17. BUSH V. GORE: A VIEW FROM ITALY
Pasquale Pasquino
Il n'est presque pas de question politique, aux Etats-Unis, qui ne se résolve tôt ou tard en question judiciaire.
ALEXIS DE TOCQUEVILLE
Electoral systems are algorithms that permit the transformation of a number of votes (N) into a smaller number of seats (n). Since the beginning of representative government, this essential mechanism of democratic states, like the right of suffrage, has been discussed extensively.[1]
Very little attention has been given, however, to each electoral system's point of departure, perhaps because the question of how to obtain N seems settled and unproblematic. This is the problem that emerged with unsuspected virulence during the very long month from November 7, 2000, election day, to December 12, 2000, when the U.S. Supreme Court decided that the American presidency should go to George W. Bush.
The question is how the votes are counted, not how they are transformed into a very small number of seats, or just one seat, as in the American presidency. It is on this seemingly banal question that the world's only superpower at the end of the second millennium got stuck, as in a giant swamp. The solution that lifted the aircraft carrier named America over the treacherous waters of the Florida count threatens, however, to discredit one of American democracy's most prestigious institutions: on December 12 the Supreme Court decided that it was too late to complete the ongoing manual vote count.
The Court's decision in Bush v. Gore is a strong candidate for being among the most incongruous and negative opinions it has issued. In Europe it has provoked a very negative opinion of American constitutional justice.
MEN, MACHINES, AND HOLES:
THE FIRST DECISION OF A FEDERAL COURT
Most attentive scholars of the American political system expected both a narrow electoral victory and a discrepancy between the popular vote and the Electoral College majority. Presidential elections are indirect elections that may result in a candidate winning the election without obtaining more votes than his closest competitor. Similar distortions may be seen in almost every method of determining the victor in an election.[2] The 1996 elections in Italy produced a comparable phenomenon: the center-Left obtained a greater number of seats in Parliament, even though the center-Right obtained a greater number of the popular votes. But no one could have expected what actually happened in Florida.
On November 8 the Florida Division of Elections gave Bush a 1,784-vote advantage over Gore. The next day, the mechanical recount reduced the Republican candidate's lead to 327 out of a total of about 6 million votes. Florida election law mandates that a mechanical recount of the votes be carried out when the leading candidate has an advantage of less than 0.5 percent of the votes. The difference between the two candidates given by the first official figure (1,784) amounted to 0.0299 percent of the total votes cast. This triggered an automatic mechanical recount. The law setting forth the procedures for protesting electoral results, Florida Statutes § 102.166 (2000), provides that within seventy-two hours of the election any candidate may request that the canvassing board in the county where the protest arises conduct a manual recount.
At this point, we Europeans discovered, with some surprise, that voting systems in the United States vary significantly from one state to another. For example, we became aware that Oregon recently introduced voting by mail for all its residents, which explains why its results were delayed for several days. Moreover, in Florida voting procedures vary from one county to the next. Of the state's 67 counties, one uses manually tabulated paper ballots (as in Italy), another uses a lever machine (as in New York City), 39 use a ballot requiring that the voter fill in the circle corresponding to the names of the candidates to be read by an optical scanner, and 26 use punchcard ballots. Experts have suggested that the margin of error in the Florida vote was on the order of 5 percent.
