On the United States, see CALIFORNIA COMM'N, supra note 9, at 91-94, 99-107. The bestdrafted American attempt at the procedure is the new Mississippi provision. MISS. CODE ANN. 23-17-29 to -37 (Supp. 1994). 157. E.g., CAL. ELEC. CODE 3502 (Supp. 1994) ("proponents" are persons presenting draft petition). 158. California proponents may formally withdraw their petition before filing if they all sign the notice. Id. 5354. (On Colorado, see infra note 172 and accompanying text.) Under the California statutes, suppose that three people claim to be proponents and two want to withdraw a proposed initiative. Section 5354 seems to say that the withdrawal is ineffective. What then if the two refuse to file petitions in their possession, or even destroy them? What if there is a dispute about who is properly called a proponent?
1995]
tory initiatives were distinctly less burdensome than for constitutional, the trend toward entrenching everything would likely reverse. And legislatures can be given power to amend statutory initiatives, especially after an intervening election. 2. The Non-Binding, Consultative Procedures in Colorado As noted above, Colorado has a mandatory but non-binding consultative procedure to improve initiative drafting enacted in 1974.159 The act followed on the heels of several unsuccessful initiatives in the 1972 election on taxation and gambling issues that were long, complicated and awkwardly drafted and anticipated ultimately successful, technically sophisticated initiatives in the 1974 election on annexation and reapportionment. 6 Originally the comments were confidential until the ballot title was established, but a 1980 constitutional amendment made the comments public. 6 ' The Colorado Supreme Court has held that the ballot title board has no jurisdiction to set a ballot title unless the public comment and review proce162 dures have been followed. The current version of this process requires that before the solicitation of petition signatures, proponents must submit proposed initiative language to the directors of the legislative council and the office of legislative legal services for comment, and the directors may request comments from other executive agencies.163 Although proponents are not bound to accept any suggested changes, the statute does have one exhortation and two requirements pertaining to drafting style: The statute encourages them "to write ... drafts in plain, nontechnical language and in a clear and coherent manner using words with common and everyday meaning which are understandable to
159. Act of May 14, 1974, ch. 66, 1974 Colo. Sess. Laws 285 (repealing and reenacting, as amended, COLO. REV. STAT. 70-1-1 (1963)), repealed and superseded by COLO. REV. STAT. 1-40-105, -107 (Supp. 1994) (in the original procedure the state attorney general rather than the directors of the legislative council and the legislative drafting office were the official commenters). 160. See supra text following note 80. 161. COLO. CONST.art. V, 1(5). 162. In re Proposed Initiative Under the Designation "Tax Reform," 797 P.2d 1283 (Colo. 1990). 163. COLO. REV. STAT. 1-40-105 (Supp. 1994). The directors are appointed by, and answerable to, joint committees of the General Assembly. Id. 2-3-301 to -508
82
[Vol. 66
the average reader," requires that "drafts shall be worded with simplicity and clarity and so that the effect of the measure will not be misleading or likely to cause confusion among voters," and requires that "[t]he draft shall not present the issue to be decided in such manner that a vote for the measure would be a vote against the proposition or viewpoint that the voter
believes that he or she is casting a vote for ....
14
Thus far, perhaps because no complainants have made the argument, Colorado courts have not viewed this statutory language as a license to rule on the clarity of initiatives. The Court has saved its time and attention for review of the language of the ballot title and summary, both the work of the ballot title board, which we explore below. 3. Comments on the Colorado Procedures Colorado solutions to the drafting problems are anemic and impractical. Legislation calls for clarity, coherency, and plain language, but the only enforcement mechanism is the consultation system, which is seriously flawed. Every petition must be submitted in draft form for official review and comment, but the process is both too broad and too shallow. It is too broad because every proposal must be reviewed, including the frivolous and whimsical. Lacking any filing fee, there is no deterrent to casual submissions. The process is too shallow because it must be done in two weeks, and in practice state officials do not comment on the merits of a draft. The directors of the legislative council and the office of legislative legal services, who review initiative drafts, are accountable to powerful legislators of both parties and take pains to be politically neutral on the merits of every proposal. As a consequence, they routinely do not point out the effects of the draft on existing law and practice, nor do they suggest alternative language to cure substantive defects. Time pressure and their caution confine their review to correcting grammar and simple internal inconsistencies.'65 Drafters are not required to adopt any of the reviewer's suggestions.
164. Id. 1-40-105(1),(3). 165. The statute requires that reviewers suggest "editorial changes to
promote compliance with the plain language provisions of this section." Id. 1-40105(1). However, this cannot be done effectively under the existing requirements
that every draft be reviewed in two weeks.
1995]
The consultation system has not affected several recent, poorly drafted initiatives, which are drafting nightmares. 6 6 The consequence of inept drafting has been a flood of litigation as disagreements over the meaning of initiative language inevitably end up in court. The tax and spending limitation initiative passed in 1992 has generated a new, lucrative subspecialty for lawyers in municipal law, at taxpayers' expense.'6 1 City and county attorneys struggle to comply, often by hiring outside counsel to advise on and defend against lawsuits. Litigation over the meaning of the language has proven to be very costly and has held up projects clearly favored by voters (school bonds, for example), adding significantly to the projects' expense. 6 It will take years of judicial precedent to work out some of69 the kinks in the more complicated of the 1
recent initiatives.
166. E.g., COLO. CONST. art. X, 20 (Taxpayer's Bill of Rights, 1992 Amendment One); 1994 Amendment 12 (rejected by Colorado voters, Nov. 8, 1994) (on file with Colorado Secretary of State). See infra notes 174-75, 318-19. 167. See COLO. CONST. art. X, 20, which has prompted over 60 lawsuits since passage in 1992. See also Court to Decide Tax Cases: Amendment 1 Violations Alleged, DENV. POST, June 28, 1994, at B3. Two cases on a technical procedural point (can a bond question and the tax to repay it be combined in a single question on the ballot) have already reached the Colorado Supreme Court. See Bickel v. City of Boulder, No. 94SA130, 1994 Colo. LEXIS 742 (Colo. Oct. 11, 1994) (consolidating the cases). See also Margot Duvall, Amendment One is Causing Schools to Suffer, RocKY MTN. NEWS (Denver), Oct. 9, 1994, at 95A (outraged at proponents motion for rehearing adding more delay on local bonds). 168. See Mary George, Green Light for Bond Issue; Court: Amendment 1 Intent Fulfilled, DENV. POST, Sept. 13, 1994, at Al (delay caused by litigation has cost taxpayers $70 million in increased interest alone; delay also increased construction costs and funds needed to buy land on which to build). See also Mary George & Janet Day, Court Rejects New Bid to Halt Sale of Bonds: Amendment 1 Ruling Clears Way for Constructionof New Schools, DENV. POST, Oct. 13, 1994, at B1 (delay will cost Boulder Valley taxpayers $12 million in added interest). An irony of the delays in implementing voter-approved projects is that proponents sold Amendment One as a citizen empowerment scheme. 169. Courts expend much more political capital when they overturn initiated measures than when they overturn legislative acts. Bickel was a rare decision in which a court had to decide between two popular votes, spending political capital either way. Bickel v. City of Boulder, No. 94SA130, 1994 Colo. LEXIS 742 (Colo. Oct. 11, 1994). The court reviewed an'attack on compliance by local governments with 1992 Amendment One, a constitutional initiative forbidding increases in taxing, borrowing, or spending without voter approval. Four such measures had been proposed by governments and approved by voters. The court rejected most of the attacks by applying the election-law rule of substantial compliance to technical aspects ofAmendment One, and by interpreting an ambiguous provision in favor of the governments. Id. One of us (Oesterle) thinks the court erred in failing to read Amendment One strictly. The other (Collins) thinks the court
84
[Vol. 66
Examples of mischievousness in initiatives are also not hard to find. How could Colorado voters understand that the words "any subject or subjects whatsoever" in the definition section of an initiative (Amendment Twelve, defeated in the 1994 election) were intended by the initiative drafter to nullify the single-subject requirement in a referendum on the ballot in the same election (Referendum A, an amendment which 17 passed)? Citizens have standing to challenge successful initiatives on the grounds that voters were misled, but the burden of proof is very high and has never been carried.' 7 ' Colorado lacks any effective procedure to promote discussions and bargaining between proponents and officials over drafting, and between proponents and opponents. As noted above, an effective procedure must allow amendment or withdrawal late in the process, in case a compromise or clarification is decided on. In Colorado, the right of amendment ends early, and the right of withdrawal is unclear.'7 2 D. Scope of Initiatives 1. Introduction to the Problem of Multi-Issue Initiatives Long and complex initiatives can be criticized on three grounds. First, constitutional stability is undermined by initiatives that alter state government too much because initia-
reached the correct result. On problems of interpreting initiatives, see Eule, supra note 2. 170. See supra note 145. 171. E.g., Glendale v. Buchanan, 578 P.2d 221 (Colo. 1978). For an example of a successful challenge in another state, see Boucher v. Bomhoff, 495 P.2d 77 (Alaska 1972). 172. On amendment, see supra note 146. On withdrawal, see COLO. REV. STAT. 1-40-105 (Supp. 1994). "Proponents" are persons who submit draft petitions and who file campaign reports. Id. 1-40-105, -121. "Proponents" are required to designate two "representatives" who can cure signature shortages or federal law defects. Id. 1-40-104, -117, -120. To be effective, signed petitions must be filed within six months after fixing title and summary, id. 1-40-108, and at least three months before the general election date. COLO. CONST. art. V, 1(2). Suppose A, B, C, D and E submit a draft petition and thus become "proponents." At the same time, they designate A and B as their "representatives." Later, A and B reach a compromise with legislators for withdrawal of the initiative in favor of a legislative substitute. But C, D, and E disagree. Can there be a lawful withdrawal? Can A and B lawfully refuse to file petitions in their possession, or destroy them?
1995]
tives can entrench a measure beyond the reach of the legislature. Second, voter confusion-and irritation-increases with the length of an initiative.'73 The ballot title or summary becomes less representative of the full text, and the full text is read and understood by fewer citizens. When a initiative is long and complicated, joining several issues, it exacerbates the problem of informing voters. The tendency to compress the debates over initiative issues into thirty second television spots means that the debate over multi-issue initiatives will focus on one or two issues out of the five or six in the text of the initiative. Voters will simply not know about, much less understand in any depth, many of the sub-issues.'74 Third, long initiatives inevitably contain positions on multiple issues, which critics say should be divided for separate votes so that citizens can say yes to one and no to another. Bundling of issues is a recognized abuse within the representative process, and it is no less so in direct democracy. Even if one assumes all voters know about and understand each of the issues in a multi-issue initiative, they can be coerced by such initiatives into placing language in the constitution that would not have alone been supported by a majority. To illustrate, assume an initiative with four separate issues. A voter may be strongly in favor of issue one but marginally against issues two, three and four. If she could vote separately on each, she would vote yes on one and no on issues two, three and four. But she does not have such a choice; it is all or nothing. Does she vote yes, absorbing the cost of supporting issues two, three and four? Or vote no, absorbing the cost of opposing an issue she deeply favors? Many voters in such a situation will vote yes. If a majority does so, the state
173. See CALIFORNIA COMM'N, supra note 9, at 182-83.
174. 1992 Amendment One added over 1700 words to the Colorado Constitution on taxation and spending limits. COLO. CONsT. art. X, 20. The Amendment requires voter approval of increases in tax and government spending
and of new government indebtedness. The theme of the campaign to pass the Amendment was the empowerment of citizens. See, e.g., Douglas Bruce, Amendment 1 Face-Off Initiative Puts Control Back in Hands of People, ROCKY
MTN. NEWS (Denver), Oct. 17, 1990, at 49. But hidden in the language were flat prohibitions on government choice, even if ratified by a citizen vote. Examples include prohibitions on new or increased transfer taxes on real property, on any new state property tax, on any new local government income tax, on a state progressive income tax, and on state general obligation bonds, and an incentive
for the sale of public land. See Dale Oesterle, Bruce's Hidden Agenda in Amendment 1, DENV. POST, Dec. 5, 1992, at B15.
86
[Vol. 66
constitution contains language favored by the majority, the text of issue one, and language disfavored by the majority, the text of issues two, three and four. In an extreme case, minority supporters of several issues could join to pass an initiative composed of parts which would all fail if offered to the electorate alone. The combination of the two disadvantages of multi-issue initiatives gives petitioners a strategic opportunity. If petitioners want language that they know will not, alone, get a majority vote, they can put the language in a multi-issue initiative with several other issues that are hot, hoping one of the hot issues will carry their language along with it into the state constitution.175 They hope that those who would vote against
175. In passing Amendment One in 1992, the voters, as experience suggests, wanted a taxation limitation but not a revenue limitation. Voters supported the proposition that government should not increase taxes without a citizen vote. But the success of many referendums to exempt local jurisdictions (and the overwhelming success of specific revenue retention measures) demonstrate that voters do not in general favor a revenue limitation, which, among other things, requires governments to give back tax revenues when revenues increase not because of increased tax rates but because, for example, of increased economic activity, unless voters approve revenue retention. Eric Anderson, Measures Try to Undo Amendment 1: Cities Want to "De-Bruce"Tax and Spending Limits, DENV. POST, Nov. 2, 1994, at B1 (forty-six de-Brucing questions on the November 8, 1994 ballot; measures do not ask for tax increases, they allow government to keep money received); Aldo Svaldi, Bond Issues Fly, While Taxes Die, DENV. Bus. J., Nov. 1117, 1994, at 1A, 38A (voters generally approved "de-Brucing" measures that allowed government to keep excess revenues from old taxes but generally did not approve requests for new or higher taxes; requests for new debt had a 62% approval rate with school districts faring the best). Several propositions on spending attempted to exempt selected local expenditures permanently from Amendment One, moves of questionable legality under the language of the Amendment. See Loveland Issue A and Issue B (both rejected by voters in Loveland, Colo., Nov. 8, 1994) (on file with Loveland, Colo. City Clerk). Many Colorado voters simply did not understand the details of the initiative beyond the basic tax limit. Sarah Ellis, Pollster Says Voters Still Back Amendment 1; But Many are in Dark About Law's Details, DENV. POST, Apr. 30, 1993, at B3 (voters did not know that measure would prevent city from keeping sales-tax revenues from new outlet mall). The mix of taxation and revenue limits in Amendment One is not just a nuisance. It has serious effects. The combination creates a "ratchet down" effect on both limits. Jeffrey A. Roberts, Tax-Limit Amendment Passes, DENY. POST, Nov. 4, 1992, at Al (spending limits will ratchet down over time without voter approval); see also Mark J. Shaw, Amendment 1: Bruce's Ratchet Creates Confusion, Uncertainty, DAILY J., Jan. 27, 1993, at 2. A humorous example is provided by Colorado's 1994 Amendment 13. In 1991, and again in 1993, voters in Manitou Springs rejected overwhelmingly an amendment to the city charter to allow gambling in their city. A frustrated potential casino owner put the issue on the state-wide ballot and, hoping to attract
1995]
the petitioners' pet language if informed and focused on the issue either will not know the language is in the initiative or will not focus on the effect of the language because the hot issue will dominate the short television spots. Those who do focus on and understand the petitioners' pet language and would vote against it if they could, may vote yes nevertheless on the hot issue. Indeed, a failure in one election to have the right mix of issues gives lessons for later elections. Eventually, on the second or third try, petitioners will get the mix of issues 7 16 right and pass their total package' In sum, multi-issue initiatives give opportunities to proponents to put language in a state constitution, either by accident or by design, that does not have majority support of the voters. Two methods are used to confine the scope of initiatives. The more common is to limit the scope to a single subject or single question. At least fifteen states, including Colorado, do this in one form or another.'77 The same concept has long been in use in many jurisdictions to limit the scope of bills in legislatures.'7 8 Whether for direct or representative democracy, this device is of limited utility because defining single 179 subject is highly subjective. The single-subject limit on initiatives has greater effect when review of the issue is made before a measure is put to a vote. Once the electorate has passed a law, courts are very reluctant to throw it out for multiple subjects. A prominent example was California's Proposition 8 in 1982, called the Victim's Bill of Rights. The measure covered a broad range of
state-wide support, tied it to a provision that would allow the operation of slot machines in public airports. The measure failed. 1994 Amendment 13 (rejected by Colorado voters Nov. 8, 1994) (on file with Colorado Secretary of State). 176. The citizens of Colorado voted on initiatives imposing taxation limitations in nine successive bi-annual elections. The last three initiatives were written by the same author, Douglas Bruce, and revised in light of circumstances. See Roberts, supra note 175, at Al (The seventh and eighth initiatives, written by Mr. Bruce, failed in 1988 and 1990). The ninth initiative, 1992's Amendment One, passed. Id. 177. See CALIFORNIA COMM'N, supra note 9, at 362 (at least 13 states; Colorado and Mississippi adopted the rule thereafter). On Colorado, see infra notes 184-94 and accompanying text. On Mississippi, see MISS. CONST. art. 15, 273(2). 178. See 73 Am. Jur. 2d Statutes 119 (1974). 179. See In re Tax Limitation, 644 So.2d 486 (Fla. 1994).
88
[Vol. 66
matters, but the state supreme court solemnly pronounced it consistent with the single-subject rule. 80 The other device to limit the scope of initiatives is a creature of judicial inventiveness. When the California Supreme Court reviewed a proposed 21,000 word initiative in 1948, it held that the state's initiative can be used to "amend" but not "revise" the state constitution, and the proposed initiative was a wrongful attempt at revision. 8 ' The state's provisions for constitutional amendments used the word amend when describing the initiative. When authorizing a constitutional convention, the provision said amend or revise. The court decided that this scheme implied a limit on the scope of initiatives, and the proposal at issue exceeded the limit.'8 2 Several other state courts have invalidated referendums or 83 initiatives on the same rationale. 2. The Colorado Single-Subject Rule Success in the 1992 Colorado election of a multiple issue tax and spending limitation initiative and the prospect of several multiple issue initiatives appearing on the ballot in the 1994 election led the General Assembly to propose a singlesubject rule by referendum. The referendum passed, in spite of bitter opposition by the most active advocate and user of the initiative process in the state. 1 84 So Colorado now has a sin-
180. Brosnahan v. Brown, 651 P.2d 274 (Cal. 1982); see also Raven v. Deukmejian, 801 P.2d 1077, 1083-85 (Cal. 1990) (upholding validity of Proposition 115, the "Crime Victims Justice Reform Act," under the single-subject rule because of its consistent theme); Steven W. Ray, Note, The CaliforniaInitiative Process: The Demise of the Single-Subject Rule, 14 PAC. L.J. 1095 (1983). 181. McFadden v. Jordan, 196 P.2d 787, 799 (Cal. 1948); see also Raven v. Deukmejian, 801 P.2d at 1085-90. 182. McFadden, 196 P.2d at 799; see also Raven, 801 P.2d at 1085-90 (invalidating parts of one other initiative on same rationale). 183. State v. Manley, 441 So. 2d 864 (Ala. 1983); Opinion of the Justices, 264 A.2d 342 (Del. 1970); Adams v. Gunter, 238 So. 2d 824 (Fla. 1970); RiveraCruz v. Gray, 104 So. 2d 501 (Fla. 1958); Ellingham v. Dye, 99 N.E. 1 (Ind. 1912), appeal dismissed, 23 U.S. 250 (1912); Miller v. Taylor, 133 N.W. 1046 (N.D. 1911); Holmes v. Appling, 392 P.2d 636 (Or. 1964). 184. COLO. CONST. art. V, 1(5.5) (amended 1994). See infra note 313. See also COLO. CONST. art X, 20(4). Ironically Mr. Bruce argues, in essence, that his successful 1992 initiative on tax reform contains a single-subject requirement. Mr. Bruce sued, and lost, arguing that the language requires two votes on a single bond issue, one vote for the bond issue and a second for the taxes that will support the repayment. For Mr. Bruce, then, a single-subject rule ought to apply to tax or spending requests by a government unit, but not to his personal initiative
1995]
gle-subject rule for initiatives. Legislation accompanying the referendum states that the single-subject rule is aimed at "forbid[ding] the treatment of incongruous subjects in the same measure, . . . for the purpose of enlisting in support of the measure the advocates of each measure, and thus securing the enactment of measures that could not be carried upon their merits" and "[t]o prevent surreptitious measures and apprise the people of the subject of each measure by the title ... to prevent surprise and fraud."'8 5 A single-subject rule has long been in force for bills,' except general appropriation bills, and case law on the requirement will now apply to all constitutional amendments, whether by initiative or by referendum.8 7 An early Colorado Supreme Court case explained the purpose of the single-subject requirement for legislation, and the arguments apply with similar force to initiatives: [I]t is important to bear in mind the evils sought to be corrected .... The practice of putting together in one bill
petitions. See Bickel v. City of Boulder, No. 94SA130, 1994 Colo. LEXIS 742 (Colo. Sept. 12, 1994). 185. COLO. REV. STAT. 1-40-106.5(1)(e)(I), (II) (Supp. 1994). 186. COLO. CONST. art. V, 21. 187. COLO. REV. STAT. 1-40-106.5(3) (Supp. 1994) ("[I]n setting titles... the initiative title setting review board . . . should apply judicial decisions construing the constitutional single-subject requirement for bills and should follow the same rules employed by the general assembly in considering titles for bills."). The statute's instruction to apply both judicial decisions and legislative "rules" for single-subject review may create some conflict. According to the current director of the office of legislative legal services, "the rules we use in our office [for singlesubject determinations] are stricter than the courts use." See Fred Brown, Keeping the Ballot Constitutional, DENVER POST, Dec. 28, 1994, at 7B (quoting Douglas Brown). This difference makes sense when one compares administrative review of a bill with judicial review of an enacted law; courts give enacted laws a strong presumption of constitutionality. E.g., In re House Bill No. 1353, 738 P.2d 371, 372 (Colo. 1987) (must show unconstitutionality "beyond a reasonable doubt"). Administrative review has an opposite incentive, to apply the single-subject standard strictly to avoid later trouble in court or legislative debate. However, the Colorado Supreme Court also review bills when referred by the General Assembly, and no presumption of constitutionality applies. E.g., Submission of Interrogatories on Senate Bill 93-74, 852 P.2d 1, 5 n.4 (Colo. 1993). Also, the Court has an established practice of deferring to the title board. See infra text accompanying note 222. Thus the statutory commands can be reconciled by applying the legislative practice in the context of single-subject review by the title board, including judicial review of the board. But when single-subject challenges are made in litigation filed after passage of a measure (and thus long after title-board approval), the judicial precedents with a presumption of constitutionality should apply.
90
[Vol. 66
subjects having no necessary or proper connection, for the purpose of enlisting in support of such bill the advocates of each measure, and thus securing the enactment of measures that could not be carried upon their merits, was undoubtedly one of the evils sought to be eradicated. Another object is to prevent surprise and fraud from being practiced upon legislators, and to apprise the people of the subjects of legislation.'8 8 Courts applying the single-subject rule to legislation speak simultaneously in the language of caution and in the language of righteousness. They, on the one hand, presume a statute to be constitutional and should not be declared invalid "unless 189 that conclusion is established beyond a reasonable doubt," and, on the other hand, state that the rule "must have a reasonable and liberal" construction.' 90 In the end, one can rely only on the specific holdings of individual cases to give meaning to the rule. An example of a bill that the court found to be invalid is House Bill No. 1353, a forty-four page, forty-six section bill. The bill, among other things, reduced state contributions to state employees' retirement funds, imposed charges on inmates for medical visits, imposed a surcharge on insurance carriers based on workers' compensation premiums, revised the statutory formula for Medicaid reimbursements to nursing homes, provided for the forfeiture of abandoned intangible property by banks, and eliminated state aid for instructional television.' 9' The Colorado Supreme Court held that the common characteristic of financial savings did not save the bill. 192 On the other hand, the court recently upheld a health care bill that prohibited certain advertising practices and specified abuses of health insurance by health care providers.' 9 3 The court held that the advertis-
188. Catron v. Board of Comm'rs of Archuleta County, 33 P. 513, 514 (Colo. 1893). 189. In re House Bill No. 1353, 738 P.2d at 371. 190. Catron, 33 P. at 514 (the court also noted that the rule is "not designed to hinder or unnecessarily obstruct legislation"). 191. In re House Bill No. 1353, 738 P.2d at 372-73. 192. Id. at 373. 193. Parrish v. Lamm, 758 P.2d 1356 (Colo. 1988). The statute was aimed at the practice of health care providers to waive their right to demand payment of the deductible and co-payment features of insurance policies, relying instead on the insurance coverage for full payment. Id. at 1359. The statute made the practice a crime and prohibited health care providers from advertising that they were willing to engage in the practice. Id. at 1360.
1995]
ing prohibited and the abuse interdicted were connected, "because the act of advertising is simply one means of alerting patients that a health care provider is willing" to engage in the abusive practice.' E. Informing Initiative Voters 1. Introduction to the Problem of Voter Education on Initiatives Voter confusion in initiative campaigns can arise from overly long initiatives, or from confusing and ambiguous drafting, the subjects of the previous sections. It can also arise from several other sources. An initiative's subject can be technical, not apparent to citizens without special knowledge. There can be so many initiatives and candidates that voters' capacity to understand them is overwhelmed. And, as in all political campaigns, there can be misleading claims made for or against a measure, particularly in advertisements. Several measures to address these problems are in common use. Many initiative states require a public official to draft ballot titles or captions for proposed initiatives. 9 s Many states also require drafting of an official summary of each measure.' 96 Some require an official assessment of the fiscal effect of a proposal.' 97 And a number of states publish and mail to each voter an official pamphlet explaining each ballot measure and outlining arguments for and against it. 198 The most important of these provisions is probably the requirement of an official ballot title. When initiators can impose their own title, the path is open to advertisers' gimmicks to sell the initiative. As it is, initiators often advertise their work under a label or title, such as "Save Our State," the name
194. Id. at 1362. More recently, the court held that a single referendum can encompass both authority for bonded indebtedness and for taxes to repay it. Bickel v. City of Boulder, No. 94SA130, 1994 Colo. LEXIS 742 (Colo. Oct. 11, 1994). See discussion supra notes 167, 169. 195. CALIFORNIA COMM'N, supra note 9, at 229, 231 (official caption in nine states; official approval of proponents' caption in five; proponent writes caption in two; no caption in eight). 196. Id. (official summary in nine states; official approval of proponents' summary in four; proponent writes summary in four; no summary in five). 197. See id. at 235 (five states require fiscal impact analysis to be printed in ballot information pamphlets); see also infra note 214. 198. See id. at 235-38.
92
[Vol. 66
adopted by initiators of California's Proposition 187, intended to forbid state services to aliens who are present in the state in violation of immigration laws.'9 9 These are often misleading. Florida initiators write their own titles and summaries, the source of some controversy there.20 0 Summaries and pamphlets are also important for voters who take the time to read them, and surveys show that some do.2 0 ' As initiative ballots grow ever longer, the contrast between American ballots and those in parliamentary nations becomes sharper. In every state, we vote separately for a long list of officials, and in initiative states, many ballot measures are added on.2 2 In parliamentary systems, a voter simply chooses a party, which will represent her in all aspects of government. 203 For a literate and informed voter, the American system has obvious advantages. For others, the question is which system represents them better. The populists who led the original push for the initiative in America were sure it would improve the lot of have-nots. 2 4 History has raised substantial doubts. 2. The Influence of Money Concern that money unduly corrupts the is widespread. The subject has also been ied.2 For this reason, it is not a point of " paper. Our comments are limited to noting initiative process thoroughly studemphasis in this some regulations
199. Proposition 187 (adopted by California voters Nov. 8, 1994) to be codified at CAL. PENAL CODE 113, 114, 834(b); CAL. WELF. & INST. CODE 10001.5, 130; CAL. HEALTH & SAFETY CODE 130; CAL. EDUC. CODE 48215, 66010.8; CAL. GOV'T CODE 53069.5. 200. See Advisory Opinion to the Attorney Gen. Re Tax Limitation, 644 So.2d 486, 497 (Fla. 1994) (Overton, J., concurring).
201. CALIFORNIA COMM'N, supranote 9, at 244-48 (pamphlets important for
educated voters).
202. See, e.g., id. at 182 (November 1990 ballot in Los Angeles had 36 ballot measures).
203. See, e.g., 15 HALSBURY'S LAWS OF ENGLAND, Elections ed. 1990) (contents of British parliamentary ballot). 204. See supra text accompanying notes 35-39.
516-17 (4th
205. See CALIFORNIA COMM'N, supra note 9, at 263-300; MAGLEBY, supra note 4, at 145-51; JOHN SHOCKLEY, THE INITIATIVE PROCESS IN COLORADO POLITICS: AN ASSESSMENT (1980); BETrY ZISK, MONEY, MEDIA AND THE GRASS
ROOTS: STATE BALLOT ISSUES AND THE ELECTORAL PROCESS (1982); Daniel H.
Lowenstein, Campaign Spending and Ballot Propositions: Recent Experience, Public Choice Theory and the FirstAmendment, 29 UCLA L. REV. 505 (1982). On
recent Colorado spending, see infra notes 265-68 and accompanying text.
1995]
that states have attempted to impose in order to limit distortions thought to arise from spending on initiative campaigns.
2 06
Some states have attempted to limit the influence of money on the initiative process by limiting contributions to initiative campaigns and by forbidding paid petition circulators.0 7 Several of these provisions have been struck down by courts under the First Amendment. 0 8 Others have proved ineffective. As a result, researchers who have studied the effect of money on the initiative process conclude that it is at least as great as on the representative process. 2 9 The only form of regulation generally sustained by the courts is requirements to disclose contributors and contributions to initiative campaigns.2
3. The Colorado Notice Procedure: The Ballot Title, the Ballot Information Booklet and Newspaper Notices Colorado relies on three types of public notice to get information out to the electorate on initiative questions: the ballot information booklet, the ballot title, and newspaper notices. Of the three, the ballot title process is the most important. a. The Ballot Title Board Procedure After their public meeting with the legislative council and legislative legal services, proponents submit a final draft of
206. More dramatic proposals have been made. For an argument that the "fairness doctrine" ought to apply to initiatives, see Robyn R. Polashuk, Protecting the Public Debate: The Validity of the Fairness Doctrine in Ballot Initiative Elections, 41 UCLA L. REV. 391 (1993) (the doctrine, among other things, would require media to give free air time to the unfunded side of a ballot initiative campaign). 207. See CALIFORNIA COMM'N, supra note 9, at 296-301; see also infra note 311 and accompanying text. 208. E.g., Meyer v. Grant, 486 U.S. 414 (1988); Federal Election Comm'n v. Massachusetts Citizens For Life, 479 U.S. 238 (1986); Citizens Against Rent Control v. City of Berkeley, 454 U.S. 290 (1981); Buckley v. Valeo, 424 U.S. 1 (1976); American Constitutional Law Found. v. Meyer, No. 93-M-1467, 1994 U.S. Dist. LEXIS 17134 (D. Colo. Nov. 23, 1994), appeal docketed, No. 94-1576 (10th Cir. Dec. 16, 1994).
209. See CALIFORNIA COMMN, supra note 9, at 265-91.
210. Id. at 291. But see Austin v. Michigan Chamber of Commerce, 494 U.S. 652 (1990) (sustaining limit on spending by for-profit corporations).
94
[Vol. 66
their initiative to the Secretary of State.21' The Secretary of State convenes a "title board" consisting of herself, the Attorney General and the director of the office of legislative services, which meets in public, when there is business, on the first and third Wednesdays of each month from the first Wednesday in December after an election until the third Wednesday in May in an election year.212 Proponents must submit their draft twelve days before one of the Wednesday meetings.13 Within two weeks, the title board, 1 4 by majority vote, must draft a title for the initiative ("which shall correctly and fairly express the true intent and meaning thereof' and which "shall unambiguously state the principle of the provision sought to be 15 added, amended, or repealed"), 2 a submission clause, and a "clear, concise summary" which "shall be true and impartial and shall not be an argument, nor likely to create prejudice, either
2 1 for or against the measure."2
The title board also has
responsibility to apply the new single-subject and clear expression requirements to ballot titles.217 If a measure contains more than one subject, no title can be set.21 The importance of the title cannot be overstated. The title and submission clause, together known as the "ballot title," appear on the ballot and on each page of the petition. Neither
211. COLO. REV. STAT. 1-40-105 (Supp. 1994). A version of the ballot title procedure has been in place since 1919. Act of Mar. 31, 1919, ch. 131, 1, 1919 Colo. Sess. Laws 431. Originally the ballot title board consisted of the Secretary of State, the Attorney General and the Reporter of the Supreme Court. Id. From 1913 to 1919, proponents wrote ballot titles. See infra note 296. 212. COLO. REV. STAT. 1-40-106(1) (Supp. 1994). 213. Id. 214. The board must request a "fiscal impact" statement from the office of state planning and budgeting or the department of local affairs "if, in the opinion of the title board, the proposed law or constitutional amendment will have a fiscal impact on the state or any of its political subdivisions." Id. 1-40-106(3)(a). The target agency "shall furnish any assistance so requested," and the information must be filed on the Friday before the Wednesday title board meeting at which'the initiative is considered. Id. 215. Id. 1-40-106(3)(b). 216. Id. 1-40-106(3)(a). If the measure proposes to increase taxes or government borrowing or spending, its title is regulated by a provision in the Colorado Constitution added by citizens' initiative in 1992. COLO. CONST. art. X, 20(3)(c). 217. COLO. CONST. art. V, 1(5.5) (amended 1994); COLO. REV. STAT. 1-40106.5(3) (Supp. 1994). 218. COLO. CONST. art. V 1(5.5) (amended 1994). In that event, proponents may revise the measure, id., or sue. See infra note 223 and accompanying text.
1995]
the summary nor the language of the initiative itself is on the 2 ballot; 19 they are on each petition and appear in the newspaper notices and ballot information booklets before the election. Since the ballot title is so important, the statute provides for an elaborate appeals procedure. Interested parties may file for a rehearing before the title board within seven days after the 22 titles and summary are set. If protesters lose on rehearing, they may, within five days, seek review in the Colorado Supreme Court, which "shall" place the matter at the "head of the calendar" and dispose of it summarily.2 2 ' Many ballot title controversies end up in the Supreme Court of Colorado, most brought by aggrieved initiative proponents, and a select few brought by petition opponents, with the court upholding the title board in all but a few cases.222 The same procedures
219. Measures referred by the General Assembly are listed by letter; measures initiated by the people are listed numerically. COLO. REV. STAT. 1-40115(2) (Supp. 1994). "A 'yes' vote on any measure is a vote in favor of changing
constitutional or statutory law
. ..
."
Id.
Only the ballot title appears on the
ballot. Id. 1-40-115(1). Colorado Constitution article V, section 1(5.5) refers to the ballot title and to "officials responsible for the fixing" of it. But there is no explicit provision to put only the title on the ballot, excluding the text. Article V, section 1 directs the Secretary of State to "submit all measures initiated by or referred to the people for adoption or rejection at the polls," COLO. CONST. art. V, 1 (emphasis added), and article XIX, section 2 directs that "amendments shall be submitted to the registered electors." COLO. CONST. art. XIX, 2 (emphasis added). One could argue that these provisions require that the full text of an initiative appear in ballots because a title is not a "measure" or an "amendment." Practice is clearly to the contrary, however. 220. COLO. REV. STAT. 1-40-107(1) (Supp. 1994). 221. Id. 1-40-107(2). 222. For the state-wide 1994 election alone, the Colorado Supreme Court issued 13 opinions on ballot titles. In re Proposed Initiated Constitutional Amendment Concerning "Fair Fishing," 877 P.2d 1355 (Colo. 1994); In re Proposed Initiated Constitutional Amendment Concerning the "Fair Treatment II," 877 P.2d 329 (Colo. 1994); In re Proposed Initiative on "Obscenity," 877 P.2d 848 (Colo. 1994); In re Proposed Initiative on Water Rights, 877 P.2d 321 (Colo. 1994); In re Petition on Campaign and Political Finance, 877 P.2d 311 (Colo. 1994); In re Proposed Initiative Concerning "Automobile Insurance Coverage," 877 P.2d 853 (Colo. 1994); In re Proposed Initiative for an Amendment to Article XVI, Section 6, Colorado Constitution, Entitled "W.A.T.E.R.," 875 P.2d 861 (Colo. 1994); In re Proposed Initiative Designated "Governmental Business," 875 P.2d 871 (Colo. 1994); In re Petition On School Finance, 875 P.2d 207 (Colo. 1994); In re Proposed Initiative on School Pilot Program, 874 P.2d 1066 (Colo. 1994); In re Proposed Initiated Constitutional Amendment Concerning the Fair Treatment of Injured Workers Amendment, 873 P.2d 718 (Colo. 1994); In re Proposed Tobacco Tax Amendment 1994, 872 P.2d 689 (Colo. 1994); In re Proposed Election Reform Amendment, 852 P.2d 28 (Colo. 1993). Only eight initiatives made the ballot.
96
[Vol. 66
should apply when the title board refuses to set a title for a 2 measure that contains more than one subject. 23 In reviewing ballot titles, the court typically notes that it "will not address the merits of the proposed initiative; instead, we give great deference to the Board's action in exercising its 2 drafting authority." 24 Courts will set aside the language only 2 when it is "clearly misleading." 25 The language "need only 2 fairly reflect the content of the measure." 26 The court is otherwise careful not to "interpret the meaning of proposed language or suggest how it will be applied if adopted by the 2 electorate." 27 "[Tihe Board is not required to include every aspect of a proposed measure in the title and submission clause ....)228 The court refuses to consider the constitutionality of the initiative in the ballot title review process 229 or otherwise consider challenges on the merits of the initiative proposal.23 The court's deference to the ballot title board's determinations is consistent with the court's interpretation of the "clear expression" requirement for titles of legislative bills. In Colorado, any given bill is limited to a single subject "which shall be clearly expressed in its title" and "if any subject shall be embraced in any act which shall not be expressed in the title, such act shall be void only as to so much thereof as shall not be so expressed."23 ' The title of legislative bills is more often the subject of litigation than the single-subject content require-
In all but two of the cases, the title board's action was affirmed. In re Proposed Initiative Designated "Governmental Business," 875 P.2d 871 (Colo. 1994); In re Proposed Election Reform Amendment, 852 P.2d 28 (Colo. 1993) (ballot title inaccurate). 223. The single-subject provisions say nothing about rehearing or judicial review. COLO. CONST. art. V, 1(5.5) (amended 1994); COLO. REV. STAT. 1-40106.5 (Supp. 1994). The statute governing rehearing and review says nothing about review of refusal to set a title or single subject, but it seems broad enough to apply. Id. 1-40-107. If it applies, opponents as well as proponents should be able to seek review when the title board finds that a measure contains more than one subject. 224. In re Petition on School Finance, 875 P.2d at 210. 225. Id. 226. In re Petition on Campaign and Political Finance, 877 P.2d at 313. 227. In re Proposed Election Reform Amendment, 852 P.2d at 31-32. 228. In re Petition on Campaign and Political Finance, 877 P.2d at 313. 229. In re Proposed Election Reform Amendment, 852 P.2d at 33 n.2. 230. In re Petition on Campaign and Political Finance, 877 P.2d at 313. 231. COLO. CONST. art. V, 21.
1995]
ment." 2 The purpose of the title requirement is identified in an early case: Another object is ... to apprise the people of the subjects of legislation by the titles of the bills, so that they might have an opportunity to be heard .... But few are able or care to take the time necessary to keep informed of all the legislation proposed at a single session, where it is necessary to examine in detail every bill in order to obtain this information. When, however, each proposed act is confined to a single subject, and that subject is clearly expressed in the title, those interested are put upon inquiry when legislation is proposed affecting such subject, without its being necessary for them to examine every bill for the purpose of seeing that nothing objectionable is coiled up within the folds of the measure. 3 In applying the language, however, the court has lost sight of the public notice aspect of the requirement and focused instead on preventing "surprise and deception to the members of the 234 As a consequence, General Assembly" itself. and in its desire to defer to the General Assembly, the court has discouraged particularity and detoothed the standard by holding that "[an appropriate general title which is broad enough to include all the subordinate matters is considered safer and wiser than an 235 enumeration of several subordinate matters in the title." The "clear expression" requirement now applies to the titles of 2 initiatives. 3 6 Perhaps this will lead courts to strenghten the 237 standard. In any event, there is ample room for better judicial protection of the public interest in both the clear expression rule and the ballot title process. For an understanding of how an officious citizen can use the process, consider the recent cases on ballot titles that were brought by Mr. Douglas Bruce against both the title board and
232. See In re House Bill No. 1353, 738 P.2d 371, 374 n.4 (Colo. 1987) ("The great majority [of cases under Article V, 21 of the Colorado Constitution] have focused principally on the sufficiency of the title to describe the contents of the bill."). 233. Catron v. Board of Comm'rs, 33 P. 513, 514 (Colo. 1893). 234. People ex rel. Dunbar v. Gilpin Inv. Co., 493 P.2d 359, 361 (Colo. 1972); see also Redmon v. Davis, 174 P.2d 945, 949 (Colo. 1946). 235. Parrish v. Lamm, 758 P.2d 1356, 1363 (Colo. 1988). 236. See supra note 217 and accompanying text. 237. The General Assembly declares that it intends the new standard to "prevent surprise and fraud from being practiced upon voters." COLO. REV. STAT. 1-40-106.5(1)(e)(II) (Supp. 1994) (emphasis added).
98
[Vol. 66
2 the proponents of two separate initiatives. 38 Mr. Bruce, the successful sponsor of a 1992 constitutional initiative and proponent of a 1994 initiative, contested the ballot title of a measure on school finance. The measure threatened to put on the ballot language that, if passed, would modify his 1992 tax amendment. Mr. Bruce also contested the ballot title of an initiative on campaign and political finance that threatened to compete with his 1994 initiative proposal. By challenging the ballot title on school finance, Mr. Bruce was able to reduce significantly the time its proponents had to gather petition signatures, with the result that they were not able to gather enough valid signatures to get their measure on the ballot. In a third hearing, Mr. Bruce supported the ballot title of 2 his own initiative. 3 9 The juxtaposition of Mr. Bruce's arguments on the ballot title of his own initiative and his arguments on the ballot titles of other initiatives demonstrate how a sophisticated student of the initiative process can abuse the system. In attacking the school finance initiative, Mr. Bruce argued that the ballot title, already over two hundred words, needed more detail2 4 and that the summary should have included more detail on fiscal impact.24 ' Included in his complaint was a request that the title board indicate in the 242 ballot title that the initiative was "internally conflicting." In his attack on the campaign and political finance initiative Mr. Bruce also argued for more detail in the ballot title and in the fiscal impact statement.24 3 In defense of his own initiative, however, Mr. Bruce argued that the ballot title was too long.2 44 Moreover, under his 1994 proposed initiative, future ballot titles would have been limited to seventy-five words, and
238. In re Petition on School Finance, 875 P.2d 207 (Colo. 1994); In re Petition on Campaign and Political Finance, 877 P.2d 311 (Colo. 1994). 239. In re Proposed Election Reform Amendment, 852 P.2d 28 (Colo. 1993). 240. In re Petition on School Finance, 875 P.2d at 210. 241. Id. at 211. 242. Id. at 212. 243. In re Petition on Campaign and Political Finance, 877 P.2d 311, 314-15 (Colo. 1994). 244. In re Proposed Election Reform Amendment, 852 P.2d 28, 33 (Colo. 1993); see also Jeffrey A. Roberts, Bruce Raps Title Assigned Election Reform Bid, DENV. POST, Feb. 4, 1993, at B5.
1995]
any requirement of a fiscal statement would have been eliminated.24 5 Mr. Bruce's activism in contesting ballot titles for the 1994 election must be compared with his success in the ballot title procedure in 1992. His initiative in 1992, Amendment One on taxation and revenue limits, passed with what is now regarded as an incomplete ballot title. The title left out many of the details in the proposal,2 46 and may have even contradicted parts of the text.24 7 b. The Ballot Information Booklet and Newspaper Notice In the 1994 election Colorado voters passed a referendum that amended the Colorado Constitution to provide for a ballot information booklet. 248 The booklet will be mailed to all registered voters before any election in which an initiated or referred measure is on the ballot, at no charge to the proponents. The booklet will contain the text and ballot title of each measure and a "fair and impartial analysis" of each measure. 249 The
analysis, to be prepared by a "nonpartisan research staff of the general assembly," will include a summary of the measure and a synopsis of the major arguments for and against the measure and "may include any other information that would assist
understanding the purpose and effect of the measure."211 Any
individual may file comments for consideration by the drafting
245. 1994 Amendment 12, 6(a) (rejected by Colorado voters Nov. 8, 1994) (on file with Colorado Secretary of State). To add further irony, Mr. Bruce's successful amendment initiative on taxing and spending powers of the government requires the government, when submitting revenue and expenditure requests to the populace, to include something akin to a fiscal impact analysis. COLO. CONST. art. X, 20(3)(b)(ii)-(iv) (1992 Amendment One). 246. See supra note 174. 247. Compare Stecher v. Northglenn, No. 93-CV-0263 (D. Ct. Adams County May 5, 1993) with Cervemy v. Wheatridge, No. 93-CV-0423 (D. Ct. Jefferson County Mar. 22, 1993) (disagreeing over whether Amendment One stops special elections on all matters or just revenue questions). Both courts found the ballot title and the text in conflict. 248. COLO. CONST. art. V, 1(7.5) (amended 1994). 249. Id. The booklet also includes the special notice required for all measures that propose to increase taxes or government borrowing or spending by article X, section 20(3)(b). Id. 1(7.5)(c). 250. Id. 1(7.5)(a). The director of research of the legislative council prepares the booklet, which is reviewed by the legislative council itself. COLO. REV. STAT. 2-3-303(1)(g) (Supp. 1994). The council consists of eighteen legislators including the leadership. Id. 2-3-301.
100
[Vol. 66
staff. Interestingly, as yet, there is no procedure in place for protesting the contents of the booklet in a judicial forum, in stark contrast to the technical review procedure in the ballot title process.2 51 When initiatives relate to issues that are uniquely in the realm of expertise of public officials, initiatives that alter the processes of government, for example, the public would seemingly be interested in the views of some of those officials on the suggested change. Our distrust of elected officials prevails, however. Colorado has enacted strict limits on the role of public officials in the initiative process, regardless of the issue. The ballot information booklet and other official summaries are the only fora for comment on initiatives by the 2 state or any political subdivisions. 5 2 No government unit can make contributions, in cash or in kind, to either side in an initiative campaign nor can they expend funds to urge electors 2 to vote in favor or against a given initiative, 53 although a unit can pass a resolution taking a position on the issue.25 4 An official may answer unsolicited questions but may spend no more than fifty dollars of public money on letters and phone calls incidental to expressing her opinion.255 Public officials can, however, use private time and money to express their personal opinions. 2" But they do so at the risk that interested parties will charge that private time was really public time.25 7 The 1994 referendum also amended the constitutional provisions on newspaper notice. 2 1 Under a decades-old provision, proposed constitutional amendments had to be published in two issues of two newspapers of "opposite political faith" in each county in the state. 2 9 The new requirement reduces the number of newspaper notices in favor of the ballot
251. This assumes that the summary prepared by the ballot title board is not the same as the summary prepared under this provision. 252. COLO. REV. STAT. 1-45-116 (Supp. 1994). 253. Can a state university hold a public debate on an initiative, using university facilities and university personnel? 254. COLO. REV. STAT. 1-45-116 (Supp. 1994). 255. Id. 1-45-116(1)(a). 256. Id. 1-45-116(1)(b)(II). 257. Can a city attorney (or a law professor) leave work and debate a proponent, during a weekday, at the invitation of a city bar association? 258. COLO. CONST. art. V, 1(7.3) (amended 1994). 259. Id. art. XXIII, 1, amended by COLO. CONST. art. V, 1(7.3).
1994]
information booklet. As a result, several newspapers editorialized against the referendum, often without disclosing their pecuniary interests, causing the vote to be close.26 The new language requires publication of the text and title of each initiated and referred measure only once in one newspaper in each county at least fifteen days before the election.26 ' Another early law, now rescinded, had added a publication requirement paid for by the proponents,26 2 which had the effect of substantially decreasing the use of initiatives.26 3 4. Comments on the Colorado Notice Procedures Voters in Colorado have traditionally had two means of informing themselves on the text of initiatives, short of requesting the text from proponents or traveling to a public library, and as noted above, voters have just added a third. First, when voters are approached by petitioners, the petitions contain ballot titles, summaries, and the full text of the initiatives. Second, a summary and the full text of each initiative is published once in every major newspaper. Third, each voter will, in the future, receive a ballot information booklet on statewide initiatives. It is easy to minimize the effect of each of the three sources of information. Petitions are signed by a small fraction of the electorate. Even those who sign do not read the summary or text of the initiative in question. Petitioners report that they can gather several thousand signatures and have not one signer read the petition language when she signs. Instead signers choose to sign or not to sign based entirely on an abbreviated oral representation made by the solicitor. A solicitor for the failed
260. The main newspaper in the city of Boulder, the Daily Camera,opposed the referendum and it lost in Boulder County. "No" on Referendums A, B, and C, DAILY CAMERA (Boulder, Colo.), Oct. 12, 1994, at C2. 261. COLO. CONST. art. V, 1(7.3) (amended 1994). 262. For over 30 years, legislation provided for another round of newspaper publications. A 1941 statute provided that after the ballot title had been fixed, the Secretary of State was required to publish, once a week for two successive weeks in each county in a newspaper of general circulation, a copy of the title and text at the expense of the proponents. Act of Mar. 31, 1941, ch. 147, 1, 1941 Colo. Sess. Laws 480, 481. The Colorado Supreme Court held the statute unconstitutional in 1972. See Colorado Project Common Cause v. Anderson, 495 P.2d 218 (Colo. 1972). 263. See Hubert D. Henry, Popular Law-Making In Colorado, 26 RocKY MTN. L. REv. 439, 449 (1954).
102
[Vol. 66
1994 Amendment Twelve, for example, might have said exclusively: "Will you sign a petition for limiting politicians' pay raises?" Newspaper publication is effective only for those who read newspapers and even then only for those who look beyond the sports page and the comics. Those who read or skim the news usually skip the omnipresent Legal Notice materials, which list sheriff sales and the like. Even vigilant newspaper readers can easily miss the one or two days that initiatives appear. Moreover, if our vigilant reader does find the text of initiatives on the one or two days they are published, she will face the daunting task of reading small typeface and turgid text, threatening a significant increase in newspaper reading time. More often than not, most in this small band of readers (and even smaller subgroup of voters) will put the section aside for later reading only to find, thankfully perhaps, that a spouse has thrown the papers out in the trash. The ballot information booklet offers the most promise for voter education, if people separate it from their junk mail and take the time to read it. The booklet is expensive and will generate litigation, once rules for appeal are established, as proponents and opponents with much at stake fight over (and appeal until exhaustion) whether the content of booklets is "fair and impartial." Our courts will be very busy around election time. Like it or not, most voters in Colorado will take their cue, first, from thirty second television spots on any initiative and, second, from the ballot title language, which they will confront when they get to mark their ballots. The importance of television spots, and to a lesser extent radio spots, cannot be overstated.264 Television spots must necessarily be very conclusory on the effect of an initiative, there is little or no time to explain detail, and spots often stretch and bend the truth; advocacy pushes out accuracy. If one side has more money, as did the opponents of 1994 Amendment One on the tobacco tax (proponents had no money for television spots at all), it can be a lopsided match in-
264. Newspaper advertisements on initiatives, which allow for more detail in the argument, are uncommon, perhaps for many of the reasons noted in the text on the ineffectiveness of required newspaper publication of the text of initiatives.
1995]
deed.265 Money counts in this battle of ads, and, more likely 266 A fast-forming lucrathan not, will determine the winner. tive business is coalescing around those who service initiative 267 activists. The trustworthiness of those who pay for and compose the television spots is an issue. It is fair game for opponents to question each other's sources of funding because it bears on their credibility. Tobacco companies' ads against the tobacco tax are self-serving and ought to be discounted. Yet, in spite of legislation regulating contributions, it is-often difficult to track down the real source of many contributions.268 The ballot title, a stale, abstracted rendition of the issues in an initiative, often with little or no indication of the effect of initiative language on the matters listed, is the only language on the ballot. For example, a ballot title of "campaign spending reform" does not specify what the reforms are, and the title would be the same if the reform is that no corporations can contribute to campaigns or is a dollar limit on all contributions from whatever source. Many voters confront initiatives for the first and only time in this sterile form. As a consequence, disputes over ballot titles are frequent, with the Colorado
265. The proponents of 1994 Amendment One spent $209,850 and the opponents spent $6.6 million. See Jerd Smith, Initiative Tabs Top $10 M, DENV. Bus. J., Nov. 4-10, 1994, at 1A, 47A. 1994 Amendment One was defeated with 62% of the voters opposing it. Aldo Svaldi, Bond Issues Fly, While Taxes Die, DENV. Bus. J., Nov. 11-17, 1994, at 1A,37A. 266. In the 1994 election, campaigns backing initiatives were outspent by forces opposing them by almost six to one. Smith, supra note 265. Of nine initiatives only one passed, Amendment 17 on term limits, and there was little or no money spent on either side of the campaign. Id. Seven of the other eight initiatives passed or failed in correlation with whether the proponents or opponents spent more. Id. The only exception, Amendment 13 on gaming in Manitou Springs, had proponents spending $183,664 and opponents spending $97,468. Id. 267. The business consists of those who provide advertising, political consultants, lawyers, and campaign managers. In the 1994 Colorado election, political consultants billed the proponents of Amendment 400 (the arts tax initiative) almost $100,000 for two months work. Id. Overheads on. campaigns, for managers and others, is estimated by professionals at 10 to 15% of the total contributions for the big ticket measures. Id. 268. In Colorado, political committees supporting or opposing initiatives must file a list of their contributors and individuals spending funds outside the political committees must file separately. COLO. REV. STAT. 1-45-104, -108, -110 (Supp. 1994). But the use of layers of front committees or the use of individuals whose occupations are not specified can make it difficult to track down funding sources. But see id. 1-45-106 (1980 & Supp. 1994) (political committees must list all supporting and affiliated organizations).
104
[Vol. 66
Supreme Court often called on to decide whether the title board 2 has accurately listed the issues. 6 9
V. RECOMMENDATIONS
A. Our PrimaryConclusion: ProceduresNot Only Determine the Mix of Representative and Direct Democracy, They Also Determine Whether the Initiative Process Works to Effect Direct Democracy The populists who gave us initiatives at the turn of the century had lofty aims: they believed that the initiative process would best empower citizens, promote good government, perfect majority rule, and control disobedient and dishonest government officials. Inherent in their optimism were beliefs that representative democracy must be supplemented with a heavy dose of direct democracy and that the initiative process was an effective mechanism to achieve direct democracy. Traditional arguments about the value of the initiative process dwell on the content of successful controversial proposals.27 These debates miss the important, indeed crucial, role played by procedural choices. Not all initiative processes are alike, and the differences affect both of the populist assumptions. We discuss the effect of procedure on both assumptions below. First, the procedural details of the initiative process control the mix of representative and direct democracy. Those who favor a maximum role for direct democracy argue for no limits on the content of initiatives, low qualifying thresholds for accessing the ballot, a simple majority of votes cast for enactment,27 ' protection from repeal or revision by elected officials, 272 government subsidies for petitions and publication,273 and limits on campaign involvement and even com-
269. See supra part IV.E.3.a. 270. See, e.g., Symposium, The Bill of Rights v. The Ballot Box: Constitutional Implications of Anti-Gay Initiatives, 55 OHIO ST. L.J. 491 (1994). 271. That is, they favor enactment by a majority of those who vote, without a minimum, even for constitutional amendments. 272. This includes protecting a successful initiative from veto by a governor or from repeal or revision by the legislature. In its extreme form this also includes limits on referendum or on government officials' own access to the ballot. 273. The government can supply petition forms and pay for the costs of all public notices. In its extreme form, the government grants subsidies for campaigns both for and against an initiative, subsidies for private counsel for help in drafting and processing an initiative through the title board and petition
1994]
ment by public officials.274 Those who believe in a more limited role for direct democracy argue primarily for limited ballot access,275 higher voting thresholds,27 6 the indirect form of the initiative, and ease of amendment or repeal. 7 The impact of the initiative process on a representative system is controlled by these procedural choices. The authors' positions on the best mix of representative and direct democracy are the following. Our primary form of government is representative. We delegate power to our elected officials under general guidelines, such as campaign platforms, and ask them to attend to the details of government. Taking our cue from shareholder voting power in publicly-held corporations,27 8 we believe that initiatives ought to be structured as
validation process, or subsidies for hiring petition circulators. 274. 1994 Amendment 12, which failed with only 23% of the vote in favor, contained a form of these provisions. Jerd Smith, Deep Pockets Fuel Election Victories, DENv. Bus. J., Nov. 11-17, 1994, at 1A; 1994 Amendment 12 (rejected by Colo. voters Nov. 8, 1994) (on file with Colorado Secretary of State). 275. Examples include higher signature requirements for petitions, prohibition of mail-in signatures, rigorous circulator qualifications, and proponent payment of the government costs of reviewing and publishing an initiative. 276. An example is to require a supermajority vote for constitutional amendments. See Lynn A. Baker, ConstitutionalChange and Direct Democracy, 66 U. CoLO. L. REV. 143 (1995). Other examples include a minimum vote requirement, so that a majority vote at a low turnout election cannot effect a major change. 277. Advocates seek to treat most initiatives as normal laws, with veto power in the governor and power in the legislature to repeal or amend. 278. The preferred structure of publicly-held firms, honed over time, is a delegation of management power to experts, with shareholder input reserved for fundamental changes. Shareholders vote only on the election of the board of directors and on ratification of major corporate transactions (mergers or dissolutions, for example). See Dale A. Oesterle & Alan R. Palmiter, Judicial Schizophrenia in ShareholderVoting Cases, 79 IOwA L. REV. 485, 495-512 (1994). Limited shareholder access to the firm's proxy machinery is provided in the Securities and Exchange Commission's controversial Rule 14a-8. 17 C.F.R. 240.14a-8 (1994). See generally Alan R. Palmiter, The Shareholder Proposal Rule: A Failed Experiment in Merit Regulation, 45 ALA. L. REV. 879 (1994). Under the current version of the Rule, shareholders may put proposals in the firm's proxy if they own at least one percent or one thousand dollars in market value of the securities entitled to vote at the meeting. See 17 C.F.R. 240.14a8(a)(1). The proposal may be omitted if, among other things, it "deals with a matter relating to the conduct of the ordinary business operations" of the firm or is not "significantly related" to the firm's business. Id. 240.14a-8(c)(7). A Rule 14a-8 proposal may not be resubmitted unless it receives specified levels of support. See id. 240.14-8(c)(12). The analogy between popular elections and corporate elections is, of course, imperfect. Disgruntled shareholders can easily sell their shares, while disgruntled citizens cannot as easily leave their domicile. The difference is reflected perhaps
106
[Vol. 66
a check on this delegation, a safety valve, but not a substitute or even an equal partner. When discontent with elected officials reaches high levels of intensity on selected issues, the initiative process provides a pressure release. The threat of an initiative, even when unused, has some effect on aligning the interests of elected officials and their electorate. Those who believe that policy-setting through the initiative process ought to dominate or be an equal partner with policy-setting through representative government run afoul of the process's basic and unavoidable paradox: The more the populace uses the initiative process, the less effectively it works.2 79 Our safety valve view is reflected in some of our recommendations below. Second, most of our recommendations are directed at the populists' second assumption: that initiatives are synonymous with direct democracy. Regardless of one's view on the proper mix of representative and direct democracy, the initiative process as a vehicle for direct democracy is problematic. An initiative process with an inadequate procedural base does not serve as a voice for the populace, it serves only the purposes of those discrete individuals and groups who can best capture the process. The major abuses that must be addressed through procedural solutions are drafting deficiencies, petition fraud, and misleading public communications in campaigns. Whatever one's view on the positive potential of initiatives, an initiative process that invites abuse may be inferior to no initiative right at all. At issue is how to structure the initiative process to minimize abuse and maximize its value, that is, its accuracy in reflecting the honest view of a majority of the populace. We offer our suggestions in three parts. First, we provide recommendations about how the initiative should be structured when there is complete freedom of choice, that is, when one is drafting from scratch constitutional and legislative provisions on the process. These proposals are relevant for those designing a proposal for a national initiative as well as those suggesting amendments to state constitutions. In our second and third
in the fact that with corporations, even the election itself is generally controlled by the board of directors through the firm's proxy solicitation machinery, a situation we would not and do not permit in popular elections. Proxy fights for control of a board are rare. See Oesterle & Palmiter, supra, at 508. 279. Experience demonstrates that a flood of initiatives inevitably results in lower vote totals as people tire of the process and abstain. See supra text accompanying note 45.
19951
parts, we offer suggestions specific to the current situation in Colorado, aimed at the state legislature under the assumption that the constitution stays as is, and at the courts under the assumption that both the constitution and state implementing legislation stay as they are. B. Writing on a Clean Slate 1. Use the Indirect Initiative Process By involving the legslature, yet allowing a vote of the people, the indirect initiative tends to get some of the best of both systems. It offers better chances to overcome the drafting problems of the direct initiative. The legislature can, or must if a state's constitution requires, give the initiative consideration in hearings and deliberations. This process can point out unintended consequences to initiators, and can at least make them aware of the points of most intense disagreement. In some instances, the problem of oppressing unpopular minorities can be reduced. Under some systems, the legislature can propose alternatives that avoid problems with the citizens' draft. Opponents of the indirect initiative often complain that it is simply a delaying device, and that American legislatures in indirect initiative states seldom do anything during the period set for legislative review. However, we think the process in many American states is too hasty anyway, so some delay is wise even if the indirect form is not used. And legislative lethargy can be reduced by requiring a minimum degree of legislative consideration, and by empowering legislatures to offer alternatives, if that power is lacking. 2. Preserve the Integrity of the American Constitution: Require Minimum Votes and Word Limits for Constitutional Amendments As we have pointed out above, frequent and easy resort to the initiative device to amend state constitutions has filled some of these constitutions with measures that are, by the traditions of American constitutionalism, matters for ordinary legislation. It also allows sweeping changes to be embedded in the constitution by the modest plurality that typically adopts initiated amendments, sometimes at serious cost to constitutional stability.
108
[Vol. 66
These effects are also true in Switzerland, where the device works well, so they are not necessarily bad. But they seriously revise the nature of a state's constitution away from the American tradition and in favor of the Swiss. The Swiss constitution is very different from ours in other ways as well. Judicial review is much more limited, and there is a much weaker sense of separation of powers. These effects will result from heavy use of the initiative here. When the state's constitution is amended, judicial review under the state's constitution is eliminated, and traditional executive and legislative functions are constricted. Even initiated statutes avoid the executive veto. We think the American system is worth preserving. Accordingly, in this section we recommend ways to reduce resorting to the initiative to amend constitutions and to encourage its use to enact statutes instead. To do this, the statutory route must be made much more attractive, and the amending process less so. There is traditionally a higher number of signatures needed for initiatives that amend constitutions than for those that enact statutes, although even that difference has been eliminated in Colorado. In any case, the petitioning industry can readily meet the constitutional threshold in most states, so this difference does not have much significance. Other proposals offer more promise. In the order of the intensity of our preference we suggest four approaches. First, we would require, as is in place in a few states, a minimum vote to enact a constitutional amendment. The present rule in most states requires simply a majority of the votes actually cast, and this is often a rather low percentage of the electorate. For a relevant example, in the few modern referendums held in Britain on constitutional matters, there has been a requirement of approval by at least forty percent of the electorate.2 Another version would require a majority of registered
280. See THE REFERENDUM DEVICE 9-10 (Austin Ranney ed., 1981). Even the most prominent constitutional initiatives in the U.S. fall short of the British rule. See supra note 18.
1995]
voters. 1 Statutory amendments would continue to require only a majority of votes cast. Second, we would limit constitutional amendments rather strictly to a single subject, but be less strict about the scope of statutory initiatives. The traditional single-subject rule is usually not a very strict limit. To make it effective, we recommend, third, combining it with a rule limiting constitutional initiatives to five hundred words.2 82 Constitutional amendments that are longer ought to be promulgated through constitutional conventions or legislative proposals. We think a limit on the length of statutory initiatives might be prudent as well, but the limit could be much higher, say 5000 words. Fourth, we recommend requiring that the final text of constitutional initiatives be drafted by the state's attorney general, legislative counsel, or other official, as is now done in New Zealand. This would of course be subject to an enforceable requirement that the text accurately carry out the initiators' purpose. Statutory initiatives would not have this limit. These measures would induce initiators of measures that would traditionally be viewed as statutory to go the statutory route, preserving the state constitution as fundamental law. Even when initiators would need to amend the constitution, they would be induced to make a simple, enabling amendment with a matching statute to carry out their program. To make the measures fully effective to restore dignity to state constitutions, it might be necessary to impose some parallel restrictions on the ability of the state legislature to propose constitutional amendments because low-level provisions are added to state constitutions that way as well. Moreover, these measures would be problematic to add to a state constitution that is already cluttered with many "legislative" provisions, which would then be made more difficult to amend or repeal
281. Cf MINN. CONST. art. 9, 1 (referred constitutional amendments require a majority of total votes cast on any matter in the election); MISS. CONST. art. 15, 273(7) (initiated constitutional amendments require 40% of total votes cast on any matter in the election). 282. The limit is analogous to the limit in the shareholder proposals rule, SEC Rule 14a-8(b)(1). See supra note 278. The Rule requires all shareholder proposals and their supporting statements to be under 500 words. 17 C.F.R. 240.14a-8(b)(1) (1994). See also CALIFORNIA COMM'N, supra note 9, at 111 (proposing limit of 5000 words for all initiatives).
110
[Vol. 66
than they were to enact. 283 To meet this problem, we propose
that an initiative to repeal a constitutional provision be subject to less severe restrictions than one proposed to add language. The scheme we propose makes best sense in the context of a state constitutional convention or in designing a national right of initiative. 3. Treat Initiated Laws No Differently Than Legislatively Passed Statutes As previously noted, many initiative states restrict the 284 In power of their legislatures to amend initiated statutes. California, changes can be done only by another initiative or referendum. This entrenches measures similarly to constitutional amendments. It makes very difficult even technical amendments to work out defects in legislation. We think these limits are based on an unrealistic fear of legislatures.285 Legislatures are in fact very cautious about trying to change anything in initiatives because of popular criticism that is aroused. However, this fear is real and needs to be accommodated. We suggest that legislatures be empowered to amend initiated statutes subject to two protections. First, such changes must not be combined with any other legislation; there should be a strict single-subject rule. Second, the procedures for a referendum to rescind this particular kind of legislation should be easier to meet than for rescinding referendums generally. The Swiss experience shows that the rescinding referendum can be a powerful check on the legislature.286
283. For a failed attempt to clean up a badly cluttered state constitution, see Arrow, supra note 47. 284. See supra note 11 and accompanying text. 285. There is no evidence that legislatures in states that allow unrestricted amendment or repeal abuse the power. Similar arguments can be made for giving a governor veto power. It will rarely be exercised, and when exercised, a governor will do so only when her extraordinary concern about the effect of a measure on the welfare of the state outweighs the risk to her political future. This is a fire alarm we may want to respect, especially since most initiatives are passed by a favorable vote of less that 33% of the total number of registered electors in the state. The major difference between an initiative and a legislative bill is, of course, there is no process for overriding a veto of an initiative. A partial solution could allow a governor to veto only initiatives that pass by less than a 67% majority vote (or, in the alternative, a favorable vote of less than 50% of the total number of registered voters). 286. See supra note 46 and accompanying text.
1995]
This kind of check would tend to induce advance negotiations between proponents of the measure to be amended and the legislature, which would work out many problems. 4. Enforce a Single-Subject Rule Many difficulties with present practices for initiatives arise out of issues about their scope. If an initiative covers unrelated subjects, voters must decide whether to swallow the bitter to get the sweet. In general, the longer an initiative, the less well voters understand it. Its title and summary become less accurate. Its problems of drafting, unintended consequences, hidden issues, and the like become more serious. In our discussion of constitutionalism and amending statutes above, we have made several proposals to address problems of scope. Here we add a few more that do not directly connect with those subjects. A single-issue rule, interpreted sensibly, is a useful limit on the scope of initiatives, both constitutional and statutory. Proponents of the initiative complain that such limits unduly restrict the initiative or make it too expensive to circulate multiple petitions. To meet this objection, we propose allowing initiatives that cover more than one subject to be joined in a common petition, but to require separate votes by the electorate. In this way, voters, like state legislators, can vote separately on discrete measures. But petition circulators can circulate multiple measures on a common petition. In our discussion above on constitutionalism, we have proposed limits on the number of words in an initiated measure, a strict limit for constitutional amendments, and a more generous limit for statutes. These limits can also be worked into a multiple petition process. As discussed below, some state courts have imposed a limit on the scope of initiatives by ruling that the scope of some initiatives is so broad that they "revise" rather than "amend" the state's constitution. This rule depends on language in the state constitution that contrasts revising, which requires a constitutional convention, with amending, that can be done by 2 initiative. 7
287. See supra notes 181-83 and accompanying text; infra notes 321-24 and accompanying text.
112
[Vol. 66
5. Provide a Qualifying Period Long Enough to Allow Discussion, Deliberation, and Amendment of Proposed Initiatives As previously noted, qualifying periods vary from ninety 288 We think mandated days to an unlimited period of time. short periods, such as Colorado's six month limit, are too hasty, especially for constitutional initiatives. Even if the indirect initiative is not adopted, some of its virtues can be achieved by extending the period for qualification. Based on the Swiss experience, one to two years seems optimal. To make the extended period useful, there must also be reasonable procedures for review of drafts, public notice and hearing, consultation among proponents, officials, and opponents, and opportunities to amend proposals or to withdraw them in favor of legislative substitutes that are satisfactory to proponents. Legislatures should be empowered to refer competing measures, as in Switzerland. There should be clear rules about who can speak for proponents to amend or withdraw proposals.28 9 Mandatory public review and comment on draft initiatives can be valuable if there is time enough for discussion, negotiation, and involvement of opponents. However, if officials must give serious and substantive review to every draft proposal, no matter how frivolous, the cost of review will cause the review of each proposal to be inadequate. Therefore we propose a substantial filing fee for initiative proposals, on the order of one thousand dollars for statewide proposals, to deter the frivolous29 and to defray some of the cost of review and comment. This recommendation may strike some advocates of direct democracy as unfair to poor proponents, but that is a shortsighted view. It takes a lot of money, much more than one
288. See supra note 105 and accompanying text. 289. Also, when initiators qualify an initiative too late to make the ballot in the year of submission, the measure should go on the next year's ballot, unless withdrawn in favor of a compromise with the legislature. This will reduce the intense pressures that surround the qualifying process in many states. 290. In the September 1994 conference, a Colorado state official related the case of parents who submitted an initiative proposal with their child as a civics lesson. Catharyn Baird, Tape of Conference on Governing by Initiative, supranote 125. However educational this was for the family, mandatory review by the state was expensive. Id.; see also CALIFORNIA COMMN, supra note 9, at 137 (similar problems in other states).
19951
thousand dollars, to qualify a statewide initiative. And it takes broad and organized support to get a favorable vote. Thus to qualify and then to win the election requires organized backing of sufficient extent that a one thousand dollar filing fee is not a significant barrier. And the need to improve drafting of ballot measures, particularly those that amend state constitutions, is very great. An alternative to the filing fee is a two-step initiative process. Initiators gather a certain number of signatures to show significant backing for their proposal. At that point, public review, hearing, and comments are required, and the proposal can be amended or withdrawn in favor of a legislative alternative. However, the cost to proponents of this procedure probably exceeds a one thousand dollar filing fee, so we think the fee the more reasonable procedure. Of course, proponents might be given a choice of either method. C. The Role of Implementing Legislation in Colorado The Colorado Constitution empowers the General Assembly to pass implementing legislation on initiatives.2 9 ' Another relevant provision states, "The general assembly shall pass laws to secure the purity of elections, and guard against abuses of the elective franchise. " "' The lead role in protecting the populace against abuses of the initiative power then rests with the legislature.29 Many of the recommendations made above are clearly beyond the power of the legislature (the minimum vote requirements for constitutional amendments, for example) and some are arguably beyond the power (the word limit, for example). In any event, the legislature can work within the structure of its existing legislation to effect some positive changes.
291. See supra note 97. 292. COLO. CONST. art. VII, 11. 293. It is interesting to speculate whether the courts would have some role if the legislature, breaching its obligations, refused to pass any legislation. At issue would be whether such a case was justiciable. Cf Pacific States Tel. & Tel. v. Oregon, 223 U.S. 118 (1912) (challenge to initiative under "republican guarantee" clause held not justiciable).
.114
[Vol. 66
1. Extend the Qualifying Periods, and Establish Fair and Effective Procedures for Ballot Title Protests, Single-Subject Review, and Negotiations About Draft Initiatives The Colorado Constitution requires review of every draft initiative within two weeks after its submission.294 The defects in this procedure have been pointed out, and the state constitution would have to be amended to cure most of them. One remedy open to the legislature is to direct that review be broader than the narrow, technical review now undertaken. It would be useful for reviewers to do more to carry out their existing duty to suggest "editorial changes to promote compli295 ance with the plain language provisions" of the statute, particularly for initiatives that would amend the state constitution.2 96 However, the short time span and requirement to review every proposal, including the trivial, limit the review procedure in any case. Other preliminary procedures are controlled by statutes, which the General Assembly can amend. First, procedures governing the new single-subject requirement297 should be clarified. Administrative review is sensibly done simultaneously with the fixing of the ballot title, but it should be clear that
294. COLO. CONST. art. V, 1(5). 295. COLO. REV. STAT. 1-40-105(1) (Supp. 1994). See supra notes 159-64 and accompanying text. 296. It would be more useful still to impose enforceable requirements of simplicity and clarity, but this would probably require a constitutional amendment. See COLO. CONST. art. V, 1(5). There is historical precedent for judicial intervention to enforce standards of minimal drafting clarity. The first statute implementing the Colorado initiative allowed proponents to write the ballot title. Act of May 8, 1913, ch. 97, 4, 1913 Colo. Sess. Laws 310, 312, repealed by Act of Mar. 31, 1919, ch. 131, 1919 Colo. Sess. Laws 431. The ballot title had to be "brief and not conflict with that selected in any petition previously filed." Id. If the ballot title was "misleading or unreasonably long," private parties could ask the state courts to enjoin its use. Id. If enjoined, "a majority of persons representing signers shall select another ballot-title that shall fairly describe the measure submitted to vote." Id. Judicial enforcement of clarity is now done by the Supreme Court of Florida. See Advisory Opinion to the Attorney Gen. re Tax Limitation, 644 So.2d 486 (Fla. 1994). 297. COLO. CONST. art. V, 1(5.5) (amended 1994); COLO. REV. STAT. 1-40106.5 (Supp. 1994). See supra notes 184-94 and accompanying text.
1995]
9 STRUCTURING BALLOT INITIATIVES
rehearing and judicial review of both can be done together in a 29 single proceeding open to opponents as well as proponents. s Second, there should be adequate notice of the pendency of titling and single-subject review both to the legislature and to the public, and review of these decisions should be fairly available to opponents who are new to the process. Third, the General Assembly should establish a regular procedure to review pending measures to identify those likely to be important to its members. These in turn should be considered for committee hearing and in appropriate cases for discussion and negotiation with proponents. In general, the General Assembly should be more actively involved in the initiative process, while initiative proponents retain the right to force a vote on their own proposal. Fourth, the authority of proponents to negotiate compromises with the legislature should be made clear. The definition of who are proponents should be made certain, and it should be clear that they can act to amend or withdraw their proposal by majority vote.299 Fifth, to facilitate other measures, the periods for qualifying initiative should be lengthened. Present periods for review an of drafts (two weeks) and title-setting (about the same) are too short. The maximum period for the entire process (about eleven months) should be longer, especially for constitutional initiatives. These changes are needed to allow more time for legislative review and possible negotiation with proponents. They would also give more time for other interested persons to participate in the qualifying process, yet give proponents ample time to gather signatures. The opportunity under present procedures for opponents to tie up an initiative proposal in court so that proponents do not have enough time to gather signatures3 should be eliminated. 2. Put the Full Text of Initiatives on the Ballot With the single-subject rule in place, initiatives will be shorter. With shorter initiatives comes an opportunity to take some of the importance away from the currently momentous
298. See supra note 223 and accompanying text. 299. Swiss practices provide a model. See supra note 156. 300. See supra notes 238-45 and accompanying text.
116
[Vol. 66
and decisive ballot title procedure.3"' The government can put the full language of any initiative on the ballot. 2 Proponents, recognizing the problem of asking for voters to pass a long measure, would be further pressured to keep their 3 measures brief." 3 The change would give voters the best chance of not being misled by a conclusory or incomplete ballot title, 30 4 avoid some of the contentious and expensive procedures currently employed, 30 5 and eliminate the disadvantage suffered by targeted groups who wake up late to the importance of the ballot title process and are, therefore, excluded.30 6 D. The Role of the Colorado Courts in Policing Initiative Procedures Assuming that neither the Colorado Constitution nor Colorado statutes are revised, the last line of defense against initiative abuse is in state courts. As noted above, state courts play two roles in the initiative process; they review administrative determinations and they rule on claims of unconstitutionality. Of the total time the Colorado courts spend on cases involving initiatives, most is spent on tasks enumerated in the implementing legislation. The Colorado Supreme Court has 0 direct review of ballot title board decisions. 3 ' And District Courts hear protests of the Secretary of State's findings on the sufficiency of petition signatures. 8 Cases on claims of unconstitutionality come infrequently and are of two types: first, frustrated proponents attack state implementing legislation on the initiative process as inconsis-
301. See supra part IV.E.3.a. 302. In Colorado, the full text of an initiative has never appeared on the ballot. 303. The legislation putting the full text on the ballot could apply only to initiatives for constitutional amendments, giving proponents two incentives: First, to consider a non-constitutional initiative, and second, to keep constitutional amendments brief. 304. For an extreme example, consider what many people would infer from a ballot title on "Election Reform" or "Taxation and Spending Limits" if they have not read the specific language (or an adequate summary) of an initiative. 305. See supra text accompanying notes 219-22. 306. See supra text accompanying notes 211-16. 307. See supra notes 221-23 and accompanying text. 308. COLO. REV. STAT. 1-40-118 (Supp. 1994). The protesting party has the burden of proof. Id. 1-40-119 (Supp. 1994). The district court is to hold hearings "as soon as is conveniently possible" and conclude hearings within thirty days. Id. Appeals go directly to the Colorado Supreme Court. Id.
1995]

9
tent with the state constitution;
and, second, unsuccessful What is missing
opponents of a successful initiative attack the content of the 3 initiative as a violation of the federal Constitution. 10 Claim-
ants have been successful in each genre.3
309. See supra note 102. Proponents have also successfully attacked state legislation under the First Amendment. Meyer v. Grant, 486 U.S. 414,428 (1988); see infra note 311. 310. See supra notes 13, 52. 311. See supra notes 52, 102. The courts have on occasion frustrated the efforts of the state legislature to impose sensible controls on the initiative procedure. In Meyer v. Grant, 486 U.S. 414 (1988), for example, the Court held that the Colorado prohibition against paying circulators of initiative petitions violated the First Amendment. The prohibition was aimed at reducing incentives for petition fraud. And, to no one's surprise, the aftermath of the opinion has been increased petition fraud in signature gathering. See supratext accompanying notes 138-40. The Supreme Court held, citing Buckley v. Valeo, 424 U.S. 1 (1976), that because the prohibition involved a limitation on political expression it was subject to "exacting scrutiny." Meyer v. Grant, 486 U.S. at 420. But the concern rejected by the Court in Buckley, the domination of the political process through expenditures on media advertising, is distinguishable from rules aimed at curbing fraud. The Meyer Court found that the prohibition limits the "number of voices" of those who will convey the message of the initiative proponents to garner the necessary signatures. Id. at 422-23. The Court held that the burdens on speech were not justified by the state's interest in curbing petition fraud because after-the-fact prosecutions of those who do forge a signature "seem adequate to the task of minimizing the risk of improper conduct in the circulation of a petition." Id. at 427. The Colorado state legislature disagreed, of course, and new evidence would appear to support their view over the Court's speculation to the contrary. It is troubling to consider where the Court's reasoning can go. Is it unconstitutional to prohibit, as Colorado does, mail solicitations of signatures? The prohibition on mail solicitations imposes the same burdens (it is more difficult to gather signatures) which concerned the Court in Meyer v. Grant, and the same alternative (after the fact prosecutions for petition fraud) is available. One could make similar arguments against any procedural rule that imposes burdens on proponents (the affidavit requirement for circulators or the Secretary of State's refusal to accept disassembled petition sections, for example). In short, the case may require courts to evaluate, under the strict scrutiny test of the First Amendment, all of the intricate procedural details of the initiative procedure. The Courts ought not assume this administrative role, whatever the constitutional doctrine ostensibly involved. For an example of further judicial meddling with initiative procedures, see American Constitutional Law Found. v. Meyer, No. 93-M-1467, 1994 U.S. Dist. LEXIS 17134 (D. Colo. Nov. 23, 1994), appeal docketed, No. 94-1576 (10th Cir. Dec. 16, 1994). The plaintiffs sought to invalidate seven technical aspects of the Colorado procedure, the identification and reporting of paid circulators, the requirement that circulators be 18 and a registered elector, the reporting of contributions, the six month circulation period, and the circulator affidavit requirement. Id. at *2-3. The court held that the Colorado requirement that paid circulators wear badges identifying themselves as "paid" is unconstitutional. Id. at *20. It also invalidated the state's requirement that the proponents of a petition must disclose the names, addresses, and county of voter registration of all paid circulators, and the state's requirement that proponents report the amount
118
[Vol. 66
in the case law is a third type of claim, a hybrid of the claims courts now entertain, a claim based on constitutionality under the state constitution of the form and content of individual initiatives. The Colorado courts have not recognized any way to enforce the minimum affirmative demands of the constitutional reservation of initiative power on the content of the initiatives themselves. This omission is unfortunate. The language and history of the 1910 amendment to the Colorado Constitution establishing the initiative right support a judicial role 2in the development of minimal standards of scope and clarity.
31
In response, perhaps, to this lacuna in the case law, the General Assembly referred the single-subject amendment to the electorate in 1994. The amendment, although attracting biting 313 comment by well known initiative activists, passed by more than a two-thirds margin. Courts will be asked to enforce the single-subject rule. Had the courts explored the necessary implications of existing language in the constitution, the single4 subject rule may not have been necessary.31 In the discussion below, we consider both the general constitutional language on initiatives and the enforcement of the new singlesubject rule. 1. Necessary Implications of the Initiative Grant Article V of the Colorado Constitution reserves for the 3 people of Colorado the initiative power." 5 The courts quite correctly protect the people's right from the state legislature by invalidating implementing legislation which frustrates the
paid to circulators. Id. at *24-25. 312. See supra notes 76-77, 296 and accompanying text. In this regard, the Colorado Supreme Court has often taken the general position that the initiative provisions of the Colorado Constitution should be liberally construed to effectuate their purpose and facilitate the exercise by electors of this most important right reserved to them by the Constitution. E.g., Colorado Project Common Cause v. Anderson, 495 P.2d 220 (Colo. 1972). 313. An active initiative proponent referred to it as "the most evil proposition... on any ballot, anywhere, at any time." Steve Lipsher, Two Days and Counting;Measure Would Cut "LogRolling," DENY. POST, Nov. 6, 1994, at C1 (quoting Douglas Bruce). 314. For a general discussion of implied or implicit powers of adjudication,
see generally CHARLES BLACK, STRUCTURE AND RELATIONSHIP IN CONSTITUTIONAL
LAW (1969). 315. COLO. CONST. art. V, 1(1)-(2).
19951
initiative process,316 but they have not protected the people's right from opportunistic proponents of initiatives. An opportunistic proponent attempts not to convince a majority of the people of the correctness of a proposal, but to put language in the constitution, via clever issue matching or confusion and deception, which does not have the support of a majority of the electorate. In most cases, the people can protect themselves from opportunistic initiative proponents by voting "no." Someone or some group will publicly oppose almost any given initiative, fund an attack, and push the proponents to justify publicly the details of their proposal. Even absent a well-funded public debate in the media, many conscientious voters will inform themselves and others in discharge of their civic responsibility. The ballot box may be protection enough if proponents are too obvious in their manipulations. Occasions arise, however, when voters are overmatched by an opportunistic strategy used by a clever proponent. In such cases, the courts may need to step in and enforce minimal standards of content aimed at preventing the more egregious cases of proponent misbehavior. a. The People's Initiative Right is Frustratedby Multi-Issue Initiatives The problems created by multi-issue initiatives are discussed extensively above.317 A multi-issue initiative can frustrate the ability of the people to voice their true views on each issue, with the result that such initiatives empower only proponents, whose strategic behavior puts voters in an unacceptable bind. b. The People's Initiative Right Is Frustratedby Initiatives That Are Devious, Vague and Opaque, or Inaccessibly Complex Initiatives that cannot be understood by reasonably informed Colorado voters ought also to fail constitutional muster. It is hard to see how putting language to the populace
316. See supra note 309 and accompanying text. The courts take great pains to construe the legislation "liberally" to facilitate the process. E.g., Montero v. Meyer, 795 P.2d 242, 245 (Colo. 1990) (three month filing requirement does not apply to "refiled" petitions). 317. See supra part IV.D.
120
[Vol. 66
that is devious, opaque or inaccessibly complex is consistent with the people's exercise of their constitutional power. Clever proponents, whose strategy is either to mislead voters into voting affirmatively or to confuse enough voters into abstaining so that an intense group of supporters can carry the vote, ought to be interdicted. Successful deception or confusion by initiative proponents is not, in any meaningful way, consistent with the people's constitutional right to propose and vote on initiatives. Of course, the standards for rejecting a devious, opaque or inaccessibly complex initiative as unconstitutional ought to be rigorous and difficult, with only those initiatives that easily and obviously fail the test being rejected. The test ought not to be that an initiative is not understood by the Colorado voters, although this is positive evidence of a problem, but that an initiative cannot be understood by even conscientious and reasonably informed citizens.31 8 And rejection should allow reasonable opportunity for proponents to cure their draft. c. Is Judicial Inaction Justified? Those who do not want the state courts more involved in the initiative process can make several arguments. First, one can argue that the people can protect themselves from even extreme forms of opportunistic behavior by using the initiative power to repeal unwanted constitutional language. History has shown, however, that it is very difficult to repeal a successful initiative. In Colorado, for example, the only repealed initiatives were those caught in a general referendum stripping the constitution of outdated language. One of the expunged provisions was a prohibition on funding for the 1976 Olympics, put in by initiative. There is typically no discrete organized constituency interested in gathering petitions to repeal a successful initiative, and the General Assembly, wary of "insulting voters," will not use its referendum authority to ask voters 3 to reconsider or even redraft a successful initiative. 19
318. See supra part IV.C.1. There was widespread agreement after Amendment One passed in 1992 that no one had even a general sense of its practical effect beyond its basic requirement of a vote on new or increased taxes. See Shaw, supra note 175, at 2 (State House Majority Leader: "You have to read the Amendment two times a day to try and understand it."). 319. One of the authors, convinced that voters wanted a tax limit in 1992 but not a ratchet down effect given them by Amendment One, see supra note 175, offered, through a state senator, referendum language that would clean up
1995]
Second, one can argue that "the people will get what they deserve" if they do not protect themselves at the polls. This is a harsh evaluation of the deserts of a basically trusting populace. Should we learn distrust? Will the costs of a painful postelection surprise on the effect of a successful initiative encourage more informed voting in the future? As noted above, the evidence is to the contrary; voters tend to weary of initiatives and abstain. Third, one can argue that judges have too little political capital to get in the business of overturning initiatives that have passed with a majority vote. As we have recently seen with the Colorado Supreme Court's decision on Amendment Two,32 a decision throwing out a successful initiative is unusually controversial, with passionate critics attacking the court's legitimacy as well as its judgment. The state court's role on Amendment Two generated support for an initiative on recalling state judges. This message is of course not lost on our judges, and gives them the proper incentive, that is, to overrule the electorate in only the most egregious cases. Moreover, a procedure empowering judges to review clarity before a vote is taken would alleviate much of the problem. 2. Comparing "Amendments" to "Revisions and Alterations": Initiatives Cannot Fundamentally Alter the Existing Constitutional Structure The scope of initiatives can also be limited by how fundamental is the change to the existing structure of Colorado government. In the Colorado Constitution, the people may propose "amendments" to the constitution by initiative (and the legislature may propose "amendments") but the people can "revise, alter and amend" the constitution only through a constitutional convention. This difference in language has a purpose. Some changes are so fundamental that they must be
Amendment One and preserve its basic effect. S. Con. Res. 7, 59th Colo. Gen. Assembly, 1st Sess. (1993). The resolution lost twice in the Senate State Affairs Committee with a majority of senators taking the position that the General Assembly would "insult" voters if they even implied that the voters did not understand the full impact of all of the language in the successful initiative. 320. Evans v. Romer, 854 P.2d 1270 (Colo. 1993); Evans v. Romer, 882 P.2d 1335 (Colo. 1994), petition for cert. filed, 63 U.S.L.W. 3477 (U.S. Dec. 12, 1994) (No. 94-1039).
122
[Vol. 66
accomplished through a constitutional convention. 321 The California Supreme Court, on similar constitutional language, has recognized the distinction. 22 At issue is whether any initiative effects such a deep fundamental change. There are two kinds of initiatives that fail this test. First, some individual constitutional changes ought not be done by initiative. They change provisions that are too basic and have too long a history. Examples include initiatives that eliminate one of the three branches of state government (elimination of the Governor's Office, for example) or effect a fundamental rebalance of power among the branches (vesting taxing power solely in the Governor's Office, for example) or eliminate an essential aspect of sovereignty (such as the power of eminent 3 domain). 23 Second, some initiatives which, through their multi-issue content, contain wholesale tinkering with numerous parts of the constitutional language, are too invasive. Any one provision in the initiative may not effect a fundamental change, but when all the provisions are added up, the change is too 3 sweeping to be done by a single initiative amendment. 24 3. The Due Process Clause and Initiative Procedures As discussed above, the implementing legislation of the initiative process can and has been ruled inconsistent with the constitutional language in the Colorado Constitution that created the initiative process. The Due Process Clause of the federal Constitution and the due process clause of the Colorado Constitution also support challenges to deficient initiative procedures in implementing legislation. If the initiative procedures are inaccessible to all but a few, because, for example, of very short review and appeal deadlines or because of nonpublic hearings before hidden tribunals, the due process
321. COLO. CONST. art. V, 1(2); Id. art. XIX, 1, 2. 322. Courts in several states have held that, although the state constitution reserves to the electorate the power to "amend" by initiative, the people may .revise" the constitution only through constitutional conventions. See supranotes 181-83 and accompanying text. 323. But see Legislature of the State of Cal. v. Eu, 816 P.2d 1309 (Cal. 1991) (ballot measure limiting California legislature's terms and budget not a "revision"); Amador Valley Joint Union High Sch. Dist. v. State Bd. of Equalization, 583 P.2d 1281, 1284-89 (Cal. 1978) (Proposition 13, a taxation reform measure, was not a "revision"). 324. E.g., McFadden v. Jordan, 196 P.2d 787 (Cal. 1948).
19951
clauses ought to come into play. The courts have rejected all such challenges to date. 32 In Montero v. Meyer, 5 for example, the Tenth Circuit heard a challenge by Spanish-speaking citizens against an initiative that designated English as Colorado's official language. The plaintiffs argued that their due process rights were violated because they had inadequate notice of the title board hearing and of their right to appeal the title board findings. The Tenth Circuit's holding, reversing the district court,32 6 was unfortunate. The Tenth Circuit held, first, that the title board was "ministerial," not judicial, and that the notice standards of judicial proceedings were inapplicable.3 27 Second, it held that the plaintiffs had no "constitutionally protected liberty interest" because "the Colorado Constitution creates no legitimate entitlement in the plaintiffs to participate in the 32 process of placing an amendment on the ballot." 8 Furthermore, the implementing legislation creates "no special rights." Any rights created by the Colorado statutes in qualified electors are rights in a narrowly-circumscribed procedural review of the actions of the title board in carrying out its responsibility to make sure that the title, submission clause and summary are fairly worded and reflect the intent of the proponents .... [A] liberty interest created by state law is by definition circumscribed by the law creating it....
... Plaintiffs missed their statutory procedural opportunity for review that they share with all qualified electors. 329 They have no further rights.
The court's holding overlooked the essence of the plaintiff's claim that given the brief, hidden nature of the ballot title review process, whatever rights they had could not be exercised. They were not arguing for special notice but that the process in
325. 13 F.3d 1444 (10th Cir. 1994). 326. The District Court granted partial summary judgment to the plaintiffs, ordering that the title board comply with specified notice and publication requirements. The court refused to order retroactive relief, however, and would not invalidate the successful initiative. Id. at 1446. 327. Id. at 1446 n.2. 328. Id. at 1448. 329. Id. at 1449-50.
124
[Vol. 66
place did not give interested parties a sufficient opportunity to 3 appeal the title board's determination." The Tenth Circuit supported both arguments using a flawed vision of the title board process. Plaintiffs' alleged expectation in this case would inject into the initiative process a debate on the merits (or at least the phrasing of a ballot measure) that would diminish, at the critical point of framing the issue to reflect the intent of the proponents, the constitutional grant of the clear right of the people of Colorado to propose laws and amendments and have them presented to the electorate. 3 ' No one disputes that the opponents of an initiative, if fortunate enough to know of the details of the title board procedure, have standing to contest the accuracy of the ballot title. So the "debate" feared by the Tenth Circuit is already an established procedure, as it should be. Opponents as well as proponents have an interest in the accuracy of the ballot title. Indeed, opponents are a check on the desire of some proponents to put inaccurate or misleading but saleable ballot title language on the ballot. The argument, misunderstood by the Tenth Circuit, is that a professional initiative proponent and opponent can use the procedure to contest ballot title board holdings, while other groups such as the Spanish-speaking population of Colorado, not normally engaged in the process but finding something dear to them on the table, find themselves closed out of the process. 4. Enforcing the Single-Subject Requirement What level of commonality among issues will satisfy Colorado's new single-subject requirement? Courts often recite vague, general standards of germaneness or relevance among 33 issues. 2 Such concepts make sense only with an eye to the intended purposes of the rule. For Colorado, these are well set
330. There is no publication requirement in Colorado until after the time has expired for a protest of the title by a registered elector before the Secretary of State. COLO. REV. STAT. 1-40-107, -124 (Supp. 1994). 331. Montero v. Meyer, 13 F.3d at 1450. 332. E.g., Perry v. Jordan, 207 P.2d 47, 50 (1949) ("all parts . . . are reasonably germane"). Colorado's recent cases on bills state the standard as "necessarily or properly connected to each other rather than disconnected or incongruous." In re House Bill 1353, 738 P.2d 371, 374 (Colo. 1987). On problems of defining a suitable test, see CALIFORNIA COMM'N, supra note 9, at 312-20; Daniel Lowenstein, CaliforniaInitiatives and the Single-Subject Rule, 30 UCLA L. REV. 936 (1983).
1995]
out in supporting legislation and in judicial opinions interpreting the state's longstanding single-subject rule applicable to legislative bills.3 33 The rule serves to avoid the logrolling mischiefs of bundling bills with partial support so that the sum has greater backing than any of its parts, and of using a popular lead provision to bring along others that could not pass alone. It improves voter information and understanding of the subjects of their vote, and it empowers voters to make more choices. When a reasonable voter would want the chance to vote on parts of a measure separately, the rule should guarantee that
right.
Two recent examples illustrate possible applications of the rule. The constitutional amendment passed as Amendment One in 1992, when no single-subject rule applied, contains at least five subjects: limits on tax increases, limits on spending increases, limits on debt increases, general election procedures, and relations between state and local governments. 334 Both public polls and voter behavior since 1992 suggest that voters would have voted no on parts of the measure if given the choice. 33 5 A second example is Amendment 13 on the 1994 ballot, which failed to pass.336 It combined authorizations for gambling in the town of Manitou Springs, and in airports. Colorado's rule is applied in the first instance by the titlesetting board.33 7 Its decision should be subject to prompt judicial review, which takes place not only before enactment by voters but before petitions are circulated. At that stage, proponents' investment in their draft is minimal and corrections can readily be made. Single-subject challenges to ballot measures can also be made after enactment. Review at that stage differs in important respects. Proponents have already incurred the expense of circulating petitions and of an elective campaign, and the measure can no longer be redrafted. For these and other reasons, the courts apply a strong presumption of validity,
333. See supra notes 186-90 and accompanying text. 334. See supra notes 166-69, 174 and accompanying text. It also forbids some taxes outright and creates an incentive to sell public land. The latter could be considered a sixth subject. 335. See supra note 175 and accompanying text. 336. Id. 337. See supra note 223 and accompanying text.
126
[Vol. 66
which does not apply before enactment.33 On the other hand, in post-enactment review a Colorado court can find the measure 3 severable and sustain its principal parts. 39 But that is a crude remedy. It is clearly better to have single-subject review completed before petitions are circulated. Without legislation, courts cannot do that directly, but they can apply different standards of review at the two stages. Review of title-setting should faithfully carry out the purposes of the single-subject rule, to combat logrolling and increase voter understanding and choice. On the other hand, review after enactment should apply the traditional presumption of constitutionality. On the single-subject rule, procedures are again of crucial importance. Colorado's procedures differ significantly from 3 0 If 4 those in California, which have been largely ineffective. the Colorado title board and courts carry out the duties that the single-subject provisions appear to impose on them, the Colorado rule should be a useful corrective to abuse of the initiative. VI. CONCLUSION Like any avenue to power, the initiative has attracted abusers. However, most critics of the device miss the mark by focusing entirely on the substantive results of the initiative votes. Abuses of the initiative are largely caused by structural and procedural flaws. If they are corrected, the initiative can work as an important check on representative and administrative government.
338.
See supra note 187 and accompanying text.
339. COLO. CONST. art. V, 1(5.5) (amended 1994). See also id., art. V 21. 340. In California, judicial review under the single-subject rule rarely occurs before enactment, and this substantially weakens the rule. See Steven W. Ray,
Note, The CaliforniaInitiative Process: The Demise of the Single-Subject Rule, 14 PAC. L.J. 1095, 1105-07 (1983). See also CALIFORNIA COMM'N, supra note 9, at
312-20. Moreover, California courts must either sustain or strike down an initiated law in its entirety; there is no authority to sever and save parts. See CAL. CONST. art. II, 8(d) ("An initiative measure embracing more than one subject may not ...have any effect."). Florida requires judicial review on single subject after 10% of needed signatures have been gathered, and the state courts apply a strict standard. If a measure contains more than one subject, proponents may then amend, and the state allows four years to qualify for the ballot. See Advisory Opinion to the Attorney General re Tax Limitation, 644 So.2d 486 (Fla. 1994); CALIFORNIA COMM'N, supra note 9, at 108-09.
1995]
127
How the initiative right is structured also affects the essential nature of a state's constitution. Where initiatives to amend state constitutions are easily available, constitutions are changed frequently and by modest pluralities of state voters. Provisions that seem legislative in character are added and placed beyond the reach of judicial review under the state constitution. We recommend designing the initiative right to enact a statute to be much more accessible than the right to amend the state constitution. We also address other procedural shortcomings of the initiative process, particularly in Colorado. The details of our recommendations aside, the most important insight of our research is that questions of structure and procedure should be an essential part of America's debate on the merits of state ballot initiatives and proposals for a national initiative.
INITIATIVE PETITION REFORMS AND THE FIRST AMENDMENT