Between November 7 and November 11 the Democratic Party, on behalf of Gore, filed protests with the canvassing boards of four counties. Three of the four immediately ordered manual recounts.[3] On November 11 Bush's lawyers appeared in the U.S. District Court for the Southern District of Florida to ask for an injunction against these manual recounts. On November 13, Judge Middlebrooks denied the request.[4] The reasoning behind
The judge had to decide on an emergency motion for a temporary restraining order and a preliminary injunction against the manual recounts on the grounds that § 102. 166(4) of the Florida election law, pursuant to which the recounts were proceeding, violated the First and Fourteenth Amendments of the U.S. Constitution, in particular the Equal Protection Clause, which roughly corresponds to "the principle of equality" under the Italian constitution. Judge Middlebrooks carefully considered the constitutional issue arising under the principle of equality:
The thrust of Plaintiffs' position is that Florida's decentralized county-bycounty electoral system can yield disparate tabulating results from county to county. For instance, similarly punched ballots in different counties may be tabulated differently in a manual recount due to the introduction of human subjectivity and error.[5]
It is well known that three of the counties at the heart of the electoral protest (Palm Beach, Broward, and Miami-Dade) use perforated ballots, where the voter uses a stylus to punch out a perforated square of paper corresponding to the name of the chosen candidate. If all goes well, a little square of paper (called a "chad") completely detaches from the rest of the ballot, resulting in a hole that the optical scanning machine counts as a vote. But it is possible that the voter, unskilled in the use of the stylus, inadvertently detaches three, or two, or just one side of the chad. The machine interprets some of these as holes and others not. How do those who recount the ballots, applying Florida law, define a "hole"?[6] Is there a standard, unambiguous, universally accepted definition of what kind of a hole, or vote, is valid in the specific case for every canvassing board charged with the manual recount? And if there is not, and equal holes may be counted differently or different holes may be counted equally, would this be a violation of the principle of equality, understood as granting each citizen an equal vote?
Judge Middlebrooks, considering these objections, which would later be tackled by the U.S. Supreme Court, reasonably asserted:
These concerns are real, and, in our view, unavoidable, given the inherent decentralization involved in state electoral and state recount procedures. For instance, at least 48 states employ recount procedures—many of which differ in their methods of tabulation. … Unless and until each electoral county in the United States uses the exact same automatic tabulation (and even then there may be system malfunctions and the like), there will be tabulating discrepancies depending on the method of tabulation. Rather than a sign of weakness or constitutional injury, some solace can be taken in the fact that
Judge Middlebrooks stated further that the plaintiffs failed to demonstrate how and why the manual recount is so unreliable as to constitute a violation of the constitutional principle of equality, especially taking account of the fact that the electronic tabulation has a margin of error of about 5 percent. In the end, Judge Middlebrooks refused to enjoin the manual recount because "we do not find sufficient evidence to declare a law authorizing the use of a manual recount to be unconstitutional on its face."[8]
It is worth dwelling on this ruling because its conclusions are the ones that were to be overturned by the U.S. Supreme Court. Three general observations may be made. The 2000 presidential election revealed some problems with democratic electoral mechanisms. The first concerns the United States in particular: namely, the extraordinary disparity of voting mechanisms, tabulation, and protest procedures dramatically revealed in Florida. This produces uncertainty concerning the fairness and clarity of the results. One hopes Congress will encourage the states, charged by the Constitution with making election law, to unify these procedures.
The second problem is a general one concerning all voting systems. In every computation of a great number of entities, whether ballots, holes, or anything else, there exists, as statisticians tell us, a certain margin of error.[9] In the case of elections, the gap between the various candidates is generally significant enough to make us forget that a situation may arise in which the numerical difference is minimal and falls within the margin of error. This opens two, not necessarily alternative paths: fix the voting and vote tabulation systems to minimize the margin of error; or accept that in some very rare cases elections will not succeed in choosing a candidate. At first glance, the second alternative seems unacceptable. The first alternative suggests using an electronic vote, which easily permits checks by both the voter and the institution in charge of the count.[10] The electronic vote naturally presents the problem of protecting the secrecy of votes.[11] But this simply demonstrates that there is no perfect solution to the problem under discussion and that it merits much more attention, discussion, and research than it has been given in the past.
The third problem is that political science research, long focused on the study of electoral systems, has not, to my knowledge, addressed the mechanics of voting and the methods of vote counting in different countries. There is a legal literature dealing with the norms regulating contests of electoral results,[12] and this ought to be developed with a comparative view to other systems. For instance, in Italy, Parliament has competence over
A COURT OR A PARLIAMENT?