EMILY CALHOUN*
Contrary to popular understanding, the initiative petition process does not embody a type of petitioning that is constitutionally guaranteed.' Colorado's initiative petition and the First Amendment petition for redress of grievances could not be
more different. In these comments, I will explain my understanding of those differences in order to shed light on how we
ought to think about proposed changes to Colorado's initiative process.' In particular, I hope to show why we should not
adopt "reforms" like those embodied in Colorado's proposed, but defeated, Amendment 12.? Such reforms would incapacitate
* Professor of Law, University of Colorado School of Law. 1. Cf, e.g., American Constitutional Law Found., Inc. v. Meyer, No. 93-1467 (D. Colo. filed July 14, 1993), denial of interlocutory appeal affd, 33 F.3d 62 (10th Cir. 1994) (affirming denial of preliminary injunction because it is not clear that state initiative process has First Amendment stature). 2. The First Amendment to the United States Constitution provides that "Congress shall make no law... abridging... the right of the people peaceably to assemble, and to petition the Government for a redress of grievances." U.S. CONST. amend. I. My evaluation of proposals to reform Colorado's initiative petition process is based on research into the First Amendment right to petition the government for redress of grievances. My comments here are based on a longstanding interest in the relationship of the First Amendment to voting rights. See, e.g., Emily Calhoun, Voice in Government: The People, 8 NoTRE DAME J.L. ETHICS & PUB. POLY 427 (1994) [hereinafter Calhoun, Voice in Government]; Emily Calhoun, The FirstAmendment and Distributional Voting Rights Controversies,52 TENN. L. REV. 549 (1985) [hereinafter Calhoun, The FirstAmendment]. The former article should be consulted for a more extensive explanation of the ideas offered in these comments. 3. Amendment 12, insofar as it relates to initiative petitions and other forms of direct democracy, was described on the ballot as: An amendment to ... restrict public resources used in ballot issue campaigns; to require a mandatory fine for willful violations of the campaign contribution, public expenditure, and petition provisions; to extend petition powers to residents of all political jurisdictions; to allow judges to be recalled and bar recalled judges from any future judicial position; to limit petition ballot titles to 75 words and to revise other procedural and substantive petition provisions for the initiative, referendum, and recall; to limit the annual number of bills that governments may exclude from referendum by petition; to limit the reasons for invalidating petition signatures; to repeal changes in state petition laws or regulations adopted after 1988 unless voter-approved; to prevent elected officials from changing certain voter-approved laws; and to
[1994
the government from fulfilling its responsibility to act as a trustee for all of the people of the state. Other articles in this issue discuss how the initiative petition process works, so I will turn immediately to the First Amendment petition for redress of grievances. The principle that there is a fundamental right to petition government for redress of grievances has a long history.4 It was recognized in the Magna Carta in 1215, and in subsequent centuries its importance was repeatedly reaffirmed. The principle was transported to America, in part via charters for colonial government, where it became firmly embedded in the American political and legal system. Petitioning was a primary source of bills in pre-constitutional America.' For example, fifty-two percent of the acts passed in Pennsylvania between 1717 and 1775 originated in petitions to the government.6 In the colonies, petitioning was used to insist that public matters that might otherwise be ignored were included on legislative agendas, to bring problems to the attention of government, and simply to provide information to government.7 Petitioning even seems to have been accorded a status superior to that of free speech and press; the latter rights received derivative protection largely because they would ensure that petitioning would be meaningful and effective. 8
authorize individual, class action, or district suits to enforce the amendment.

LEGISLATIVE COUNCIL OF COLORADO GENERAL ASSEMBLY, AN ANALYSIS OF 1994 BALLOT PROPOSALS, Res. Pub. No. 392, 20 (1994). See Richard B. Collins & Dale
Oesterle, Structuringthe Ballot Initiative:Procedures That Do and Don't Work, 66 U. COLO. L. REV. 47, 75 n.142, 77 n.145, 99 n.245 (1995), for a more detailed description of the Amendment's provisions. 4. My brief historical discussion is generally based on the accounts of
EDMUND S. MORGAN, INVENTING THE PEOPLE: THE RISE OF POPULAR SOVEREIGNTY IN ENGLAND AND AMERICA (1988); David C. Frederick, John Quincy Adams,
Slavery, and the Disappearanceof the Right of Petition, 9 LAW & HIST. REV. 113 (1991); Norman B. Smith, "ShallMake No Law Abridging...": An Analysis of the Neglected, but Nearly Absolute, Right of Petition, 54 U. CIN. L. REV. 1153 (1986); Stephen A. Higginson, Note, A Short History of the Right to Petition Government for the Redress of Grievances, 96 YALE L.J. 142 (1986). Where appropriate, below, I have indicated specific page references to these authorities. 5. Higginson, supra note 4, at 144.
6. MORGAN, supra note 4, at 229.
7. Higginson, supra note 4, at 144-55. 8. See, e.g., Frederick, supra note 4, at 115-16 (asserting the general point); Sheldon M. Novick, The Unrevised Holmes and Freedom of Expression, 1991 SUP. CT. REV. 303, 333 n.135 (noting that some constitutional commentaries used stronger wording for petition rights than for speech rights generally); Smith, supra
Vol. 66]
INITIATIVE PETITION REFORMS
Despite the fact that petitioning rights (and the corollary right of the people to assemble) were historically extremely important, there was debate about whether a petition right should be included in the Constitution and, if so, what form that petition right should take.9 Some believed that a form of petitioning should be adopted that would enable voters to "instruct" (or bind) their representatives to vote in a particular way on a given issue.' James Madison, however, strongly disagreed. He advocated a version of the First Amendment right to petition government that he believed would not undermine the purposes of the representative form of government generally embodied in the Constitution." Madison's version of the right of petition preserved the citizen's right of direct access to government but gave to elected representatives the ultimate power and responsibility for debating, adopting, or rejecting a particular petition. Madison rejected a petition right of instruction in favor of the First Amendment right we now enjoy because he adhered to a familiar theory of government and understanding of human nature. Madison believed that representative government (for all practical purposes an inevitable feature of decisionmaking, 2 including initiative decisionmaking)" should be structured to guard against the possibility that "factions" of citizens might take control of government and abuse its powers. According to Madison, factions are composed of "a number of citizens, whether amounting to a majority or a minority of the whole, who are united and actuated by some common impulse of passion, or of interest, adverse to the rights of other citizens, or to the permanent and aggregate interests of the community." 3
note 4, at 1153 (asserting the general point); Higginson, supra note 4, at 166 n.98 (referring to Madison's beliefs). 9. Frederick, supra note 4, at 117 (noting that "[d]espite the importance of petitions in the colonies, the inclusion of a right of petition in the Constitution was no certainty"). 10. For general discussions of the proposed right of instruction, see THOMAS
E. CRONIN, DIRECT DEMOCRACY: THE POLITICS OF INITIATIVE, REFERENDUM, AND
RECALL 24-26 (1989); 2 BERNARD SCHWARTZ, THE BILL OF RIGHTS: A DOCUMENTARY HISTORY 1091-1103 (1971); GARRY WILLS, EXPLAINING AMERICA: THE FEDERALIST 42-43, 223-24 (1981). 11. Madison drafted the original version of the First Amendment right of petition. For futher background, see Smith, supra note 4, at 1175; Higginson, supra note 4, at 155. 12. See infra p. 309. 13. WILLS, supra note 10, at 193-94.
[1995
Madison feared any group of citizens-minority or majority-driven by self-interest to the exclusion of the interests of other citizens or the interests of the public good and the community as a whole. In the system of government advocated by Madison, representatives were obligated by their office to exercise judgment on behalf of the common good, after deliberation and debate. Representatives could not fulfill these obligations if a faction-even a majority faction-of voters had a petition right of instruction that would bind representatives to a particular position.' 4 These arguments were grounded in part in a principle of government trusteeship. 5 According to the trusteeship principle, government has a trust relationship with all of its citizens. As trustee of the res publica, government is obligated to act for the general public good rather than in a partial or self-serving manner. An instruction right would make it impossible for representatives to take into account the public good before adopting public policy; it would enable factions to bind representatives to positions that served only the interests of some voters, contrary to the trusteeship principle. The trusteeship principle has firm roots in political philosophy.'" It was an essential part of John Locke's writings on government. 7 It was incorporated into many post-Revolution charters of state government. 8 Most importantly, the principle of government trusteeship was familiar to and accepted by a broad range of people instrumental to the framing of the Constitution. Even those who were more skeptical of elected representative government than Madison-for example, Thomas Jefferson-believed in the principle."9 One cannot doubt that this principle influenced the decision to favor the
14. CRONIN, supra note 10, at 28-29. 15. Id. at 27. 16. For example, Jean-Jacques Rousseau and Edmund Burke are both associated with the idea, see CRONIN, supra note 10, at 27, 39, as are the classical philosophers influencing advocates of contemporary civic republicanism. See, e.g., Cass R. Sunstein, Beyond the Republican Revival, 97 YALE L.J. 1539, 1560-61 (1988); cf. Steven L. Winter, The Meaning of "Under Color of' Law, 91 MICH. L. REV. 323 (1992) (discussing the concept of betrayals of trust responsibilities that occur when public officials act under color of authority).
17. PETER C. HOFFER, THE LAW'S CONSCIENCE: EQUITABLE CONSTITUTION-
ALISM IN AMERICA 73-74 (1990). 18. Id. at 78-79. 19. See Calhoun, Voice in Government, supra note 2, at 432-34.
Vol. 66]
First Amendment version of the petition right over a petition right of instruction. During the early years of our country's existence, the nature of the right to petition Congress was vigorously and publicly debated, particularly when abolitionists began heavily using the petition in the early 1800s.2 0 In the last one hundred and fifty years, however, there has not been much public debate in the United States about the First Amendment right of petition. There are only a few Supreme Court decisions that discuss the kind of petitioning the Framers thought worthy of constitutional protection.2 ' What does exist in Supreme Court opinions, however, confirms ideas that recur in the historical record of public debate.22 3 The Supreme Court Petitions Clause cases 2 tell us that the constitutional petition right was adopted to ensure that citizens could make their wishes known to government and that government, in turn, could be responsive to the will of the people. In Petitions Clause parlance, the will of the people is not a phrase that describes how a majority of people feel at any given moment about any given issue. Rather, the will of the people is an aspirational concept, posited as something that government strives to discern in the multitude of views-majority, minority, or special interest-presented to it.
20. For a discussion of abolitionist petitioning, see Calhoun, Voice in Government, supra note 2, at 436-39; Frederick, supra note 4, at 120. The debaters argued about issues such as whether Congress had an obligation to reply to petitions or even to receive them. 21. Only the following Supreme Court opinions include what I would describe as discussions of the right of petition, and most of these discussions tend to be rather brief: Meyer v. Grant, 486 U.S. 414 (1988); McDonald v. Smith, 472 U.S. 479 (1985); Minnesota State Bd. for Community Colleges v. Knight, 465 U.S. 271 (1984); Bill Johnson's Restaurants, Inc. v. NLRB, 461 U.S. 731 (1983); Smith v. Arkansas State Highway Employees, 441 U.S. 463 (1979); California Motor Transp. Co. v. Trucking Unlimited, 404 U.S. 508 (1972); Adderley v. Florida, 385 U.S. 39 (1966); Eastern R.R. Presidents Conference v. Noerr Motor Freight, Inc., 365 U.S. 127 (1961); Thomas v. Collins, 323 U.S. 516 (1945); Bridges v. California, 314 U.S. 252 (1941); De Jonge v. Oregon, 299 U.S. 353 (1937); United States v. Cruikshank, 92 U.S. 542 (1875). 22. In addition to the ideas that informed constitutional debate on petitioning and the right of instruction, see supra notes 9-20 and accompanying text, the ideas put forth during the early nineteenth-century debates on abolitionist petitions are especially interesting. Regarding the latter debates, see sources cited supra, note 20. 23. An extended analysis of the Petitions Clause cases can be found in Calhoun, Voice in Government, supra note 2, at 440-52.
[1995
The Supreme Court's Petitions Clause cases tell us that a will of the people cannot be formed if some speakers are excluded from government processes or public debate. A will of the people cannot be formed even if those who are excluded are in an extreme minority that can never hope to win a political contest and can never hope to persuade a majority of citizens to accept its views. Because the Petitions Clause will of the people should transcend mere majority preference, this will cannot be formed when minority views are excluded. It is because the Petitions Clause will of the people reflects the trusteeship principle that government is accountable to all the people-not merely to a majority or minority faction.24 This view of the will of the people is reinforced by the imagery and metaphors used by the Supreme Court in discussing the constitutional right of petition. The Petitions Clause cases are strikingly devoid of the adversarial, competitive marketplace metaphors used in other First Amendment cases. Instead, the Petitions Clause cases talk about how petitioning works to promote collective action, assembly, communication for common achievement, consultation with others respecting public affairs, and the ability of all people to enter into transactions and pacts with other people. The right to petition the government is more than an adversarial right of access to or control of government. Rather, according to the Supreme Court, First Amendment rights, presumably including the petition right, comprise a "charter for government."25 The petition right ensures that government will act as it is supposed to act, with the integrity and attitude of a trustee for the people. It protects against government processes that are corrupted because captured by minority or majority factions.
24. Compare the Supreme Court's treatment of representative obligations in contemporary voting rights analysis. Because in our constitutional system an elected representative is charged with taking into account the interests of all of the people, not merely the interests of the majority that elected her, political structures that send a message that elected officials need represent only the interests of some voters are constitutionally disfavored. See, e.g., Emily Calhoun, Shaw v. Reno: On the Borderline, 65 U. COLO. L. REV. 137 (1993) (discussing Shaw v. Reno, 113 S.Ct. 2816 (1993), and Davis v. Bandemer, 478 U.S. 109 (1986)). 25. Thomas v. Collins, 323 U.S. 516, 537 (1945).
Vol. 66]
The First Amendment petition right is a necessary incident of citizenship in a government that can compel all citizens to act to support its decisions-not merely those who voted for its decision. It flows inexorably from the principle that government is created as a trusteeship for more than mere fluid majorities. It helps ensure that a will of the people will be reflected in the adoption of policies that can be justified-and are accountable-to all of the people. Colorado's initiative petition process is designed to serve values fundamentally different from the values and principles served by the First Amendment petition right.26 At the risk of repeating what may be obvious, I will elaborate on a few differences that I find especially significant in thinking about proposals for reform of the initiative process. First, the Colorado initiative process is not designed to facilitate cooperative transactions or pacts among citizens. Rather, it is quintessentially adversarial from its beginning to its end. It is not structured to promote the formation of a will of the people that transcends factional interests. Unless initiative proponents voluntarily choose to consult with others early on, the initiative petition will embody proposals for public policy driven by a single point of view. Colorado law does not require deliberative debate in the formation or refinement of proposals prior to their presentation to voters. There is always, of course, debate about whether voters should adopt a given proposal,, but the public debate-conducted primarily through broadcast media-has the quality of salesmanship. It is not the kind of deliberative debate that classical political philosophers believed was essential to a discussion of public policy and "the fate of peoples."27 I am addressing an issue of structure. Madison's design for representative government and the First Amendment petition right took into account the tendency of people to organize themselves into factions and interposed structures that would
26. See Collins & Oesterle, supranote 3, at 55-63; David B. Magleby, Let the Voters Decide? An Assessment of the Initiative and Referendum Process, 66 U.
COLO. L. REV. 13, 15-21 (1995). 27. See EDWARD P. CORBETT, CLASsIcAL RHETORIC FOR THE MODERN STUDENT 516 (3d ed. 1990) ("Of the three types of oratory defined by the ancients
. the political was esteemed most highly, because it dealt with the loftiest of issues; namely, the fate of peoples, rather than of individuals.").
[1995
operate as a check on that tendency even when-perhaps especially when-the faction consisted of majorities. The First Amendment petition right was intended to help ensure that government would not neglect its trust relationship to all of the people. The Colorado initiative process, however, was not designed with this end in mind. To be sure, an initiative process can protect groups of voters from being entirely ignored by a legislature. When used with discretion, it can be a powerful check on a legislature determined to ignore its trust responsibilities to all of the people. But there is a significant difference between initiative petitioning and First Amendment petitioning. Even if a First Amendment petition itself embodies the views of powerful factions (which it well may),28 First Amendment petitioning leaves the final decision on public policy to elected officials formally and informally constrained to act with due regard to the trusteeship principle. In contrast, an initiative petition (which is frequently captured by powerful, special interest factions) gives a final decision on public policy to voters who have no reason to take anything but private, factional interests into account. The initiative process is not internally structured to guard against the possibility that public policy will be captured by factions and voters oriented only by their own self-interest. Unlike representatives, initiative voters who make final decisions about public policy are neither formally nor informally constrained by the responsibilities of a trustee. Initiative voters
28. It is important to recognize that the initiative petition itself is no more and no less subject to exploitation by powerful interests than the petition for redress of grievances. For example, the petition for redress of grievance came to be used in England as "afavorite weapon in contests among the few for control of the government. . . . [They] came from the top down, from parties within the government contending against each other and seeking public support in the contest." MORGAN, supra note 4, at 227. According to Morgan, "[P]etitions ... [could be used as] one of the rituals of popular government, messages in which it ...[would be] difficult to distinguish the giver from the receiver, the supplicant from the sovereign." Id. at 230. In the United States, petitions were used less frequently by representatives themselves to gain leverage with colleagues, but nonetheless the powerful often used them. The Virginia Statute of Religious Liberty, drafted by Jefferson, was adopted only after Madison solicited petitions from citizens for its approval (in response to a bill calling for a general state tax to support religion). Id. at 229. Petitions to Congress became a weapon exploited by powerful political interests and speakers in the debates over slavery preceding the Civil War. See sources cited supra note 20.
Vol. 661
are not constrained by oath of office to think of all the people or about whether measures will be sufficiently impartial that the people in general will consent to their implementation. They are not practically constrained by a need to explain or justify their vote to their constituents. They can adopt policies without worrying whether other members of government will be able to implement those policies consistent with their oaths of office and trust responsibilities. As Jane Mansbridge has commented in a slightly different context, it would be remarkable were preferences formulated under these conditions to tend toward anything other than the private.29 As a practical matter (and contrary to the myth favored by its extreme advocates), the initiative process is a form of representative decisionmaking about public policy. Through the initiative, only some of all registered voters decide on policies affecting themselves and other voters of the jurisdiction. But the policies also affect significant numbers of non-voters who live, work, or own property within the jurisdiction. These nonvoters might include, among others, people under the age of 18, citizens like the mentally disabled who cannot qualify to vote, people qualified to vote who for some reason do not vote (perhaps many elderly or low income citizens), people who live in the community for a significant portion of the year but who choose to vote elsewhere, people whose work brings them to the community and on whom the community depends for its wellbeing, or people who own commercial property in the community but vote elsewhere. All of these persons are part of the community: they contribute to it and their well-being depends on the policies adopted for it. In a real-world sense, they are involved in the res publica. In the initiative process, however, their interests
29. JANE J. MANSBRIDGE, BEYOND ADvERSARY DEMOCRACY 301 (1983). As conducted today, voting is essentially a very private activity. Mansbridge describes the essentially private nature of voting the following way (and if we are honest, most of us will recognize our own behavior in this description): Citizens file into a curtained box, mark a preference, and file out. In special circumstances, if a big-city political machine is at work or if the community is small, they may see someone they know on the way in and out of the box, smile, and exchange a triviality. Most voters see no one they know. They sit in their homes; they consume information; they determine a preference; they go to the polling place; they register their preference; they return to their homes.
[1994
will be represented-if at all-only by the small percentage of eligible voters who actually vote.3 The trusteeship principle and governing structures that promote it are-as the Constitution's Framers understood-a necessary prerequisite to the legitimacy of representative government. The initiative form of representative decisionmaking, however, ignores these principled and structural prerequisites of legitimacy. Initiative structures place final decisions respecting public policy in the hands of some (representative) voters who are not constrained to take any account of the trust obligations that a representative owes to the collective people. Indeed, the initiative process actually enhances the likelihood that final decisions on public policy will ignore trustee responsibilities. Were initiatives the only-or even the primary-way of governing in Colorado, we would have a system of representative government devoid of essential, legitimizing structures and operating principles. Proposals for reforming the initiative process need to be evaluated with this fact in mind. To date, Colorado's initiative process has not been the dominant process for making final decisions about public policy. The initiative process has generally functioned in tandem with conventional structures of representative government without abrogating the trusteeship principle of the First Amendment. But what would proposed reforms do both to the existing balance between conventional and initiative forms of representative decisionmaking and to the trusteeship principle? To restate Professor Magleby's concern about the effects of the initiative on representative government: Would these reforms enhance or diminish the possibilities for government to act as a trustee?3 ' Some reforms would simply make the initiative process a better or a fairer adversarial process, without compromising other governmental structures. Reforms that promote voter understanding of the substance of an initiative and of the people, organizations, and money that support it, and that limit the ability of special interests to hide self-serving measures deep in the verbiage of lengthy proposals make the adversarial process fairer. These sorts of reforms seem perfectly appropri-
30. Collins & Oesterle, supra note 3, at 52; Magleby, supranote 26, at 32-34. 31. Magleby, supra note 26, at 45.
Vol. 66]
ate and there should be no objection to them. If citizens find that the legislature is ignoring its trusteeship responsibilities, or that their voices and issues of importance to them are systematically excluded from the legislative agenda, they may well need an adversary process like the initiative. And if some citizens need an adversary process in such circumstances, we should give them-and all citizens-the fairest adversary process possible. Other reforms would basically transform the initiative process. That would be the result, for example, of changing the Colorado initiative process to conform to the Swiss model that Professors Collins and Oesterle describe.32 Reforms of this sort would suit someone who objects to the initiative petition process per se, which I do not. I see no reason to eliminate the initiative entirely or to change its essentially adversarial character. Used sparingly, in tandem with conventional processes and as a check against the possibility that elected officials might themselves ignore trustee responsibilities to all of the people, the initiative process can promote realization of the trusteeship principle. Among proposed reforms to the initiative process, however, are those that threaten to disrupt the balance between the initiative and other representative processes of government. At their worst, these reforms would enable factions to dominate the formation of public policy, eclipsing other processes essential to having a government that can act as a trustee for all the people. The most worrisome of this sort of reform was embodied in proposed Amendment 12."
32. Collins & Oesterle, supra note 3, at 80. 33. Another, less troublesome, proposal is the suggestion that the legislature should become involved in "drafting" initiative petitions proposed by others. Collins & Oesterle, supra note 3, at 79. Legislative involvement in "drafting" is no substitute for the type of legislative involvement that would serve First Amendment petition values and the trusteeship principle. Drafting will likely not reflect a broad range of views, will probably not be done publicly, and legislators will not be asked to take a public position to which they will be held accountable by their constituents. The very term--"drafting"-accurately describes in what legislators will be involved: a bureaucratic task. Unfortunately, although legislative involvement in drafting does not actually serve the trusteeship principle or First Amendment petition values, it may very well be taken to serve them. Voters may be encouraged to believe that a proposal emanating from legislative halls has some kind of legislative imprimatur when, in fact, it has none. Voters may mistakenly believe that initiative measures have been endorsed or reviewed with the interests of all the people in mind. Whatever
[1995
My purpose, here, is not to describe the provisions of Amendment 12. Others have done so. I simply recommend an evaluation of Amendment 12's proposals in light of the following assertions. Under Amendment 12, conventional processes of public policy decisionmaking-processes structured to help ensure that public policy will reflect more than factional interests-would have been subordinated to the initiative, a process most frequently used without regard to the trusteeship principle. The provisions of Amendment 12 would have made it almost impossible for representative bodies to formulate or implement public policy that serves all of the people and fulfills the trust obligation of government. They would have made it easier for factions to capture public policy. They would have deprived citizens of an ability to hold accountable the promoters of policies that serve only majority or minority factions. Even judicial checks on special interest factions would have been compromised by new recall provisions. The bulk of Amendment 12's "reforms" would not have served values that had constitutional significance for the Framers of our Constitution. As Jane Mansbridge said, "To maintain its legitimacy, a democracy must have both a unitary and an adversary face."34 Proposed Amendment 12 contained provisions that would have removed our government's legitimating, unitary face, a face represented by conventional governing structures that offer at least some realistic hope that public policy will be based on the trusteeship principle. If these conventional structures are not working well, the solution is not to adopt so-called reforms like those of Amendment 12 which will only exacerbate the situation; the solution is to become serious about reforming conventional structures of representative government. As for reforming the initiative process itself, let me add to my colleagues' suggestions one additional, simple reform proposal. I propose that each initiative measure on the election 3 ballot should be preceded by the following reminder. 5
clarity might arguably be gained through legislative drafting, that gain does not compensate for the danger in the proposal.
34. MANSBRIDGE, supra note 29, at 300.
35. The proposed reminder is intended to alert citizens who act as direct lawmakers to their trust responsibilities. Cf. JOHN RAWLS, POLITICAL LIBERALISM 219-20 (1993) (arguing that public reason demands that we reconceptualize voting
Vol. 66]
REMINDER WHEN YOU CAST YOUR VOTE YOU WILL ACT AS A REPRESENTATIVE OF OTHER CITIZENS AFFECTED BY THIS ISSUE OF PUBLIC POLICY. PLEASE TAKE SERIOUSLY YOUR REPRESENTATIVE RESPONSIBILITY TO ACT ON BEHALF OF ALL OF THE PEOPLE WHEN YOU VOTE. YOUR VOTE SHOULD NOT BE CAST IN FAVOR OF THIS PROPOSAL UNTIL YOU HAVE FIRST CONSIDERED WHETHER IT WORKS FOR THE COMMON GOOD OR ONLY PARTIALLY, FOR SOME FACTION OF THE PEOPLE. This initiative measure was formulated without
formal representative deliberation by financial support comes primarily from whose
You should vote against this measure EITHER if you believe it does not serve the common good OR if you believe that alternatives to this proposal-alternatives that might better work for the people as a whole-have not been fully considered. If you vote no because you believe alternatives have not been sufficiently considered and you wish to have the proposal referred to the legislature for such consideration, please check the box below. A reminder of this sort will not guarantee that people who participate in an initiative election will act appropriately as representative voters. It will, however, offer at least some hope that the initiative process will work better, both to provide citizen access when government ignores its trust responsibilities and in tandem with conventional processes that-unlike the initiative-are inherently structured to fulfill trust responsibilities.
to be something other than a purely private activity which can be engaged in without regard to the interests of the public good).
CONSTITUTIONAL CHANGE AND DIRECT DEMOCRACY