The U.S. Supreme Court's first intervention in the electoral contest seemed to me ambiguous and, to a certain extent, inconclusive. It is a fact that the Court decided to accept the case on November 24, surprising many observers (and notwithstanding the opposition of some of its members) and transforming a political question—the choice of the president of the United States—into a legal one. It is also a fact that from November 24 until the Court's final decision on December 12, the Court became the center of the attention of the country, the press, and the mass media, even as the conflicts in Florida were intensifying.
Bush's lawyers aimed with their petition to overturn both the decision of Judge Middlebrooks and that of the Florida Supreme Court, which seemed too favorable to Gore.[14] The Supreme Court's per curiam December 4 decision paid the price of opacity for painfully achieved unanimity.[15] The Court vacated the decision of the Florida Supreme Court and asked it to justify its decision in a more satisfactory way, apparently maintaining an interlocutory stance.[16]
This gave the Florida Supreme Court one week to clarify its position. On December 6, the Eleventh Circuit Court of Appeals rejected Bush's appeal of Judge Middlebrooks's decision. By November 26 Secretary of State Harris had already certified Bush the winner of the Florida election. On December 8 the Florida Supreme Court—this time split four justices to three[17]—handed down a new opinion, insisting that the manual recount of the nine thousand ballots in Miami-Dade County be continued and requiring Harris to add votes identified by the manual recounts to those officially attributed to Gore. Bush's lawyers then filed an emergency application for a stay of the manual recounts, and the U.S. Supreme Court, in a 5–4 decision, issued its December 9 stay suspending the recounts.[18]
At that point the alignments within the Supreme Court became clear. What most surprises an observer of constitutional courts, which usually deliberate in secret and without dissenting opinions, like the French and Italian courts, is that the U.S. Supreme Court, like a nineteenth-century parliament, is divided into a liberal Left, a conservative Right, and a powerful "swamp" of the center, the median justices in the language of political science, on whom the essential decisions depend. The oldest justice on the Court, Stevens, in his dissenting opinion of December 9, declared that the majority's decision "departs from … venerable rules of judicial restraint
Faced with such an explicit attack, which threatened to disqualify the majority's decision in the eyes of the public and accused it above all of abuse of power, Justice Scalia, ideologue of the conservative wing of the Court, felt obliged to defend the majority decision, a highly unusual practice in the context of granting a stay. In his brief concurrence, Justice Scalia tried to counter the legitimacy of counting every valid vote by arguing that the important issue is the very validity of such votes and that they cannot be counted first, leaving for later what the criteria of validity ought to be. It is this way of proceeding, he stressed, that would undermine the public acceptance of the results that democratic stability requires. In contrast to Judge Middlebrooks's opinion, Justice Scalia insisted on the fact that the standards for the determination of the intention of the voter must be the same in each county (which provides a ground for declaring Florida law unconstitutional for violating the principle of equality).
It is clear that the majority and the dissenters were divided on an important and complex question: May the contested votes be manually counted, leaving the criteria of validity (or, more concretely, when a hole is a hole) to the individual county canvassing boards, or must such criteria rather be fixed and objective and binding on all boards before counting begins? And who must fix these criteria?
Naturally, it is doubtful that there are such objective standards; certainly, homogeneous criteria may be authoritatively fixed, but unless one trusts the canvassing boards, what way is there to make sure, for example, that the homogeneous standards are uniformly applied? The Court may have had other ways out of this problem, to which I shall return in the conclusion.
The December 12 per curiam opinion probably expressed the positions of Justices O'Connor and Kennedy—the two justices who de facto decided the forty-third president of the United States. This decision turned on the Equal Protection Clause, a constitutional provision introduced after the Civil War to protect the "new" American citizens created by the abolition of slavery. The point again is that of the need for equal standards to be used in ascertaining the voter's intention. The majority opinion argues that it is possible to set these general criteria and, simultaneously, that there is no time to do so. These, essentially, are the reasons that led to overturning the decision of the Florida Supreme Court and the final acceptance of Secretary of State Harris's November 26 certification.