LYNN A. BAKER*
At the state level, unlike the federal level, the American people have long had a direct and determinative role to play in altering their constitutions.' In forty-nine states, any constitutional amendment proposed by the legislature must be submitted to the People and approved by at least a majority of those voting.2 Only in seventeen (thirty-five percent) of these states, however, can the People also propose and adopt constitutional amendments without any legislative participation.3
* Professor, University of Arizona College of Law. J.D., 1985, Yale University; B.A., 1982, Oxford University; B.A., 1978, Yale University. I am grateful to Dean Gene Nichol of the University of Colorado School of Law for inviting me to participate in the "Governing by Initiative" Conference, and to Danielle Johnson, Quinn Roy, and their colleagues at the University of Colorado Law Review for assistance with the publication of this essay. Conversations with Toni Massaro, Ted Schneyer, David Golove, and, especially, Samuel Dinkin, made this essay better and made writing it more fun. 1. As Article V makes clear, the only role granted the People in amending the U.S. Constitution is electing their representatives to Congress and to their state legislatures. U.S. CONST. art. V. Some scholars contend, however, that our federal Constitution may be-and, indeed, has already been-legitimately amended though procedures other than those set out in Article V. See, e.g., BRUCE ACKERMAN, WE THE PEOPLE: FOUNDATIONS (1991); Akhil Reed Amar, The Consent of the Governed: ConstitutionalAmendment Outside Article V, 94 COLUM. L. REV. 457 (1994) [hereinafter Consent of the Governed];Akhil Reed Amar, Philadelphia Revisited: Amending the Constitution Outside Article V, 55 U. CHI. L. REV. 1043 (1988). 2. See THE COUNCIL OF STATE GOVERNMENTS, THE BOOK OF THE STATES, 1994-95, at 21-22 tbl. 1.2 (1994) [hereinafter THE BOOK OF THE STATES]. Delaware is the exception. An amendment to the Delaware constitution is adopted if approved by two-thirds of each house of the legislature in two consecutive sessions; no popular vote is required for ratification. Id. 3. Id. at 24 tbl. 1.3. The 17 states are Arizona, Arkansas, California, Colorado, Florida, Illinois, Michigan, Mississippi, Missouri, Montana, Nebraska, Nevada, North Dakota, Ohio, Oklahoma, Oregon, and South Dakota. Id. Massachusetts also permits initiative amendments to its constitution, but no amendment may be submitted to the electorate for ratification without prior approval at two sessions of a successively elected legislature by not less than onefourth of all members elected, sitting in joint session. Id.
144
[Vol. 66
Often heralded as the purest and highest form of democracy,4 procedures for direct state constitutional change have come under increased scrutiny and attack since the passage of Colorado's Amendment Two. In November 1992, by a vote of 53.4% to 46.6%, the people of Colorado adopted an initiative amendment to their constitution, which prohibits the state, its
agents, and its subdivisions from:

[Elnact[ing], adopt[ing] or enforc[ing] 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.5 Whatever one's views of the morality or constitutionality of this amendment to Colorado's constitution,6 its passage by a
4. See, e.g., THOMAS E. CRONIN, DIRECT DEMOCRACY: THE POLITICS OF INITIATIVE, REFERENDUM, AND RECALL 7-20 (1989); DAVID B. MAGLEBY, DIRECT
LEGISLATION: VOTING ON BALLOT PROPOSITIONS IN THE UNITED STATES 2-34 (1984); DAVID B. SCHMIDT, CITIZEN LAWMAKERS: THE BALLOT INITIATIVE
Opinion polls recently conducted in a wide range of states reveal that direct democracy is enormously popular in the states that currently permit it and is much desired in the states that do not. CRONIN, supra, at 203 n.10; PHILIP L. DUBOIS & FLOYD F. FEENEY, IMPROVING THE CALIFORNIA INITIATIVE PROCESS: OPTIONS FOR CHANGE 8 n.5 (1992). Indeed, during the past 20 years, 12 states that do not currently provide for plebiscites considered direct democracy devices at constitutional conventions or in legislative debates and hearings, while no state has seriously considered the abolition of plebiscites. CRONIN, supra, at 51 tbl. 3.1. In 1992, Mississippi amended its constitution to permit initiative amendments, although with certain subject matter exceptions. MISS. CONST. art. 15, 273. 5. Evans v. Romer, 882 P.2d 1335, 1338 (Colo. 1994), petition for cert. filed, 63 U.S.L.W. 3477 (U.S. Dec. 12, 1994) (No. 94-1039). 6. For a sampling of views reported in the local and national press, see, e.g., Louis Aguilar, Boycotts Beget Boycott Over ColoradoGay Issue, PHOENIX GAZETTE, Jan. 6, 1993, at A2; Colorado Anti-Gay Law Dead?, SACRAMENTO BEE, Dec. 15, 1993, at A7; George de Lama, Colorado Springs Showdown: Gays Facing Fundamentalists, CHI. TRIB., Apr. 27, 1993, at 1; Gay-Bashing, DemocracyBashing,N.Y. TIMES, Dec. 18, 1993, at 22; Molly Ivins, ColoradoLoses Its Balance on the Slippery Slope of Intolerance, SACRAMENTO BEE, Aug. 1, 1993, at F03; Dirk Johnson, Colorado Judge Overturns Initiative Banning Gay Rights Laws, N.Y. TIMES, Dec. 15, 1993, at A22; Sue Lindsay, Amendment 2 Gets Another Day in Court; State's Attorney Argues that Gays Haven't Shown Need for Protection, ROCKY MTN. NEWS (Denver), July 1, 1994, at 4A; Sue Lindsay, State Appeals Ruling; ColoradoSolicitor to Ask Court to Bring Back Amendment 2, ROCKY MTN. NEWS (Denver), June 30, 1994, at 25A; Pacy Markman, Famous Victories, but Nothing Won; Rights: The Defeat of Oregon's Measure Nine and Others of Its Ilk Only Keeps Homosexuals from Losing Ground, L.A. TIMES, Nov. 13, 1992, at B7; Sue Anne Pressley, Hearing Opens on Challenge to Colorado Gay Rights Ban;
REVOLUTION 3-40 (1989).
1995]
DIRECT DEMOCRACY
relatively slim majority, and the presence of a related initiative amendment on the Oregon ballot this fall,' provoke thought on a more general, procedural question: should the same simple majority of the popular vote that is required to enact a statute by initiative continue to be all that is required to adopt an initiative amendment to a state constitution?' In its investigation, this essay takes the perspective of those seeking to maximize the constitutional protection of individual rights.9 This normative premise was chosen because it embodies a central issue in the recent debates over Colorado's Amendment
Trial May Become Forum on Status of U.S. Homosexuals, WASH. POST, Oct. 13, 1993, at A2; Louis Sahagun, Judge Hears Challenge to Law on Gays; Colorado: An Initiative that Bans Protected Status for Homosexuals Will Take Effect Friday if Not Overturned,L.A. TIMES, Jan. 12, 1993, at A3; David Tuller, Gays Win Some, Lose Some; More Homosexuals Elected, but ColoradoRestricts Rights Laws, S.F. CHRON., Nov. 5, 1992, at A12. 7. The proposed Oregon initiative stated in part: "Children, students and employees shall not be advised, instructed or taught by any government agency, department or political unit ... that homosexuality is the legal or social equivalent of race, color, religion, gender, age or national origin." Bettina Boxall, Social Issues; Anti-Gay-Rights Measures Ignite Aggressive Battles in 7 States; Proposed Ballot Initiatives in Washington, Oregon, Nevada, Arizona, Idaho, Missouri and Michigan Have Rallied Forces on Both Sides, L.A. TIMES, June 9, 1994, at A5. See also Gay Rights Groups "Relieved" but Wary as 8 InitiativesFail, ARIZ. REPUBLIC, July 17, 1994, at A19. The Oregon measure was defeated on November 8, 1994, by a 52% majority. Bettina Boxall, CaliforniaElections;Key Victories Hearten Gay Activists; Politics: First Openly Gay Legislator is Elected in California; Oregon and Idaho AntiRights Initiatives Lose, L.A. TIMES, Nov. 10, 1994, at A37. On the same day, a similar statutory initiative lost in Idaho "by only a few thousand votes." Id.; AntiGay Rights Measures Appear to be Defeated in Idaho, Oregon, BNA Management Briefing, Nov. 10, 1994, available in LEXIS, BNA Library, BNAMB File. 8. These facts may also provoke thought on two questions that I have previously discussed at substantial length: Are plebiscites more likely than representative processes to produce laws that disadvantage numerical minorities, such as homosexuals? And should the courts, which have not historically varied their analyses because of a law's popular origin, begin to employ different standards when reviewing the enactments of plebiscites and legislatures? Although scholars have largely agreed that the answer to both questions is "yes," I have argued that the answer to both questions is, in fact, "no." Lynn A. Baker, DirectDemocracy and Discrimination:A Public Choice Perspective, 67 CHI.-KENT L. REV. 707 (1991). 9. It is, of course, not always obvious when a constitutional change increases rather than decreases the protection accorded individual rights. Some changes, for example, may be viewed as increasing the constitutional protection of certain individuals' rights at the expense of others'. Happily, this essay need not first resolve this timeless difficulty of political theory: the analysis that follows is unaffected by one's preferred definition of "maximizing the constitutional protection accorded individual rights," so long as that definition remains constant throughout.
146
[Vol. 66
Two' as well as in the ongoing academic discussion of the costs and benefits of lawmaking by plebiscite." Thus, the central inquiry of this essay is whether the states that currently provide for initiative amendments would be more likely to maximize the constitutional protection accorded individual rights if the number of votes required for adoption were increased from a simple majority to a super-majority, perhaps two-thirds.
I. SIGNATURE GATHERING VS. VOTE GATHERING
Intuition suggests that it should be significantly more difficult to adopt a constitutional amendment than a Statute. A state constitution is, after all, a fundamental law, which we therefore expect-and want-to be more permanent than the ever-changing, ordinary, statutory law. In the fifteen states that permit use of the initiative for both statutes and constitutional amendments, however, none requires more votes to adopt an amendment than to enact a statute. Although thirteen of these fifteen states have different procedures for enacting initiative statutes and amendments, those differences lie not in the number of votes necessary for adoption, but in the number of signatures necessary to place the proposed law on the ballot. 2 Five states currently require about twice as many
10. See supra note 6. 11. Compare Baker, supra note 8,at 710 (demonstrating that "a rationally self-interested racial minority's preference between representative and direct lawmaking processes is a difficult empirical question and cannot confidently be resolved on the strength of a priori reasoning"); Ronald J. Allen, The National Initiative Proposal: A Preliminary Analysis, 58 NEB. L. REV. 965 (1979) (evaluating the potential contributions and risks of a national initiative procedure, and observing that "[tihe history of the initiative is remarkably free of the enactment of abusive legislation"); and Clayton P. Gillette, Plebiscites,Participation, and Collective Action in Local Government Law, 86 MICH. L. REV. 930 (1988) (arguing that "popular votes on issues that share certain salient characteristics can be constrained in ways that minimize deviation from public interest") with Derrick A. Bell, Jr., The Referendum: Democracy's Barrierto Racial Equality, 54 WASH. L. REV. 1 (1978) (contending that, from the perspective of racial minorities, representative democracy has many important advantages over direct democracy); and Julian N. Eule, JudicialReview of DirectDemocracy, 99 YALE L.J. 1503 (1990) (making similar contentions and arguing for heightened judicial scrutiny of plebiscitary enactments). 12. DUBOIs & FEENEY, supra note 4, at 20-21 tbl. 5. The thirteen states are Arizona, Arkansas, California, Massachusetts, Michigan, Missouri, Montana, Nebraska, North Dakota, Ohio, Oklahoma, Oregon, and South Dakota. Id. Colorado and Nevada require the same number of signatures to put a statute or a constitutional amendment on the ballot. Id. Florida and Illinois permit
1995]
DIRECT DEMOCRACY
signatures for constitutional amendments as for statutes, and seven states require from one-third to one-half more signatures. 13
Although it is obviously harder to gather more rather than fewer signatures, this means only that it will be somewhat more difficult to place an amendment than a statute on the ballot. This difference in signature requirements, however, does not necessarily mean that it will also be significantly more difficult to adopt a constitutional amendment than a statute. Because the existing empirical evidence suggests that money plays very different and significant roles in the signaturegathering and vote-gathering processes, even the largest differences in the number of signatures necessary to place a constitutional amendment and a statute on the ballot are not likely to pose a barrier to adoption comparable to that of requiring a super-majority of the popular vote.' 4 The widespread use of paid signature gatherers 5 as well as the prevalence of voters who "are willing to sign petitions without evaluating or even paying much attention to the proposals that they concern," 6 has yielded a consensus that "the signature-gathering process today is more a test of how much money the proponent has than of citizen support for the proponent's legislative ideas." v Some scholars have gone so far as to conclude that "anyone willing to put up the funds can buy a place on the ballot." 8 Although well-financed interests have been found to have a clear advantage in qualifying measures for the ballot, much empirical evidence suggests that heavy campaign spending alone cannot "buy" the subsequent election.' 9 One-sided
initiative amendments but not statutory initiatives. Id. 13. Id. 14. See infra notes 15-22 and accompanying text. 15. See, e.g., CRONIN, supra note 4, at 62-66; MAGLEBY, supra note 4, at 6164; DUBOIS & FEENEY, supra note 4, at 74, 164; Daniel H. Lowenstein & Robert M. Stern, The First Amendment and Paid Initiative Petition Circulators: A Dissenting View and a Proposal, 17 HASTINGS CONST. L.Q. 175, 194-200 (1989). 16. DUBoIs & FEENEY, supra note 4, at 164. See also CRONIN, supra note 4, at 62-66; MAGLEBY, supra note 4, at 61-64; Lowenstein & Stern, supra note 15, at 198-200. 17. DUBoIs & FEENEY, supra note 4, at 164. 18. Lowenstein & Stern, supra note 15, at 200. 19. Allen, supra note 11, at 1036. See also DUBOIS & FEENEY, supra note 4, at 147-51. In the words of one author:
148
[Vol. 66
spending has been successful in persuading people to vote against an initiative, but has had a negligible effect on obtaining affirmative votes.2 Even when spending is more equalized, those seeking to block an initiative are much more likely to succeed than its proponents.2 1 And there is no demonstrated relationship between the amount of money spent on a successful initiative campaign and the margin of victory.22 Thus, if one believes that it ought to be significantly more difficult to adopt a constitutional amendment than a statute, further increasing the number of signatures necessary to place an initiative amendment on the ballot is not likely to accomplish this end. A more promising alternative may be to increase the number of votes required for the adoption of an initiative amendment from a simple majority to a supermajority, perhaps two-thirds.
II.
VIRTUES AND VICES OF A SUPER-MAJORITY REQUIREMENT
It would surely be more difficult to adopt initiative amendments if the seventeen states that currently provide for them required two-thirds of the popular vote, rather than a
[Tihere does not appear to be a consistent relationship between spending and voter behavior. Most heavily contested measures were voted up or down by large majorities and, while bigger expenditures might have narrowed the margins of victory or defeat in some instances, the end result would almost certainly have been the same. DUBOIs & FEENEY, supra note 4, at 151 (quoting MICHAEL D. MEYERS, A STUDY OF CALIFORNIA INITIATIVES, 1976-1986, at 33 (1988)); John R. Owens & Larry L. Wade, Campaign Spending on California Ballot Propositions,1924-1984: Trends and Voting Effects, 39 W. POL. Q. 675, 687 (1986) ("It is obvious that other undefined, unspecified factors are much more important than money in shaping electoral outcomes in direct legislation campaigns."). One must be careful when evaluating these empirical studies, however, to note that they typically do not distinguish statutory initiatives from initiative amendments. Nonetheless, there seems little reason to expect that the findings would be significantly different if only initiative amendments were considered. More problematic for purposes of the present analysis is the fact that these studies do not always separate signaturegathering expenses and vote-gathering (i.e., post-petition) expenses, perhaps because the two categories are not clearly distinguishable. 20. See, e.g., DUBOIs & FEENEY, supra note 4, at 147-49; MAGLEBY, supra note 4, at 147-48; SCHMIDT, supra note 4, at 35-37; Allen, supra note 11, at 103436; Daniel H. Lowenstein, Campaign Spending and Ballot Propositions: Recent Experience, PublicChoice Theory and the FirstAmendment, 29 UCLA L. REV. 505, 518-43, 545, 609-32 (1982); Owens & Wade, supra note 19, at 684. 21. Owens & Wade, supra note 19, at 684, 686-87; DUBOIS & FEENEY, supra
note 4, at 149-50 (citing ABT ASSOCIATES, INC., FACTORS AFFECTING POLLUTION
REFERENDA 5, 61, 72 (1971)). 22. DUBOIS & FEENEY, supra note 4, at 150.
1995]
DIRECT DEMOCRACY
simple majority for adoption.23 Under such a requirement, Colorado's Amendment Two, for example, would not have passed. Indeed, of the 100 initiative amendments that appeared on state ballots between 1978 and 1988, only nine would have been adopted under a two-thirds rule as opposed to the thirty-six that were in fact adopted under the prevailing simple majority rules.2 4 To begin, then, a super-majority requirement might well be attractive to those disappointed by the outcome of the popular vote on Amendment Two or on any of the other initiative amendments that have been adopted over the years by a slim majority. A super-majority requirement might also be attractive to those who believe that initiative amendments are (unjustifiably) easier to adopt than legislatively proposed amendments.2 5 Although legislatively proposed amendments, in all but one state, similarly require no more than a simple majority of the popular vote for ratification,2 6 they must be approved by a
23. See THE BOOK OF THE STATES, supra note 2, at 23 tbl. 1.3. Thirteen of the 17 states unambiguously require a simple majority of the popular vote for adoption. Massachusetts and Nebraska further require that the majority vote on the amendment be at least 30% and 35%, respectively, of the total vote at the election. Nevada requires a majority vote on the amendment in two consecutive general elections, and Illinois requires a majority voting in the election or threefifths of those voting on the amendment. Although the Illinois requirement could be construed as a mild super-majority requirement, it is better understood, I believe, as a simple majority requirement that, in an attempt to reduce voter "drop off," simply counts as "no" votes the non-votes of those who come to the polls but fail to express a preference on the proposed initiative amendment. 24. See DUBOIS & FEENEY, supra note 4, app. at 1-32. 25. An examination of the voter approval rates for initiative amendments and legislatively proposed amendments suggests that this concern is unfounded. Indeed, a study of eight states from 1898 to 1978 revealed that the approval rates for legislatively proposed amendments ranged from 47% to 69%, while those for initiative amendments ranged from 21% to 48%. Only in Arkansas did the approval rate for initiative amendments (48%) even begin to approximate that for legislatively proposed amendments (47%). MAGLEBY, supra note 4, at 73. Magleby posits three explanations for these findings. First, in the case of initiatives, "voters may believe that the 'burden of proof is on the advocate,'" and may be much less willing to vote yes when an initiative "is met by an organized and powerful opposition." Id. at 72. Second, because "legislative referendums have arisen from compromise and bargaining," they may be less likely to face strong opposition. Id. Finally, voters may be "more trusting of elected representatives than of the special-interest or public-interest groups that generally proposeinitiative amendments." Id. at 72-73. 26. New Hampshire requires two-thirds of the vote on the amendment for ratification. See THE BOOK OF THE STATES, supra note 2, at 21-22 tbl. 1.2. Fortythree of the remaining 49 states unambiguously require a simple majority of the popular vote for adoption. Tennessee requires a majority of those voting for
150
[Vol. 66
majority to two-thirds of each house of the legislature before being placed on the ballot." There is no comparable requirement in the case of initiative amendments.2 8 It is important to note, however, that even where the approval of two-thirds of each chamber is required for legislatively proposed amendments,2 9 this requirement will not necessarily be equivalent to approval by a super-majority of voters. As an illustration, imagine the hypothetical state portrayed in Diagram One. The state legislature consists of three senators (A, B, and C) and three representatives (1, 2, and 3), each of whom represents forty-eight voters.3 0 Further assume that: (1) all state legislators are elected on a geographical basis; (2) each voter gets to elect a representative in both the House and the Senate; (3) Senators A and B (exactly two-thirds of the Senate) and Representatives 1 and 2 (exactly two-thirds of the House) support a proposed amendment while Senator C and Representative 3 do not; (4) exactly fifty-percentplus-one (25 of 48) of the constituents of Senators A and B and of Representatives 1 and 2 support the proposed amendment (the shaded area of squares Al, A2, B1, and B2 in Diagram One); and (5) none of the constituents of Senator C or Represen-
governor. Hawaii and Nebraska require that the majority vote on the amendment must be at least 50% of the total votes cast at the election; or, in the case of a special election, a majority of the votes counted, which must be at least 30% of the total number of registered voters. Illinois requires a majority of those voting in the election or three-fifths of those voting on the amendment. Louisiana requires a majority of votes in the state as a whole and also in affected subdivisions if the amendment affects five or fewer political subdivisions of the state. And, for amendments concerning certain elective franchise and education matters, New Mexico requires ratification by three-fourths of electors voting in the state and two-thirds of those voting in each county. Id. 27. Id. Sixteen states require two-thirds of each house, 2 require two-thirds of both houses, 17 require a simple majority, 9 require three-fifths, and 6 states have some other requirement. Id. 28. As the discussion in Part I above shows, the existing signature requirements for placing an initiative amendment on the ballot are not a comparable substitute. See supra notes 12-22 and accompanying text. 29. THE BOOK OF THE STATES, supra note 2, at 21 tbl. 1.2. The 16 states are Alaska, California, Colorado, Delaware, Georgia, Idaho, Kansas, Louisiana, Michigan, Mississippi, South Carolina, Texas, Utah, Washington, West Virginia, and Wyoming. In Maine and Montana, two-thirds of both houses, rather than of each house, is required. Id. 30. The Senate districts are drawn horizontally and the House districts vertically such that Senator A, for example, represents the voters in squares Al, A2, and A3, while Representative 1 represents the voters in squares Al, B1, and C1.
1995]
DIRECT DEMOCRACY
tative 3 support the proposed amendment. If all of these conditions are met, an amendment proposed by two-thirds of each chamber of the legislature theoretically could have the support of as few as 34.7% (50 of 144) of the state's voters. Diagram One
Representatives
0 0 00 0 0O0 00 D DD000000 0 0 0 0DDDD 0 0 D 0 0] 0 0- D D 0 0 0 00 00 0 0 DCC~L_ 0 fLI
0 0 0] 0 0 D 0 0 0I 0 0 C
0 0 0 00 0 0 00.0
0 0 00 0 00 0 00
In any of the existing fifty American states, of course, only the first two of these conditions will always be met. And as the relevant numbers of representatives, senators, and voters increase from the theoretical minimums stated in assumptions three through five above, the actual number of voters that support an amendment proposed by the required two-thirds of each chamber also increases from the theoretical minimum (34.7%) portrayed in Diagram One. Nonetheless, the requirement that a majority of those voting ratify any proposed amendments to the state constitution is, perhaps surprisingly, the only guarantee that even amendments approved by two-thirds of each house of the legislature in fact represent the will of the majority. And, far from having the desired equalizing effect, a requirement that initiative amendments be ratified by two-thirds of those voting may well
152
[Vol. 66
make them much more difficult to adopt than legislatively proposed amendments. The reality of the constitutional amendment process in the states indicates that advocates of a super-majority requirement take their inspiration from some other source, perhaps the federal government. For both the text of Article V and the history of successful amendments to our federal Constitution suggest that a "good" constitution should be insulated from, or somehow "above," everyday majoritarian politics. 3 ' And in the case of the federal Constitution in a federal structure of government this may indeed be desirable. 2 But even in the federal context, the attractiveness of a super-majority requirement for adopting constitutional amendments cannot rest on the notion that it provides greater protection for individual rights than a simple majority requirement would. To be sure, it would be more difficult to repeal the U.S. Constitution's guarantee of freedom of speech, for example, under the existing requirements of Article V than it would be under a simple majority ratification rule. On the other hand, we would today have an Equal Rights Amendment to the U.S. Constitution if the approval of only a majority, sixty percent, or even seventy percent of the states were necessary for ratification rather than three-fourths. 3 In brief, at both the federal and state levels,
31. U.S. CONST. art. V states in relevant part: The Congress, whenever two thirds of both Houses shall deem it necessary, shall propose Amendments to this Constitution, or, on the Application of the Legislatures of two thirds of the several States, shall call a Convention for proposing Amendments, which, in either Case, shall be valid to all Intents and Purposes, as Part of this Constitution, when ratified by the Legislatures of three fourths of the several States, or by Conventions in three fourths thereof, as the one or the other Mode of Ratification may be proposed by the Congress .... A reading of the 27 amendments to the U.S. Constitution reveals the continued expansion of protection for individual rights. 32. Akhil Amar, however, has argued at substantial length that "[wie the People of the United States-more specifically, a majority of voters-retain an unemumerated, constitutional right to alter our Government and revise our Constitution in a way not explicitly set out in Article V." Amar, Consent of the Governed, supra note 1, at 458-59 (arguing that, via a petition of a majority of American voters, Congress could be required to call a constitutional convention, and that a simple majority of the electorate could ratify an amendment or a new constitution). 33. Only 35 of the required 38 states had ratified the Equal Rights Amendment by the June 30, 1982 deadline, even though from 1972 to 1982 'a majority of Americans consistently told interviewers that they favored this amendment to the Constitution." JANE J. MANSBRIDGE, WHY WE LOST THE ERA 1 (1986).
1995]
DIRECT DEMOCRACY
153
any change in the constitutional amendment procedure that makes it more difficult to repeal or constrain existing individual rights simultaneously makes it more difficult to adopt new 4 rights and expand existing ones. 3 Especially at the federal level, this crucial symmetry of effect is all too easily and frequently overlooked. Thus, one might rationally prefer that initiative amendments to a state's constitution be adopted by a simple majority if one is eager either to add new rights or to repeal existing ones. If, however, one is reasonably satisfied with the state constitution's current bill of rights or is simply fearful of change, one would likely prefer that the constitution be more difficult to amend. Although a super-majority requirement would make it more difficult to add new rights, it would also-and, from this perspective, more importantly-make it more difficult to repeal or otherwise constrain existing individual rights. There is another variable to consider in the case of state constitutions, however: amendments that violate the federal Constitution will not be enforced. That is, the individual rights guaranteed each state's citizens by the U.S. Constitution cannot be constrained or diminished by any state's constitution. 35 And this fact importantly affects one's assessment of whether a super-majority of the popular vote should be required for the adoption of initiative amendments. To the extent that the protection for individual rights provided by a state constitution is no greaterthan that afforded by the federal Constitution, either because a state constitution is silent or because it mirrors the federal Constitution on an issue, a super-majority requirement for the adoption of initiative amendments will have some economic benefits but seemingly greater non-economic costs. To be sure, a super-majority requirement makes it more difficult to enact state constitutional
34. For a discussion of this reciprocity of effect in the context of blocking and passing proposed statutory initiatives, see Baker, supra note 8, at 712-15. 35. Under the Due Process Clause of the 14th Amendment, virtually all of the rights guaranteed by the first eight Amendments have been selectively absorbed into the 14th Amendment and, therefore, applied to the states. The U.S. Supreme Court has justified this incorporation by holding the rights involved to be "implicit in the concept of ordered liberty," Palko v. Connecticut, 302 U.S. 319, 325 (1937), and "fundamental to the American scheme of justice," Duncan v. Louisiana, 391 U.S. 145, 148-49, reh'g denied, 392 U.S. 947 (1968). See generally LAURENCE H. TRIBE, AMERICAN CONSTITUTIONAL LAW 772-73 (2d ed. 1988).
154
[Vol. 66
amendments that constrain or repeal individual rights in violation of federal constitutionalguarantees. But even if such amendments pass, they are likely to be invalidated if challenged under the federal Constitution, 6 rendering the entire enactment process a deadweight social loss.37 Because it poses a higher barrier to the adoption of such amendments, a supermajority requirement is likely to reduce the number that are proposed and therefore also the resulting deadweight social loss. An additional, related benefit of a super-majority requirement may be some reduction in litigation costs: since fewer amendments, including amendments that would be invalidated under the federal Constitution, will be adopted under a supermajority requirement, there is likely to be less litigation concerning the constitutionality of adopted amendments. Against these economic benefits of a super-majority requirement, however, must be weighed the non-economic costs. Most significantly, a super-majority requirement increases the difficulty of adopting state constitutional amendments that would expand the protection of individual rights beyond that provided by the federal Constitution, and fewer such amendments are therefore likely to be enacted. For those seeking to maximize the protection of individual rights, then, these costs of a super-majority requirement are likely to outweigh the economic benefits. To the extent that the state constitution, explicitly or as interpreted, may provide greaterprotection for individual rights than that afforded by the federal Constitution,3 8 the benefits
36. Colorado's Amendment Two appears to be taking this route. See Evans v. Romer, 854 P.2d 1270, 1285-86 (Colo. 1993) (finding that "Amendment Two singles out and prohibits [homosexuals, lesbians, and those of bisexual orientation] from seeking governmental action favorable to [them] and thus, from participating equally in the political process .... In short, Amendment Two, to a reasonable probability, infringes on a fundamental right protected by the Equal Protection Clause of the United States Constitution."); Evans v. Romer, 63 Fair Empl. Prac. Cas. (BNA) 753 (Dec. 14, 1993) (concluding that Amendment Two is unconstitutional under the 14th Amendment because it is "violative of the fundamental right of an identifiable group to participate in the political process without being supported by a compelling state interest."); Evans v. Romer, 882 P.2d 1335, 1349 (Colo. 1994) (applying strict scrutiny under the 14th Amendment and finding that "[n]one of the interests identified by the state is a necessary, compelling governmental interest which Amendment 2 is narrowly tailored to advance."). 37. In some cases, it might be argued that the enactment process, although constitutionally doomed, had "symbolic value." 38. "Examples abound where state courts have independently considered the merits of constitutional arguments and declined to follow opinions of the
1995]
DIRECT DEMOCRACY
of a super-majority rule appear, if only marginally, to exceed the costs. Insofar as a super-majority requirement makes it more difficult to adopt amendments that constrain or repeal state constitutional protections for individual rights which exceed the federal Constitution's guarantees, it provides a significant benefit to persons seeking to maximize these protections. So long as such amendments do not seek to reduce below the federal minimums the protection provided individual rights, they are unlikely to be invalidated if challenged under the federal Constitution. Against this substantial benefit, however, must be weighed the most obvious and significant cost of a super-majority requirement: proposed amendments that would expand existing individual rights or add new ones are less likely to be adopted. Although these corollary costs and benefits might well be expected to balance out over time, we have seen above that a super-majority rule provides the additional benefit of a reduction in litigation costs. In summary, to the extent that the state constitution provides protection for individual rights no greater than that afforded by the federal Constitution, the non-economic costs of a super-majority requirement are likely to outweigh the economic benefits-at least from the perspective of those seeking to maximize the constitutional protection accorded individual rights. But insofar as the state constitution already provides protection for individual rights which exceeds the federal Constitution's guarantees, the benefits of a super-majorUnited States Supreme Court they find unconvincing even where the state and federal constitutions are similarly or identically phrased." William J. Brennan, Jr., State Constitutionsand the Protectionof Individual Rights, 90 HARV. L. REV. 489, 500 (1977). Among the individual rights explicitly contained in some state constitutions but not the U.S. Constitution are: "open courts" or "right to remedy" provisions, see, e.g., FLA. CONST. art. V, 3(b); David Schuman, The Right to a Remedy, 65 TEMPLE L. REV. 1197 (1992); prisoners' rights, see, e.g., OR. CONsT. art. I, 88 13, 15, 16, 23; Ann I. Park, Human Rights and Basic Needs: Using International Human Rights Norms to Inform ConstitutionalInterpretation,34 UCLA L. REV. 1195 (1987); the right of privacy, see, e.g., ALASKA CONST.' art. I, 22; Gerald B. Cope, Jr., Toward a Right of Privacy as a Matter of State ConstitutionalLaw, 5 FLA. ST. U. L. REV. 631 (1977); the right to education, see, e.g., N.J. CONST. art. IV, 7; Robert F. Williams, Equality Guaranteesin State Constitutional Law, 63 TEX. L. REV. 1195 (1985); the right to public assistance, see, e.g., N.Y. CONST. art. XVII, 1; Burt Neuborne, Foreword: State Constitutions and the Evolution of Positive Rights, 20 RUTGERS L.J. 881 (1989). See ROBERT F. WILLIAMS, STATE CONSTITUTIONAL LAW: CASES AND MATERIALS 423-477, 857-884 (2d ed. 1993).
156
[Vol. 66
ity requirement could reasonably be expected to outweigh, however marginally, the costs. Thus, if at least half of the time a state constitution provides protection for individual rights no greater than that afforded by the federal Constitution, a simple majority requirement is likely to have benefits that exceed its economic costs, and therefore would be preferable to any supermajority requirement. And, in fact, state constitutions only rarely provide protection for individual rights which exceeds that guaranteed by the federal Constitution. 39 One more consideration remains: the ability to amend a constitution enables the People to "overrule" judicial decisions, including those interpreting constitutional provisions.4 This historically has been an important function of our federal amendment process.4 1 A super-majority requirement would clearly make it more difficult to amend a state constitution in order to invalidate a judicial decision interpreting a constitutional provision. Would this be an improvement? To the extent that a state constitution provides protection for individual rights no greater than that afforded by the federal Constitution, persons desiring to maximize that protection need not fear that the courts will "dilute" it through their interpretations. For any judicial decision that would reduce the state constitutional protections for individual rights below the minimum guaranteed by the federal Constitution would likely be overturned on appeal,4 2 and no amendment to the state constitution would therefore be necessary to "overrule" such a decision. In this regard, then, the requirements for adopting initiative amendments are simply irrelevant. Nonetheless, a super-majority requirement has the benefit in this
39. See supra note 38. 40. "Where the judges are carrying out the function of constitutional review, the final responsibility of the people is appropriately guaranteed by the provisions for amending the Constitution itself ... ." Eugene V. Rostow, The Democratic Character of Judicial Review, 66 HARV. L. REV. 193, 197 (1952). Of course, no constitutional amendment is necessary to "overrule" a judicial decision interpreting a statute: one need simply enact a corrective statute. 41. Four of the 27 amendments to the U.S. Constitution were adopted to overturn U.S. Supreme Court decisions: the 11th Amendment, ratified in 1795, overturned Chisholm v. Georgia, 2 U.S. 419 (1793); Section One of the 14th Amendment, ratified in 1868, overruled Scott v. Sandford, 60 U.S. 393 (1856); the 16th Amendment, ratified in 1913, overturned Pollock v. Farmers' Loan & Trust Co., 157 U.S. 429 (1895); and the 26th Amendment, ratified in 1971, overturned Oregon v. Mitchell, 400 U.S. 112 (1970). 42. See supra note 36.
1995]
DIRECT DEMOCRACY
context of making it more difficult to adopt an amendment that overrules a decision expanding the state constitution's protection of individual rights. To the extent that a state constitution provides protection for individual rights which exceeds that afforded by the federal Constitution, some might fear that the courts, through their interpretations, will reduce those guarantees to the federal minimums. Thus, one obvious and significant cost of a supermajority requirement is the increased difficulty of adopting a constitutional amendment to "overrule" such decisions and thereby reinstate, or even expand, the original constitutional guarantees. Against this cost, however, must be weighed the corollary benefit: a super-majority requirement also makes it more difficult to adopt an amendment that overrules a decision in which a court expands the state constitution's protection of individual rights. If these costs and benefits can be expected to balance out over time, those desiring to maximize the protection of individual rights have no reason to prefer either a simple majority or a super-majority requirement for the adoption of initiative amendments. In summary, when the initiative amendment process is used to overrule judicial interpretations of a state constitution, a super-majority requirement has substantial benefits and no obvious costs to the extent that the state constitution provides protection for individual rights no greater than that afforded by the federal Constitution. And, insofar as a state constitution provides protection for individual rights which exceeds the federal Constitution's guarantees, a super-majority requirement has costs and benefits that, in this context, are likely to balance out. Thus, those seeking to maximize the protection of individual rights should prefer a super-majority requirement to the extent that it increases the difficulty of adopting amendments intended to overrule specific judicial decisions interpreting the state constitution. So what are we now to conclude given our earlier, seemingly contradictory determination that a simple majority requirement is preferable to any super-majority requirement in the 4 general context of initiative amendments? 3 Because initiative amendments intended to overrule specific judicial decisions will be proposed less frequently than other types of initiative
43.
See supra text accompanying notes 25-37.
158
[Vol. 66
amendments, the answer is clear: persons desiring to maximize the constitutional protection of individual rights should expect the benefits of a simple majority requirement greatly to exceed the costs more often than not.
III. CONCLUSION
Given the advantages that a simple majority requirement has over any super-majority requirement, it is scarcely surprising that none of the states that provide for initiative amendments has ever required a super-majority of the popular vote for adoption. More startling is the implication of the above discussion for those concerned about the passage of Colorado's Amendment Two and, more generally, the constitutional protection of individual rights. For this analysis ironically suggests that those most distraught about Amendment Two's success at the polls should be the most vigorous defenders of a simple majority requirement for the adoption of initiative amendments-a requirement without which Amendment Two would not have passed.
CRITICAL RACE THEORY: AN ANNOTATED BIBLIOGRAPHY 1993, A YEAR OF TRANSITION