In the course of oral argument, the justices did interrogate Gore's lawyers on the possible homogeneous criteria for a manual recount of the ballots, though without obtaining a response. It also seems that Justices Breyer and Souter attempted to strike a compromise within the Court on this point but without success. My impression is that the Court's use and abuse of dissenting opinions discourages discussion and internal deliberation. It also dedicates very little time to the "Council Chamber,"[20] another factor making compromise very difficult. More than a judicial body, the U.S. Supreme Court seems, at least to European eyes, like a miniparliament, with its majorities, minorities, and ideological contests.
Applying the Humean principle of tolerance,[21] one may argue that the majority's choice was guided by the will to avoid a constitutional crisis.[22] But this crisis would have been a perfectly manageable one. Effective laws and a certain number of historical precedents[23] enable us to imagine an alternative scenario to that imposed by the Supreme Court majority. First of all, it is not at all evident that the final result of the manual count would have been unfavorable to Bush. With their decision, the (conservative) majority justices gave the unpleasant impression of wanting to protect Bush from every risk. Suppose that the Florida vote favored Gore in the end. On December 18 the Electoral College would probably have received two lists of electors from Tallahassee, the Republican one sent by Secretary of State Harris and a Democratic list certified by the Florida Supreme Court. In this case, the Electoral College would not have been able to reach a clear majority and would have had to defer the final decision to Congress.
Certainly this solution would have kept the country in relative uncertainty for yet another month. And it would have assured the White House for Bush, given that the House of Representatives, which had a clear Republican majority, decides the presidency. But perhaps it would have led to Lieberman's selection as vice president, since the vice presidency is decided by the Senate, which at that point had fifty Republicans and fifty Democrats and would have been subject to a tie-breaking vote by the vice president, Gore.
But the vote of just one justice was enough to elect Bush the forty-third president of the United States. The arguments used to justify this choice are highly contestable and have been contested. But the fact remains: the majority of the Court arrogated to itself the right to choose the president, who is the one who chooses (with the consent of a simple majority of the Senate) the members of the Court. Beyond the difficulty in considering the majority's arguments irrefutable, it is difficult to say that the justices behaved in an elegant fashion. Leaving the choice of the president to Congress, the justices would have given to Caesar that which is Caesar's: to a politically responsible body, the burden of a choice that falls, in principle, to the
The U.S. Supreme Court, in deciding to choose he who will choose its own future members, consumed an enormous amount of credibility as an impartial organ. "Millions of Americans," wrote Cass Sunstein at the end of the struggle (and surely millions of Europeans think this as well), "believed that the court had acted in an unacceptably partisan manner. … Bush v. Gore raised widespread doubts about the neutrality of the Supreme Court."[24] Considering the very high institutional cost of this likely consequence, Justice Breyer wrote in his dissenting opinion that "the Court was wrong to take this case."[25] In Bush v. Gore everything leads to the thought, at least on the other side of the Atlantic, that the Supreme Court abused its power, substantially harming its image as an institution super partes. Unless there is a more profound misunderstanding, Americans are in fact disposed to accept the partisan political character of their judicial power.
In a recent book on justice, a prominent scholar of the Italian judicial system writes aptly that the judge, a neutral third with the job of resolving a conflict between two parties, must convince them that he or she is not taking sides. "His or her impartiality—and above all his or her image of impartiality—is thus preserved and, moreover, strengthened as much as possible."[26]
Whatever the reasons motivating the individual justices, the majority's decision cannot be seen as illustrating the indisputable prestige and impartiality of the supreme organ of judicial power in the United States. It seems, unfortunately, to have behaved like a politically irresponsible miniparliament rather than as a high court of justice.
NOTES
Epigraph: De la démocratie en Amérique, vol. 1 (Paris: Gallimard, 1986), 401 (De l'esprit légiste aux Etats-Unis, et comment il sert de contrepoids àla démocratie): "Scarcely any political question arises in the United States that is not resolved, sooner or later, into a judicial question" (Democracy in America [New York: Vintage, 1945], 290). I owe this quotation to John Ferejohn, whom I thank.