RICHARD DELGADO*
JEAN STEFANCIC**
INTRODUCTION
A little more than one year ago, we published CriticalRace Theory: An Annotated Bibliography.' In it, we (1) traced the history and development of the Critical Race Theory (CRT) movement;2 (2) identified ten themes within that movement's corpus that seemed to us central and characteristic;3 and (3) annotated over 200 of the most important Critical Race works, employing a numbering system corresponding to the main themes.4 As we were working on the original bibliography, we speculated about a number of matters that we were not able to address in the original article: Would the prodigious output of the Critical Race scholars that we were struggling to describe continue unabated? Would new voices enter the arena? Would the movement change direction, or begin to include new themes other than the ones we found in the original 217 works? Would it attract new criticism, the bibliography perhaps serving as a lightning rod for less-than-friendly commentary? Would the movement stall, having reached its apogee? One of us had written about a similar phenomenon in legal scholarship-the symposium movement'-and so we wondered whether a oneyear retrospective might begin to supply answers to questions such as these. In particular, we wondered if there would be
* Charles Inglis Thomson Professor of Law, University of Colorado School of Law. ** Research Associate, University of Colorado School of Law. 1. Richard Delgado & Jean Stefancic, CriticalRace Theory: An Annotated Bibliography, 79 VA. L. REV. 461 (1993). 2. Id. at 461. 3. Id. at 462-63. 4. Id. at 464-516. 5. Jean Stefancic, The Law Review Symposium Issue: Community of Meaning or Re-Inscription of Hierarchy?, 63 U. COLO. L. REV. 651 (1992).
160
[Vol. 66
anything like an "observer" effect, in which the attention drawn by the annotation or symposium changes the phenomenon described.6 This update provides (a) annotations of the principal works of Critical Race scholarship that appeared during a period of approximately one year7 following the cut-off for the original bibliography, and (b) commentary and analysis aimed at shedding light on the above questions. Our methodology for choosing, annotating, and headnoting items was similar to that employed in the original bibliography,' with a single exception. Responding to a substantial increase in writing by and about women of color, we added Theme number 11, Critical Race Feminism, to our list.9 To be included, a work needed to address one or more themes we considered to fall within Critical Race thought. The eleven themes, together with the numbering system we employ, follow: 1. Critique of liberalism. Most, if not all, CRT writers are discontent with liberalism as a means of addressing the American race problem. Sometimes this discontent is only implicit in an article's structure or focus. At other times, the author takes as his or her target a mainstay of liberal jurispru-
6. See infra text accompanying note 40 (addressing this and related questions). 7. This bibliography covers works which came out after the publication deadline for the original bibliography (late 1992) and before January 1, 1994. 8. Delgado & Stefancic, supra note 1, at 461-62, 463-64. We entered into a master database every work published by a recognized Critical Race scholar during the period in question. We then read each item to make sure that it did indeed incorporate one or more clearly recognizable Critical themes or ideas. We noted additional works cited or discussed by these authors for possible inclusion. Then we sent a preliminary draft of this update to each author requesting comments and corrections. We also invited suggestions for additional authors and works we might have overlooked. In addition to our methodology, the numbering scheme for classifying the books and articles is identical to that used in the original bibliography, id. at 46263, with the exception of Theme number 11, which is new to this bibliography. The theme descriptions, which are identical to those used in the original bibliography, have been reprinted with the permission of the authors and the Virginia Law Review. 9. Will there one day be a headnote: Critical Race Masculinism (Theme 12)? We have chosen not to create one at this time, although there is a discernible body of emerging writing on this subject. E.g., Brown, The Social Construction of a Rape Victim: Stories of African-American Males About the Rape of Desiree Washington, infra pp. 165-66; Culp, Notes From California: Rodney King and the Race Question, infra p. 169.
1994]
CRITICAL RACE THEORY
dence such as affirmative action, neutrality, color blindness, role modeling, or the merit principle. (Theme number 1). 2. StorytellingI counterstorytellingand "naming one's own reality." Many Critical Race theorists consider the majoritarian mindset-the bundle of presuppositions, received wisdom, and shared cultural understandings of persons in the dominant group-to be a principal obstacle to racial reform. To analyze and challenge these power-laden beliefs, some writers employ counterstories, parables, chronicles, and anecdotes aimed at revealing their contingency, cruelty, and self-serving nature. (Theme number 2). 3. Revisionist interpretationsof American civil rights law and progress. One recurring question for Critical scholars is why American antidiscrimination law has proven so ineffective in redressing racial inequality-or why progress has been cyclical, consisting of alternating periods of advance followed by ones of retrenchment. Some Critical scholars address this question by seeking answers in the psychology of race, white self-interest, the politics of colonialism and anticolonialism, or other sources. (Theme number 3). 4. A greater understandingof the underpinnings of race and racism. A number of Critical writers seek to apply insights from social science writing on race and racism to legal problems. For example, understanding how majoritarian society sees black sexuality helps explain the law's treatment of interracial sex, marriage, and adoption; knowing how different settings encourage or discourage discrimination helps in deciding whether the movement toward Alternative Dispute Resolution is likely to help or hurt disempowered disputants. (Theme number 4). 5. Structural determinism. A number of CRT writers focus on ways in which the structure of legal thought or culture influences its content, frequently in a status quo-maintaining direction. Understanding these constraints results in working more effectively towards racial and other types of reform. (Theme number 5). 6. Race, sex, class, and their intersections. Other scholars explore the intersections of race, sex, and class, pursuing such questions as whether race and class are separate disadvantaging factors, or the extent to which black women's interests are or are not adequately represented in the contemporary women's movement. (Theme number 6).
162
[Vol. 66
7. Essentialism and anti-essentialism. Scholars who write about these issues are concerned with the appropriate unit for analysis: Is the black community one, or many, communities? Do middle- and working-class African-Americans have different interests and needs? Do all oppressed peoples have something in common? (Theme number 7). 8. Cultural nationalismIseparatism. An emerging strain within CRT holds that people of color can best promote their interests through separation from the American mainstream. Some believe that preserving diversity and separateness will benefit all, not just groups of color. We include here, as well, articles encouraging black nationalism, power, or insurrection. (Theme number 8). 9. Legal institutions, Criticalpedagogy, and minorities in the bar. Women and scholars of color have long been concerned about representation in law school and the bar. Recently, a number of authors have begun to search for new approaches to these questions and to develop an alternative, Critical pedagogy. (Theme number 9). 10. Criticism and self-criticism; responses. Under this heading we include works of significant criticism addressed at CRT, either by outsiders or persons within the movement, together with responses to such criticism. (Theme number 10). 11. Critical Race feminism. Here we include works addressing the unique situation of women of color (other than intersectionality (Theme number 6) and essentialism (Theme number 7)), such as reproductive freedom and the social construction of women of color. (Theme number 11). A final note about our methodology: What follows is a list of articles and books-not of persons, much less members of a school. CRT lacks a formal organizational structure; there are no members as such. Many of the authors whose works we include devote only a portion of their output to Critical jurisprudence. Others are flatly critical of CRT (see Theme number 10) or discuss its place within legal thought in general. Our annotations cannot begin to do justice to the rich body of Critical thought they summarize. We warrant only that the works that follow can shed light on one or more aspects of Critical Race Theory, and we invite readers to study the complete works for themselves.
1994]
ANNOTATED BIBLIOGRAPHY
Aleinikoff, T. Alexander, Re-Reading Justice Harlan's Dissent in Plessy v. Ferguson: Freedom, Antiracism, and Citizenship, 1992 U. Ill. L. Rev. 961. (Themes 3, 5). Re-examines Justice Harlan's "our Constitution is color1 blind" dissent in Plessy v. Ferguson.0 Responds to conservatives' misappropriation of the phrase, showing that it was not intended as an enshrinement of colorblindness, but as a prohibition against a caste system characterized by white supremacy. Suggests new attention to the concept of "liberty," which has deep roots in the civil rights movement, and may be understood to encompass positive rights as well as anti-caste protection. Barnes, Robin D., Standing Guard for the P.C. Militia, or, Fighting Hatred and Indifference: Some Thoughts on Expressive Hate-Conduct and Political Correctness, 1992 U. Ill. L. Rev. 979. (Themes 1, 4). Urges more stringent penalties for perpetrators of race crimes and hate speech. Rejects the view that those who call for such a response are engaging in censorship, since persons who utter targeted racial vilification are not inviting their victims to engage in a dialogue, but are hoping to silence and harass them. Asserts that campus hate speech impairs the university's ability to serve its broad social functions and, ultimately, threatens national productivity if promising students of color leave or become demoralized. Rejects the view that the debate is only about feelings or about merely symbolic concessions to campus activists. Bell, Derrick, Political Reality Testing: 1993, 61 Fordham L. Rev. 1033 (1993). (Themes 1, 2, 3, 4, 8). Applies themes from his recent book, Faces at the Bottom of the Well," to the current political scene. While applauding President Clinton's intention, declared
10.
163 U.S. 537 (1896).
11. DERRICK BELL, FACES AT THE BOTTOM OF THE WELL: THE PERMANENCE OF RACISM (1992).
164
[Vol. 66
in the 1993 State of the Union address, to bring about social reforms, points out and discusses two matters Clinton neglected: (1) the status of black Americans is far worse than that of whites, and (2) white racism blocks reforms that would benefit whites as well as blacks. Discusses scapegoating, tokenism, code words, and the rallying of antiblack sentiment among lowincome whites. Illustrates the ineffectiveness of antidiscrimination laws and policies by means of a hypothetical Racial Preference Licensing Act, and shows the fragility of black rights with a hypothetical "Chronicle of the Space Traders." Bell, Derrick & Erin Edmonds, Students as Teachers, Teachers as Learners, 91 Mich. L. Rev. 2025 (1993). (Themes 2, 9, 10). Critiques Judge Harry Edwards' recent article 12 which charges that law schools have abandoned their mission of teaching practical scholarship. Contends that Edwards fails to acknowledge the contributions of new approaches such as feminism and Critical Race Theory because they depart from traditional legal pedagogy. Proposes that interdisciplinary and critical scholarship offer alternative approaches that address the crisis of ethics in the practice of law, a situation that Judge Edwards finds so disturbing. (Foreword by Derrick Bell, article by Bell and Erin Edmonds). Brooks, Roy L., A Critical Race Theory Critique of the Right to a Jury Trial Under Title VII, 5 U. Fla. J.L. & Pub. Pol'y 159 (1993). (Themes 1, 4). Analyzes a recent amendment to Title VII of the 1964 Civil Rights Act affording a jury trial in cases that include a demand for compensatory or punitive damages. Warns that the jury-trial provision may not be a positive development because racism is an ingrained, ordinary and normal feature of American life and experience. As a result, jurors are apt to be no fairer than judges in setting damages, and likely worse. Asserts that procedural devices, such as voir dire and post-trial motions, are unlikely to be able to control prejudice on the part of
12. Harry T. Edwards, The Growing Disjunction Between Legal Education and the Legal Profession, 91 MICH. L. REV. 34 (1992).
1994]
jurors because of the cost of employing them and the difficulty of making racism manifest. Consequently, minority plaintiffs now have an even greater chance of losing Title VII lawsuits. Brown, Kevin, Do African-Americans Need Immersion Schools?: The Paradoxes Created by Legal Conceptualization of Race and Public Education, 78 Iowa L. Rev. 813 (1993). (Themes 1, 5, 8). Argues that the current debate over African-American immersion schools (schools that emphasize Black pride and an Afrocentric perspective) does not take account of the real justification for those schools, namely equipping black students to overcome racial obstacles contributing to their own repression at the hands of the dominant culture. Once we lose sight of this justification, there is no good way for the law to resolve the constitutionality of these schools-any solution will lead to a paradox. Brown, Kevin, The Legal Rhetorical Structure for the Conversion of Desegregation Lawsuits to Quality Education Lawsuits, 42 Emory L.J. 791 (1993). (Themes 1, 3, 5, 8). 13 Shows how the Supreme Court in Freeman v. Pitts gives federal courts discretion to relinquish supervision over student assignments yet insist that predominantly black and predominantly white schools are given roughly equal funding. Discusses the historical debate among African-Americans over integrated versus separate education. Highlights the paradoxical nature of the Supreme Court's framework for resolving de jure segregation, namely, that the needs for segregation in the past and desegregation now are both premised on African-American inferiority. Points out that advocates of racial separation today are able to urge that mutual respect and African-American equality support their position. Brown, Kevin, The Social Construction of a Rape Victim: Stories of African-American Males About the Rape of Desiree Washington, 1992 U. Ill. L. Rev. 997. (Themes 2, 4, 6, 7). Uses conversations with African-American males in Indianapolis to show how belief systems operating in the
13.
112 S. Ct. 1430 (1992).
166
[Vol. 66
black community helped construct Mike Tyson as a victim, even though he was accused and convicted of raping Desiree Washington. Points out that loyalty to the black community demands that racism trump sexism as the first fight to be won, due to the fundamental belief that justice is white. Argues that this view victimizes African-American women indefinitely, because the racial problem will never be solved. Further argues that African-American men are victimized by the reinforced stereotypes of them as violent and oversexed. Calmore, John 0., Spatial Equality and the Kerner Commission Report: A Back-to-the-Future Essay, 71 N.C. L. Rev. 1487 (1993). (Themes 1, 3, 8). Re-examines the 1968 Kerner Commission's conclusion that the "urban crisis" needed two kinds of solutions: (1) programs to encourage the integration of blacks into white neighborhoods, and (2) programs to improve the quality of life in existing black communities. Urges attention to a "spatial equality" approach, namely redirecting resources to improve housing and enrich life in black communities. Critiques the traditional civil rights focus on integrated housing, showing that Whites resist it strongly beyond a certain point, and that the token blacks who are accepted suffer constant stress as well as a loss of community connection. Shows how federal policies, especially after World War II, such as FHA and VA mortgage qualifications, together with other practices such as redlining and neighborhood covenants, made it difficult or impossible for blacks to buy adequate housing. Suggests strategies to assure affordable housing in communities of color as well as integrated housing opportunities. Chang, Robert S., Toward an Asian American Legal Scholarship: Critical Race Theory, Post-Structuralism, and Narrative Space, 81 Cal. L. Rev. 1241 (1993). (Themes 2, 3, 5, 7, 10). Examines historical and present nativist violence and discrimination along with the Model Minority Myth (which holds that Asian Americans-the Model Minority-adapt to the dominant social groups and achieve success quickly and easily) as mechanisms of oppression targeted specifically against Asian Americans. Demon-
19941
strates how acts of violence are dismissed as isolated incidents, while the Model Minority Myth's denial of such discrimination serves to perpetuate its existence. Argues that the latter device not only disadvantages Asian Americans, but harms other minority groups as well, when their valid claims of racial bias are discounted as failure to strive. Discusses the use and justification of narrative in Asian American outsider scholarship generally. Argues that Critical scholarship must expand to include Asians' narratives and experiences, and that without these, CRT cannot give an adequate account of the experience of any group of color. Concludes by proposing a post-structuralist critique as the most effective locus for the emerging and much-needed Asian American legal scholarship. Cook, Anthony, Cultural Racism and the Limits of Rationality in the Saga of Rodney King, 70 Denv. U. L. Rev. 297 (1993). (Themes 1, 3, 5). Discusses three levels of conscious and unconscious racism-individual, institutional, and cultural-that Cook believes were present in the minds ofjurors who acquitted the police officers who beat Rodney King. Shows that intentional individual racism embraces and stems from a rationalist model. Examines the way institutions, through bureaucratic group-think, develop facially neutral norms which can result in racist actions and decisions. Describes how cultural racism is rooted in assumptions of white racial superiority found in JudeoChristian beliefs, Newtonian science, and the Enlightenment. Argues that the jurors, despite indisputable videotape evidence, could not escape these forces and so voted for acquittal. Cook, Anthony E., The Spiritual Movement Towards Justice, 1992 U. Ill. L. Rev. 1007. (Themes 1, 2, 6, 7, 8). Shows how black strategies for fighting racial oppression shift from accentuating sameness (during Abolition and the Civil Rights Movement) when whites emphasized race-conscious particularism, to praising difference (during the Black Power struggle) when whites insisted on colorblind universalism. Calls for a spiritual movement that will balance the tension between unifying
168
[Vol. 66
universalism and a parochial particularism. Hopes that Critical Race Theory will be able to develop such a synthesis. Crenshaw, Kimberl W. & Gary Peller, Reel Time/Real Justice, 70 Denv. U. L. Rev. 283 (1993). (Themes 1, 3, 4, 5) Describes how defense counsel broke down the videotape of Rodney King's beating into individual still pictures, none of which conclusively showed "excessive force," thereby presenting Rodney King as a threat to police rather than a victim of their brutality. Shows that this process (of divorcing meaning from context) is similar to that which the Supreme Court used in disaggregating the evidence presented by the City of Richmond in City of Richmond v. Croson,14 thereby converting a raceconscious affirmative action case into one of colorblindness and formal legal equality. Declares that in each arena a struggle ensued over which narrative would prevail in interpreting events, exposing the inadequacy of a simplistic "rule of law" to mediate these conflicts. Crenshaw, Kimberlb W., Words That Wound: Critical Race Theory, Assaultive Speech, and the First Amendment (1993). See Matsuda, Mari J., et al., Words that Wound: Critical Race Theory, Assaultive Speech and the First Amendment (1993). 15 Culp, Jerome M., Jr., Neutrality, the Race Question, and the 1991 Civil Rights Act: The "Impossibility" of Permanent Reform, 45 Rutgers L. Rev. 965 (1993). (Themes 1, 3). Reviews assumptions in the debate surrounding the passage of the 1991 Civil Rights Act, identifying three fundamental errors: (1) the fixity assumption (belief in a permanent and fixed legal standard), (2) the neutrality assumption (belief that it is possible to develop neutral standards), and (3) the failure to ask "the race question'"-viz., take seriously the reality of racism in America. Examines how the Supreme Court deemed race an irrelevant issue in Wygant v. Jackson Board of
14. 488 U.S. 469 (1989). 15. See infra p. 182.
19941
Education" and other employment discrimination cases, thus enshrining an official view that is blind to the history of past wrongs against blacks and vigilant against special benefits toward them. Shows how a law and economics approach to Title VII incorporates many of the errors mentioned above, and demonstrates how recent Supreme Court decisions frustrate Title VII's intention to increase the number of black employees. Discusses how the 1991 Civil Rights Act leaves unresolved two critical issues: coverage of small employers and temporary workers, and hiring minorities in professional settings such as the academic marketplace. The combination of a sluggish Congress and a refractory Court implies that blacks' disadvantage in the workplace may well be permanent. Culp, Jerome M., Jr., Notes from California: Rodney King and the Race Question, 70 Denv. U. L. Rev. 199 (1993). (Themes 1, 3, 4). Points out that California, the most multiracial state, has failed to fulfill its promise of racial justice in the wake of the Rodney King case and the Los Angeles insurrection. Explains the "rules of engagement" by which African-American males are taught to survive in the face of a hostile criminal justice system. Argues that failing to acknowledge race in matters of criminal justice increases racial subordination. Examines the case of a sixty-five-year-old white woman receiving worker's compensation due to her aversion toward black men after she was attacked by an unidentified male she insisted was black. Urges that law and legal remedies should not be racially blind or silent and that "the race question" should be constantly and insistently asked. Culp, Jerome M., Jr., You Can Take Them to Water but You Can't Make Them Drink: Black Legal Scholarship and White Legal Scholars, 1992 U. Ill. L. Rev. 1021. (Themes 2, 4, 6, 9). Analyzes and compares stories by scholars of color that white legal scholars find comprehensible with those they generally do not. Notes the wide public acceptance of
16.
476 U.S. 267 (1986).
170
[Vol. 66
Stephen Carter's Reflections of an Affirmative Action Baby 7 and shows that Patricia Williams' The Alchemy 1 of Race and Rights 8 is less understood and accepted by whites because it is unconventional, oppositional, and tells stories whites do not like to hear. Concludes that Williams' approach, if heeded by majority race legal scholars, can transform the legal system into one that will take seriously the lives and experiences of black people. Davis, Peggy C., Neglected Stories and the Lawfulness of Roe v. Wade, 28 Harv. C.R.-C.L. L. Rev. 299 (1993). (Themes 3, 6). Traces the impact of slavery and abolition on the development of family law. Shows that attention to these neglected stories can illuminate recent legal struggles over such issues as parental rights, abortion, contraception, sterilization, and the status of children. Demonstrates that the ideal of family liberty was refined and strengthened through the struggle against slavery and caste oppression. Concludes that the same principles of respect for individuals and the inviolability of life that animated emancipation and Reconstruction can and should shape the developing law of intimate communities. Delgado, Richard, Five Months Later (The Trial Court Opinion), 71 Tex. L. Rev. 1011 (1993). (Themes 1, 2). Presents a fictional trial court opinion written by a federal judge upholding the minority hiring programs of several law schools and the American Association of Law Schools. Follows an earlier article by Michael Paulsen 9 in which the same judge overruled a motion for summary judgment brought by the schools when sued by a disappointed white applicant for a law teaching position. Shows how case law and social policies can easily be marshalled to support race-conscious hiring, indeed, that the law schools' "two-pile" procedure might
17.
STEPHEN L. CARTER, REFLECTIONS OF AN AFFIRMATIVE ACTION BABY
(1991).
18. PATRICIA J. WILLIAMS, THE ALCHEMY OF RACE AND RIGHTS (1991). 19. Michael S. Paulsen, Reverse Discrimination and Law School Faculty Hiring: The Undiscovered Opinion, 71 Tex. L. Rev. 993 (1993).
19941
well have been instituted to protect applicants such as the disappointed white. Delgado, Richard, The Inward Turn in Outsider Jurisprudence, 34 Wm. & Mary L. Rev. 741 (1993). (Themes 7, 10). Discusses two movements respecting outsider jurisprudence-the call for standards by which radical feminist or Critical Race scholarship can be judged, and antiessentialism, the tendency of these groups successively to identify smaller subgroups within themselves. Argues that the first movement is premature and potentially harmful, while the second (antiessentialism) is normal and not to be deplored. Argues that both developments are responses to power shifts and concern over paradigm change. Delgado, Richard, On Telling Stories in School: A Reply to Farber and Sherry, 46 Vand. L. Rev. 665 (1993). (Theme 10). Replies to Farber and Sherry's critique of narrative jurisprudence. Points out that critics overlook that majoritarians tell stories too (although these seem not at all like stories, but the truth). Points out a number of errors and misstatements in Farber and Sherry's critique, including their mistaken belief that the validity of CRT storytelling rests on the claim of a distinct minority "voice" and that narrative writing poses greater than usual risks of lack of authenticity, typicality, truth, and reasoned analysis. Delgado, Richard, Rodrigo's Second Chronicle: The Economics and Politics of Race, 91 Mich. L. Rev. 1183 (1993). (Themes 1, 2, 4). Rodrigo Crenshaw (the author's fictional alter ego) returns from exile and engages in a dialogue with his mentor about issues of race, in particular the question of whether the economic marketplace is likely over time to eliminate racism because nonracist businesses will drive out ones that discriminate. Rodrigo argues that racism will tend to increase, not decrease, in a capitalist system because of the operation of a widely shared construction or stereotype and because of Western myths
20. See infra p. 175.
172
[Vol. 66
about salvation and superiority, all operating together in a "neo-crypto theologico double feedback loop" to keep African-Americans and others in a subordinate position. Delgado, Richard, Rodrigo's Third Chronicle: Care, Competition, and the Redemptive Tragedy of Race, 81 Cal. L. Rev. 387 (1993). (Themes 2, 3, 5). Rodrigo outlines a new civil rights strategy based on love. In his ideal society, caregiving would be socialized, while the productive sector would be relegated to competition and the free market economy. Concentrated caregiving would redeem many whom our society currently ignores. Crossovers from the two sectors would assure a vibrant, productive, and well-equipped caregiving sector, effectively providing for those who now exist "beyond love." Suggests a new mythic framework and a new form of coercion in order to begin moving toward such a society. Delgado, Richard, Rodrigo's Fourth Chronicle: Neutrality and Stasis in Antidiscrimination Law, 45 Stan. L. Rev. 1133 (1993). (Themes 2, 3, 5). Rodrigo and the professor use an incident at Rodrigo's law school as a springboard for discussing the difficulty of redressing social harms through neutral legal and social rules. Rodrigo argues that in a society such as ours, even a decision maker with the most pristine social conscience and no traces of overt racism will still end up making decisions adverse to minorities of color. Analyzes the cultural background against which our society and legal system render decisions, concluding that it has much more efficacy than law on the books. Notes that even race-conscious rules intended to protect and benefit minorities generally end up making matters worse. Gives a number of cultural and structural reasons for the persistence of racism even in a society whose formal rules condemn it. Delgado, Richard, Rodrigo's Fifth Chronicle: Civitas, Civil Wrongs, and the Politics of Denial, 45 Stan. L. Rev. 1581 (1993). (Themes 1, 2, 9). Rodrigo and the professor discuss the highly normative turn legal education has taken. They decide that the
1994]
intense preoccupation with professional responsibility and vague, aspirational catch phrases like "the role of the good lawyer" are aspects of the "Owl of Minerva" phenomenon and by-products of a profession and a culture in decline. Instead of engaging in denial and normative discussion, they conclude, we would do better to confront our racial and economic problems directly. In this fashion, outsider discourse and insights can break the circle of nonproductive, predictable, normative dialogue that rarely goes anywhere. Delgado, Richard, A Shifting Balance: Freedom of Expression and Hate-Speech Restriction, 78 Iowa L. Rev. 737 (1993). See Stefancic, Jean & Richard Delgado, A Shifting Balance: Freedom of Expression and Hate-Speech Restriction, 78 Iowa L. Rev. 737 (1993).21 Delgado, Richard, Words That Wound: Critical Race Theory, Assaultive Speech, and the First Amendment (1993). See Matsuda, Marl et al., Words That Wound: Critical Race Theory, Assaultive Speech and the First Amendment (1993).22 Edmonds, Erin, Students as Teachers, Teachers as Learners, 91 Mich. L. Rev. 2025 (1993). See Bell, Derrick & Erin Edmonds, Students as Teachers, Teachers as Learners, 91 Mich. L. Rev. 2025
(1993).23
Espinoza, Leslie G., The LSAT: Narratives and Bias, 1 Am. U. J. Gender & L. 121 (1993). (Themes 4, 9). Examines actual LSAT questions for test bias, finding that many incorporate stories that are offensive-that degrade women, reinforce stereotypes, or remind outsiders that they are outsiders. Such items bring out socialized self-doubt and hinder performance. Other items encode cultural views, presuppositions, information, or preferences that test-takers of color or from disadvantaged backgrounds may not share. Describes test agencies' resistance to mandated disclosure of tests
21. 22. 23.
See infra p. 187. See infra p. 182. See supra p. 164.
174
[Vol. 66
and test items, and argues for continuing narrative analysis of test items through public exposure. Evans, Monica J., Stealing Away: Black Women, Outlaw Culture and the Rhetoric of Rights, 28 Harv. C.R.-C.L. L. Rev. 263 (1993). (Themes 1, 2, 8, 11). Compares the predicament of black women to that described in early African-American escape songs and spirituals. Urges that the messages of those songs, which included resistance and subversion (themes heard only by the slaves) have special force today, when many communities of color live a constructed identity as outlaws. Examines the role of black women as shapers of a positive, outlaw culture-which included theft, passive resistance, noncooperation with the draft, and cheating-as an alternative to their continued subjugation. Uses example of nineteenth century black women's clubs to illustrate how assertion and resistance can subvert the dominant definition of woman and shows how rights discourse is fully consistent with a community-based model of reconstructive theory and practice. Fair, Bryan K., Foreword: Rethinking the Colorblindness Model, 13 Nat'l Black L.J. 1 (1993). (Themes 1, 2, 3, 5). Shows that racial inequality and color awareness, including government-sponsored or -sanctioned discrimination against blacks and others, are dominant themes in U.S. racial history. Examines the evolution of the antidiscrimination principle and the competing visions of the colorblindness model to reveal the means by which colorblindness today permits the same racial subordination allowed by the separate-but-equal doctrine 2 of Plessy v. Ferguson.4 Demonstrates that the Supreme Court analyzes race and gender classifications differently, and suggests that the Court's different treatment of these discrimination cases is unwarranted. Advocates applying the gender classification standard to racial cases for a fairer and more consistent result. Supports his argument by applying to the facts of City 2 of Richmond v. Croson 1 the Court's ground of differ-
24. 163 U.S. 537 (1896). 25. 488 U.S. 469 (1989).
1994]
ence rationale usually applied in sex discrimination cases. Farber, Daniel A. & Suzanna Sherry, Telling Stores Out of School: An Essay on Legal Narratives, 45 Stan. L. Rev. 807 (1993). (Themes 2, 10). Evaluates the legal storytelling movement and its claims. Asserts that women and people of color do not write or speak in a special or different voice inaccessible to others. Holds that legal storytelling may be a useful and valid form of scholarship, but that it also has dangers. Stories may be inauthentic, atypical, or untrue. Moreover, they may be used to support a particular interpretation of an event (for example, discrimination in a department store) when other interpretations are equally plausible. Moreover, stories must be adequately tied to legal doctrine and analysis to be useful to lawyers; many that have been offered so far are not. Concludes with a call for a standard for evaluating the new narrative scholarship so that the legal academy will be able to recognize high quality narratives and condemn those that fail to meet the standard. Flagg, Barbara J., Was Blind But Now I See: White Race Consciousness and the Requirement of Discriminatory Intent, 91 Mich. L. Rev. 953 (1993). (Themes 1, 4, 5). Argues for greater Critical focus on whiteness, and in particular the way in which ostensibly race-neutral legal rules end up favoring whites and disadvantaging blacks. Urges that "transparency" be recognized as a defining characteristic of whiteness and that skepticism toward whiteness is overdue. This new skeptical stance toward whiteness can ultimately lead to development of a positive white racial identity, one that sees whiteness not as a racial norm but as merely one racial identity among many. Government decisionmakers (including courts) should adopt the same skepticism, taking responsibility for eradicating the racism that arises from unquestioned application of neutral rules and norms that perpetuate white advantage.
176
[Vol. 66
Gordon, James W., Did the First Justice Harlan Have a Black Brother? 15 W. New Eng. L. Rev. 159 (1993). (Theme 3). In 1848, James Harlan, the father of future Supreme Court Justice John Marshall Harlan, appeared in a Kentucky courtroom for the purpose of freeing his thirtytwo-year-old slave, Robert Harlan. Robert, a lightskinned, blue-eyed man, had been raised in the household of the Harlan family, given an informal education, and afforded opportunities to travel and make money. Argues that the special treatment accorded Robert by his influential master-family has a simple explanation: Robert Harlan was James Harlan's son, and the first Justice Harlan (the author of the famous dissent in 26 Plessy v. Ferguson ) had a black half-brother. Traces the lives of Robert and the other Harlans; reviews the history of miscegenation in the antebellum South; and concludes that the special relation of Robert and the young Justice-to-be may have shaped his dream of a color-blind future-"and by writing about it [he] ... began the process of creating it." Greene, Dwight L., Justice Scalia and Tonto: Judicial Pluralistic Ignorance and the Myth of Colorless Individualism in Bostick v. Florida, 67 Tul. L. Rev. 1979 (1993). (Themes 1, 8). Challenges Justice Antonin Scalia's view that ethnic whites are not privileged and cannot be held responsible for the legacy of discrimination against blacks. Uses stories to illustrate ways class and color privilege whites and burden people of color. Shows how twenty years of Republican judicial appointments produced demographically distorted courts. Demonstrates how judges, by virtue of their similar class, ideology, and race, form a cognitive reference group sharing erroneous beliefs about individuals not like them ("pluralistic ignorance"). Shows how judges draw and build on shared myths (e.g., the Lone Ranger and Tonto) to reconcile disparities between the treatment of blacks and whites, particularly in police encounters. Suggests alternatives to "colorless individualism," including developing a more diverse judiciary and using a broader reasonable person stan-
26.
163 U.S. 537 (1896).
1994]
dard. Advocates rights training in communities of color so that people can learn how to exercise their rights in encounters with police and "just say no." Guinier, Lani, Groups, Representation, and Race-Conscious Districting: A Case of the Emperor's Clothes, 71 Tex. L. Rev. 1589 (1993). (Themes 1, 5). Counters arguments by critics of the Voting Rights Act that race-conscious districting and group representation diminish democracy, cause fixation on racial identity, and unfairly dilute the white vote. Defends group representation while questioning whether race-conscious or territorial districting is the most effective way of assuring that all have a voice. Suggests "one-vote-onevalue" as an alternative to gerrymandering and winnertake-all elections. Illustrates all these points by means of an analysis of a recent election in a Brooklyn/Queens district. Guinier, Lani, The Representation of Minority Interests: The Question of Single-Member Districts, 14 Cardozo L. Rev. 1135 (1993). (Themes 1, 5). Focuses on the limits of districting as a method for representing minority interests. Develops two critical assumptions that should underlie a political fairness standard: representing interest constituencies and insuring these constituencies' proportional influence. Discusses three generations of voting rights issues: (1) gaining direct access to the ballot, (2) increasing the number of minorities in elected office, and (3) analyzing how minority group interests are marginalized and diluted even when minorities are elected. Argues that winner-take-all single-member districting has failed to mobilize sustained voter participation, foster useful debate about issues, or go beyond tokenism in promoting opportunities for minorities. Discusses alternative election systems to single-member majority black districts, such as minority influence districts or cumulative voting (which abandons districting altogether). Declares that proportionate interest representation can reduce the legislative power of a prejudiced majority and assure that minority votes are not diluted.
178
[Vol. 66
Harris, Cheryl I., Law Professors of Color and the Academy: Of Poets and Kings, 68 Chi.-Kent L. Rev. 331 (1992). (Themes 2, 9, 11). Incorporates Chinua Achebe's stories of African kings regarding power and its associated risks. Illustrates similarities between the tribulations Achebe recounts and those of scholars of color today. Argues that it is the duty of scholars of color to expose law's hypocrisy and oppression, despite the risk that doing so may make "the king" angry. Recalls her own ambivalence and doubleconsciousness as she entered law teaching and the many problems she encountered as a lone black attempting to infuse students with a sense of critical analysis. Harris, Cheryl I., Whiteness as Property, 106 Harv. L. Rev. 1707 (1993). (Themes 1, 2, 3, 4, 5). Investigates the relationship between race and property. Examines how "whiteness," originally a form of racial identity, evolved into a property interest for those who could define themselves as white. Shows how the origins of whiteness as a property interest lie in systems of domination by whites of black and Native American peoples. Argues that the common premise "whiteness" and "property" share-a right to exclude-enabled white identity to become the basis of racial privilege which allocated societal benefits, these arrangements then becoming legitimated in law as a kind of status property. Illustrates how property interest in whiteness was not discarded in Plessy v. Ferguson2' and Brown v. Board 2 of Education,8 but rather transformed, and that recent affirmative action cases mask the privileging of whiteness in the guise of enshrining the status quo as a neutral baseline. Concludes that focusing on distortions created by property interest in whiteness opens alternative perspectives on the affirmative action debate. Iglesias, Elizabeth M., Structures of Subordination: Women of Color at the Intersection of Title VII and the NLRA. Not!, 28 Harv. C.R.-C.L. L. Rev. 395 (1993). (Themes 1, 6, 7, 10, 11).
27. 163 U.S. 537 (1896). 28. 347 U.S. 483 (1954).
19941
Takes issue with recent writing that focuses excessively on legal meanings and stories of the oppressed in hopes of improving their condition. Argues instead for attention to questions of institutional power and the structural violence of daily life, as abetted by law. Borrows from liberation theology its commitment to material empowerment of subordinated people. Argues that the potential of approaches such as "sensitivity training" and appeals to empathy has largely been exhausted. Suggests that a better approach for advancing the objectives of Title VII and national labor laws would focus on collective organization of workers and the way in which the legal system operates to keep workers, especially women of color, fragmented and powerless. Ikemoto, Lisa C., Traces of the Master Narrative in the Story How We of African-American/Korean-American Conflict: Constructed "Los Angeles," 66 S. Cal. L. Rev. 1581 (1993). (Themes 2, 3, 4, 5, 6, 7). Analyzes how society constructed, explained, and gave meaning to "Los Angeles"-the wave of disturbances that followed the first Rodney King verdict-and especially to the intergroup conflict that supposedly underlay those events. Explores the dynamic behind the many accounts of the L.A. events, noting the way a master narrative defines race oppositionally and in terms of power-laden, interest-serving categories and narratives, including claims of entitlement, racial positioning, constructed identities and racial pairing. Shows that "Los Angeles" has become a metaphor for multiculturalism's failure and impossibility, as well as a justification for racial distancing. Johnson, Alex M., Jr., Defending the Use of Quotas in Affirmative Action: Attacking Racism in the Nineties, 1992 U. Ill. L. Rev. 1043. (Themes 1, 3, 4, 5, 9). Shows how benign affirmative action programs have failed to improve the lot of African-Americans, and defends racial quotas as a superior means of effecting change. Refutes the myth that quotas benefit unqualified blacks, and argues that the "objective standards" created and applied by the white meritocracy are not objective at all. Insists that, given the history of
180
[Vol. 66
discrimination against blacks in the U.S. and the ascription of merit to society's dominant group, mandatory admissions quotas for colleges and graduate schools remain necessary. Points out that the "pool problem" and other arguments against affirmative action programs may themselves often be unconscious expressions of simple racism. Johnson, Sheri L., The Language and Culture (Not to Say Race) of Peremptory Challenges, 35 Wm. & Mary L. Rev. 21 (1993). (Themes 1, 4). Focuses on the defendant's claim that race-based jury selection violates the guarantee of equal protection of the law. Reviews cases concerning the use of peremptory challenges to eliminate jurors of different races in criminal cases, leading up to Hernandez v. New York, 29 the "court interpreter" case. Argues that courts are insufficiently vigilant in cases where a trait, such as language ability, culture, appearance, or demeanor serves as a pretext for juror exclusion that is in reality race-based. Points out that the makeup of a jury powerfully determines verdicts, especially where the defendant is a member of a racial minority group. Urges that the current approach be replaced by a "Strauder" model of jury selection that would recognize the realities and dynamics of race in the use of jury strikes. Johnson, Sheri L., Racial Imagery in Criminal Cases, 67 Tul. L. Rev. 1739 (1993). (Themes 4, 5). Examines the use of racial stereotypes, both blatant and subtle, in criminal cases. Shows how negative racial imagery and slurs often permeate criminal trials from pretrial publicity, voir dire and opening statements to testimony, closing arguments, jury instructions, and deliberations. Discusses specific stereotypes and fears. Reviews existing remedies that can be used to restrict racial imagery in trials and shows, through examples, how they have been for the most part ineffective. Proposes other measures that could be taken to control racial stereotyping in criminal cases such as code-of-
29.
50 U.S. 352 (1991).
1994]
ethics provisions forbidding the use of negative racial imagery, and racial imagery shield laws similar to the rape shield laws. Warns that society may be slow to adopt such measures because self-interest and selfrighteousness stand in the way. Kennedy, Randall, In Praise of the Struggle for Diversity on Law School Faculties, 22 Seton Hall L. Rev. 1389 (1992). (Themes 1, 6, 9). Praises the movement toward diversity. Uses example of his own institution (Harvard), which did not hire its first black tenured professor until 1969, to show the slow pace of racial progress. Points out that diversification is particularly crucial in the professoriate, since the newcomers can challenge power and conventional notions of authority. Characterizes the core aspiration of the diversity effort as "morally, politically, and intellectually compelling," and urges that schools avoid criteria for selecting candidates that are themselves tainted with racism (such as Supreme Court clerkships). Encourages diversity proponents not to exaggerate the role of racial discrimination in explaining the paucity of minorities on law faculties, nor to succumb to the credentializing of race and racial voice as criteria for selection. Lawrence, Charles R., III, Crossburning and the Sound of Silence: Antisubordination Theory and the First Amendment, 37 Vill. L. Rev. 787 (1992). (Themes 1, 2). Criticizes the Supreme Court's decision in R.A. V v. City 3 of St. Paul" as focusing almost entirely on the freespeech rights of cross burners while the victims (the Russ Jones family) are almost entirely absent from the opinion. Notes that courts often ignore the historical significance of cross burning and other forms of hate speech as instruments of terror and repression, transforming an act designed to silence its victims and put them in their place into an invitation to join a public discussion. Discusses the upsurge in racist hate crimes and speech, and argues that the equality principle of
30.
112 S. Ct. 2538 (1992).
182
[Vol. 66
3 Brown v. Board of Education ' supplies a means of fashioning remedies to control it.
Lawrence, Charles R., III, Words That Wound: Critical Race Theory, Assaultive Speech, and the First Amendment (1993). See Matsuda, Mari J., et al., Words That Wound: Critical Race Theory, Assaultive Speech, and the First Amendment (1993).32 Matsuda, Mari J., Charles R. Lawrence III, Richard Delgado & Kimberl6 W. Crenshaw, Words That Wound: Critical Race Theory, Assaultive Speech, and the First Amendment (1993). (Themes 1, 2, 3, 4). Collection of essays on the question of hate speech, embracing tort, regulatory, and constitutional aspects of this problem. Points out that the current near-absolutist paradigm of free-speech protection inadequately protects minorities against the harms of hate speech. Includes a lengthy analytical introduction explaining Critical Race Theory and showing the relation of the emerging anti-hate speech scholarship to that broader movement. Nunn, Kenneth B., Rights Held Hostage: Race, Ideology and the Peremptory Challenge, 28 Harv. C.R.-C.L. L. Rev. 63 (1993). (Themes 1, 4, 5). Argues that the Supreme Court's recent jury selection cases have failed abjectly to control racism in criminal trials. Attributes this failure to reliance on colorblind principles which substitute procedural for substantive equality, an approach that ends up privileging white over black interests. Concludes with a counternarrative demonstrating that black defendants are situated differently from whites with respect to jury discrimination and that ostensibly neutral treatment silently and effectively contributes to black subjugation. Urges that the use of the peremptory challenge to strike black jurors be prohibited in cases where there is a danger that the trial will be tainted by racism.
31. 32.
347 U.S. 483 (1954). See infra p. 182.
1994]
Ontiveros, Maria L., Three Perspectives on Workplace Harassment of Women of Color, 23 Golden Gate U. L. Rev. 817 (1993). (Themes 4, 6). Provides a framework for understanding the roles of race and culture in workplace harassment of women of color. Describes how racism and sexism combine to form an indistinguishable whole from the perspectives of the harasser, the victim, and the legal system. Argues for reform through either a modification of the rules governing sexual harassment, or the creation of a new cause of action for women of color who are harassed at the workplace. Peller, Gary, Criminal Law, Race, and the Ideology of Bias: Transcending the Critical Tools of the Sixties, 67 Tl. L. Rev. 2231 (1993). (Themes 1, 3, 6, 9). Examines racial bias of the criminal justice system and the way criminal law interacts with American racism. Discusses the contemporary emphasis on procedure rather than substance in criminal law. Shows how that emphasis, combined with the integrationist model of racial harmony (which discounts race as an arbitrary attribute), limits the range of discourse about race and criminal law. Reviews Black Nationalist position that police conduct in African-American communities resembles colonialism, and calls for community control. Suggests rejection of premises of neutrality in procedural criminal law and an analysis of substantive criminal law as "embodying a culturally particular way of ascribing responsibility and punishment." Peller, Gary, The Discourse of Constitutional Degradation, 81 Geo. L.J. 313 (1992). (Theme 10). Replies to Mark Tushnet's critique of contemporary legal scholarship in his article, The Degradationof Constitu33 tional Discourse. Takes particular issue with its critique of the integrity of certain authors, such as Derrick Bell, Catharine MacKinnon, and Patricia Williams, and the article's assertion that their writing is lacking in irony and a sense of proportion, thus contrib-
33.
See infra pp. 187-88.
184
[Vol. 66
uting to a "degradation" of legal discourse. Argues that Tushnet's attack itself is not unbiased and is a manifestation of generational and cultural tensions at work in the legal academy. Peller, Gary, Notes Toward a Postmodern Nationalism, 1992 U. Ill. L. Rev. 1095. (Themes 1, 5, 8). Discusses tensions between black nationalism and integration, pointing out potential pitfalls in each. Proposes a role for white scholars exploring the roots and voice of the dominant culture, with a view toward coexisting with blacks as different and sometimes alien groups, moving toward a cosmopolitan postmodern nationalism. Peller, Gary, Reel Time/Real Justice, 70 Denv. U. L. Rev. 283 (1993). Reel See Crenshaw, Kimberl6 W. & Gary Peller, 34 Time/Real Justice, 70 Denv. U. L. Rev. 283 (1993). Perea, Juan F., Demography and Distrust: An Essay on American Languages, Cultural Pluralism, and Official English, 77 Minn. L. Rev. 269 (1992). (Themes 1, 3). Recounts history of U.S. "nativism"-anti-immigrant and anti-minority sentiment aimed at reinforcing the core culture-and shows that the recent English-only movement is a similar demand for conformity. Shows that this movement rests on and disseminates a series of myths such as that national unity depends on a common language and that the only true language of American identity is English. Shows how legal history documenting the relation between the dominant and other cultures with respect to language casts doubt on these myths, indeed that early in our history a variety of languages other than English were deemed official. Argues that the intense nativism of the official English movement and its predecessors is merely a device to exclude unpopular groups from full citizenship, and deserves to be resisted as an unwarranted attack on liberty and the right to equal treatment under the law.
34. See supra p. 168.
1994]
powell, john a., Race and Poverty: A New Focus for Legal Services, 27 Clearinghouse Rev. 299 (1993). (Themes 1, 5, 6). Points out that the work of lawyers for the poor and that of civil rights activists, while valuable and gallant, has not done much to address the unique problems of the black poor. Urges greater attention to the way in which race and poverty intersect. Challenges the assumption that addressing problems of racial injustice will cause poverty to recede; and conversely, that redressing poverty will solve all the problems of the community of color. Shows that inadequate schools, housing, and criminal-justice policies trap many blacks in the lower reaches of income and class. Warns that these problems will require special treatment if they are not to become permanent features of our national landscape. Ramirez, Deborah A., Excluded Voices: The Disenfranchisement of Ethnic Groups from Jury Service, 1993 Wis. L. Rev. 761. (Themes 3, 7). Criticizes the United States Supreme Court decision, Hernandez v. New York,"6 in which the Court held that it was constitutional to exclude bilinguals from a jury where witnesses were to testify in Spanish. Responds to the plurality's assumption that the potential jurors were challenged only because of deficient responses and not because of their status as bilinguals. Uses psycholinguistic evidence to explain why it is impossible for any bilingual to comply with an instruction to acknowledge only the English interpretation of foreign language testimony when the juror comprehends that foreign language. Argues that Hernandez effectively discriminates against Latinos because of the super-correlation between race and language in the Latino community. Roberts, Dorothy E., Crime, Race, and Reproduction, 67 Tul. L. Rev. 1945 (1993). (Themes 1, 11). Shows, through numerous examples, the way race is embedded in the foundation of criminal law (viz., who criminals are, what conduct constitutes a crime, which crimes are most serious). Discusses the eugenics
35.
500 U.S. 352 (1991).
186
[Vol. 66
movement and the long history in the U.S. of biological solutions (e.g., sterilization and castration) to control and punish blacks and others deemed "socially inadequate." Focuses on the convergence of race, crime, and reproduction to highlight how new "technologies of power" (e.g., Norplant, biochemical therapies) are being used further to oppress and dominate blacks. Roberts, Dorothy E., Rust v. Sullivan and the Control of Knowledge, 61 Geo. Wash. L. Rev. 587 (1993). (Themes 1, 11). Discusses the impact of the Supreme Court decision in Rust v. Sullivan 6 which upheld a regulation prohibiting federally funded family planning clinics from providing information about abortion. Describes the plight of poor and low-income women dependent on these clinics for health care and maintains that the Court inflicted a kind of violence on them. Examines how oppressors use knowledge to maintain their superior position by determining social meanings, controlling educational content, stifling sources of information needed by oppressed groups, and, in the case of the Supreme Court, manipulating purportedly neutral legal doctrine. Explores how the oppressed can use knowledge as a means of liberation. Suggests that a proper constitutional vision would recognize the importance of knowledge for self-determination and affirm a governmental duty to provide it to those dependent on government funds. Russell, Jennifer M., On Being a Gorilla in Your Midst, or The Life of One Blackwoman in the Legal Academy, 28 Harv. C.R.C.L. L. Rev. 259 (1993). (Themes 2, 3, 9). Describes her thoughts and feelings on being confronted by a photograph of a menacing gorilla left anonymously in her faculty mailbox. Attributes the incident to white resentment over the presence of a black woman in the academy and the sense that her appointment must have come at the expense of a white individual who was not selected. Describes her reaction to treatment, as the beneficiary of a "clemency" appointment, at the hands of
36.
500 U.S. 173 (1991).
1994]
colleagues and students. Points out that diversity appointments, unaccompanied by structural and attitudinal change on the part of the institution, exact a severe toll on their beneficiaries. Spann, Girardeau A., Race Against the Court (1993). (Themes 1,3,5,8). Book-length expansion of his earlier article, Pure Poli7 tics, 3 on the political strength and prospects of minorities. Argues that blacks and other minority people have relatively little to gain from Supreme Court litigation, but much to lose. Urges that a habit of bringing grievances to the courts breeds dependency and is apt to lead to disillusionment as courts inevitably turn aside claims that will too radically change the status quo. Suggests that extrajudicial means, including legislative lobbying and grassroots activism, will often offer greater possibilities for advancing the black agenda. Stefancic, Jean & Richard Delgado, A Shifting Balance: Freedom of Expression and Hate-Speech Restriction, 78 Iowa L. Rev. 737 (1993). (Themes 1, 3). Reviews Striking a Balance: Hate Speech, Freedom of 38 Expression, and Non-discrimination, a recent collection of essays about the treatment of hate speech and hate crimes in many societies. Argues that the experiences of other nations in coping with these problems can shed light on arguments and counter-arguments commonly made in the U.S. debate, includirg that hatespeech rules are ineffective and that suppressing hateful expression only makes matters worse. Concludes that speech and inclusion, dialogue and community are inseparable-each presupposing, but also threatening, the other. Tushnet, Mark, The Degradation of Constitutional Discourse, 81 Geo. L.J. 251 (1992). (Theme 10). Criticizes recent uses of storytelling in the law as presenting problems of integrity and judgment, reflected
37. Girardeau A. Spann, Pure Politics, 88 MICH. L. REV. 1971 (1990).

38. STRIKING A BALANCE: HATE SPEECH, FREEDOM OF EXPRESSION, AND NONDISCRIMINATION (Sandra Coliver ed., 1992).
188
[Vol. 66
in missteps in literary style and presentation. Concedes that properly employed, stories and narratives serve to connect particular cases and general legal rules. Concludes, however, that this mediation may easily fall short of its promise by overworking authorial authority, by an excess of "style," by self-righteousness and moral superiority, and by cavalier dismissal of alternative interpretations of the event the writer is describing. Further illustrates the pitfalls of story construction by means of the Hill-Thomas hearings and recent Supreme Court opinions. (Reply by Gary Peller follows.) West, Cornel, Race Matters (1993). (Themes 1, 3, 6, 8). Eight essays dealing with the poverty of current discourse about race. Urges that the "race problem" is not that black people bother and endanger whites through crime and other social pathology (as conservatives believe), nor that blacks somehow fail or refuse to be included or integrated into white society (as many liberals believe). Rather, concludes that the problem of color is constitutive and symptomatic of our national life; its solution will require that all must change, not just blacks. Depicts the rise of both black nationalism and neo-conservatism as responses to failed national policies based on hedonism and racial xenophobia. Calls for visionary leadership, black and white, to rescue U.S. society from its downward spiral and loss of hope.
COMMENTARY
We offer the following observations on the 1993 Critical Race Theory literature: 1. Quantity. Readers of the original bibliography might well be curious to know whether the productivity reflected in that collection has continued or declined. If anything, it seems to have increased. Assuming that CRT began as a more or less formal movement around 1980, it has produced about fifteen to twenty significant articles and books each year during the 39 period covered by the first bibliography. Last year, however,
39. See supra note 1. The first bibliography included 217 items, most of which appeared over a 13-year period, 1980-1992, for an average of about 17 per year.
1994]
we counted fifty-two items using the same criteria as employed for the first bibliography. The concern that symposia and bibliographies may signal the beginning of a movement's decline4 seems not to be borne out (so far, at any rate) in the case of CRT. 2. Quality of the output. It is difficult to devise any sort of objective measure of the quality of a large body of work. CRT articles and books appear to be cited regularly by mainstream and non-mainstream writers. Of the forty-nine law review articles annotated, eighteen appeared in journals ranked in the top twenty by the latest Chicago-Kent survey of faculty scholarship. 4 ' Three books appeared, each published by a major press.4 2 3. Criticismand criticalcommentary. The movement drew more criticism than it did in the period covered by the earlier bibliography.43 Much of the criticism centered around narrative jurisprudence (storytelling), with a number of writers raising doubts about the validity, authenticity, and usefulness of that type of scholarship, as well as questioning how storybased writing could be evaluated for purposes of promotion and tenure. 44 The literature also contained spirited replies to all these queries and challenges. 45 4. Assimilation into the mainstream. At the same time that the movement was attracting and responding to criticism, it was earning a number of measures of acceptance into the mainstream. Several law review symposia and collections of
40. Stefancic, supra note 5, at 673 (raising similar possibility in connection with scholarly symposia). 41. Janet M. Gumm ed., Chicago-Kent Law Review Faculty Scholarship Survey, 66 CHI.-KENT L. REv. 509 (1990).
42. MARI J. MATSUDA ET AL., WORDS THAT WOUND (1993) (published by Westview Press); GIRARDEAU A. SPANN, RACE AGAINST THE COURT (1993) (published by New York University Press); and CORNEL WEST, RACE MATTERS
(1993) (published by Beacon Press). 43. In this bibliography, there are seven works which fall under Theme 10 (criticism and self-criticism; responses). In the earlier bibliography, covering 13 years, there were 45, or approximately three per year. 44. See, e.g., Farber & Sherry, supra p. 175; see also Edward L. Rubin, On Beyond Truth: A Theory for Evaluating Legal Scholarship, 80 CAL. L. REV. 889 (1992). 45. See, e.g., Chang, supra pp. 166-67; Culp, You Can Take Them to Water But You Can't Make Them Drink: Black Legal Scholarship and White Legal Scholars, supra pp. 169-70; Delgado, On Telling Stories in School: A Reply to Farber and Sherry, supra p. 171; Peller, The Discourse of Constitutional Degradation,supra pp. 183-84.

66 UColo LRevi

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

66 UColo LRevi

Uploaded by

Copyright:

Available Formats

Citation: 66 U. Colo. L. Rev. i 1994-1995 Content downloaded/printed from HeinOnline (http://heinonline.

University of Colorado Law Review

HeinOnline -- 66 U. Colo. L. Rev. [i] 1994-1995

HeinOnline -- 66 U. Colo. L. Rev. [ii] 1994-1995

COLORADO LAW REVIEW

HeinOnline -- 66 U. Colo. L. Rev. [iii] 1994-1995

HeinOnline -- 66 U. Colo. L. Rev. [iv] 1994-1995

COLORADO LAW REVIEW

HeinOnline -- 66 U. Colo. L. Rev. [v] 1994-1995

HeinOnline -- 66 U. Colo. L. Rev. [vi] 1994-1995

COLORADO LAW REVIEW

HeinOnline -- 66 U. Colo. L. Rev. [vii] 1994-1995

HeinOnline -- 66 U. Colo. L. Rev. [viii] 1994-1995

COLORADO LAW REVIEW

HeinOnline -- 66 U. Colo. L. Rev. [ix] 1994-1995

HeinOnline -- 66 U. Colo. L. Rev. [x] 1994-1995

COLORADO LAW REVIEW

HeinOnline -- 66 U. Colo. L. Rev. 1 1994-1995

UNIVERSITY OF COLORADO LAW REVIEW

* Associate Justice, United States Supreme Court.

HeinOnline -- 66 U. Colo. L. Rev. 2 1994-1995

JUSTICE BYRON R. WHITE

Barksdale, supra note 1, at 4.

HeinOnline -- 66 U. Colo. L. Rev. 3 1994-1995

UNIVERSITY OF COLORADO LAW REVIEW

HeinOnline -- 66 U. Colo. L. Rev. 4 1994-1995

JUSTICE BYRON R. WHITE

HeinOnline -- 66 U. Colo. L. Rev. 5 1994-1995

UNIVERSITY OF COLORADO LAW REVIEW

HeinOnline -- 66 U. Colo. L. Rev. 6 1994-1995

JUSTICE BYRON R. WHITE

HeinOnline -- 66 U. Colo. L. Rev. 7 1994-1995

UNIVERSITY OF COLORADO LAW REVIEW

HeinOnline -- 66 U. Colo. L. Rev. 8 1994-1995

JUSTICE BYRON R. WHITE

HeinOnline -- 66 U. Colo. L. Rev. 9 1994-1995

UNIVERSITY OF COLORADO LAW REVIEW

L.J. 5, 16-17 (1993).

HeinOnline -- 66 U. Colo. L. Rev. 10 1994-1995

JUSTICE BYRON R. WHITE

HeinOnline -- 66 U. Colo. L. Rev. 11 1994-1995

HeinOnline -- 66 U. Colo. L. Rev. 12 1994-1995

HeinOnline -- 66 U. Colo. L. Rev. 13 1994-1995

UNIVERSITY OF COLORADO LAW REVIEW

HeinOnline -- 66 U. Colo. L. Rev. 14 1994-1995

INITIATIVE AND REFERENDUM PROCESS

HeinOnline -- 66 U. Colo. L. Rev. 15 1994-1995

UNIVERSITY OF COLORADO LAW REVIEW

J. ALLEN SMITH, THE SPIRIT OF AMERICAN GOVERNMENT 352 (1965). THE

the commission form of local government. Id.

HeinOnline -- 66 U. Colo. L. Rev. 16 1994-1995

INITIATIVE AND REFERENDUM PROCESS

RICHARD HOFSTADTER, THE AGE OF REFORM: FROM BRYAN TO F.D.R.

HeinOnline -- 66 U. Colo. L. Rev. 17 1994-1995

UNIVERSITY OF COLORADO LAW REVIEW

24. U.S. CONST. art. IV, 4.

HeinOnline -- 66 U. Colo. L. Rev. 18 1994-1995

INITIATIVE AND REFERENDUM PROCESS

See infra notes 98-101 and accompanying text.

HeinOnline -- 66 U. Colo. L. Rev. 19 1994-1995

UNIVERSITY OF COLORADO LAW REVIEW

HeinOnline -- 66 U. Colo. L. Rev. 20 1994-1995