1. See, e.g., Pierre Rosanvallon, La rivoluzione dell'uguaglianza (Milan: Anabasi, 1994); and Pierre Martin, Les systèmes électoraux et les modes de scrutin (Paris: Montchrestien, 1994).
2. Taking this into account, the diffuse criticism of the institution of the Electoral College (I am thinking of the pieces by Ronald Dworkin, chap. 4, this volume; and Akhil Amar, New York Times, November 9, 2000) does not seem particularly pertinent, especially a posteriori. Furthermore, to abolish it and introduce the direct election of the president would require a constitutional amendment, very difficult to obtain given the opposition of the small states that have the right to veto any constitutional reform (see the fine interview of Kathleen Sullivan on the editorial page of the New York Times, November 15, 2000).
3. In particular, on November 13, as we shall see in looking at the first important legal decision, a manual recount had been authorized in three districts in Broward County. The decision was being awaited in Miami-Dade County. In Palm Beach County the authorization had been given for all the districts. And finally in Volusia County, after the confirmation in district 216 of the poor functioning of the floppy disk containing the program for the electronic count of the votes and its replacement, a manual recount of all the ballots in the county was authorized.
4. Siegel v. LePore, 120 F. Supp. 2d 1041 (S.D. Fla. 2000).
5. 120 F. Supp. 2d at 1051.
6. On the not so simple ontology of the same, see R. Casati and A. Varzi, Holes and Other Superficialities (Cambridge, Mass.: MIT Press, 1994).
7. 120 F. Supp. 2d at 1051.
8. Id.
9. I am indebted to Molly Przeworski for this observation.
10. The voter may check the accuracy of his or her vote on the screen before finally confirming it; furthermore, a computer system can quickly recount the electoral results many times, reducing the margin of error.
11. Electronic voting has been instituted in Belgium, and the German government proposes its introduction on a national scale in 2002. On some of the problems posed by this type of voting, see "County in California Touches Future of Voting," New York Times, February 12, 2001 (discussing the experience of Riverside County). Note that 9 percent of American citizens already vote with an electronic touch screen. On electronic democracy more generally, see S. Rodotà, La démocratie électronique (Paris: Apogée, 1999).
12. I am familiar with the work of Maria Vicenta Garcia Soriano, Jueces y magistrados en el proceso electoral (Madrid: Centro de Estudios Politicos y Constitucionales, 2000).
13. In regional elections, administrative courts and the Council of State are competent.
14. Italians were negatively struck by the fact that all seven justices of the Florida Supreme Court are Democratic nominees and, moreover, by the fact that in America all federal judges are chosen by a simple political majority—in contrast to Kelsenian systems of constitutional justice, in which the nominations for a super partes organ are exempt from the majoritarian principle.
15. Bush v. Palm Beach County Canvassing Board, 121 S. Ct. 471.
16. See, e.g., the comment on this decision published by Cass Sunstein in the New York Times, December 5, 2000.
17. Gore v. Harris, 772 So. 2d 1243 (2000).
18. Bush v. Gore, 521 U.S. 98 (2000).
19. Id. at 513 (Stevens, J., dissenting).
20. The Italian Constitutional Court discusses all of its decisions collectively, and the most important ones at length, in collegial meetings (called the jurisdictional chamber councils) that occupy the justices for twenty-one weeks a year.
21. As did Burt Neuborne so elegantly in a series of fictitious opinions that develop the justices' opinions; see chap. 9, this volume.
22. This is the argument made by Richard Pildes, in chap. 3, this volume.
23. For example, the Tilden-Hayes presidential election of 1876.
24. Chronicle of Higher Education, January 5, 2001 (reprinted in E. J. Dionne and William Kristol, eds., Bush v. Gore. The Court Cases and the Commentary [Washington, D.C.: Brookings Institution Press, 2001], 339–40).
25. In Bush v. Gore. The Court Cases and the Commentary, 132.
26. Carlo Guarnieri, La giustizia in Italia (Bologna: Il Mulino, 2001), 14.